Sequence Stratigraphy and Depositional Response To Eustatic, Tectonic and Climatic Forcing

SEQUENCE STRATIGRAPHY AND DEPOSITIONAL RESPONSE TO EUSTATIC,
TECTONIC AND CLIMATIC FORCING

Coastal Systems and Continental Margins
VOLUME 1
Sequence Stratigraphy and
Depositional Response to Eustatic,
Tectonic and Climatic Forcing
Edited by
Bilal U. Haq
Springer-Science+Business Media, B.V.

Library of Congress Cataloging-in-Publication Data
Sequence <tratigraphy and depositional response to eustatic, tectonic,

and c 11m ate for c i n g / e d i ted by B i 1a 1 U, Ha q .
p. cm. -- (Coasta 1 systems and cont i nanta 1 marg i ns ; v. 1)
Inc 1udes index.
ISBN 0-7923-3780-8 (acid-free)
1. Geology, Stratigraphic. 2. Sedimentation and depositlon.
3. Sea level. I. Haq, Bilal U. II. Series.
QE651.S4588 1995
551.7--dc20 95-41541
ISBN 978-90-481-4633-8 ISBN 978-94-015-8583-5 (eBook)

DOI 10.1007/978-94-015-8583-5
Printed on acid-free paper
All Rights Reserved

1995 Springer Science+Business Media Dordrecht
Originally published by Kluwer Academic Publishers in 1995.
Softcover reprint of the hardcover 1st edition 1995
No part of the material protected by this copyright notice may be reproduced or
utilized in any form or by any means, electronic or mechanical,
including photocopying, recording or by any information storage and
retrieval system, without written permission from the copyright owner.
Table of contents
Preface B. U. Haq ix
1 Depositional Sequences, Facies Control and the Distribution of Fossils S.M. Holland 1
1. Introduction . . . . . . . . . . . . . . I
2. The Problem of Facies Control . . . . . 2
2.1. The Perfect Stratigraphic Record 2
3. The Effect of Sampling Bias . . . . . . 2
3.1. The Effects of Facies and Depositional Sequences 3
4. A Sensitivity Analysis of Facies Control 7
4.1. Preferred Depth 9
4.2. Depth Tolerance . . . . . 9
4.3. Peak Abundance . . . . . 9
4.4. Extinction and Origination 10
4.5. Depth Range of Parasequence . 10
5. Sequence Control on First and Last Occurrences 10
5.1. A No-Lowstand Sequence 10
5.2. A Complete Sequence 13
5.3. Generalizations...... 15
6. Preliminary Tests of the Model . . 16
6.1. The Upper Ordovician of the Cincinnati Arch 16
6.2. Other Evidence. . . . . . . . . . . . . . . 18
7. Conclusions . . . . . . . . . . . . . . . . . . . . 19
2 Mixed Carbonatc-Siliciclastic Sequence Stratigraphy Utilizing Strontium Isotopes: Deciphering
the Miocene Sea-Level History of the Florida Platform D.l. Mallinson. l.S. Complon 25
1. Introduction . . . 25
2. Geologic Setting . . . . . 26
3. Methods. . . . . . . . . 28
4. Results and Interpretations 32
4.1. Facies DislIibutions In Relation To Sea Level 32
4.2. Interpreting Sr-Isotope Data. . . . . . . . . 32
4.3. Identification of Sequences and their Components 37
4.4. Episodic Phosphogenesis and Sea-level Highstands 47
5. Discussion. . . . . . . . . . . . . . . . . . 49
5.1. Sequence Interpretations . . . . . . . 49
5.2. Comparison to Other Sea-level Studies 52
6. Conclusions . . . . . . . . . . . . . . . . . 54
3 Using Ecostratigraphic Trends in Sequence Stratigraphy F. Olariz. l.E. Caracuel. F.l. Rodrfguez 59
I. Introduction . . . . . . . . . . . . . . . . . . . . . . . 59
2. The Ecostratigraphic Procedure in Upper Jurassic Outcrops . . 60
3. Ecostratigraphy and Sequence Stratigraphy . . . . . . . . . . 64
4. A Theoretical Ca~e as Deduced from the Iberian Upper Jura~sic 65
5. Applications to Upper Jura~sic Outcrops in the Western Tethys 70
5.1. Example I: Ecosedimentary Evolution at the Stage-substage Level 73
5.2. Example 2: Ecostratigraphic Interpretations at the Ammonite Biochronozone Level 75
5.3. Example 3: Comparison of Faunal Assemblages from Distant Epicontinental Areas 76
5.4. Example 4: Ecostratigraphy and Trends in Abiotic Components . . . . . . . . . 77
5.5. Example 5: Influence of Tectono-eustatic Interactions on Trends in the Composition of
Fossil Assemblages . 78
6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 79
VI
4 Sea-Level Changes, Carbonate Production and Platform Architecture: The L1ucmajor Platform,
Mallorca, Spain L. Pomar. We. Ward 87
l. Introduction . . . . . . . . . . 88
2. The L1ucmajor Platform . . . . 88
2.1. Reef Platform Lithofacies 90
2.2. The Basic Building Block. 91
2.3. Larger-Scale Accretional Units 94
2.4. L1ucmajor Platform 3-D Architecture 96
3. Sequence Stratigraphic Implications . . . . 96
3.1. High-frequency Reefal Platform Depositional Sequences 96
3.2. Systems Tract Boundaries . . . . . . . . . . . . 102
3.3. Accommodation vs. Carbonate Production. . . . 102
3.4. Basin Floor Topography vs. Carbonate Production 105
4. Conclusions . . . . . . . . . . . . . . . . . . . . . . 105
5 Geometric Responses in Neogene Sediments of Offshore New Zealand: Simulated as Products of
Changes in Depositional Base Level Driven by Eustasy and/or Tectonics C. G.SI. C. Kendall. G.L Whittle,
C.S. Fullhorpe, Ph. Moore. T.D. Hickey, R.Cannon, D. Hellman 113
l. Introduction . . . . . . . . . . 113
2. Sedpak . . . . . . . . . . . . 114
3. Examples of Sedpak Simulations 116
4. Testing the Exxon Model. . . . 119
5. Isolating the Effects of Sediment Supply 119
6. Sedpak Simulation of the Canterbury Ba~in 121
7. Conclusions . . . . . . . . . . . . . . . 127
6 Controls on Long-Term Global Ratcs of Coal Deposition, and the Link between Eustasy and Global
Geochemistry L.B. Railsback 137
l. Introduction . . . . . . . . . . . . . . . . . 137
2. The Record of Global Coal Depositional Rates 138
3. Controls on Coal Deposition . . 140
3.l. Climate and Geography 140
3.2. Tectonics........ 142
3.3. Changes in Sea Level . . 142
3.4. Evolution of Land Plants 143
3.5. Hypotheses. 145
4. Methods and Results 145
4.1. Parameters. 145
4.2. Results... 147
4.3. Numerical Model. 148
5. Discussion. . . . . . . . 151
5.1. Paleogeographic Implications 151
5.2. Coal Accumulation and Plant Evolution 152
5.3. Stratigraphic Considerations 152
5.4. Eusta~y and Global Geochemical Cycles. 154
6. Conclusions . . . . . . . . . . . . . . . . . . 155
7 Sequence Stratigraphy and Sea-Level Changes in the Early to Middle Triassic of the Alps: A Global
Comparison Th. Riiffer, R. Ziihlke 161
2. Geological Setting 163
3. Methods. . . . . 164
4. Dolomites . . . . 165
4.1. Previous Studies 165
4.2. Regional Geology 166
4.3. Facies and Sequence Stratigraphy of Selected Sections 168
4.4. Controls on Sedimentation in the Dolomites . . . . . 176
5. Northern Calcareous Alps . . . . . . . . . . . . . . . . . 180
5.1. Late Scythian to Anisian Depositional Environments. 181
vii
5.2. Ladinian to Early Carnian Depositional Environments . . . . . . . . . 183

5.3. Facies and Sequence Stratigraphy of Selected Sections . . . . . . . . . 184
5.4. Middle Triassic Sequence Stratigraphy of the Northern Calcareous Alps 189
6. Depositional Sequences in the Northwestern Tethys 192
6.1. Biostratigraphy........... 192
6.2. Sequence-Stratigraphic Correlations . 195
7. Global Depositional Sequences . 198
7.1. Scythian...... 199
7.2. Anisian . . . . . . 200
7.3. Ladinian to Carnian. 200
8. Conclusions . . . . . . . . 201
8 Reconstructing Eustatic and Epeirogenic Trends Crom Paleozoic Continental Flooding Records
rI. Algeo. K.B. Seslavinsky 209
1. Introduction . . . . 209
1.1. Eusta~y 210
1.2. Epeirogeny. 210
2. Analysis of Paleozoic Eustasy and Epeirogeny 211
2.1. Paleozoic Continental Flooding . . . . 211
2.2. Paleozoic Continental Hypsometry .. 215
2.3. Paleozoic Sea-Level and Epeirogenic Trends. 222
3. Discussion. . . . . . . . . . 229
3.1. Eustasy . . . . . . . 229
3.2. Continental Epeirogeny 231
4. Conclusions . . . . . . . . . 238
9 Depositional Sequences on Upper Cambrian Carbonate PlatCorms: Variable Sedimentologic
Responses to Allogenic Forcing D.A. Osleger 247
I. Introduction . . . . . . . . . . . . . . . . . 248
2. Tectonic and Stratigraphic Settings . . . . . . 248
2.1. Time Control and Accumulation Rates. 250
3. Upper Cambrian Depositional Sequences . . . 251
3.1. Paleobathymetric Trends and Sequence Correlation 253
4. Late Cambrian Accommodation History . . . . 256
4.1. Stacking Patterns of Meter-Scale Cycles . . . . . 256
4.2. Subsidence Analysis . . . . . . . . . . . . . . 259
4.3. Compruison of Late Cambrian Sea-Level Curves. 262
4.4. Variable Sedimentary Response to Accommodation Change 264
5. Sequence Stratigraphy . . . . . . . . . . . . . . . . . . 265
5.1. Interbasinal Comparison of Sequence Components. . . . . 265
6. Summary and Conclusions . . . . . . . . . . . . . . . . . . . . 269
10 Sequence-Stratigraphy oC Early-Middle Cambrian Grand Cycles in the Carrara Formation,
Southwest Basin and Range, CaliCornia and Nevada R.D. Adams 277
I. Introduction . . . . . . . . . . . . . . . . . . . 278
2. Previous Work and Regional Geology . . . . . . 279
3. Lithofacies and Meter-Scale Cycles/Parasequences 283
4. Correlation Strategy . . . . . . . . . . . . . . . 289
5. Carrara Lithostratigraphy. . . . . . . . . . . . . 289
5.1. First Sequence/Grand Cycle: Zabriskie Quartzite-Eagle Mountain Shale-Thimble
Limestone . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
5.2. Second Sequence/Grand Cycle: Echo Shale-Gold Ace Limestone . . . 300
5.3. Third Sequence/Grand Cycle: Pyramid Shale-Red Pass Limestone . . 301
5.4. Fourth Sequence/Grand Cycle: Pahrump Hills Shale-Jangle Limestone 302
5.5. Fifth Sequence/Grand Cycle: Desert Range Limestone-Papoose Lake Member of the
Bonanza King Formation . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
6. Carrara Sequence Stratigraphy . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
6.1. First Sequence/Grand Cycle: Zabriskie Quartzite-Eagle Mountain Shale-Thimble
Limestone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
Vlll
6.2.Second Sequence/Grand Cycle: Echo Shale-Gold Ace Limestone . . . 308

6.3.Third Sequence/Grand Cycle: Pyramid Shale-Red Pass Limestone . . 309
6.4.Fourth Sequence/Grand Cycle: Pahrump Hills Shale-Jangle Limestone 312
6.5.Fifth Sequence/Grand Cycle: Desert Range Limestone-Papoose Lake Member of the
Bonanza King Formation . . . . . . . 314
6.6. Overview . . . . . . . . . . . . . . 315
7. Sequence Stratigraphic Model of Grand Cycles 317
7.1. Introduction..... 317
7.2. Comprehensive Model . . . . . . . . 318
8. Summary . . . . . . . . . . . . . . . . . . 322
11 Milankovitch Fluctuations in Sea Level and Recent Trends in Sea-Level Change: Ice may not
always be the answer D.K. Jacobs, D.L. Sahagian 329
2. Ice . . . . . . . . . . . . . . . . . . . . 330
3. Quaternary Proxy . . . . . . . . . . . . . 332
3.1. Evidence for Increased Early Holocene Monsoonal Precipitation 332
3.2. Quantification of Large Basin Volumes 335
3.3. Importance of Groundwater . . . . . . . . . . . . . . . . . . 336
3.4. Context of the Volume Calculations . . . . . . . . . . . . . . 337
4. Orbital Variation, Continental Configuration and the Monsoons: Theoretical Considerations . 337
4.1. Precession, Eccentricity, and Obliquity and Idealized Continents 338
4.2. Null Monsoons and Double Precession Beats 339
4.3. Orographic Effects . . . . . . . 340
5. Evolution of Internal Drainage . . . . . 340
6. Carbonate Models and Sea Level Change 345
6.1. Carbonate Production Rate . . . 345
6.2. Rate of Pluvially-Induced Sea Level Changes 346
7. The Late Triassic . . . . . . 347
8. Other Time Periods 349
8.1. Cambrian--Ordovician 349
8.2. Devonian....... 349
8.3. Late Permian-Early Tria~sic . 350
8.4. Jura~sic ......... 351
8.5. Early Cretaceous . . . . . 351
9. Anthropogenic Sea Level Change. 353
9.1. Fluxes of Water to the Sea 354
9.2. Dams . . . . . . . . . . 356
9.3. Toward more Comprehensive Analyses 357
10. Lakes, Climate Models and Environmental Concerns 357
10.1. Lakes and Climate Models . . . . . . . 357
10.2. Lakes, Fauna~ and Environmental Policy " 358
11. Conclusions . . . . . . . . . . . . . . . . . . . . 359
12 The Enigma of Third-Order Sea Level Cycles: A Cosmic Connection? c.G.St.c. Kendal/,
Ph.A. Levine, R. Ehrlich 367
1. Introduction . . . . . . . . . . . . . . . . . . 367
2. Third-Order Sequences Driven by Glaciation . . 368
3. Third-Order Sequences Driven by Stress Relea~e 368
4. Third-Order Sequences Driven by Sediment Supply 370
5. Third-Order Sequences Driven by Geoidal Distortion 370
6. Hypothesis: Extraterrestrial Impacts Cause Third-Order Sequences. 370
7. Evidence for Impacts. 373
8. Conclusions . . . 374
List of Contributors 377
Index 379
Preface
Sequence stratigraphy has advanced considerably since the early applications of

the concepts on seismic data. It attempts to discern the migration of facies re-
sulting from changes in a combination of factors such as, sea level, tectonics,
climate and sediment flux, and integrates it with a meaningful chronostratigraphy.
The stratigraphic record is envisioned as a framework of repetitive packages of
genetically-related strata, formed in response to the shifting base level, in which
the locus of deposition of various sediment types may be anticipated. This attribute
is rapidly promoting sequence stratigraphy as an indispensable tool for prediction
of facies in exploration and production geology.
In hydrocarbon exploration the application of sequence stratigraphy has ranged
from anticipating reservoir- and source-rock distribution to predicting carbonate
diagenesis, porosity and permeability. The capability to anticipate vertical and
lateral distribution of facies and reservoir sands in the basinal, shoreface, incised
valley-fill and regressive settings alone has been a great asset for exploration. In
frontier areas, where data are often limited to seismic lines, sequence-stratigraphic
methodology has helped determine the timing and of types of unconformities
and anticipate transgressive- and regressive-prone intervals. In production it is
aiding in field development by providing improved source and seal predictions
for secondary oil recovery. A recognition of stratigraphic causes of poor recovery
through improved understanding of internal stratal architecture can lead to new
well recompletions and enhanced exploitation in existing fields.
The sequence-stratigraphic discipline is in a state of rapid expansion. New
case studies of its application are axiomatic to appreciate its full potential and to
further advance depositional models so that valid variations to the basic theme
can be realized. The concepts have been applied in increasing number of sections
and in a variety of depositional settings. The methodology seems to work best in
shallow marine environments, but offshore exploration targets have also benefited
from the predictive capabilities that are inherent in the approach.
High-resolution sequence-stratigraphic analysis in field-size settings, in which
individual depositional sites and facies related to various phases of the higher-
frequency sea-level cycles can be identified and/or predicted, is becoming impor-
tant as an exploration tool. Depositional environments where its potential has not
as yet been fully appreciated are in fluvial, coastal plain and lucustrine settings,
though some examples of its use in the latter environment are beginning to appear.
Fluvial/coastal plain environments, in particular, represent potentially vast explo-
ration targets and adaptation of the models to these settings could be a major step
forward.
ix
x Preface
The case studies appearing in this volume represent a variety of depositional

environments and time frameworks. The articles included largely resulted from
presentations at the session on 'Global Sea-Level Change' in a meeting on "Strati-
graphic Record of Global Change" held in August, 1993 at the Pennsylvania State
University under the aegis of Society of Sedimentary Geology. They exemplify the
response of the depositional systems to various governing factors, particularly, eu-
stasy (global sea-level changes that can produce or destroy accommodation space),
tectonics (local and regional variations in subsidence that also affect accommo-
dation), and climate (which can cause both local and long-distance variations in
sediment supply and affect dissolution and diagenesis). Another major factor that
influences the resultant stratal patterns is the topography of the depositional profile,
especially in carbonate systems.
The first article, by Steve Holland, models the occurrence of biofacies within a
sequence stratigraphic framework. The results have important implications for the
apparent patterns of distribution of fossil taxa. They imply that the stratigraphic
record may show clustering of first and last occurrences at surfaces that represent
major flooding or basin ward shift of facies, and may also show offsets of these
events by up to one full sequence from the true times of origination or extinction.
Such patterns would be more strongly expressed in highly facies-limited taxa,
but may also be evident for less facies-limited forms. Field data from the Upper
Ordovician (Cincinnatian) of southeastern Indiana verify the existence of such
distribution patterns.
In the second case study David Mallison and John Compton reconstruct a re-
gional sequence-stratigraphic framework of the Miocene of northeastern Florida
Platform. They use strontium-isotopic composition of phosphoritic crusts and
dolostones (condensed sections) and reworked phosphorite, dolostone sand and
gravel (at erosional surfaces associated with sequence boundaries and transgres-
sive surfaces) to constrain the depositional ages of associated lithofacies. The Sr
isotope-derived ages are used to date the highstands and recognize seven major
sequences with a maximum flooding of Florida Platform between 17 and 15 Ma.
Federico 016riz and his colleagues demonstrate the application of ecostrati-
graphic trends within a sequence-stratigraphic framework. The authors point out a
direct relationship between accommodation and ecospace, the latter shifting with
the migrating facies in response to changes in the base level. These applications are
exemplified by megainvertebrates sampled bed-by-bed from Middle Oxfordian to
Lower Tithonian sections of several basins in southeastern Spain. The authors con-
clude that ecospace (as characterized by averaged faunal spectra, ecostratigraphic
trends and events) is intimately related to accommodation, stratal patterns and
systems tracts and their boundaries. The application of the methodology requires
adequate sampling programs in which both biotic and abiotic signals are accounted
for.
Another case study from Spain concerns the stratal architecture of the Upper
Miocene reefal platform exposed on the island of Majorca. Here the sections, ac-
cessable on land and cliff faces, are relatively undisturbed with much of the reefal
morphology still in its growth position, making these exposures well suited for
closeup study of the platform architecture. From such a detailed analysis Luis Po-
Preface xi
mar and Bill Ward conclude that stratal patterns, from the basic unit of a prograding
sigmoid to megasets of sigmoids, are all controlled by changes in the accommoda-
tion and sediment flux (or carbonate production). Both of these factors are in tum
governed by changes in sea level and the morphology of the depositional profile
itself. A gentle depositional profile permits greater progradation than profiles with
steeper gradients. The system experiences higher production during sea-level rise
and lower production during stillstands and falls. Carbonate sediment accumula-
tion is high when platform is flooded and lagoons reestablish production, which
continues to be high as long as the system can keep pace with increasing accom-
modation. Thus, instead of a transgressive retrogression, aggradation takes place
in all depositional settings of the system. This is an important point of departure
between carbonate and siliciclastics depositional system.
Chris Kendall and his colleagues demonstrate the usefulness of computer
simulation techniques to fathom the significance of changes in the base level in
creating sequence-bounding unconformities. Their program graphically simulates
sedimentary fill in a basin by changing variables, such as, the sea-level position,
subsidence and sedimentation rates. The simulation is applied to a Neogene sec-
tion from offshore South Island, New Zealand. To recreate the interpreted seismic
section from the Canterbury Basin it was necessary to manipulate the depositional
base level (whether by changing eustasy or subsidence). Varying the rate of sedi-
ment supply alone did not produce sequence-bounding unconformities as observed
on the seismic section. Such simulations may be a valid iterative technique to work
backwards with model assumptions to decipher the relative importance of various
forcing factors in producing the observed stratal architecture.
In a thought-provoking paper on the relationship of eustasy to coal accu-
mulation, Bruce Railsback suggests that global rates of coal deposition correlate
positively with long-term sea-Ievellowerings, perhaps due to the excess coastal
plain areas that become available during regressions. Contrary to earlier belief,
paleogeography may not be as important for global coal accumulation rates, al-
though it may control the sites of coal deposition. Abundance or type of plant
species also seem to play little role in the overall long-term global rates. This im-
plies that although coal deposition is sensitive to climate (latitude), such sensitivity
may diminish during periods of major regressions. This in tum may have been an
important controlling factor for atmospheric oxygen and carbon-dioxide composi-
tion. The author suggests that these patterns emphasize the importance of eustasy
as a controlling factor for the changing composition of the earth's atmosphere in
the Phanerozoic.
The study by Thomas Ruffer and Rainer Zuhlke on the Middle Triassic of the
northern Calcareous Alps and the Dolomites illustrates the combined importance
of tectonics and sea-level changes in controlling resultant sequences. Changing
subsidence rates, as well as base level, and regional tectonics that affected the
basinal morphologies, all seem to have influenced resulting sequences at varying
degrees during different times in the evolution of these basins. However, The
eustatic signal is discernable through much of the duration except in the Ladinian,
when the tectonic signal overwhelms the sea-level signal. The authors come to the
xii Preface
conclusion that in the Alpine Tethys eustasy played a major role in the development
of the depositional sequences, especially in the Scythian and early Carnian times.
The next series of three studies deal with eustasy, tectonics, and the develop-
ment of sequences in the Paleozoic. Thomas Algeo and Kirill Seslavnisky observe
that co-existing landmasses are likely to experience the same range of eustatic
changes, and the extent of continental flooding may primarily be a function of
coastal hypsometries. If this observation is valid then differences in estimated
sea levels between individual continents may be largely due to epeirogeny. By
applying these premises to their estimates of the extent of Paleozoic continental
floodings, the authors conclude that eustatic high stands of that time may have
been only 100 to 225 m above present day base level, which is substantially lower
than previous estimates of 300 to 600 m above the present sea level. Paleozoic
continents experienced as much as 100 m of independent vertical motions on time
scales of 10 to 40 m.y., which may have coincided with major tectonic events. The
continental landmass of Gondwana exhibits large epeiorgenic excursions in the
Silurian-Carboniferous interval that are antithetic to global sea-level trends and
may have a link to eustasy through mid-ocean ridge spreading rates.
In the second study David Osleger demonstrates that intrabasinal correlations
of sequences of various Upper Cambrian carbonate platforms from North America
can allow deciphering of the history of accommodation changes at second- and
third-order scales. Two additional techniques, Fischer plots of changes in stacking
patterns of cycles and subsidence analyses, also yield direct information about
changes in accommodation. Comparison of accommodational curves from various
platforms and sea-level curves from other North American regions provide an
overall eustatic signal that is valid for the Late Cambrian of much of North Amer-
ica. The author maintains that similarities in internal architecture of individual
sequences suggest that continent-wide depositional patterns were predominantly
controlled by eustasy.
The third case study, by Roy Adams, documents sequences from a mixed
siliciclastic-carbonate system of the Lower-Middle Cambrian Carrara Formation
of California and Nevada. These represent large-scale depositional couplets of
siliciclastics overlain by carbonates, also called the "grand cycles". The exam-
ple is typical of the way mixed systems behave, with siliciclastics preferentially
accumulating during the lowstands and carbonates during the highstands. Car-
rara Formation deposition took place on a mature passive margin with thermal
subsidence as the dominant long-term tectonic activity and accommodation pro-
ducer. Lowstand times, when shelf-margin wedges accumulated, were typically
dominated by non-marine and subtidal siliciclastics, transgressive times by mixed
facies, and the highstands by subtidal and intertidal carbonates. Offlapping of these
facies is seen toward the craton.
David Jacobs and Dork Sahagian raise yet another issue concerning eustasy.
Could human activities of the recent past have helped to permanently transfer
enough water from continental reservoirs to the oceans to significantly influence
sea-level rise? These authors estimate that a third of the sea-level rise inferred
from tide-gauge data may be ascribable to this factor. They also argue that non-ice
related water storage on land can produce significant changes in the base level on
Preface xiii
the Milankovitch frequency time scales. They contend that this mechanism may
have been dominant source of eustatic fluctuations during times when the earth
lacked continental ice sheets.
The last article, by Chris Kendall and his colleagues, is also the most specula-
tive. The authors hypothesize that periodic asteroidal and/or cometary impacts on
earth may be responsible for most of the third-order sequences and their bound-
ing unconformities. Collisions of 10 km or larger bodies with earth can generate
enough energy to cause massive release of stress built up at plate boundaries,
leading to isostatic readjustment of continental margins and global transgressions.
These authors contend that frequency of impact craters, that are often coincident
with eustatic events, lends weight to their argument. Other mechanisms of eustatic
change, such as glaciations, interplate stress, tectonics or geoidal distortions, do
not fully explain all third-order cycles. The alternative explanation posed by the
authors, which is testable independently, may therefore deserve serious consider-
ation.
The case studies presented in this volume advance our knowledge of the fac-
tors controlling the genesis of sequences at different physical and temporal scales.
They also enhance our overall understanding of the discipline and the usefulness
of the paradigm as a predictive methology in sedimentary geology. Many col-
leagues gave generously of their time to review the included papers. The editor
greatly appreciates the help of the reviewers who were chosen from amongst the
contributors to the volume, as well as externally. The external reviewers included,
Gerard Bond, Martin Farley, Bill Hay, Susan Kidwell, Ken Miller, Greg Mountain,
Allison Palmer, Linc Pratson, and Mike Steckler. Many thanks to these reviewers
and the authors for ensuring the quality of the papers. The opinions, findings and
conclusions herein are those of the authors and do not necessarily reflect the views
of the National Science Foundation.
BILAL U. HAQ
Bethesda, Maryland
CHAPTER 1
Depositional Sequences, Facies Control

and the Distribution of Fossils
Steven M. Holland
ABSTRACT: Three increasingly realistic models of the stratigraphic distribution of

fossils are presented. The first assumes a perfect stratigraphic record, the second al-
lows sampling effects, and the third combines sampling effects, facies control, and
stratigraphic cyclicity. When this third model is applied to depositional sequences ex-
hibiting realistic stacking patterns, several patterns emerge. Taxa will typically display a
scarce-common-scarce pattern of occurrence within a parasequence. This pattern may be
abruptly truncated at a flooding surface, and will be separated by a long gap in adjacent
parasequences provided both parasequences contain the proper facies for the collection
of the taxon. Even longer gaps will occur if one of the parasequences does not contain the
correct facies. For these reasons, first and last occurrences can be offset by up to one full
sequence from the true times of origination and extinction. Clusters of first occurrences
and last occurrences will tend to occur at major flooding surfaces (particularly within
the transgressive systems tract) and at major basinward shifts of facies, such as at the
base of the lowstand systems tracts. These clusters will have a predictable composition
of taxa with respect to preferred water depth and facies tolerance. All of these patterns
will be expressed most strongly for highly facies-limited benthic taxa, but will also be
evident for less facies-limited benthic taxa as well as nektic and planktic forms that are
ecologically tied to the benthos. Limited field data verify the existence ofthese patterns
in the fossil record.
1. Introduction
Sequence stratigraphy is forcing a reinterpretation and re-evaluation of many

previous stratigraphically-based studies. Because critical assumptions of bios-
tratigraphy and paleobiology hinge on the stratigraphic distribution of fossils, the
predicted distributions of fossils in stratigraphic sections must be re-evaluated as
well. The most obvious implication of sequence stratigraphy is that depositional
environments are constantly shifting and that an assumption of constant deposi-
tional environment in a section is probably erroneous to some degree. Although
facies control of fossil taxa has long been recognized, relatively few approaches
B. u. Haq (ed.),
Sequence Stratigraphy and Depositional Response to Eustatic, Tectonic and Climatic Forcing, 1-23.
1995 Kluwer Academic Publishers.
2 S.M. Holland
have been sufficiently developed to recognize its effects or incorporate this data
into interpretations (Shaw, 1964; Ludvigsen and others, 1986). In this paper, I will
present the results of a numeric model that combines facies control and depositional
sequences to predict how fossils might be distributed in a stratigraphic section.
The final model presented here incorporates facies control, sampling effects, and
depositional sequences. In particular, I focus on the distribution of first and last
occurrences in a single stratigraphic section, and at what horizons clusters of first
and last occurrences should form. Although more field data will be collected in the
near future to test this model, some limited field data are presented that suggest
the basic aspects of this model are correct.
2. The Problem of Facies Control
2.1. THE PERFECT STRATIGRAPHIC RECORD
The simplest approach to predicting the stratigraphic distribution of fossils would

assume that if a taxon was extant in a sedimentary basin at a particular time, that
taxon would be collected from rocks deposited during that time. Although probably
very few paleontologists accept this simple model, it serves as a useful basis to
compare to other more sophisticated models.
Graphically, this model can be depicted as a 100% chance of collection of
a fossil taxon, regardless of depositional water depth, which will be used here
as a proxy for facies control (Fig. lA). Setting the probability of collection to
100% reflects the absence of sampling effects. The 50 taxa in this simulation were
allowed to originate and go extinct any time during the simulation (see Appendix).
Each of the 80 time steps represents 50,000 years, giving a total model duration of
4 million years. As expected. the fossil record in this single stratigraphic section
perfectly reproduces the time in which a taxon was extant (Fig. 2).
3. The Effect of Sampling Bias
The previous model can be made more realistic by reducing the probability of col-
lection to some constant valueless than 100%, which introduces random sampling
effects (Fig. IE). Using the same taxa as in the previous model, but with a 50%
probability of collection for all taxa, a fossil record with gaps is produced (Fig. 3A).
Since the occurrence of a taxon at any horizon is independent of its occurrence at
any other horizon, gap lengths follow a Dirichlet distribution (Strauss and Sadler,
1989), roughly similar to an exponential distribution with many short gaps and
relatively fewer long gaps. This model satisfies the assumptions necessary to apply
confidence limits to fossil ranges (Paul, 1982; Springer and Lilje, 1988; Strauss and
Sadler, 1989; Marshall, 1990). When the probability of collection is low (Fig. 3B),
the preserved range of a taxon is likely to be considerably shorter than the actual
time in which it lived. Therefore, first and last occurrences are distinguished from
times of origination and extinction in this paper, as in typical usage of these terms.
Origination and extinction refer to the times in which an organism first arose or
last lived. regardless of whether it was preserved or not. First and last occurrences
Sequences and Fossil Distribution 3
100%-+----------
100 % ~~~~danCe
Probability Probability
of of
Collection Collection
O%~---- _____________
O%~--~------~~~~~----
Water Depth
A -Perfect Stratigraphic
C - Facies /
Record Model Cycle Model
J\
I \
I \
I ,
100% I ,
I \
Probability I \
100%
of
Collection
Probability
of
O%~-- _______________ Collection
Water Depth
B - Sampling Effects Model Water Depth
Fig. 1. Assumptions of three models of the stratigraphic distribution of fossils. For the Perfect
Stratigraphic Record Model (A, top left) the distribution of each taxon in the model is represented
by a solid horizontal line at a level of 100% chance of collection. The probability of collection does
not change with environment of deposition, measured as water depth. Because the probability of
collection is 100%, any taxon extant will be collected. For the Sampling Effects Model (B, bottom
left) the distribution of all taxa in the model is a solid horizontal line, but at some level less than 100%.
Sampling effects become important, but because the probability of collection does not change with
depth, facies and sequences are unimportant. For the Facies/Cycle Model (C, right), the probability
of collection for any taxon is given as a Gaussian distribution with respect to depth. Three parameters
need to be chosen for each taxon: (I) the depth of maximum abundance, the preferred environment of
the organism, (2) the amount of facies control or depth restriction, set by the standard deviation of the
abundance/depth distribution, and (3) the maximum probability of collection, that is, the probability
of collection at the depth of maximum abundance. With these three variables, a wide variety of
distributions are possible, including some nearly facies-insensitive distributions (see distribution A
in the lower right). Where the abundance of a taxon rises above the 100% mark (distribution C in the
lower right), the probability of collection in the model is 100% over that part of the depth range.
represent the preserved fossil record, are the lowest and highest horizons at which
a taxon is collected, and can be poor approximations of the times of origination
and extinction.
3.1. THE EFFECTS OF FACIES AND DEPOSITIONAL SEQUENCES
A more realistic approach to the previous models would include facies control and
cyclical facies change, such as that expressed in parasequences and sequences. It
4 S.M. Holland
I I I I
I
80 -
I
I I IIIIIII~
oc: 60-
.~
I..;;
~ -
20 r rI -
1.1
O~~--~~~~~~~~~~L-----~ ""
10 20 30 40 50
Taxa
Fig. 2. Stratigraphic ranges in a single measured section produced by the Perfect Stratigraphic
Record Model. Note that because probability of collection is 100% regardless of environment. the
occurrences of any taxon (in black) reflect the actual true range of the taxon in time.
is necessary to specify both the nature of facies control and the form of cyclical
facies changes.
Most organisms have an optimum relationship with some environmental fac-
tor (Whittaker, 1970), such that the abundance of an organism peaks at one value
of this factor, and its abundance decreases at higher or lower values. Theoretical
ecologists have frequently treated this relationship as a Gaussian distribution of
abundance with respect to the environmental factor (Whittaker, 1970). Many ma-
rine organisms show such a relationship with water depth. Although water depth
itself rarely controls the distribution of an organism, many environmental factors,
such as light, temperature, wave energy, turbidity, and substrate, are important to
organisms and correlate to a high degree with water depth. Consequently, water
depth can be treated as a composite variable against which an organism will display
this optimality relationship. Although facies control of almost all benthic taxa is
widely acknowledged (Dodd and Stanton, 1990), facies control has been repeatedly
demonstrated for many pelagic and nektobenthic forms, such as conodonts (e.g,
Clark, 1984; Sweet and Bergstrom, 1984; Geitgey and Carr, 1987; Sweet, 1988),
graptolites (e.g., Berry and Boucot, 1972; Erdtmann, 1976; Lenz and Xu, 1985;
Finney, 1986), and ammonoids (e.g., Scott, 1940; Bayer and McGhee, 1985).
Since the probability of collection of a taxon is closely tied to its abundance
(McKinney, 1986a, 1986b), the probability of collection as a function of depth is
modeled here as a Gaussian distribution. This distribution for any specific taxon
can be described with three parameters: (1) preferred depth, (2) depth tolerance,
and (3) peak abundance (Fig. 1C). The preferred depth corresponds to the optimal
water depth in which the taxon lived. The depth tolerance corresponds to the
sensitivity of a taxon to water depth, and is measured as the standard deviation of
the distribution. Peak abundance reflects the abundance of the taxon at the preferred
depth, and sets the height of the Gaussian curve. The probability of collection for
any taxon as a function of depth can be read as the height of the Gaussian curve
at that depth. Where the peak abundance forces the curve to rise above 100% (as
for an extremely abundant taxon), the probability of collection is capped at 100%
(Fig. 1C, distribution C in lower right). With only these three parameters, a wide
variety of distributions can be simulated.
In addition to specifying the form of facies control, the form of facies change
through a stratigraphic section must be specified; in this model, shallowing-upward
cycles bounded by sharp flooding surfaces (e.g., parasequences) are simulated.
Two 40 time-step shallowing-upward cycles are modeled here; in each cycle, wa-
ter depth shallows up linearly from the base to the top of the cycle. Water depth
increases back to its original value at each flooding surface, such that both parase-
quences have identical water-depth histories. Two versions of this facies/cycle
model are presented. In the first (Fig. 4A), all taxa share the same peak abundance
and depth tolerance, but differ in their preferred depths. In the second version, all
taxa differ in preferred depth, peak abundance, and depth tolerance (Fig. 4B).
Several patterns emerge in this model. The fossil record is still incomplete, but
fossil occurrences are no longer randomly distributed as in the sampling effects
model (compare Fig. 3 and Fig. 4A). The combination of facies control and
changing water depths within a cycle produces a characteristic fossil distribution
consisting of a few spotty occurrences low in the parasequence, followed by a
zone in which the fossil achieves a peak abundance, followed by a return to spotty
occurrences (e.g., taxon 37 in Fig. 4A). This pattern of occurrences is repeated
in the following parasequence if the taxon is still extant (e.g., taxa 36 and 39
in Fig. 4A). Cycle boundaries (flooding surfaces) can truncate this characteristic
pattern if the preferred environment of a taxon lies close to the cycle boundary
(e.g., taxon 17 in Fig. 4A, taxon 41 in Fig. 4B).
The stratigraphic separation of clusters of occurrences of a taxon is produced
by the recurrence of the preferred facies in successive cycles and generates a rather
long gap separating zones in which the fossil occurs frequently. Abnormally long
gaps have frequently been cited as the major non-random feature of fossil gap
distributions (Paul, 1982; McKinney, 1986b).
Originations and extinctions that occur near cycle boundaries can lead to first
and last occurrences that are displaced vertically from the cycle boundary because
of facies control (taxon 11 in Fig. 4A). In many cases, this displacement represents
only a small fraction of the range of a taxon, but in some cases, the difference
6 S.M. Holland
A-50 % sampling B-10 % sampling

80 I I'11:1:I:~~i' 80
:::1 11!~i!I.I~I
.....
:...... j"..j!II~::
-. 1
.'
a
.~
60 1 .I.::! ~I: i. I :::
!.:d: 11 ::. :'11
a 60
.~
'1 II .. .11
~ .1:1 . :111 11 '1 II i:1 ~
4D 1::!I.:j::II= II' :.1 I:I'!' 40
II 11'1' I I : 1= 1
;:1 i: . i! ~I' . . !:II!.!
1:_.- 1-'.: I I.; :.. i
!.: ":.
20 .1' 1,
I. I, I .' I
I
I r'
:.. ' 11 20
i=i 1'1. ':: i:j..:~ .::
I! .:! I:'!!I III
oL-~~~j~:-L~~~~-L__~~-L__~
10 20 30 40 5D 10 20 30 40 50
Taxa Taxa
Fig. 3. Stratigraphic ranges in a single measured section produced by the Sampling Effects Model
using same taxa as in the Perfect Stratigraphic Record Model. Assuming a 50% probability of
collection (A, left) for all taxa at any given horizon produces a fossil record with gaps. Assuming
a 10% probability of collection (8, right) increases the number and size of the gaps, so that the
preserved range indicated by the black symbols can be much less than the actual range in which the
taxon was alive, indicated by the gray bar.
A - Same facies control Water Depth B - Different facies control

~ ~----------------~~r-m ~ r---------------70nr~~
c:
c
.~
~
20
10 20 30 40 so 10 20 30
Taxa Taxa
Fig. 4. Stratigraphic ranges in a single measured sectIOn produced by the Facies/Cycle Model using
same taxa as in the previous two models. The 80 time steps are divided into two identical shallowing
upward cycles bounded by a sharp flooding surface. In A (left), all taxa share the same depth tolerance
and peak abundance, but differ in their preferred depths. Shallow water forms occur, as expected,
at the tops of parasequences, whereas deeper water forms occur at the bases of parasequences. In
B (right), all taxa have unique values of depth tolerance, preferred depth, and peak abundance.
Eurytopic forms can be recognized by their occurrence in all water depths; stenotopic forms can be
recognized by their occurrence in relatively narrow ranges of water depth.
between the true range and the preserved range can be significant (for example,
taxon 9 in Fig. 4A; taxon 33 in Fig. 4B). The maximum extent of this error in the
timing of extinction or origination is one cycle, provided the preferred facies is
preserved in the overlying and underlying cycles.
There is also a tendency for first and last occurrences to be somewhat concen-
trated near cycle boundaries, and this will be explored further later in the paper. If
a shallow water taxon becomes extinct shortly after a cycle boundary (but before
it is preserved later in the cycle), its last occurrence will lie immediately below the
sharp facies change occurring at the cycle boundary prior to its time of extinction
(e.g., taxon 19 in Fig. 4A). Likewise, deeper water taxa that originate late in a
cycle will appear abruptly at the base of the next cycle, where the proper facies is
preserved (e.g., taxon 15 in Fig. 4A).
It should be emphasized that all of these patterns result solely from abrupt
facies changes at cycle boundaries and that no depositional hiatus is required to
produce these patterns. Any hiatus present will further enhance some patterns,
particularly clustering of first and last occurrences (Bambach and Gilinsky, 1988).
4. A Sensitivity Analysis of Facies Control
Because the patterns that emerge from this model depend so strongly on the
values of preferred depth, depth tolerance, and peak abundance, as well as the
depths present within a cycle, it is useful to non-dimensionalize these variables
and systematically explore their effects on fossil distributions. Preferred depth can
be non-dimensionalized as
*PD = PD - SD (1)
DD-SD'
where *PD is the non-dimensionalized preferred depth, PD is the preferred depth
of a taxon, SD is the shallowest depth within a cycle and DD is the deepest depth
within a cycle. Smaller values of *PD indicate a shallow water taxon, and larger
values of *PD indicate a deeper water taxon. If a taxon has a "'PD between 0 and
1, its preferred depth occurs within the cycle.
Depth tolerance can be non-dimensionalized as
*T= T (2)
DD - SD'
where *T is the non-dimensionalized depth tolerance, and T is the depth tolerance.
Larger values of "'T indicate a eurytopic taxon and smaller values indicate a
stenotopic taxon.
Peak abundance can be non-dimensionalized as
"'A=~ (3)
100%'
where "'A is the non-dimensionalized peak abundance normalized to a peak abun-
dance (A) of 100%. Larger values of "'A indicate an abundant taxon and smaller
values indicate a rare taxon.
8 S.M. Holland
0.25 0.50 0.75 1.00
shallow Preferred Depth ----+ deep
0.58 0.36 0.20 0.10 0.06
eurytopic ...- - - Depth Tolerance----+' stenotopic
1.50 1.00 0.75 0.50 0.25

abundant -Peak Abundance---- rare
LJ
Fig. 5. Sensitivity analysis of preferred depth, depth tolerance, and peak abundance. Note that the
boxed set of runs in the center indicate three runs under the same conditions (preferred depth =
0.50, depth tolerance =0.20, and peak abundance =0.75). Within each row, only one parameter was
varied while the other two were held constant; for example, the bottom row of runs were made with
preferred depth =0.50 and depth tolerance =0.20, while peak abundance was varied from 1.50 to
0.25. All runs in all rows simulate the same, linearly shallowing-upward parasequence.
A single shallowing-upward parasequence was simulated (Fig. 5). Because the

taxon is assumed to be alive throughout the parasequence, and because preferred
depth and depth tolerance are scaled here to the depth range of the parasequence,
no time-scale is assigned to this simulation. Although numerous combinations
of the three parameters were tested, thirteen are presented here to illustrate the
effects of each parameter. Three runs of a single baseline combination of preferred
depth (0.50), depth tolerance (0.20), and peak abundance (0.75) is presented in
the bold box in Fig. 5. The differences in the three runs reflect the importance of
the stochastic component of the model. Within each row, a single parameter was
varied while holding the other two parameters constant to isolate their effects. For
example, in the first row, preferred depth was varied from 0 to 1.0 while holding
depth tolerance at 0.20 and peak abundance at 0.75.
4.1. PREFERRED DEPTH
As preferred depth is increased from 0 to 1 (Fig. 5, top row), the taxon progressively
appears from the top of the parasequence towards the bottom of the parasequence.
At values of 0.5, the characteristic pattern of facies control on the distribution of at
taxon is apparent: spotty occurrences low in the parasequence, followed by a zone
of frequent occurrences, again followed by a zone of spotty occurrences. As the
value of preferred depth becomes closer to either 0 or 1, this pattern of occurrences
is truncated at the flooding surface. For instance, the deep water taxon (*PD =
1.0) appears frequently and then sparingly, without an initial lower zone of scarce
occurrences.
4.2. DEPTH TOLERANCE
As depth tolerance is decreased from 0.58 to 0.06 (Fig. 5, middle row), the taxon
can be found over an increasingly smaller portion of the parasequence. The proba-
bility of collection is little affected in the center of the range, despite the restriction
of the taxon to a smaller interval in the parasequence. The run at 0.06 also illustrates
the effect of random chance in this model; in this case, the taxon lacks the spotty
occurrences above and below the zone of common occurrences. In another run un-
der the same conditions, those spotty occurrences might be present; again, compare
the three cases in the black box - all were run under identical conditions. Small
depth tolerances contract the range of a taxon within a parasequence, and thereby
favor anomalously long gaps between occurrences in successive parasequences.
4.3. PEAK ABUNDANCE
As peak abundance is decreased from 1.50 to 0.25 (Fig. 5, bottom row), the taxon
can be found over approximately the same range of the parasequence, but the
frequency with which it is collected drops over the entire range. The effect of
changes in peak abundance can be distinguished from changes in depth tolerance
by the lack of systematic range contraction as peak abundance decreases, and by
the lack of decreased collection probabilities in the center of the range as depth
tolerance decreases.
10 S.M. Holland
4.4. EXTINCTION AND ORIGINATION
Origination or extinction within a parasequence will have the effect of truncating

the characteristic scarce-common-scarce pattern of facies control. Except in the
case of highly eurytopic taxa, the first occurrence datum is likely to occur well
above the horizon at which a taxon originates, and the last occurrence datum is
similarly likely to occur well below the horizon at which a taxon becomes extinct.
Only in the cases of strong eurytopy or when a species originates or becomes
extinct while the proper facies is present is the time or origination or extinction
likely to be known, a well-accepted observation (Shaw, 1964). The magnitude of
this error increases with smaller values of depth tolerance and smaller values peak
abundance.
4.5. DEPTH RANGE OF PARASEQUENCE
Since non-dimensionalized preferred depth and depth tolerance are scaled to the
depth range of a cycle, changes in the depth range of a cycle can mimic changes
in preferred depth or depth tolerance of a taxon. For example, high-amplitude
parasequences will make taxa appear to be relatively stenotopic. Therefore, high-
amplitude parasequences will favor anomalously long gaps between clusters of
occurrences in successive parasequences. A greater depth range within a parase-
quence also makes it more likely that any given taxon will be collected within a
parasequence, and thereby also tends to minimize the discrepancy between first
occurrence and the time of origination and between last occurrence and the time
of extinction.
5. Sequence Control on First and Last Occurrences
In Fig. 4, there is a subtle tendency for first and last occurrences to cluster at
cycle boundaries. Because so few taxa are simulated in that model, it is difficult
to evaluate the strength of that tendency. The cycles in Fig. 4 are also highly
stylized relative to the more complicated parasequence stacking patterns observed
in depositional sequences, so more realistic sequence architectures are treated
below.
5.1. A No-LOWSTAND SEQUENCE

A common type of depositional sequence in upramp and cratonic areas is charac-
terized by a relatively thin transgressive systems tract, a thicker highstand systems
tract, and no preserved lowstand systems tract. Such depositional sequences have
been described from Paleozoic cratonic settings (e.g., Brett and others, 1990; Hol-
land, 1993) and from landward areas in the Neogene of the Atlantic coast (e.g.,
Kidwell, 1984; Greenlee and others, 1992), for example. These areas presumably
remained relatively shallow through much of a sequence and were subaerially ex-
posed for the entire lowstand systems tract, resulting in a lack oflowstand deposits.
These sequences are characterized by one or a few relatively major flooding sur-
faces during the transgressive systems tract and several relatively minor flooding
surfaces during the highstand systems tract. Such a case was modeled (Fig. 6, and
appendix for details of model) with a constant diversity of 1000 taxa. The rate of
shallowing within a parasequence and the duration of a parasequence was held
constant for all parasequences. A single stratigraphic section through two identical
sequences was modelled.
First occurrences of taxa (FAD) in this model generally fall into a relatively
low but fluctuating background level, with the exception of several major spikes.
These major FAD spikes occur at each of the three transgressive systems tract
flooding surfaces, and each successive spike decreases in magnitude. The first
spike, corresponding to the combined transgressive surface/sequence boundary
(again, no lowstand is present) is dominated by shallow water taxa and eurytopic
taxa that originated during the time of the lowstand. The second spike, correspond-
ing to the second transgressive flooding surface, is composed of deeper water taxa
that originated in the previous highstand and lowstand systems tracts when water
depths were too shallow for the taxa to be encountered in the measured section. The
third spike, corresponding to the third transgressive flooding surface is relatively
subdued, and reflects the relatively few deep-water stenotopic taxa that originated
since the previous sequence. The relatively small size of this peak is attributed
to the model constraints which demanded that all taxa have their preferred depth
within the total depth range of the sequence; if more deeper water taxa were gener-
ated in the model, this third spike would be much larger. Each of the transgressive
FAD spikes is followed immediately by a horizon of higher-than-average number
of FADs, reflecting the low probability of collecting a rarer taxon in the first horizon
in which the proper facies is present. The height of these three transgressive spikes
can be several times the background FAD level, but their height will depend on the
relative mix of shallow versus deep water taxa, eurytopes versus stenotopes, rare
versus abundant taxa, the magnitudes of flooding, and the duration of the sequence.
Therefore, no generalization about the intensity of these spikes can be made for all
situations.
Last occurrences of taxa (LAD) in this model also fall into a relatively low but
fluctuating background level with the exception of several spikes. Three situations
consistently produce a LAD spike in this model: the horizons just before the second
and third transgressive systems tract flooding surfaces and the horizon just below
subaerial exposure. The horizons just beneath the second and third transgressive
flooding surfaces produce a cluster of LADs dominated by shallow water taxa that
go extinct prior to the next occurrence of shallow water depths in the following
highstand. The horizon just beneath the surface of subaerial exposure (sequence
boundary) is also marked by a cluster of LADs dominated by shallow water and
eurytopic taxa that go extinct during the lowstand. The intensity of the LAD peaks
may be several times the height of background LAD levels, but their height will
depend on the same factors as the FAD spikes described above. In addition, the
LAD spike in the horizon beneath the subaerial exposure surface can be further
enhanced by both a longer period of subaerial exposure (hiatus of Wheeler, 1964)
tv
-
70 60 50 40 30 20 10 0 65 o 40 0 10 20 30 40 50 60 70
First Occurrences Water Depth (m) Last Occurrences

Fig. 6. Clusters of first and last occurrences produced in a single measured section through two identical missing-lowstand ttl
sequences. typical of many cratonic and upramp areas. The shaded portion of the water depth curve represents submarine
deposition; whereas the white portion represents subaerial exposure. No fossil preservation occurs during subaerial exposure.
~
Spikes of last occurrences occur immediately beneath the sequence boundary and immediately beneath the TST flooding ~
surfaces. Spikes of first occurrences occur immediately following all of the TST flooding surfaces. including those at the
sequence boundaries. ~
Q.
and by any erosion that takes place during the lowstand (degradation vacuity of
Wheeler, 1964). Note that erosion was not simulated in these models.
Although immigration into the basin or emigration out of the basin were not
simulated here, their effects in the model would be identical to that of origination
or extinction, respectively. That is, an "origination" within the model could be
viewed as either a true origination or merely an immigration into the basin of a
taxon that had previously originated elsewhere. Originations and extinctions in
these models could represent either global events or merely basin-wide events; in
model terms, they would be indistinguishable.
5.2. A COMPLETE SEQUENCE

A second common type of sequence is characterized by relatively deep water
through much of the transgressive and highstand systems tract, and a lowstand
systems tract that does not result in subaerial exposure. Such a sequence could be
developed in a basinal setting during a type 1 sequence boundary, or a shelf/ramp
edge during a type 2 sequence boundary (Van Wagoner and others, 1990). Forced
regressions in a deep-water setting (Po sam en tier and others, 1992; Posamentier
and Allen, 1993) result in this type of sequence. The sequence modeled here
is characterized by major transgressive flooding surfaces, minor highstand and
lowstand flooding surfaces, and a major basinward shift of facies at the base of
the lowstand (Fig. 7). No subaerial exposure or erosion was incorporated in this
model, but their effects will be discussed below. As in the previous model, diversity
was constant at 1000 taxa, all parasequences have the same duration and rate of
shoaling, and a single stratigraphic section through two identical sequences was
simulated.
As in the previous model, FADs generally form a low but fluctuating back-
ground level punctuated by several spikes. The transgressive systems tract flood-
ings produce spikes several times as high as the background level, and are com-
posed of deep water taxa that have originated since the deep-water facies of the
underlying sequence. These transgressive FAD spikes are dwarfed in this run by
the prominent FAD spike corresponding to the horizon immediately above the bas-
inward shift in facies. Unlike the other FAD spikes, this sequence boundary spike
is dominated by shallow water taxa that have originated since the shallow-water
facies of the previous sequence.
LADs also generally form a low but fluctuating background level punctuated
by spikes. As in the previous model, several of the LAD spikes occur immediately
below the transgressive systems tract flooding surfaces and represent the last
occurrences of shallow water taxa that go extinct before shallow water facies are
encountered again upsection. Several short LAD spikes occur immediately below
the basinward shift in facies. Unlike other LAD spikes, these basinward shift
spikes represent the last occurrences of deep-water taxa that go extinct prior to the
deep-water facies of the next sequence.
.....
~
80 70 60 50 40 30 20 10 0 140 o 0 10 20 30 40 50 60 70 80
CI)
Water Depth (m)
First Occurrences Last Occurrences ~
Fig. 7. Clusters of first and last occurrences in a single measured section through two identical complete sequences, typical ~
of type 1 sequences in deep-water settings, and type 2 sequences near the shelf edge. Conventions same as in Fig. 6. Clusters
of last occurrences immediately beneath the sequence boundary and all TST flooding surfaces. Clusters of first occurrences
~
:::s
Q.
occur immediately above the sequence boundary and at alI TST flooding surfaces.
5.3. GENERALIZATIONS
From these simulations, several generalizations can be drawn about the predicted
distribution and composition of FAD and LAD spikes in the stratigraphic record.
All of these spikes are enhanced by increases in the proportion of stenotopic and
abundant taxa.
5.3.1. Sequence Boundary and Lowstand Systems Tract

Sequence boundaries, where they are marked by an abrupt basin ward shift in
facies and a preserved lowstand systems tract, should be characterized by a spike
of first occurrences immediately above the sequence boundary. This spike of first
occurrences should be dominated by shallow-water taxa. Such a sequence boundary
should be immediately underlain by a spike of last occurrences dominated by deep-
water taxa.
Where no lowstand systems tract is present and the sequence boundary and
transgressive surface are combined into the same physical surface, the sequence
boundary should be overlain by a spike of first occurrences dominated by deeper
water taxa. The sequence boundary in this case should be underlain by a spike of
last occurrences of shallow-water taxa.
Erosion at a sequence boundary and increased duration of erosion or subaerial
exposure at a sequence boundary will enhance any FAD or LAD spikes developed
at a sequence boundary.
5.3.2. Transgressive Systems Tract

Flooding surfaces within the transgressive systems tract should be overlain by a
spike of first occurrences of relatively deep water taxa. These flooding surfaces
should overlie a spike of last occurrences dominated by relatively shallow water
taxa.
If the transgressi ve systems tract flooding surfaces are accompanied by signif-
icant condensation and enhanced preservation of fossils (Loutit and others, 1988;
Kidwell, 1989, 1991), the spike of first occurrences dominated by deep-water taxa
should be even further enhanced.
5.3.3. Highstand Systems Tract

Unless the highstand systems tract is composed of high-amplitude parasequences
with significant condensation at flooding surfaces, no significant spikes of FAD or
LAD are predicted within the highstand systems tract. A LAD spike may occur at
the top of the highstand systems tract where it is overlain by a lowstand deposit
marked by an abrupt basin ward shift of facies.
16 S.M. Holland
6. Preliminary Tests of the Model
6.1. THE UPPER ORDOVICIAN OF THE CINCINNATI ARCH
The Upper Ordovician of the Cincinnati Arch consists of a thick assemblage of

skeletal carbonates and shales deposited on a gently dipping cratonic ramp. Fos-
sil preservation in the type Cincinnatian is legendary both for its quality and its
abundance. Most limestone beds are densely packed with well-preserved whole to
broken shelly marine invertebrates, and many hundreds of species have been de-
scribed (Dalve, 1948). Cumings and Galloway (1913) present stratigraphic ranges
for over 230 species through the nearly continuous Tanner's Creek section in
southeastern Indiana. The vast majority of these taxa are benthic, and the list is
dominated by a typical open marine Ordovician assemblage of bryozoans, bra-
chiopods, trilobites, and molluscs, with lesser numbers of polychaetes and corals.
The sequence stratigraphy of the Cincinnatian has been recently studied (Hol-
land 1992, 1993; Holland and others, 1993). As is typical of many cratonic regions,
depositional sequences in the Cincinnatian generally consist of alternating tran-
gressive and highstand systems tracts; for most sequences, the lowstand is entirely
missing and is represented by a subaerial unconformity. Water depths and facies
in the Tanner's Creek section ranged from shale-dominated storm beds in the
offshore, to mixed packstone-shale storm beds in the transition zone, to highly
burrowed wavy to nodular-bedded packstones in the shoreface. Lagoonal micritic
carbonates occur at the very top of the section within the Saluda Dolomite. The
Tanner's Creek Section, now heavily overgrown in places, spans all but the lowest
portion of the Cincinnatian Series.
In Cumings and Galloway'S compilation, individual taxa rarely show the char-
acteristic pattern of rare occurrences near the limits of their range and abundant
occurrences near the center of their range. Nonetheless, many of these taxa are
clearly facies-controlled because they reoccur either at the species or genus level
within a particular facies in several sequences (Holland, 1995). This type of dis-
tribution in which taxa are facies controlled, yet lack the rare occurrences at the
edges of their range could be generated by very large peak abundances and narrow
depth ranges.
First and last occurrences within this data set are clearly non-randomly dis-
tributed, and their distribution contains many more spikes than should occur ran-
domly. The 99.9% peak lines (Fig. 8) represent the maximum peak sizes to be
expected 99.9% of the time, if the total number of first and last occurrences were
randomly distributed among the total number of horizons. Thus, first and last oc-
currences are far more clustered in this data set than would be expected from a
random distribution.
Some of this clustering occurs near combined sequence boundaries / transgres-
sive surfaces (Fig. 8). Other large peaks occur within the transgressive systems
tracts, such as the FAD spike during the C3 sequence, and the LAD spike during
the C2 sequence. Several anomalous spikes occur within the C5 highstand and
are unexplained by the models presented here; these spikes may represent true
biologic events.
Upper Ordovician - Southeast Indiana

.,
99.9% Lithostrat. '"
Ol
l!! Sequences Water Depth
99.9%
peak
peak C/)
1ST 0
C6
I
\
TST
Saluda
Whitewater c:
co HST
'0
c:
0
C5
Libert
E
.r=
Waynesville TST
a:
()
Arnheim
C4
HST
TST ")
~
u urn
Corryville c:
C3 TST
Bellevue ~ HST
.~
>-
co
::;;
C2 TST
Fairview lST
Kope c:
co
.,
c
"0
W
C1 HST
\
50 40 30 20 10 0 0 10 20 30 40 50
First Occurrences Last Occurrences
Fig. 8. First and last occurrences in the Upper Ordovician Tanner's Creek section in southeastern
Indiana. Cumings and Galloway (1913) divided the section into approximately 1.5 m (5 foot)
intervals and pooled all identifiable fossils from each interval; the vertical thickness of each bar
equals their sampling interval. Locations of lithostratigraphic contacts, stage boundaries, sequence
boundaries, and systems tracts are plotted. Sequences are numbered from Cl through C6; note that
this provisional numbering has been modified since Holland (1993), based on the recognition of a
sequence boundary at the base of the Fairview Formation (Holland and others, 1993). Water depths
within the section are shown diagramatically; I indicates intertidal depths, S indicates shoreface, T
indicates transition zone, and 0 indicates offshore. Gaps between water depth segments indicate the
presence of a sequence-bounding unconformity. Line at 99.9% peak indicates the maximum number
of first or last occurrences expected in 999 out of 1000 trials, gi ven a random distribution of first and
last occurrences among all horizons.
Some combined sequence boundaries/transgressive surfaces involve relatively

smaller water depth changes (such as the basal C4 sequence boundary), whereas
other boundaries involve much larger water depth changes (the basal Ml and C5
sequence boundaries). Boundaries with large depth changes tend to have greater
amounts of clustering of first and last occurrences than boundaries with smaller
depth changes, as expected.
The North American Upper Ordovician Edenian, Maysvillian, and Richmon-
dian stages were defined in the Cincinnati area, and the bases of all three correspond
to major faunal and lithologic changes in the area (Holland, 1993). The base of
the Edenian stage corresponds to the base of the Kope Formation at Cincinnati. In
this area, the base of the Kope represents one of the most prominent transgressive
systems tract flooding surfaces in the entire Cincinnatian and is characterized by
18 S.M. Holland
the disappearance of numerous shallow water taxa (e.g, brachiopods Platystrophia,

Hebertella and Rhynchotrema) and the appearance of many deeper water taxa (e.g,
trilobite Triarthrus). The base of the Maysvillian stage corresponds to the contact
between the Kope and Fairview Formations in the Cincinnati area. Recent work has
indicated that this is a previously undetected type 2 sequence boundary (Holland
and others, 1993); this surface is characterized by the disappearance of many deep-
water Kope Formation taxa and the sudden appearance of many shallow water taxa
typical of the Maysvillian stage (e.g, return of the brachiopods Platystrophia and
Hebertella). The base of the Richmondian stage corresponds to the base of the
Arnheim Formation in the Cincinnati area. This horizon marks a transgressive
systems tract flooding surface and is also characterized by the introduction of
many new taxa (Holland, 1995). Thus, evidence from the Upper Ordovician of the
Cincinnati Arch suggests that major faunal changes that define stage boundaries
correspond closely to either transgressi ve systems tract flooding surfaces or to type
2 sequence boundaries, both of which are predicted by modelling to be surfaces
of major faunal change. Stage boundaries in other regions may also correspond to
such surfaces.
6.2. OTHER EVIDENCE
In his study of the evolution of the bryozoan Metrarabdotos, Cheetham (1986)

lists three lines of evidence that suggested to him that the first occurrences of
various species reflected their true origination in time. He reported that the first
occurrences were spread throughout the section rather than concentrated at discrete
horizons. The species initially occur in one section rather than in several sections
simultaneously. The first occurrences are generally characterized by markedly
lower abundances than occurrences higher in the section. Note that all three of
these observations are consistent with fossil distributions controlled by facies
and depositional sequences. These observations do not necessarily contradict the
conclusions of his study, but they do point out the difference between one worker's
conception of the effect of facies control and the effects illustrated by the models
presented here.
Many biostratigraphic studies that illustrate the occurrences of taxa through a
measured section reflect the patterns described here. For example, in a recent study
of ammonites below the K-T boundary in Spain and France (Ward and Kennedy,
1993), clustering of last occurrences near lithologically recognizable flooding
surfaces, the scarce-common-scarce pattern of occurrences within lithologically
definable parasequences, and the truncation of this pattern at lithologically sharp
flooding surfaces can all be observed in their measured sections (see their Figs. 4
and 8, for example).
The sequence models presented here suggest that there should exist horizons
characterized by the abrupt last occurrences of many relatively shallow water
taxa and the abrupt first occurrences of many relatively deep water taxa. If the
fossil record were read literally, this pattern might be misinterpreted as a mass
extinction followed by a major radiation of deep water taxa. Cambrian trilobite
biomeres share these characteristics (Palmer, 1984), and recent studies of Cambrian
sequence stratigraphy indicate that the bases of the Pterocephalid, Ptychaspid, and
Symphysurinid biomeres correspond to major transgressive systems tract flooding
surfaces (Osleger and Read, 1993; Patzkowsky and Holland, 1993; Westrop and
Ludvigsen, 1987). The possibility that these major biostratigraphic divisions are
enhanced by sequences and facies control needs to be tested.
Recent studies in the Lower Paleozoic of New York State by Carlton Brett and
co-workers have revealed a phenomenon known as Coordinated Stasis, in which a
fauna is relati vely stable both morphologically and taxonomically for a long period,
and then suddenly undergoes both major morphologic and taxonomic change. The
horizons at which these changes occur typically correspond with major flooding
surfaces in the early transgressive systems tract (Brett and Baird, 1992). Again,
the possibility must be tested that the expression of these events is enhanced by the
combined effects of facies control and the stacking of facies within depositional
sequence.
Recent detailed studies of extinction at the Permo-Triassic boundary have
emphasized that the extinction is concentrated into several "pulses" (Hallam, 1993;
Wignall and Hallam, 1992; Wignall, 1993). All of the "extinction" pulses coincide
with major floodings in the early transgressive systems tract. The question again
arises as to whether these extinctions appear to be pulsed because true pulsed
extinction or because of sequence and facies control on the faunas?
7. Conclusions
1) Facies control and the abrupt juxtaposition of facies within depositional se-
quences will cause first and last occurrences to be displaced away from the horizons
of origination and extinction. Provided that the proper facies is preserved in the
overlying or underlying sequence, the maximum extent of this error is equivalent
to the duration of one depositional sequence.
2) Clusters of first and last occurrences can be produced by taxa that are abun-
dant and/or stenotopic, large facies changes at flooding surfaces and sequence
boundaries, long sequences, long durations of non-deposition, and significant ero-
sion.
3) Major flooding surfaces in the transgressive systems tract are expected to
produce a cluster oflast occurrences of shallow water taxa immediately beneath the
flooding surface and a cluster of last occurrences of shallow water taxa immediately
above the flooding surface.
4) Where sequence boundaries are characterized by a sharp basinward shift
of facies and preservation of a lowstand systems tract, they should be expected
to produce a cluster of last occurrences of deep water taxa immediately beneath
the sequence boundary and a cluster of first occurrences of shallow water taxa
immediately above the sequence boundary. Where no lowstand systems tract is
present and the sequence boundary is combined with a transgressive surface, the
sequence boundary is expected to take on the faunal characteristics of a flooding
surface in a transgressive systems tract. Erosion and increased duration of hiatus
at a sequence boundary will enhance the magnitude of FAD and LAD spikes.
20 S.M. Holland
5) In the absence of high-amplitude parasequences, no clustering of first or

last occurrences produced by facies control and sequences is expected within the
highstand systems tract. Such clustering found in the HST may represent a true
biologic event and not an artifact of facies control and sequences.
Acknowledgments
I thank Mark Patzkowsky, Tomasz Baumiller, Mike Foote, Charles Marshall, edi-
torial reviewer Mike Steckler, and an anonymous reviewer for helpful discussions,
comments, and reviews. I also thank numerous but unnamed workers for shar-
ing their fossil patterns and their enthusiasm for this work. Acknowledgement is
made to the Donors of the Petroleum Research Fund, administered by the Ameri-
can Chemical Society, for support of this research. Programs written in Symantec
Think C 5.0 programs for the Apple MacIntosh that were used in this study are
available on request from the author.
,Appendix
All models in this paper folIow essentially the same procedure. The model steps
through time, and at each time step, calculates the current water depth, and cal-
culates the probability of collection (always between 0 and 1) for each taxon. A
random number between 0 and 1 is generated, and if it is less than the probability
of collection, the program records the occurrence. The program can generate either
a list of occurrences for each taxon or simply the first and last occurrences for all
taxa.
In the perfect stratigraphic record model, the probability of collection is 1 at all
horizons. In the sampling effects model, the probability of collection is set to some
constant value less than zero (0.5 and 0.1 in the simulations presented here). In all
facies control models, the probability of collection is generated from the following
equation:
P(collection) = A e-(d-PD)2/ 2T2,

where A is the peak abundance, PO is the preferred depth, T is the depth tolerance,
and d is water depth at that horizon. The form of the curve is Gaussian-like
(symmetrically bell-shaped), with PD analogous to the mean and T analogous to
the standard deviation.
In the no-lowstand sequence and complete sequence simulations, the values
of A, PD and T were randomly selected for all taxa. All were selected from flat
probability distributions between two endpoints, that is, the 0 to 1 distribution
supplied by the random number generator was rescaled to fit two endpoints. For
*A, the endpoints were 1.5 and 0.25. For PD, the endpoints were equal to the
deepest depth (65 in the no-lowstand simulation, and 140 in the complete sequence
simulation) and zero. For SO, the endpoints were equal to 0.32 x DD (the deepest
depth) and 1. This arbitrary cutoff for SD (approximately 1/3 the deepest depth)
was obtained empirically through the sensitivity analysis to prevent the simulation
from being overwhelmed by eurytopic taxa that display few of the effects of the
model.
References
BAMBACH, R.K. and GILINSKY, N.L., 1988, Artifacts in the apparent timing of
macroevolutionary "events": Geological Society of America Abstracts with Programs,
v. 20, p. AI04.
BAYER, U. and MCGHEE, G.R., 1985, Evolution in marginal epicontinental basins: The
role of phylogenetic and ecologic factors (Ammonite replacements in the German Lower
and Middle Jurassic), in Bayer, U., and Seilacher, A., ed., Sedimentary and Evolutionary
Cycles: Springer-Verlag, New York, p. 164-220.
BERRY, W.B.N. and BOUCOT, A.I., 1972, Silurian graptolite depth zonation: International
Geological Congress 24th Session, section 7, Paleontology, p. 59-65.
BRETT, C.E. and BAIRD, G.c., 1992, Coordinated stasis and evolutionary ecology of
Silurian-Devonian marine biotas in the Appalachian Basin: Geological Society of Amer-
ica Abstracts with Programs, v. 24, p. A139.
BRETT, C.E., GOODMAN, W.M. and LODUCA, S.T., 1990, Sequences, cycles, and basin
dynamics in the Silurian of the Appalachian Foreland Basin: Sedimentary Geology, v.
69, p. 191-244.
CHEETHAM, A.H., 1986, Tempo of evolution in a Neogene bryozoan: Rates of morpho-
logic change within and across species boundaries: Paleobiology, v. 12, p. 190-202.
CLARK, D.L., 1984, Conodont biofacies and provincialism: Geological Society of Amer-
ica Special Paper, v. 196, p. 1-340.
CUMINGS, E.R. and GALLOWAY, J.J., 1913, The stratigraphy and paleontology of
the Tanners Creek section of the Cincinnati Series of Indiana: Indiana Department of
Geology and Natural Resources Annual Report, v. 37, p. 353-479.
DALVE, E., 1948, The fossil fauna of the Ordovician in the Cincinnati region: University
Museum, Department of Geology and Geography, Cincinnati, Ohio, 56 p.
DODD, J.R. and STANTON Jr., R.I., 1990, Paleoecology: Concepts and applications: John
Wiley & Sons, New York, 502 p.
ERDTMANN, B.-D., 1976, Ecostratigraphy of Ordovician graptoloids, in Bassett, M.G.,
ed., The Ordovician System: Proceedings of a Palaeontological Association Symposium,
Birmingham, September, 1975: University of Wales and National Museum of Wales,
Cardiff, p. 621-643.
FINNEY, S.C., 1986, Graptolite biofacies and correlation of eustatic, subsidence, and
tectonic events in the Middle to Upper Ordovician of North America: Palaios, v. 1, p.
435-461.
GEITGEY, J.E. and CARR, T.R., 1987, Temperature as a factor affecting conodont di-
versity and distribution, in Austin, R.L., ed., Conodonts: investigative techniques and
applications: Ellis Horwood Ltd., Chichester, p. 241-255.
GREENLEE, S.M., DEVLIN, w.J., MILLER, K.G., MOUNTAIN, G.S. and FLEMINGS,
P.B., 1992, Integrated sequence stratigraphy of Neogene deposits, New Jersey continen-
tal shelf and slope: Comparison with the Exxon model: Geological Society of America
Bulletin, v. 104, p. 1403-1411.
HALLAM, A., 1993, Nature of the delayed Triassic marine radiation after the end-
Paleozoic mass extinction: Geological Society of America Abstracts with Programs,
v. 25, p. A156.
HOLLAND, S.M., 1992, Sequence stratigraphy of the Cincinnatian Series, in Davis,
R.A., and Cuffey, R.I., ed., Sampling the layer-cake that isn't: The stratigraphy and
22 S.M. Holland
paleontology of the "type Cincinnatian": Ohio Di vision of Geological Survey Guidebook

No. 13, Columbus, Ohio, p. 199-227.
HOLLAND, S.M., 1993, Sequence stratigraphy of a carbonate-clastic ramp: The Cincinna-
tian Series (Upper Ordovician) in its type area: Geological Society of America Bulletin,
v. 105, p. 306-322.
HOLLAND, S.M., 1995, Using time-environment analysis to recognize faunal events in
the Upper Ordovician of the Cincinnati Arch, in Brett, C.E., ed., Paleontological event
horizons: ecological and evolutionary implications: Columbia University Press, New
York.
HOLLAND, S.M., DATTILO, B.F., MILLER, A.I., MEYER, D.L. and DIEKMEYER,
S.c., 1993, Anatomy of a mixed carbonate-clastic depositional sequence: Kope For-
mation (Upper Ordovician: Edenian) of the Cincinnati Arch: Geological Society of
America Abstracts with Programs, v. 25, p. 338.
KIDWELL, S.M., 1984, Outcrop features and origin of basin margin unconformities in
the Lower Chesapeake Group (Miocene), Atlantic Coastal Plain, in Schlee, J.S., ed.,
Interregional unconformities and hydrocarbon accumulation: American Association of
Petroleum Geologists, Tulsa, Oklahoma, p. 37-58.
KIDWELL, S.M., 1989, Stratigraphic condensation of marine transgressive records: Origin
of major shell deposits in the Miocene of Maryland: Journal of Geology, v. 97, p. 1-24.
KIDWELL, S.M., 1991, Condensed deposits in siliciclastic sequences: Expected and
observed features, in G. Einsele and others, ed., Cycles and Events in Stratigraphy:
Springer-Verlag, Berlin, p. 682-695.
LENZ, A.C. and XU, C., 1985, Graptolite distribution and lithofacies: Some case histories:
Journal of Paleontology, v. 59, p. 636-642.
LOUTIT, T.S., HARDENBOL, J., VAIL, P.R. and BAUM, G.R., 1988, Condensed sections:
The key to age determination and correlation of continental margin sequences, in Wilgus,
c.K., and others, ed., Sea-level changes: An integrated approach: Society of Economic
Paleontologists and Mineralogists, Tulsa, Oklahoma, p. 183-213.
LUDVIGSEN, R., WESTROP, S.R., PRATT, B.R., TUFFNELL, P.A. and YOUNG, G.A.,
1986, Dual biostratigraphy: Zones and biofacies: Geoscience Canada, v. 13, p. 139-154.
MARSHALL, C.R., 1990, Confidence intervals on stratigraphic ranges: Paleobiology, v.
16, p. 1-10.
MCKINNEY, M.L., 1986a, Biostratigraphic gap analysis: Geology, v. 14, p. 36-38.
MCKINNEY, M.L., 1986b, How biostratigraphic gaps form: Journal of Geology, v. 94, p.
875-884.
OSLEGER, D. and READ, J.F., 1993, Comparative analysis of methods used to define
eustatic variations in outcrop: Late Cambrian interbasinal sequence development: Amer-
ican Journal of Science, v. 293, p. 157-216.
PALMER, A.R., 1984, The biomere problem: Evolution of an idea: Journal of Paleontology,
v. 58, p. 599-611.
PATZKOWSKY, M.E. and HOLLAND, S.M., 1993, Ecologic and environmental pat-
terns of extinction: comparison of Late Cambrian and Middle Ordovician extinctions:
Geological Society of America Abstracts with Programs, v. 25, p. 332.
PAUL, C.R.C., 1982, The adequacy of the fossil record, in Joysey, K.A., and Friday, A.E.,
ed., Problems of Phylogenetic Reconstruction: Academic Press, New York, p. 75-117.
POSAMENTIER, H.W. and ALLEN, G.P., 1993, Variability of the sequence stratigraphic
model: effects of local basin factors: Sedimentary Geology, v. 86, p. 91-109.
POSAMENTIER, H.W., ALLEN, G.P., JAMES, D.P. and TESSON, M., 1992, Forced
regressions in a sequence stratigraphic framework: Concepts, examples, and exploration
significance: American Association of Petroleum Geologists Bulletin, v. 76, p. 1687-
1709.
SCOTI, G., 1940, Paleoecological factors controlling the distribution and mode of life of
Cretaceous ammonoids in the Texas area: Journal of Paleontology, v. 14, p. 299-323.
SHAW, A.B., 1964, Time in stratigraphy: McGraw-Hill, New York, 365 p.
SPRINGER, M. and LILJE, A., 1988, Biostratigraphy and gap analysis: The expected
sequence of biostratigraphic events: Journal of Geology, v. 96, p. 228-236.
STRAUSS, D. and SADLER, P.M., 1989, Classical confidence intervals and the Bayesian
probability estimates for the ends of local taxon ranges: Mathematical Geology, v. 21,
p.411-427.
SWEET, W.e., 1988, The Conodonta: Morphology, taxonomy, paleoecology, and evo-
lutionary history of a long-extinct animal phylum: Clarendon Press, New York, 212
p.
SWEET, W.C. and BERGSTROM, S.M., 1984, Conodont provinces and biofacies of the
Late Ordovician: Geological Society of America Special Paper, v. 196, p. 69-87.
VAN WAGONER, I.e., MITCHUM, R.M., CAMPION, K.M. and RAHMANIAN, V.D.,
1990, Siliciclastic sequence stratigraphy in well logs, cores, and outcrops: American
Association of Petroleum Geologists Methods in Exploration Series, No.7, Tulsa,
Oklahoma, 55 p.
WARD, P.D. and KENNEDY, w.J., 1993, Maastrichtian ammonites from the Biscay Region
(France, Spain): Paleontological Society Memoir, v. 34, p. 1-58.
WESTROP, S.R. and LUDVIGSEN, R., 1987, Biogeographic control of trilobite mass
extinction at an Upper Cambrian "biomere" boundary: Paleobiology, v. 13, p. 84-99.
WHEELER, H.E., 1964, Baselevel, lithosphere surface, and time stratigraphy: Geological
Society of America Bullctin, v. 75, p. 599-610.
WHITIAKER, R.H., 1970, Communities and ecosystems: Macmillan, New York, 162 p.
WIGNALL, P.B., 1993, Anoxia as the cause of the end-Permian mass extinction: Geolog-
ical Society of America Abstracts with Programs, v. 25, p. A155.
WIGNALL, P.B. and HALLAM, A., 1992, Anoxia as a cause of the Permianrrriassic
extinction: facies evidence from northern Italy and the western United States: Palaeo-
geography, palaeoclimatology, Palaeoecology, v. 93, p. 21-46.
CHAPTER 2
Mixed Carbonate-Siliciclastic
Sequence Stratigraphy Utilizing
Strontium Isotopes: Deciphering the
Miocene Sea-Level History of the
Florida Platform
David J. Mallinson and John S. Compton
ABSTRACT: Miocene depositional sequences are identified based on sequence strati-

graphic concepts applied to sixteen cores that transect the Hawthorn Group on the north-
east Florida Platform. Sequence components which represent various stages of sea-level
cycles are identified and interpreted in a mixed carbonate-siliciclastic platform setting
that has been subjected to multiple depositional and erosional events. The 87 Sr/86 Sr com-
position of phosphorite and dolomite is used to determine the age of in-place phosphorite
crusts and dolostone beds (condensed sections) and reworked phosphorite and dolostone
sand and gravel (unconformities and transgressive surfaces) and to constrain the depo-
sitional age of associated lithofacies. A regional sequence stratigraphic framework is
constructed and the depositional and sea-level history of the region is interpreted. The
Sr-derived ages are used to document the age of highstands because the phosphorite
formed from the early diagenesis of organic-rich sediments deposited during periods of
high productivity that resulted from intensified and persistent upwelling associated with
rising and maximum sea level. Seven major depositional sequences are documented that
correspond to local, and possibly eustatic, sea-level fluctuations. At least seven high-
stands occurred between 25 and 6 Ma with maximum flooding of the Florida Platform
from 17-15 Ma.
1. Introduction
The Miocene deposits of Florida present a unique and challenging opportunity for
the application of sequence stratigraphic concepts. The Florida Platform records the
transition from an isolated carbonate platform to a mixed carbonate-siliciclastic
shelf margin after siliciclastics inundated the platform in the Neogene (Scott,
1988; Locker and Doyle, 1992). This long-term transition from a predominantly
25
B. U. Haq (ed.),
26 D.l. Mallinson, I.S. Compton
carbonate depositional environment to a predominantly siliciclastic depositional

environment was interspersed with episodes of carbonate versus siliciclastic dom-
inance responding to variable sea-level positions (Scott, 1988; Locker and Doyle,
1992). Miocene sediments elsewhere on the southeastern U.S. continental margin
preserve the record of high-amplitude, high-frequency sea-level fluctuations (Sny-
der and others, 1982; 1990; Riggs, 1984; Compton and others, 1990; Mallinson
and others, 1994). Deposition, early diagenesis and reworking of these sediments
have produced a complex stratigraphy in the Miocene Hawthorn Group of Florida,
Georgia and South Carolina. These relatively thin, condensed stratigraphic units
were deposited on the extreme updip edge of a broad shallow-water carbonate
platform and provide a novel setting for the application of sequence stratigraphic
principles.
Phosphorite, a rock type that consists predominantly of the carbonate fluorap-
atite mineral francolite, is abundant within the Miocene deposits of the southeast
U.S. The phosphorite is interpreted to have formed during early diagenesis of
organic-rich sediment (Compton and others, 1993). These organic-rich sediments
were deposited during periods of rising sea level and highstand in areas of persis-
tent upwelling and low siliciclastic influx (Snyder, 1982; 1990; Riggs, 1984; Riggs
and Mallette, 1990; Mallinson and others, 1994). Dolomite silt and dolostone beds
and clasts are another early diagenetic rock type that is abundant in the Miocene
deposits of the southeast U.S. Much of the dolomite appears to have formed nearly
concurrently with the phosphorite (Baker and Allen, 1990; Compton and others,
1994). Unaltered phosphorite and dolostone records the Sr-isotopic signature of
the ambient pore-water at the time of their formation (McArthur, 1985; McArthur
and others, 1990). The 87Sr/86Sr composition of phosphorite and dolomite may
therefore be used to date r.hosphogenesis and dolomitization using the established
Neogene seawater 87 Sr/8 Sr record (DePaolo and Ingram, 1985; Hess and others,
1986; Hodell and others, 1991). The 87 Sr/86 Sr ages provide a record of rising sea
level and highstand and may be used to constrain the depositional age of various
associated lithofacies. Utilization of 87Sr/86Sr data for chronostratigraphic pur-
poses represents a major advance in our ability to decipher the depositional history
of Neogene deposits, particularly where biostratigraphic data are poor or absent.
The primary objectives of this paper are (1) to illustrate that sequence strati-
graphic concepts can be successfully applied to the thin, updip Miocene sediments
on the Florida Platform, (2) to demonstrate the utility of Sr-isotopes in sequence
stratigraphy, and (3) to decipher the depositional history and paleoceanographic
significance of these deposits (Riggs, 1984; Compton and others, 1990). Study of
the Florida Platform is complicated by a general absence of outcrops due to the
low relief of Florida, a paucity of biochronostratigraphic control, and inadequate
seismic data. Consequently, we rely primarily on Sr-derived ages and core analysis
for sequence stratigraphic interpretations.
2. Geologic Setting
The Florida Platform is an extensive (2 x 105 km 2 ), tectonically stable carbonate

platform bordered to the west by the Gulf of Mexico, to the east by the Atlantic
Sequence Stratigraphy of the Florida Platform 27
~~~.Study Area
(Fig. 2)
a _ 50 _ _100
IIIIi /
nautical miles
iami
a 50 100 26 0
....... 1
kilometers
.-- .- -
.- .,
Fig. 1. Map of Florida illustrating the depth to the Miocene depositional surface and topographic
features which affected the deposition of Miocene sediments. Structure contours are in meters below
sea level (modi tied from Popenoe, 1990).
28 D.I. Mallinson, I.S. Compton
Ocean, to the south by the Straits of Florida and to the north by coastal plain
sediments of southeast Georgia which prograde into a Paleogene topographic low
called the Gulf Trough (Popenoe and others, 1987). The State of Florida currently
occupies the eastern half of the Florida Platform. The platform is bounded by
escarpments which are enhanced by various chemical and physical submarine ero-
sional processes including interactions with the Florida Current and Gulf Stream
(Mullins and others, 1987; Dillon and Popenoe, 1988). The platform is primarily
composed of Cretaceous, Eocene and Oligocene carbonates and was originally
isolated from the North American continent by the Suwanee Straits in the Cre-
taceous to Eocene, and the Gulf Trough seaway in the Oligocene (Popenoe and
others, 1987; Dillon and Popenoe, 1988). Siliciclastic sediments began invading
the Platform from the north during the late Oligocene to early Miocene as the Gulf
Trough was filled (Popenoe and others, 1987; Scott, 1988; 1990; Locker and Doyle,
1992). The carbonate to siliciclastic transition in the Hawthorn Group of northeast
Florida is recorded in the predominantly carbonate sediments of the Penney Farms
Formation, that are overlain by interbedded carbonates and siliciclastics of the
Marks Head Formation, which are in tum overlain by predominantly siliciclastic
sediments of the Coosawhatchie Formation (Scott, 1988).
Neogene paleotopography of northeast Florida includes several negative and
positive structural features which affected the deposition of Miocene sediments
(Fig. 1). The Jacksonville Basin is a paleotopographic low defined at the un-
conformable contact between the Upper Eocene Ocala Group limestone and the
overlying Miocene Hawthorn Group (Scott, 1988), and is considered a sub-basin
of the Southeast Georgia Embayment (Riggs, 1979; Scott, 1988; Popenoe, 1990).
To the south of the Jacksonville Basin is a broad, low-gradient area referred to as
the St. John's Platform. Positive structural features in the area include the Sanford
High to the south and the Ocala High to the west. The Jacksonville Basin is isolated
from the Atlantic Basin to the east by a depositional topographic high underlying
the continental shelf (Popenoe and others, 1987; Popenoe, 1990). The Hawthorn
Group is absent from the structural highs, maintains a thickness of 30 to 60 meters
across the St. John's Platform and attains a thickness of>150 meters in the center
of the Jacksonville Basin (Riggs, 1979; Scott, 1988).
The Florida Platform is considered to be tectonically stable, although there is
evidence to indicate localized, minor isostatic adjustments from sediment load-
ing and carbonate dissolution during the Plio-Pleistocene (Opdyke and others,
1984). Subsidence rates during the Neogene in northeast Florida are considered
insignificant in their influence on lithofacies migration when compared to rates of
sea-level rise (Mallinson and others, 1994). Therefore, the shallow-water platform
should serve as a sensitive sea-level indicator with lithofacies migrations closely
approximating global sea-level fluctuations coupled with local changes in sediment
sources or fI uxes.
3. Methods
Sixteen cores from northeast Florida were measured and described. These cores
transect Jacksonville Basin and St. John's Platform (Fig. 2). The texture, min-
TABLE 1 Cr.!
(l)
Lithofacies descriptions and interpretations ...0
c:
(l)
Lithofacies Dominant Component Descriptions Other Characteristics Interpretation :::s
n
(l)
LFI phosphorite gravel Dolostone and phosphorite intraclasts reworked condensed
Cr.!
section; palimpsest
~
.....
LF2 quartz sand Tan to gray, unconsolidated, slightly clayey, may contain J--5% phosphorite shoreface palimpsest,
~.
dolosilty, fine to coarse quartz sand sand and shell hash or lowstand wedge
LF3 limestone Gray to white, indurated to semi-indurated shallow shelf
.g
:::s-
moldic to fossiliferous (pelecypods, echinoids), '<:
0
....,
dolomitic (10-20%) limestone
LF4 dolostone Gray, phosphorite and quartz sandy commonly with phosphorite shallow to
::r
(l)
moldic dolostone intraclasts; bored and intermediate shelf ::!1

0
encrusted ::!.
LF5 dolosilt Gray to white, unconsolidabd to commonly burrowed; shallow to fr
semiconsolidated, sandy, clayey dolosilt diatomaceous intermediate shelf ~
to
LF6 peloidal phosphorite Dark gray to black, unconsolidated, muddy, foraminifera-bearing deep shelf, condensed S.
0
sand fine quartz and phosphorite sand section
LF7 phosphorite crust common on dolostone minor glauconite deep shelf, marine
~
flooding surface
LF8 mud Gray-green, unconsolidabd mud diatomaceous; deep shelf
planktonic foraminifera-bearing
N
\Q
30 D.J. Mallinson, 1.S. Compton
":'6~~AAAAAAAAAA
:'SY'~ATLANTIC
OCEAN
W14641
FIG. 6
- - - -
o 10 20 30 40
SANFORD
kilometers
HIGH
Fig. 2. Map of the study area showing depth to the Miocene depositional surface and location of
cores, seismic transect (Fig. 5), and lithostratigraphic transects (Figs. 6 and 7).
eralogy, and isotopic and elemental compositions were determined for selected
samples from the cores. The mineralogy of bulk sediment, grain-size separates,
and phosphorite grain-types (peloidal sands, nodules, intraclasts, and crusts) was
determined using a Scintag XDS 2000 x-ray diffractometer. These data were used
to define the eight major lithofacies presented in Table 1. Sequence boundaries
were identified based on various recognition criteria presented by Reinson (1992),
Van Wagoner and others (1990), Posamentier and others (1988), Baum and Vail
(1988), Esteban and Klappa (1983) and others. These criteria include the presence
of gravellags, shallow-water bioclastics associated with well-sorted quartz sands,
relative
'" . ]5i '"
relative
'"
t
'" 1-
!
water
.!;; Kuhrt#l oS; water Scott #2
i Wl4476 core
~
~ ' ;)5
3 W13751
depth g. II
8." depth
..,
II dePlhl ~
~~ i II
MBSl
(Ieel) .. 1'MBSl
30
20
40
6 5
50
_/0 ~ phosphorite/dolomite gravel (LF1)

CJ quartz sand (LF2)
~ limestone (LF3)
~ dolostone (LF4)
~ dolosllt (LF5)
P ~ phosphorite sand (LF6)
_ phosphorite crust (LF7)
S mud (LF8)
@ phosphorite age (see Table 2) ~ burrows
126.71 dolomite age (see Table 2) GIl diatoms @ echlnolds

/'"V'"V'"' sequence boundary
GIl 1!i. l - 13.9 biostratigraphic age
Fig. 3. Detailed lithologic logs of cores WI4476 and W13751 illustrating interpreted relative water
depths associated with the deposition of sediments, interpreted depositional sequences and sequence
boundaries, Sr-derived ages of phosphorite or dolomite, and correlations between cores (modified
from Mallinson and others, 1994). Sr-derived ages <24 Ma are from the HodeIl and others (1991)
regression. Sr-derived ages >24 Ma are from the Oslick and others (1993) regression.
32 D.J. Mallinson, I.S. Compton
truncated burrows, caliche-like horizons, and hiatuses recognized by Sr-isotope

age estimates or biostratigraphicaIly.
87Sr/86Sr analyses were performed at the University of Florida by D.A. Hodell
on phosphorite and dolostone grains foIlowing the procedure described in Compton
and others (1993). 87Sr/86Sr measurements were normalized to a 86Sr/88Sr of
0.1194 and to a Standard Reference Material 987 87Sr/86Sr value of 0.710235.
Sample ages were determined by using the 87Sr/86Sr regression equations for
Miocene seawater as determined by Hodell and others (1991) and Oslick and
others (1994). The age resolution using the 87 Sr/86 Sr ratio (Tables 2 and 3) is a
function of the slope of the seawater Sr-isotope curve (DePaolo and Ingram, 1985;
Hess and others, 1986; HodeIl and others, 1991; Oslick and others, 1994).
4. Results and Interpretations
4.1. FACIES DISTRIBUTIONS IN RELATION To SEA LEVEL
The interpreted relation between lithofacies and sea level is presented in Table 1
and Fig. 3. The timing of phosphogenesis and dolomitization relative to sea-level
position is most important to this investigation as these are the minerals from
which the 87Sr/86Sr data are obtained. The phosphorite facies is proposed to form
in association with persistent upwelling regimes (Riggs, 1984; Snyder and oth-
ers, 1990; Compton and others, 1993; Mallinson and others, 1994) occurring at
times of maximum rates of sea-level rise and sea-level highstands coincident with
a prolonged landward shift of sedimentation (Mitchum and others, 1977; Riggs,
1984; Haq and others, 1987, 1988; Loutit and others, 1988). Textural associa-
tions and Sr-isotope data indicate that dolomite and phosphorite generally formed
concurrently during early diagenesis, with dolomite forming at somewhat deeper
burial depths in deep-water organic-rich facies, and separately from phosphorite
in adjacent shallow-water lagoonal environments (Compton and others, 1994).
4.2. INTERPRETING SR-IsOTOPE DATA
In order to use the Sr-derived age dates for chronostratigraphic purposes, it must
be shown that the 87 Sr/86 Sr composition of the mineral is unaltered. Evidence that
the 87Sr/86Sr composition of tightly-cemented, dark phosphorite grains used in
this study are not significantly altered includes: (l) high concentrations of sr2+,
Mg2+, Na+ and CO~- in the francolite structure which would otherwise be lost
during weathering (McArthur, 1985; McArthur and others, 1990; Compton and
others, 1993; Mallinson and others, 1994); (2) the presence of included organic
matter and pyrite in phosphorite which would be oxidized during alteration; (3)
the Sr-derived ages are Miocene and become progressively older down-section
(Table 2), with very few exceptions; (4) the 87 Sr/86 Sr composition of phosphorite
is roughly correlative across the Basin and Platform, as well as to phosphorite
in southwest Florida (Compton and others, 1993); (5) Sr-isotopic ages are sup-
ported by biostratigraphic ages where age-diagnostic microfossils co-occur with
phosphorite (Fig. 4) (Mallinson and others, 1994).
TABLE 2
Sr isotopic ratios of phosphorite and dolomite samples from northeast Florida Agel is
derived from Hodell and others (1991) regressions; Age2 is derived from Oslick et al.
(1994) regressions. PP = phosphorite pebble, MPP = multiphosphorite pebbles, PS =
phosphorite sand, PC =phosphorite crust, DP =dolostone pebble, DS =dolostone bed,
Ds =dolosilt
Core Depth (ft) Sample type 87Srf!6Sr Agel (Ma) Age2 (Ma)
WI4477 105 PP 0.708878 9.31.36 11.51.17

WI4477 105 PP 0.708881 9.11.36 11.41.17
WI4477 171 DP 0.708729 15.61.36 16.50.61
WI4477 192 PS 0.708486 20.00.74 20.00.61
W14619 112 PP 0.708922 8 to 6 9.61.42
W14619 151 MPP 0.708964 8 to6 7.7?
W14619 177 MPP 0.708861 1O.01.36 12.31.17
W14619 315 MPP 0.708850 1O.41.36 12.81.17
W14619 358 PP 0.708848 10.51.36 12.91.17
W14619 423 PC 0.708727 16.00.74 16.50.61
W14619 438 PS 0.708674 16.90.74 17.30.61
W14619 463 PS 0.708691 16.60.74 17.00.61
W14376 136 PSG 0.708840 1O.91.36 13.21.17
W14376 143 PS 0.708766 14.01.36 15.90.61
W14376 174 PS 0.708629 17.60.74 17.90.61
W14376 199 PP 0.708583 18.40.74 18.60.61
W14354 46 MPP 0.708830 11.31.36 13.61.17
W14354 88 MPP 0.708692 16.60.74 17.00.61
W14354 103 DS 0.708592 18.30.74 18.50.61
WI4476 141 PP 0.708938 8 to 6 8.8?
WI4476 168 MPP 0.708864 9.81.36 12.21.17
WI4476 230 PP 0.708786 13.21.36 15.60.61
WI4476 230 DP 0.708765 14.01.36 16.00.61
WI4476 230 DP 0.708804 12.41.36 14.81.17
WI4476 247 PS 0.708730 15.51.36 16.50.61
WI4476 302 PS 0.708564 18.70.74 18.90.61
WI4476 321 PC 0.708463 20.40.74 20.40.61
WI4476 382 PP 0.708233 24.20.74 24.71.08
WI4476 382 PP 0.708189 24.9? 25.51.08
W14353 81 MPP 0.708907 8 to 6 1O.21.17
W14353 112 DS 0.708803 12.41.36 14.91.17
W14353 137 PP 0.708568 18.60.74 18.80.61
W14521 172 PP 0.708930 8 to 6 9.21.42
W14521 187 PS 0.708897 8.41.36 1O.71.17
W14521 300 PS 0.708614 17.90.74 18.20.61
W14521 323 MPP 0.708148 25.6? 26.31.08
continued on next page
34 D.l. Mallinson, I.S. Compton
TABLE 2
Continued
Core Depth (ft) Sample type 87 Sr/86Sr Agel (Ma) Age2 (Ma)
WI4413 117 PP 0.708845 10.61.36 13.01.17

WI4413 218 PS 0.708407 21.30.74 2l.20.61
WI4413 218 DP 0.708606 18.00.74 18.30.61
WI4413 218 Ds 0.708601 18.10.74 18.40.61
Wl4413 242 DS 0.708317 22.80.74 22.50.61
W13744 61 PS 0.708922 8 to 6 9.6l.42
W13744 178 PS 0.708590 18.30.74 18.50.61
W13744 205 PS 0.708472 20.20.74 20.20.61
W13751 102 PP 0.708910 8 to6 1O.11.17
W13751 107 PP 0.708869 9.61.36 1l.91.17
W13751 240 PC 0.708691 16.60.74 17.00.61
W13751 275 PS 0.708535 19.20.74 19.30.61
W13751 301 DS 0.708413 21.20.74 21.10.61
W13751 319 DS 0.708129 25.9? 26.7l.08
W14283 90 PP 0.708936 8 to 6 8.9?
W14283 90 PP 0.708904 8 to 6 1O.41.17
W14283 99 PP 0.708888 8 to6 11.11.17
W14283 99 PP 0.708920 8 to6 9.61.42
W14283 174 PP 0.708673 16.90.74 17.30.61
W14283 174 DP 0.708703 16.40.74 16.90.61
W14283 229 PP 0.708467 20.30.74 20.30.61
W14283 229 PP 0.708435 20.80.74 20.80.61
W14193 93 PP 0.708891 8 to6 1O.91.17
W14193 93 PP 0.708858 1O.11.36 12.41.17
W14193 120 PP 0.708851 1O.41.36 12.71.17
The 87Sr/86Sr composition of individual phosphorite grains and crusts from

throughout the study area were found to fall into at least six distinct groups
(Table 3). More than six groups are suggested by the sequence stratigraphy, but
the resolution of Sr-derived ages in the late Miocene prevent us from further
subdividing these samples into distinct groups. The analysis of multiple grains
(e.g. phosphorite peloids) was done when individual grains were not large enough
to provide sufficient material for analysis. Samples composed of multiple grains
commonly yield slightly older Sr-derived ages than individual grain samples from
the same sequence. We interpret this to indicate that some older phosphorite grains
are reworked into younger deposits through mUltiple depositional and erosional
cycles, a process which is most apparent in the cores from topographically higher
areas such as the Sanford High. Reworking of older phosphorite grains into younger
TABLE 3
Sr-derived ages of single phosphorite grains and resulting age groups
87 Sr/86Sr Age (Ma) Error@ Age Group Age (Ma) Error@ Age Group
Hodell 95%CI Mean Oslick 95%CI Mean
0.708938 8 to 6 NR 8 to 6 8.8- NR 10.9

0.708936 8 to 6 NR 8.9- NR
0.708930 8 to 6 NR 9.2 1.42
0.708922 8 to 6 NR 9.6 1.42
0.708920 8 to 6 NR 9.6 1.42
0.708910 8 to 6 NR 10.1 1.17
0.708904 8 to 6 NR 10.4 1.17
0.708891 8 to 6 NR 10.9 1.l7
0.708888 8 to 6 NR 11.1 1.l7
0.708881 9.1 1.36 9.9 11.4 1.17

0.708878 9.3 1.36 11.5 1.17
0.708869 9.6 1.36 11.9 1.l7
0.708858 10.1 1.36 12.3 1.17
0.708851 10.4 1.36 12.8 1.l7
0.708848 10.5 1.36 12.9 1.17
0.708845 10.6 1.36 13 1.17
0.708786 13.2 1.36 13.2 15.5 0.61 15.5
0.708727 16 0.74 16.5 16.5 0.61 16.9

0.708691 16.6 0.74 17 0.61
0.708673 16.9 0.74 17.3 0.61
0.708583 18.4 0.74 18.5 18.6 0.61 18.7

0.708568 18.6 0.74 18.8 0.61
0.708467 20.3 0.74 20.5 20.3 0.61 20.5

0.708463 20.4 0.74 20.4 0.61
0.708435 20.8 0.74 20.8 0.61
0.708233 24.2- NR 24.5 24.7 1.08 25.1

0.708189 24.9- NR 25.5 1.08
- indicates age by extrapolation of nearest Sr-curve regression NR indicates no resolution
Scott #2
W13751 core
MBSL '::.:== depth

(feel)
'Age 2Age
87Sr~Sr (Ma) (Ma)
12.5"""J'i'''I"'''t!?-'li''l''~- 100 0.708910 86 10.1
110 0.708869 9.6 11.9
120
130
Dialoms Silicoflagellates Proposed Age
140 (J. Barron , pers. comm.)
Acrinocyclus ellipricus Mesocene elliprica
I
150
Denriculopsis husredrii
160 Delphinels engusrara
Delphinels pennelliplica 13.3Ma
170 Rhaphonels elegans
R. magnapuncrara
180 R. parilis
190 / Acrinocyclus ingens

Delphinels angusrala
D. pannelliplice
\
R. magnapunctala 15.1 10 13.9Ma
R. parilis
210 Rossiella mediopuncrara
Thalassiosira eccenrrica
220
1~~~}230 Actinocyclus /ngens

~ 240
\
0.708691 16.6 17.0 Annellus calitornicus?
Craspedodiscus coscinodiscus
250 Cymarogonia arrillyocerus 17.01015.2 Ma
Delph/nels ct. angusrara
260 D. ovata
Rhaphoneis adamanlea
270 R. margarirata?
0.708535 19.2 19.3
0.708413 21.2 21.1
0.708129 25.9 26.7
Fig. 4. Core log of core W13751 showing the agreement between ages derived from the diatom
assemblages (1. Barron, personal communication, 1994), and from the 87Sr/86Sr composition of
phosphorite samples.
deposits can also result in age inversions where the Sr-derived age of reworked
intraclasts are older than the Sr-derived age of an underlying bed. Therefore, it is
imperative to evaluate the 87Sr/86Sr composition of beds or clasts in the context
of their reworking history. Sr-derived ages of phosphorite or dolostone gravel
are maximum ages for the surrounding sediment because gravel is commonly
reworked into younger sediment.
SOUTH NORTH
A SEISMiC LINE #, 4
0
area shown
Vl in Fig. 46
'"t;;w 10
~
UJ
u
<i
V) 20
...J
<5
i=
'"
UJ
> 30
UJ
I-
::::;:
x OCALA
~ 40
n.
0 2 3
50 KILOMETERS
Fig. SA. Interpreted seismic stratigraphic profile from offshore northeast Florida (see Fig. 2 for
location) showing sequence boundaries.
4.3. IDENTIFICATION OF SEQUENCES AND THEIR COMPONENTS
Seismic reflection data were acquired along the St. John's River and from the
inner shelf, approximately 40 km east of the cores studied (Fig. 2). These data
are discussed in greater detail by Snyder and others (1989) and Mallinson and
others (1994). It is difficult to make a direct correlation between the lithologic and
seismic data due to the spatial separation. Nevertheless, the seismic data verify
the presence of multiple unconformities defining the boundaries of five to six
depositional sequences within the Miocene Hawthorn Group (Fig. 5). The seismic
profiles illustrate the seaward expression and continuity of the disconformities that
are recognized in the onshore cores.
The basal reflector (al) exhibits a high-amplitude signal throughout the area
and represents a type 1 unconformity on top of the Eocene Ocala Limestone
(Fig. 5). Overlying seismic units are characterized by moderately continuous,
parallel to subparallel, low-amplitude reflectors. Seismic units are separated by
high-amplitude continuous reflectors. Seismic discontinuities (}:2, (3z, 1'1 and 81
are interpreted as type 1 unconformities and ravinement surfaces because of the
presence of fluvial channel scars along these surfaces (Hine and Snyder, 1985).
Seismic discontinuities a3 and (31 are interpreted as type 2 sequence boundaries
(Posamentier and Vail, 1988) or perhaps marine flooding surfaces because of a
lack of any erosional features (Snyder and others, 1989). The upper Miocene seis-
mic unit between reflectors 1'1 and 81 is thin and characterized by discontinuous,
subparallel low- to moderate amplitude reflectors and multiple channel scars indi-
o 20~N~o~rt~h ______________________________~S~ou~t~o
5
10
15
25
60 60
50 80 80
o
5
10
15
25
Fig. 58. Raw seismic data (top) with interpretations (bottom) of a portion of the profile line shown
in Fig. 5a.
eating several episodes of deposition and erosion. Approximately 10 to 12 m of

undifferentiated Plio-Pleistocene sands overlie the Hawthorn Group in this area.
Eight sequence boundaries defining seven depositional sequences are recog-
nized from lithologic and Sr-isotope data in the Hawthorn Group of northeast
Florida (Table 4; Figs. 3, 6 and 7). These depositional sequences in ascending
stratigraphic order are referred to as Hawthorn Group Sequences (HGS) 1 through
7. The sequence boundaries and associated sequence components are discussed in
detail below.
4.3.1. Type 1 Sequence Boundaries

Type 1 Sequence boundaries in our study area are readily apparent by the presence
of a physically eroded and concentrated gravel lag (Fig. 8). The gravel intra-
clasts are composed of the dominant rock present in the original eroded sequence.
SOUTH NORTH
~---------------- MSL----------------------------4 0
10
~
20
m
~
m
:D
(J) 30
40
SO
post-Miocene
solution
collapse
ST. JOHN'S
o 10 20 30 40 PLATFORM
I
KILOMETERS
Fig. 6. Correlations of sequence boundaries in cores from the Sanford High in the south to the
Jacksonville Basin in the north (refer to Fig. 2 for core locations and Fig. 3 for legend). Sr-derived
ages less than 24 Ma are from the Hodell and others (1991) regression. Sr-derived ages >24 Ma are
from the Os lick and others (1993) regression.
In northeast Florida the intraclasts are invariably phosphorite and/or dolostone.

The gravel may be deposited upon a laminated caliche-like surface (Esteban and
Klappa, 1983) probably indicating subaerial exposure. In other places, the gravel
occurs superjacent to a well-indurated dolostone bed that may have impeded ero-
sion. Gravel horizons are generally overlain by medium to coarse quartz sands indi-
cating a high-energy, shallow-water depositional environment. In situ phosphorite
or dolomite beds separated by the suspected sequence boundary have distinctly
different 87Sr/86Sr compositions reflecting formation during widely separate time
intervals. However, gravel intraclasts may have the same 87 Sr/86 Sr composition as
underlying beds reflecting upward reworking of material (Fig. 5). Gravel lags as-
sociated with type I unconformities are best developed in the basal and uppermost
Miocene section (sequence boundaries 1, 2, 6, 7 and 8; Table 4, Figs. 3, 6 and 7),
consistent with the presence of type 1 unconformities (reflectors a2, (3z, II and 81 ;
Fig. 5) as identified in the seismic data.
WEST EAST
o 10 20 30 40
KILOMETERS
10
s:: 20
m
-l
m
JJ
(J) 30
OCALA
HIGH
40
50
@
ST. JOHN'S
PLATFORM
JACKSONVILLE BASIN
Fig. 7. Correlations of sequence boundaries in cores from the edge of the Ocala High in the west,
crossing a shallow reentrant of the Jacksonville Basin (center) and then onto the St. John's Platform
in the east (refer to Fig. 2 for core locations and Fig. 3 for legend). Sr-derived ages <24 Ma are from
the Hodell and others (1991) regression. Sr-derived ages >24 Ma are from the Oslick and others
(1993) regression.
TABLE 4
Sequence boundaries with diagnostic criteria used for identifcation and the dominant
lithofacies above and below sequence boundaries
Sequence Below Above Diagnostic

boundary boundary boundary criteria
SB8 LF2,5 LF2,3,5 gravel lag

SB7 LF5 LF2,5 gravel lag truncated
Thalassinoides burrows
SB6 LF2,5 LF2 coarse quartz sand, gravel and shell hash
SB5 LF2 LF4 coarse quartz sand, gravel and shell hash;
sharp contact between lithofacies
SB4 LF2,5 LFl,2,5 coarse quartz sand and gravel;
sharp contact between lithofacies
SB3 LF5 LFl,2,4 truncated sand-filled
Thalassinoides burrows, gravel
SB2 LF4 LFI,4 gravel lag, possible duricrust
SBI LF3 LFl,4 gra vel lag, sharp contact
4.3.2. Type 2 Sequence Boundaries

Type 2 boundaries are more subtle than type 1 boundaries and are indicated by
an abrupt upward lithofacies transition. A concentrated gravel lag may not be
present but there is generally an abrupt increase in grain size of the sediment
upward across the boundary corresponding to a basinward shift in coastal onlap
and reflecting higher-energy conditions. Additionally, Thalassinoides burrows in
middle shoreface sandy muds to muddy sands of a highstand systems tract may be
truncated and filled with medium to coarse quartz and phosphorite sand or gravel.
Furthermore, like type 1 unconformities, in situ phosphorite or dolomite beds
which are separated by a sequence boundary have distinctly different 87Sr/86Sr
compositions. Phosphorite or dolomite sand and gravel infilling burrows exhibit a
lower (older) or similar 87Sr/86Sr composition as compared to beds just below the
horizon due to erosion of the beds upslope and seaward transport of clasts. Sequence
boundaries 3, 4 and 5 (Table 4, Figs. 3, 6 and 7) may be type 2 unconformities,
consistent with interpretations of the seismic reflectors a3 and f31 (Fig. 5).
4.3.3. Lowstand Systems Tracts (LST)

Where some degree of relief exists on the depositional surface as a result of fluvial
incisement or karstic processes, the sequence boundary may be overlain by a LST
comprised of a thick (5-10 m), muddy quartz sand unit occasionally interspersed
Fig. 8. Sample WI4476-137'; Intraclastic phosphorite gravel interpreted as a transgressive lag

associated with sequence boundary 8. The phosphorite has a Sr-derived age of 8-6 Ma. Bar scale =
2cm.
with bioclastic debris. In areas with minor topographic relief, the sequence bound-
ary and transgressive surface are coincident with thin to no lowstand deposits
separating them.
The sands of the LST may indicate lowstand to transgressive deposition in
an incised fluvial valley, in which case the gravel lag is deposited at the base
of the thalweg. These sand bodies can be predicted to occur along preexisting
paleotopographic lows on the depositional surface. The sand body in HGS-2, cores
W14476 (Fig. 3) and W14193, is interpreted as valley fill because it follows the axis
of the St. John's River solution valley. Alternatively, a sand unit may be peritidal
in origin, where repeated strandline migration and ravinement during mUltiple
high-frequency (fourth and fifth-order) sea-level fluctuations completely eroded
pre-existing phosphatic or dolomitic sequences resulting in multiple superimposed
sequence boundaries and ravinement surfaces. Sand units may also be barrier-inlet
sand bodies associated with a laterally migrating tidal inlet (Reinson, 1992). These
last two mechanisms are probably responsible for the highly reworked sandy
sediments of the upper sequences, especially around the paleotopographic highs.
Regardless of the origin of the sand body, phosphorite associated with the LST
and upper-bounding transgressive lag has a 87Sr/86Sr composition lower (older)
than overlying phosphorite or dolomite beds. The age distribution simply indicates
reworking of phosphorite grains from an older sequence. The quartz sandy unit
is commonly bounded by an overlying transgressive lag on a ravinement surface
consisting of coarse phosphorite and quartz sand. This ravinement surface separates
the LST from the TST.
4.3.4. Transgressive Systems Tracts (TST)

The TSTs in the study area are predominantly thin (0.l-2m) dolosilt or dolostone
beds. The dolostone is moldic and contains significant amounts of quartz sand
as well as dolostone and phosphorite clasts reworked from underlying sequences
where high energy conditions prevailed (generally eastward on the platform). In
other areas (westward and southward) where lower energy conditions prevailed,
the dolostone is nearly homogeneous reflecting cementation of silt-sized dolomite
(dolosilt) in the shallow subsurface. These thin, homogeneous beds are heavily
bored indicating intermittent exposure and sediment bypass (Fig. 9a and b) (Brom-
ley, 1975). Although they are probably not completely continuous and correlative
across the region, dolostone beds having similar 87 Sr/86 Sr compositions indicate
formation at nearly the same time. The presence of these dolomite horizons is
consistent with a sudden increase of precursor shallow-water biogenic carbonates
consisting predominantly of aragonite and high-Mg calcite during initial flooding
of the margin. In cores from the Sanford High and Ocala High areas (Fig. 2), the
TSTs consist of fine silt- to very fine sand-sized limpid dolomite with neoformed
palygorskite (Fig. 10). These dolosilt beds are interpreted to have formed from
precursor aragonite and high-Mg calcite mud in a lagoonal environment isolated
from major siliciclastic influence from the north during initial flooding of the area.
4.3.5. Maximum Flooding Surfaces/Condensed Sections

The maximum flooding surface separates the thin TST from the highstand systems
tract (HST) and is indicated by the presence of a phosphorite crust precipitated
on a dolostone bed (Figs. 11 a and b), a foraminifera- and diatom-bearing, muddy,
very fine to fine quartz and peloidal phosphorite sand (Fig. lIe), or bioeroded
phosphorite and dolomite gravel immediately overlain by deepwater muds. The
phosphorite structural carbonate has a negative oI3C composition consistent with
a genesis in organic-rich sediments (Compton and others, 1993; Mallinson and
others, 1994). Crusts or single intraclasts have similar 87Sr/86Sr compositions
throughout the area and provide an important chronostratigraphic marker for the
age of maximum flooding and formation of a major condensed section. In some
areas there is an abrupt upward lithofacies transition from a dolostone bed to a
deepwater diatomaceous mud with no intervening phosphatic bed.
Fig. 9. a) Sample from the Brooks Sink exposure; heavily bored dolostone hardground indicative
of sediment bypass during platform flooding. Bar scale =2 cm. b) Sample W13751-301 '; Heavily
bored dolostone hardground and intraclasts indicating sediment bypass during a rapid transgression.
Inverted borings (borings on the underside) indicate undermining of the hardground horizon by
burrowers and borers, ultimately resulting in the collapse and brecciation of the surface. Bar scale =
3cm.
Fig. 10. Sample Wl4413-21S'; Dolomite rhombohedra with neoformed palygorskite. Bar scale is
5 microns.
These phosphatic and diatomaceous sediments indicate high productivity con-

ditions in the overlying water mass, high organic carbon flux to the sediments, and
low siliciclastic sediment influx (Riggs, 1984; Riggs and Mallette, 1990; Mallinson
and others, 1994). Carbonate sedimentation and preservation may also have been
temporarily curtailed by an upwelling-induced nutrification, temperature decrease,
and increased turbidity of the overlying water mass.
4.3.6. Highstand Systems Tracts (HST)

The highstand systems tracts in our study area vary considerably in thickness due
to intense lowstand erosion. HSTs commonly consist of a mixture of clay minerals,
silt- to sand-sized quartz and feldspar, and dolosilt. Siliciclastic mud consisting of
smectite, palygorskite, sepiolite and quartz and feldspar silt tend to dominate the
lower portion of the HST while dolosilt dominates the upper portion. Planktonic
diatoms and siliceous sponge spicules are common and indicate a mesotrophic to
eutrophic environment (Fig. 12a). The diatoms appear to be an important silica
source for the neoformation of palygorskite and sepiolite (Fig. 12b). Planktonic
foraminifers indicating pelagic-water influence are present in some samples. The
HST generally coarsens upward from a dolosilty mud to a dolosilty sand, or sandy
46 D.1. Mallinson, I.S. Compton
Fig. l1A. Sample W13751-240'; Phosphorite crust containing bioclastic debris occurring on a
sandy dolostone and interpreted as the maximum flooding surface for Sequence 4. The phosphorite
has a Sr-derived age of 16.6 Ma. Bar scale = 2 cm.
Fig. 11 B. Sample W 14619-423'; Phosphorite crust and bioclastic debris interpreted as correlative
with Fig. l1A. The phosphorite has a Sr-derived age of 16.0 Ma. Bar scale =1 cm.
Fig. IIC. Sample Wl4476-302'; Light photomicrograph of concentrated phosphorite peloids

(dark grains) which have a Sr-derived age of 18.7 Ma. White grains are primarily recrystallized
foraminifera. Magnification is 30x.
dolosiltldolostone (carbonate caprock). The upper portion of the HST is commonly

burrowed with what appear to be Thalassinoides burrows. Although the majority
of these muds appear to have been deposited in a lower shoreface/outer shelf-type
setting as indicated by the diatoms and planktonic foraminifers, some of these
muds, particularly near the top of an HST, may be lagoonal or estuarine in origin.
4.4. EPISODIC PHOSPHOGENESIS AND SEA-LEVEL HIGHSTANDS
Based on the analyses of single grains or crusts, and using the regressions from
Hodell (1991), the data reveal phosphorite age groups corresponding to major
phosphogenic events associated with highstands in northeast Florida at approxi-
mately 24.5, 20.5, 18.5, 16.5 and 13.2 Ma. Two more age groups are tentatively
defined at 9.9, and 8-6 Ma (Table 3). By using the regressions from Oslick (1994)
age groups are clearly defined with averages at 25.1,20.5, 18.7, 16.9 and 15.5 Ma,
and poorly defined between 13 and 8.8 Ma (Table 3). It is apparent that the lower
age resolution during the late Miocene limits the ability to define age groups be-
tween approximately 15 and 6 Ma. Nevertheless, the lithologic data exhibit at least
three depositional sequences in this time range.
Fig. 12. A) Sample W13751-236'; Scanning electron photomicrograph of Actinoptychus senarius

(Ehrenburg), a meroplanktonic diatom characteristic of coastal upwelling areas (Andrews, 1988;
Powers, 1987). Bar scale is 50 microns. B) Sample W13751-206';Scanning electron micrograph
revealing neoformed palygorskite replacing a diatom fragment. Bar scale is 1 micron.
s. Discussion
5.1. SEQUENCE INTERPRETATIONS
The recognition of multiple sequence boundaries within the Miocene Hawthorn

Group and the analyses of subjacent and superjacent lithofacies within cores has
enabled us to determine the depositional and erosional processes that led to the
development of various types of sequences. The sequences vary in their lithofacies
patterns largely because they are near the updip feather edge of deposition where
lateral facies changes may be extremely variable, aggradational processes are
minimized and erosional processes are maximized. Fig. 13 summarizes the char-
acteristics of interpreted depositional sequences and the mechanisms responsible
for their development.
Given maximum preservation, an idealized depositional sequence on a nondis-
sec ted portion of the platform appears to consist of 1) a basal unconformity and
coincident transgressive gravel or bioclastic lag and quartz sand palimpsest, 2) a
dolomitic hardground with obvious sediment bypass features (borings, encrusta-
tions, phosphorite crust), 3) a condensed phosphatic interval (either a crust, peloidal
sand, or bioeroded gravel), 4) a deepwater diatomaceous mud grading upward to
5) a shallower water Thalassinoides-burrowed sandy dolosilt or dolostone. Given
the variable erosive processes provided by fluvial incisement during lowstands,
boundary current erosion during highstands, and ravinement throughout a sea-level
cycle, all, part or none of an idealized sequence may be preserved (Fig. 13). Further
complicating the picture is the fact that an idealized sequence may not be deposited
initially because of lateral variability in depositional facies patterns corresponding
to variable topography and depositional energy levels.
The lithofacies successions indicate terrigenous siliciclastic input from the
north and northwest during lowstands (Fig. 14a). This siliciclastic influx increases
upward through the Hawthorn Group and was most important during the late
Miocene when the depositional regime was siliciclastic dominated. Prior to the
late Miocene, northeast Florida was a true mixed carbonate-siliciclastic system.
Siliciclastics were transported from source regions in the north and northwest.
The quartz sands were incorporated into lowstand and early transgressive deposits
while they were being actively transported within the littoral and upper shoreface
zone. Once transgression was well underway the platform was flooded, the fluvial
and littoral supply line from the north was interrupted, and the system quickly
evolved into a carbonate mud and shell producing environment (Fig. 14b) perhaps
similar to the modern Florida Bay environment (Enos and Perkins, 1977; Davis
and others, 1992).
At some point between the maximum rate of sea-level rise and the highstand,
boundary current interactions with local bathymetry produced the upwelling pat-
terns which greatly increased primary production and the flux of organic matter to
the sediments resulting in phosphogenesis (Fig. 14c) (Riggs, 1984; Compton and
others, 1990; Mallinson and others, 1994). Aragonitic and high-Mg calcitic muds
directly underlying high organic matter concentrations may have additionally en-
hanced phosphorite and dolomite genesis in the shallow subsurface by providing
a nucleus for francolite crystallization (Berner, 1990) and an unstable carbonate
VI
o
deep shelf muds 1!!

carbonate mud
~~~~~:b ~
upwelling and sediment bypass
shallow shelf carbonates ~
shoreface ravinement ~
deep shell muds
estuary backfill ~
upwelling and sediment bypass
shallow shelf carbonates .R
palimpsest subaerial exposure caliche .~'
DEPOSITIONAL SEQUENCE
NONDISSECTED PLATFORM DEPOSITIONAL SEQUENCE
DISSECTED PLATFORM
(
lowstand stillstand
fluvial
incisement equence boo ndary
- .~ ... .
.' ..
"""",,~ ~
-,: .. ..
~,"': .""""~,, .. ..""""
( " "
.......
. . . sequence boundary
. .... PLATFORM HIGH
~BASINWARD
tl
~
DEPOSITIONAL SEQUENCE ~
SEMIRESTRICTEDILAGOONAL ::::
ERODED SEOUENCE sb = sequence boundary
/ower shoreface 1st lowstand ~tems tract
S
en
muds hst Is = transgressivelravinernant surface
lst = transgressive ~tems tract ~
~1s sediment bypass mfslcs mfs = maximum flooding surface
lagoonal muds 1St cs condensed section ~
hst = highstand systems tract ~
Fig. 13. Diagram illustrating a generalized crosssection of the northeast Rorida Platfonn and the varieties and interpretations
of depositional sequences recognized in this investigation (refer to Fig. 3 for legend).
j
g
ST. JOHN'S R1YeA
I
SOlUTfON VAllY
I
HIGH
0CAl.A HIGH _
LOW
100
EUSTASY
"'" V\J

LOW
EUSTASY
shoreljne SANFORD
HIGH
OCN.AHIGH

'00
EUSTASY
I
HIGH
LOll
EUSTASY
Fig. 14. Diagram illustrating the relationship or the dominant lithofacies to sea-level position. a)
During low stands, siliclastics are transported onto the Florida Platform via fluvial sources from the
northwest and littoral transport from the north-northeast (white arrows). b) During a sea-level rise
the majority of the Florida Platform is quickly inundated, siliciclastic sources are displaced to the
north and carbonate sedimentation dominates. c) During mid-transgression to highstand, boundary
current interactions with paleotopographic highs result in sustained upwelling causing high organic
matter flux to the sediment and resulting in phosphogenesis. d) During late highstand to regression,
seaward movement of the boundary current, progradation of siliciclastic sources in the north and
shedding of carbonates from the south and west shut down major phosphogenesis.
52 D.l. Mallinson, 1.S. Compton
Sea-level Fluctuations on the Eustatic Sea-level Curve

Northeast Florida Platform (Haq and others, 1987)
STAGE Age relative amplitudes meters above mean sea level
(Ma) +Iandward basinward+ +landward basinward+200 150 100 50 0
MESSINIAN 5 Ages 'rom - f ',om
SliCK and oth.tI (1994) NJandNC - 5
Margins -
7'--+--+-
SERRA
E
VALllAN
1-----l15
LANGHlAN 15'
1-
BURDI 16.9
GALIAN 1 T
18.7-
,
1-----120
AQUI
TANIAN
1--_---1"5 ~25.1
_ .&Vet.g. age of l~uaJ phosphorite 9fMfU.
CHATTIAN 1 range 01 ages 01' IndMIiJai pho$phonte gt'i.'UftJ
1 \dh error bars 0 95% conhdence level
1-----1:30
RUPELIAN ~ biost/Ol'll'oohi).go toogo
Fig. 15. Comparison of the local sea-level curves constructed for the northeast Rorida Platform
from the average Sr-derived ages of single-grain phosphorite samples using the Hadell and others
(1991) regressions and the Oslick and others (1994) regressions, to the eustatic curve proposed by
Haq and others (I 987). Amplitudes are relative and are estimated from the lithologies, thickness and
updip extent of sequences.
precursor for dolomite formation. Persistent upwelling may have curtailed carbon-
ate production and preservation during highstand to early regression and enhanced
the concentration of diatomaceous siliciclastic muds. As the Gulf Stream moved
offshore during regression, upwelling in the region decreased and coarser grained
siliciclastic, biogenic, and intraclastic grains were concentrated along with re-
worked dolosilt in shoaling environments. The increased flux of siliciclastics from
the north and cessation of upwelling associated with lower sea level shut down
phosphogenesis (Fig. 14d).
5.2. COMPARISON TO OTHER SEA-LEVEL STUDIES
A local sea-level curve has been constructed for the NE Florida Platform (Fig. 15)
by integrating the depositional sequence interpretations and the average values
of Sr-derived ages for single phosphorite grains or crusts which occur within the
sequences (Table 3) (Mallinson and others, 1994). As a result of the variable
preservation of sequences, analyses of the changes in onlap are not sufficient to
decipher sea-level cycle amplitudes. Furthermore, such an analysis from only core
data is extremely difficult and suspect. A qualitative, relative estimate of ampli-

tudes was derived by consideration of lithofacies patterns in conjunction with the
regional extent and thickness of sequences. For example, the thick and region-
ally extensive diatomaceous muds of HGS-4 are strongly indicative of sustained,
moderately-deep, open-shelf sedimentation. HGS-4 is therefore interpreted as be-
ing deposited at a time of maximum sea level during the late Burdigalian-early
Langhian ('" 17-15 Ma; N7-N9). The extensive reworking and homogenization
of Upper Miocene sequences in platform areas, and predominance of siliciclastics
and gravel intraclasts suggest that HGS-l, 5, 6, and 7 were deposited during lower
magnitude high stands and extensively reworked during extreme low stands. The
occurrence of type 1 unconformities identified from the seismic data in the low-
ermost and uppermost Miocene section and type 2 unconformities in the middle
Miocene section corroborate the sea-level interpretation. Other regional and global
investigations support a sea-level maximum during the middle Miocene and lower
overall sea-levels in the early and late Miocene (Haq and others, 1987; Greenlee
and Moore, 1988; Dillon and Popenoe, 1988; Popenoe, 1990; Snyder and others,
1990).
The data show some consistency with the sea-level curve and condensed section
ages presented by Haq and others (1987) (Fig. 15). However, the significance of this
correlation is questionable. Given the number and frequency of sea-level events
and condensed sections presented by Haq and others (1987), and the limited age
resolution of the Sr-isotope age dating technique, particularly in the middle to late
Miocene (younger than 16 Ma), it is difficult to say with certainty that a correlation
exists.
The Sr-derived ages of around 25 Ma for phosphorite and dolomite of the
lowermost sequence in our study, HGS-l, agree well with the age of the downlap
surface of Haq and others (1987) at 24.8 Ma. The data indicate, therefore, that
HGS-l is correlative to the third-order sea-level event TB 1.4 (Haq and others,
1987).
The most obvious discrepancy between our data and the Haq and others (1987)
eustatic curve occurs during the Aquitanian (Fig. 15). Our data indicate a type 1
unconformity between approximately 24 and 23 Ma and deposition ofHGS-2 dur-
ing the time interval 22.8 (maximum) to 20.0 Ma with phosphogenesis occurring
at 20.5 Ma. The data of Sugarman and others (1993) also indicate sequence depo-
sition on the New Jersey margin (Kirkwood sequence 1) from 22.6(1) to 19.2 Ma
(Fig. 15). These data contrast with the Haq and others (1987) sea-level curve which
shows a slowly rising sea level following a major lowstand and type 1 unconfor-
mity at 21 Ma. Although more data are needed, these preliminary results suggest
that perhaps the 21 Ma sequence boundary of Haq and others (1987) should be
adjusted downward to approximately 23 Ma.
Our data and the data of Snyder and others (1990) from the North Carolina
margin indicate an unconformity at approximately 20 to 19 Ma separating HGS-2
and 3. A corresponding unconformity is not indicated on the sea-level curve of Haq
and others (1987). The data of Sugarman and others (1993) are equivocal in regards
to an unconformity at 20 to 19 Ma. Their data show a hiatus from approximately
19.2 to 17.5 Ma which they correlate to the type 2 sequence boundary at the base
54 D.l. Ma11inson, I.S. Compton
of TB2.2 at approximately 18 Ma (Haq and others, 1987). It is possible that the

unconformity noted in our study area is the result of boundary current erosion as
opposed to a sea-level fall. However, it is unclear whether this explanation would
hold for the data of Snyder and others (1990). We tentatively correlate HGS-2 with
the lower portion of TB2.1.
Phosphorite and dolomite of HGS-3 correlate well with the 18.5 Ma (TB2.1)
condensed section of Haq and others (1987), and the data of Snyder and others
(1990) also indicate sequence deposition beginning at 18.5 Ma. Corresponding
deposits in the study area of Sugarman and others (1993) are absent, possibly as a
result of erosion associated with the hiatus at 19.2 to 17.5 Ma.
HGS-4 phosphorite correlates well with the 17 and 16 Ma condensed sections
of TB2.2 and TB2.3 (Haq and others, 1987). However, unlike Haq and others
(1987), our data do not indicate unconformities at 16.5 and 15.5 Ma. Snyder and
others (1990), and Sugarman and others (1993) also show a major hiatus between
17 and 16 Ma. Within the range of biostratigraphic and Sr-isotope age resolution,
there is temporal agreement between the 16.5 Ma phosphorite horizon and the
immediately overlying diatomaceous muds comprising the highstand systems tract
of HGS-4 (Fig. 3). Furthermore, phosphorite from HGS-4 has highly substituted
francolite with high concentrations of Mg2+, Na+ and S~+, as well as high CO~
and F- contents indicating negligible meteoric-water diagenesis. It is possible
that all record of TB2.3 and TB2.4 was removed by erosion associated with the
TB2.4rrB2.5 unconformity.
Correlation of phosphorite and dolomite samples younger than 16 Ma to previ-
ous studies is difficult because of the low gradient of the 87 Sr/86 Sr curve during the
middle to late Miocene (Hodell and others, 1991; Oslick and others, 1994). How-
ever, the biostratigraphic data available in this study corroborate Sr-isotope data
and are most consistent with the regression of Hodell and others (1991) during the
middle to late Miocene. These data suggest that the phosphorite and dolomite of
HGS-5 is roughly correlative to TB2.5 of Haq and others (1987). In summary, the
limited age resolution possible using Sr-isotopes from 16 to 6 Ma, and reworking
of the phosphorite sediments make it difficult to determine the number and age of
condensed sections and to construct accurate sea-level curves for the late Miocene.
Correlation to the proposed eustatic curves may be improved when the 87Sr/86Sr
ratio corresponding to the condensed intervals defined by Haq and others (1987)
are measured.
6. Conclusions
Depositional sequences and their components can be identified in the thin updip
platform deposits of the Miocene Hawthorn Group in northeast Florida. In ascend-
ing stratigraphic order, an idealized sequence consists of: 1) a basal lowstand wedge
or transgressive shallow-shelf siliciclastic palimpsest unit with reworked phospho-
rite, quartz sand, gravel and bioclastic material, 2) a transgressive shallow-lagoon to
intermediate-shelf dolomitized carbonate unit, 3) a mid-transgressive to highstand
omission surface commonly characterized by a bored dolomitic hardground with
associated phosphorite, 4) a lower shoreface diatomaceous mud grading upward
into 5) a burrowed quartz sandy dolosilt or dolostone. These lithofacies succes-

sions reveal the influence of sea-level position on the dominant sediment-type
in this mixed carbonate-siliciclastic system. Siliciclastic deposition dominated
during relative lowstands when sediments were transported into the area from the
continental provinces to the north via longshore currents and fluvial processes. Car-
bonate deposition dominated during platform flooding when siliciclastic sources
were no longer connected to the platform by a contiguous littoral zone, fluvial point
sources were displaced northward, and upwelling waters were not influencing the
area. Phosphorite deposition occurred when sedimentation rates were low during
rapid flooding and boundary current interactions with local topography produced
upwelling and high productivity conditions. Siliceous biogenic sedimentation was
important as nutrient levels associated with upwelling were maintained and sedi-
mentation rates began to increase as sediments were reworked from local highs or
fluvial sources prograded from the north during high to early falling sea-level con-
ditions. The platform deposits, though thin and eroded, readily recorded sea-level
conditions as accommodation space was quickly created or eliminated altogether.
Sr-isotope age determinations of phosphorite and dolomite horizons have pro-
vided new and valuable data for the evaluation of the chronostratigraphic range
of the Hawthorn Group and the episodic nature of phosphogenesis and dolomiti-
zation. When considered in the context of sequence stratigraphy and lithofacies
patterns, the Sr-isotope data provide the chronostratigraphy to aid in the regional
correlation of depositional sequences in the Miocene Hawthorn Group of northeast
Florida. Furthermore, the ages of early diagenetic, unaltered phosphorites allow
us to determine the timing of sea-level highstands and construct a local sea-level
curve. At least seven high stands occurred in the study area between 25 and 6 Ma
with a maximum highstand at 17 to 15 Ma. There is mixed agreement between
our data and the data of Haq and others (1987), Snyder and others (1990), and
Sugarman and others (1993).
Acknowledgments
We would like to thank editor Bilal Haq and reviewers Kenneth Miller and Christo-
pher Kendall for their suggestions. Al Hine also offered significant constructive
discussions and we thank John Barron for his help with the diatom biostratig-
raphy. Core material was obtained from the Florida Geological Survey with the
help of Tom Scott. This investigation was supported by National Science Founda-
tion Grant #EAR-900-S0S4, Donors to the Petroleum Research Fund administered
by the American Chemical Society, the Gulf Oceanographic Fellowship and the
William and Elsie Knight Fellowship.
References
ANDREWS, G.w., 1988, A Revised Marine Diatom Zonation for Miocene Strata of the
Southeastern United States: U.S. Geological Survey Professional Paper 1481.
BAUM, G.R., and VAIL, P.R., 1988, Sequence stratigraphic concepts applied to Paleogene
outcrops, Gulf and Atlantic Basins: in Wilgus, C.K. and others, eds., Society of Economic
Paleontologists and Mineralogists Special Publication No. 42, Sea-Level Changes: An

Integrated Approach, p.309-327.
BERNER, R.A., 1990, Diagenesis of phosphorus in sediments from non-upwelling areas,
in Burnett, W.C., and Riggs, S.R., eds., World phosphate deposits, Volume 3, Neogene
phosphorites of the southeastern United States: Cambridge University Press, p. 27-32.
BROMLEY, R.G., 1975, Trace fossils at omission surfaces, in Frey, R.W., ed., The Study
of Trace Fossils, Springer-Verlag, New York, p.399-428.
COMPTON, J.S., SNYDER, S.w., and HODELL, D.A., 1990, Phosphogenesis and weath-
ering of shelf sediments from the southeastern United States: Implications for Miocene
613C excursions and global cooling: Geology, v. 18, p. 1227-1230.
COMPTON, J.S., HODELL, D.A., GARRIDO, J., AND MALLINSON, D.J., 1993, Origin
and age of phosphorite from the south-central Florida Platform: Relation of phospho-
genesis to sea-level fluctuations and bI3C excursions: Geochimica et Cosmochimica
Acta, v. 57, p. 131-146.
COMPTON, J.S., HALL, D.L., MALLINSON, D.J., AND HODELL, D.A., 1994, Origin of
dolomite in the phosphatic Miocene Hawthorn Group of Florida: Journal of Sedimentary
Petrology, v. B64.
DAVIS, R.A. JR., HINE, A.C. AND SHINN, E.A., 1992, Holocene coastal development
on the Florida Peninsula: in Fletcher, C.H. III, and Wehmiller, J.F. eds., Society of
Economic Paleontologists and Mineralogists Special Publication No. 48, Quaternary
coasts of the United States: marine and lacustrine systems, pI93-212.
DEPAOLO, DJ. AND INGRAM, B.L., 1985, High resolution stratigraphy with strontium
isotopes: Science, v. 227, p. 938-941.
DILLON, w.P. AND POPENOE, P., 1988, The Blake Plateau Basin and Carolina Trough:
in Sheridan, R.E. and Grow, J.A. eds., The Geology of North America vol. 1-2, The
Atlantic Continental Margin, p. 291-328.
ENOS, P. AND PERKINS, R.D., 1977, Quaternary sedimentation in south Florida: Geo-
logical Society of America Memoir 147, 198 p.
ESTEBAN, M. AND KLAPPA, C.F., 1983, Subaerial exposure environment: in Scholle,
P.A., Bebout, D.G. and Moore, C.H., eds., Carbonate depositional environments, Amer-
ican Association of Petroleum Geologists Memoir 33, pI-54.
GINSBURG, R.N., 1971, Landward movement of carbonate mud: new model for regressive
cycles in carbonates: American Association of Petroleum Geologists Bulletin, v. 55, p.
340, Abstract.
GREENLEE, S.M., AND MOORE, T.C., 1988, Recognition and interpretation of deposi-
tional sequences and calculation of sea-level changes from stratigraphic data - offshore
New Jersey and Alabama Tertiary: in Wilgus, C.K. and others, eds., Society of Economic
Paleontologists and Mineralogists Special Publication No. 42, Sea-Level Changes: An
Integrated Approach, p. 329-356.
HAQ, B.U., HARDENBOL, J., AND VAIL, P.R., 1987, Chronology of fluctuating sea
levels since the Triassic: Science, v.235, p. 1156-1166.
HESS, J., BENDER, M.L., AND SCHILLING, J.G., 1986, Seawater 87Sr/86Sr evolution
from Cretaceous to present - Applications to paleoceanography: Science, v. 231, p.
979-984.
HINE, A.C., AND SNYDER, S.w., 1985, Coastallithosome preservation: Evidence from
the shoreface and inner continental shelf off Bogue Banks, North Carolina: Marine
Geology, v. 63, p. 307-330.
HODELL, D.A., MUELLER, P.A., AND GARRIDO, J.R., 1991, Variations in the stron-
tium isotopic composition of seawater during the Neogene: Geology, v. 19, p. 24-27.
LOCKER, S.D. AND DOYLE, L. J., 1992, Neogene to Recent stratigraphy and deposi-
tional regimes of the northwest Florida inner continental shelf: Marine Geology, v. 104,
p. 123-138.
LOUTIT, T.S., HARDENBOL, J., VAIL, P.R., AND BAUM, G.R., 1988, Condensed
sections: The key to age dating and correlation of continental margin sequences: in
Wilgus, C.K. and others, eds., Sea-level Changes: An Integrated Approach, Society of
Economic Paleontologists and Mineralogists Special Publication No. 42, p.183-216.
MALLINSON, OJ., COMPTON, J.S., SNYDER, S.W., AND HODELL, D.A., 1994,
Strontium isotopes and sequence stratigraphy across the northeast Florida Platform:
Journal of Sedimentary Research, v. B64, No.8.
MCARTHUR, J.M., 1985, Francolite geochemistry - compositional controls during forma-
tion, diagenesis, metamorphism and weathering: Geochimica et Cosmochimica Acta, v.
49, p. 23-35.
MCARTHUR, J.M., SAHAMI, A.R., THIRWALL, M., HAMILTON, P.J., AND OSBORN,
A.O., 1990, Dating phosphogenesis with strontium isotopes: Geochimica et Cosmochim-
ica Acta, v. 54, p. 1343-1351.
MITCHUM, R.M., VAIL, P.R., AND THOMPSON, S., 1977, Seismic stratigraphy and
global changes of sea level, Part 2: The depositional sequence as a basic unit for
stratigraphic analysis: in Clayton, C.E., ed., Seismic Stratigraphy - Applications to
Hydrocarbon Exploration: American Association of Petroleum Geologists Memoir 26,
Tulsa, Oklahoma, p.53-62.
MULLINS, H.T., GARDULSKI, A.F., WISE, S.w. JR., AND APPLEGATE, J., 1987,
Middle Miocene oceanographic event in the Gulf of Mexico: Implications for seismic
stratigraphic succession and Loop Current / Gulf Stream circulation: Geological Society
of America Bulletin, v. 98, p 702-713.
OPDYKE, N.D., SPANGLER, D.P., SMITH, D.L., JONES, D.S., AND LINDQUIST,
R.C., 1984, Origin of the epeirogenic uplift of Pliocene-Pleistocene beach ridges in
Florida and development of the Florida karst: Geology, v. 12, p. 226-228.
OSLICK, J.S., MILLER, K.G., FEIGENSON, M.D., AND WRIGHT, J.D., in press,
Oligocene-Miocene strontium isotopes: stratigraphic revisions and correlation to an
inferred glacioeustatic record: Paleoceanography
PIETRAFESA, L., 1990, Upwelling processes associated with Western Boundary Currents:
in Burnett, W.c., and Riggs, S.R., eds., World phosphate deposits, Volume 3, Neogene
phosphorites of the southeastern United States: Cambridge University Press, p.3-26.
POPENOE, P., HENRY, VJ., AND IDRIS, F.M., 1987, Gulf Trough - the Atlantic connec-
tion: Geology, v.l5, p.327-332.
POPENOE, P., 1990, Paleoceanography and paleogeography of the Miocene of the south-
eastern United States: in Burnett, W.C., and Riggs, S.R., eds., World phosphate deposits,
Volume 3, Neogene phosphorites of the southeastern United States: Cambridge Univer-
sity Press, p.352-380.
POSAMENTIER, H.W., AND VAIL, P.R., 1988, Eustatic controls on clastic deposition 11-
Sequence and systems tract models: in Wilgus, C.K. and others, eds., Sea-level Changes:
An Integrated Approach, Society of Economic Paleontologists and Mineralogists Special
Publication No. 42, 125-154.
POWERS, E.R., 1987, Diatom biostratigraphy and paleoecology of the Miocene Pungo
River Formation, North Carolina Continental Margin [unpublished M.S. thesis]: East
Carolina University, 240 p.
REINSON, G.E., 1992, Transgressive barrier island and estuarine systems: in Walker,
R.G. and James, N.P. eds., Facies Models Response to Sea Level Change, Geological
Association of Canada, p. 179-194.
RIGGS, S.R., 1979, Petrology of the Tertiary phosphorite system of Florida: Economic
Geology, v. 74, p. 195-220.
RIGGS, S.R., 1984, Paleoceanographic model of Neogene phosphorite deposition, U.S.
Atlantic continental margin: Science, v.223, p.123-131.
RIGGS, S.R., AND MALLETTE, P.M., 1990, Patterns of phosphate deposition and lithofa-
cies relationships within the Miocene Pungo River Formation, North Carolina continen-
tal margin, in Burnett, W.C., and Riggs, S.R., eds., World phosphate deposits, Volume 3,
Neogene phosphorites of the southeastern United States: Cambridge University Press,
p.424-443.
RIGGS, S.R., AND SHELDON, R.P., 1990, Paleoceanographic and paleoclimatic controls
of the temporal and geographic distribution of Upper Cenozoic continental margin
phosphorites, in Burnett, W.C., and Riggs, S.R., eds., World phosphate deposits, Volume
3, Neogene phosphorites of the southeastern United States: Cambridge University Press,
p.207-222.
SCOTT, T.M., 1988, The Lithostratigraphy of the Hawthorn Group (Miocene) of Florida:
Florida Geological Survey Bulletin No. 59, 148 pp.
SCOTT, T. M., 1990, A brief overview of Miocene lithostratigraphy - northern Florida and
eastern Georgia: Open File Report 31, Florida Geological Survey, 6 p.
SNYDER, S.W, 1982, Miocene seismic stratigraphy within the Carolina Phospho-
genic Province: chronostratigraphy, paleotopographic controls, sea-level cyclicity, Gulf
Stream dynamics, and the resulting depositional framework [unpublished M.S. thesis]:
University of North Carolina - Chapel Hill, 182 pp.
SNYDER, S.W, EVANS, WE., HINE, A.C., AND COMPTON, I.S., 1989, Seismic ex-
pression of solution collapse features from the Florida Platform: in Beck, B.F., ed.,
Engineering and Environmental Impacts of Sinkholes and Karst: Proceedings of the
Third Multidisciplinary Conference on Sinkholes and the Engineering and Environ-
mental Impacts of Karst, St. Petersburg Beach, Florida, 2-4 October 1989, p.281-298.
SNYDER, S.W, HINE, A.c., AND RIGGS, S.R., 1990, Carolina continental margin: Part
2 - The seismic stratigraphic record of shifting Gulf Stream flow paths in response
to Miocene glacio-eustasy: implications for phosphogenesis along the North Carolina
continental margin, in Burnett, WC., and Riggs, S.R., eds., World phosphate deposits,
Volume 3, Neogene phosphorites of the southeastern United States: Cambridge Univer-
sity Press, p. 396-423.
STAINFORTH, R.M., LAMB, 1.L., LUTERBACHER, H., BEARD, I.H., AND JEF-
FORDS, R.M., 1975, Cenozoic planktonic foraminiferal zonation and characteristic
index forms: University of Kansas, Paleontological Contributions, v.62, Lawrence,
Kansas, 425 p.
SUGARMAN, PJ., MILLER, K.G., OWENS, J.P. AND FEIGENSON, M.D., 1993,
Strontium-isotope and sequence stratigraphy of the Miocene Kirkwood Formation,
southern New Jersey: Geological Society of America Bulletin, v. 105, p. 423-436.
VAIL, P.R., MITCHUM, R.M., JR., TODD, R.G., WIDMIER, J.M., THOMPSON, S., III,
SANGREE, 1.B., BUBB, J.N., AND HATLELID, WG., 1977, Seismic stratigraphy and
global changes in sea level: in Clayton, C.E., ed., Seismic Stratigraphy - Applications to
Hydrocarbon Exploration: American Association of Petroleum Geologists Memoir 26,
Tulsa, Oklahoma, p.49-212.
VANWAGONER, 1.C., MITCHUM, R.M., CAMPION, K.M., AND RAHMANIAN, V.D.,
1990, Siliciclastic sequence stratigraphy in well logs, cores, and outcrops: American
Association of Petroleum Geologists Methods in Exploration Series, No.7., 55 p.
WEAVER, C.E. AND BECK, K.C., 1977, Developments in Sedimentology 22: Miocene
of the S.E. United States: A Model for Chemical Sedimentation in a Peri-Marine Envi-
ronment: reprinted from Sedimentary Geology, VoU7, Nos. 112.
CHAPTER 3
Using Ecostratigraphic Trends in

Sequence Stratigraphy
Federico Ol6riz, Jesus E. Caracuel and Francisco J. Rodriguez-Tovar
ABSTRACT: Recent advances in basin analysis based on surface data advance sedi-
mentary and paleoecologic research by taking into account the stratigraphic framework.
Sequence stratigraphy and high-resolution event stratigraphy are well-known approaches
that aid these advances. Ecostratigraphic interpretations have proved to be valuable tools
in high-resolution event stratigraphy through the recognition of ecostratigraphic events.
We propose the application of ecostratigraphy as a complement in sequence stratigraphy
using ecostratigraphic trends. The conceptual basis consists of the assumption of close,
though complex, relationships between the accommodation and the ecospace. Systems
tracts, a key to sequence stratigraphy, are assumed to be related to shifting ecospaces
and ecostratigraphic trends related to eustasy. Ecostratigraphic trends can also provide
information about local ecospace deviations.
Appropriate ecostratigraphic sampling programs are of prime importance for ecostrati-
graphic interpretations based on the combined analysis of the stratigraphic features and
the recorded fossil assemblages of megainvertebrates. Some applications demonstrated
here are based on 7,000 megainvertebrates sampled bed-by-bed in sections belonging
to the Subbetic Zone, Prebetic Zone, Algarve Basin, Iberian Cordillera, and the is-
land of Mallorca on the Iberian Subplate. The five cases studied deal with condensed
(ammonitico rosso) and expanded (rhythmic marly-limestones) facies from the Middle
Oxfordian to the Lower Tithonian, and concern: a) the ecosedimentary evolution at the
stage and substage levels; b) ecostratigraphic interpretations at the ammonite biochrono-
zone level; c) comparison of faunal assemblages from distant epicontinental areas; d)
relationship between ecostratigraphic interpretations and trends in abiotic components;
and e) the influence of tectono-eustatic interactions on trends in the composition of fossil
assemblages. We conclude that shifting bio- and lithofacies can be more adequately
interpreted by combining ecostratigraphic and sequence stratigraphic approaches.
1. Introduction
Project Ecostratigraphy (IGCP) is a research program for increasing our under-

standing of combined ecologic and sedimentary evolution. The so-called ecostrati-
graphic events (Kauffman, 1986, 1988; Kauffman and others, 1991) have been
59
B. U. Haq (ed.),
60 F. 016riz et a1.
considered as significant changes in community structure in response to allocyclic

factors determining environmental fluctuations. Therefore ecostratigraphic events
are considered time markers in High-Resolution Event Stratigraphy (HIRES),
and ecostratigraphic trends should provide information about progressive shifts in
ecosedimentary conditions (Ol6riz and others, 1991, 1993). From autochthonous
benthos-rich deposits in sensitive (changing) lithofacies, to comparatively poor
planktic-pelagic biotic content in relatively monotonous lithofacies, and all the in-
termediate combinations of litho- and biofacies, should be influenced by changing
ecosedimentary conditions.
Transgressive-regressive cycles have been traditionally considered the forcing
factors for changes in bio- and lithofacies (Hallam, 1969, 1975, 1985; Donovan,
1985; Gygi, 1986, and others). Developments in sequence stratigraphy during
the last decade (Haq and others, 1987; Vail and others, 1987; Wilgus and others,
1988, among others) have provided new interpretations in which global fluctua-
tions in relative sea level were invoked to explain shifting lithofacies and even
biofacies (see Rey, 1992). Thus, ecosedimentary evolution in a given basin could
relate to eustasy, or tectono-eustasy during ice-free periods (e.g., the Jurassic),
thereby providing a new framework for correlations. Accordingly, we can inter-
pret shifting bio-lithofacies in the framework of both ecostratigraphic and sequence
stratigraphic approaches.
Sequence stratigraphic interpretations, based on lithofacies analysis, are pos-
sible using available outcrops, but serious limitations may occur, owing to poorly
informative facies (mudstones, fine-grained mixed clastic-carbonate facies) and
restricted and/or disturbed sections. Ol6riz and others (1991, 1993) considered
some of the usual limitations of sequence stratigraphic analyses in outcrops, and
emphasized ecostratigraphic analyses as a valuable complement in basin analysis
within the conceptual frame of sequence stratigraphy. Eco-evolutionary responses
of marine megainvertebrate assemblages to fluctuations in relative sea-level can
be identified on the basis of ecostratigraphic analyses. Thus, eco-evolutionary re-
sponses have been demonstrated in ammonites (Gygi, 1986; Atrops and Ferry,
1989; Hantzpergue, 1991; El Hariri and others, 1992; Hoedemaeker, 1992; Re-
boulet and others, 1992; Ol6riz and others, 1993, 1994; Rodriguez-Tovar, 1993).
Ecostratigraphic approaches should be applied in all the cases for which the fossil
record is favourable. Ol6riz and others (1991) recognized several orders of geo-
biological interactions through the analysis of the ratio of ammonites to megainver-
tebrates (other than ammonites) recorded in the Upper Jurassic of Southern Iberia.
Thus supported by adequate taphonomic studies, ecostratigraphic approaches need
not be restricted to benthic faunas.
2. The Ecostratigraphic Procedure in Upper Jurassic Outcrops
Upper Jurassic outcrops in Europe are frequently represented by more or less

expanded sections relative to their paleogeographic setting. In epicontinental de-
posits, mid- to outer-shelf sites are represented by rhythmic alternations of lime-
stones and marls with variable fossils content. The ratio of ammonites to megain-
vertebrates (other than ammonites) fluctuates in response to local ecosedimen-
Using Ecostratigraphic Trends in Sequence Stratigraphy 61
tary conditions, and a high-resolution biostratigraphy is usually available from

the ammonite records. Therefore, traditional research has focused on ammonite
biostratigraphy and evolution, and bio-chronostratigraphic correlation has progres-
sively improved. Where the record of benthic faunas is favourable, paleoecologi-
cal interpretations have been proposed (Fiirsich, 1977; Fiirsich and Sykes, 1977;
Fiirsich and Werner, 1986). By contrast, in epioceanic sites (i.e. distal pelagic swell
areas with ammonitico rosso facies) condensed deposits developed with a variable
ammonite record, but horizons with autochthonous benthic faunas are generally
scarce. Consequently, paleontological research in ammonitico rosso facies has fo-
cused on improving biostratigraphy and an understanding of ammonite evolution.
On the whole, research in integrated biostratigraphy is one way to improve strati-
graphic interpretations based on paleontologic information (Rasplus and Fourcade,
1987; Galbrum and others, 1991). As mentioned previously, the investigation of
combined records of ammonites and megainvertebrates (other than ammonites) is
comparatively poorly developed, due perhaps to two main factors: a) ammonite
researchers are normally specialists; and b) benthic faunas are usually poor for in-
depth studies from ammonite-rich beds. Moreover, sampling procedures and goals
differ sharply according to whether the focus is on benthos- or ammonite-rich
horizons.
In contrast to the traditional biostratigraphic approach centered on correla-
tion, a stratigraphic paleontologic analysis can also be based on paleobiological
and paleoecological observations in their stratigraphic context aimed at an envi-
ronmental understanding. Ecostratigraphic analyses rely on detailed bed-by-bed
sampling. The precise range control of taxa involved is interpreted, taking into
account the ecosedimentary conditions.
Working on Upper Jurassic sections surrounding Iberia (Fig. 1), we employ
an ecostratigraphic sampling program based on the standardization of the sample
size at a minimum of 50 specimens of megainvertebrates per sampled bed. We do
not attempt to determine fossil abundance per rock volume, because of sampling
difficulties (rock hardness, fossil size) and excessive time required (see, Tipper,
1976). When a lar~e fossil-rich bed surface is exposed, we count the fossil remains
per quadrat (1 m ). When we need a volumetric estimate of fossil abundance
from surface observations, we follow McKinney's approach (1986). Middle- to
high-rank taxonomic data (genus, family, family-groups) are used for ammonites;
megainvertebrates (other than ammonites), which are mainly benthic faunas in
the cases studied (bivalves, brachiopods, but locally also belemnites), might be
considered as a whole or segregated in major groups (e.g. bivalves). Favorable
preservation and advanced phases in our research program should extend to the
study at the species level and the identification and structure of paleocommunities.
Therefore, the components selected in the faunal assemblages will be determined
by the objective of the research, the relative abundance of the recognized fossil
groups, and preservation. Consequently, taphonomic observations are of highest
importance, and ichnology is a valuable complement for revealing ecosedimentary
conditions of the substrate.
In the cases discussed here, the sampling strategy was mainly determined by
lithology because of consistent types of fossils (mainly ammonites, belemnites,
62 F. OI6riz et al.
mERIA -
lOOKm
'-----"
AFRICA
Fig. I. Outcrops of studied Upper Jurassic sections in the South and East Iberia. Epicontinental
shelves (Vertical ruling), epioceanic fringe (black), Intermediate Units belonging to the epicontinental
slope (dotted). AL (Algarve), IB (Iberian Cordillera), MA (Mal\orca), PB (Prebetic Zone), SB
(Subbetic Zone).
bivalves and brachiopods). Epicontinental successions of marly limestones and

marls facilitate the bed-by-bed sampling procedure, but bedding in ammonitico-
rosso facies is frequently masked by burrowing and extreme bed thinning. Thus,
in nodular and condensed limestones we sampled at constant thickness intervals of
10-20 cm, thinner sampling intervals being used only when necessary (e.g., high
time-averaged horizons with or without reworking and hardgrounds). Intrabed
sampling (bottom, middle, top) was made wherever uneven, irregular distributions
of fossils occurred. Characteristics of our standard sampling program, other than
size, are: I) random-sampling effect; 2) consistent distribution of faunal compo-
nents, and/or counterbalanced changes in faunal composition, affecting all except
the unstable-component (records ranging from 0 to 10%) of the faunal content;
and 3) precise demarcation of samples and sampling stations in the sections inves-
tigated. Moreover, we worked with single and averaged faunal spectra, which are
the quantified images of registered fossil assemblages. The first are used mainly
to obtain ecostratigraphic interpretations from faunal assemblages recorded in a

significant bed with a relatively poor to very poor fossil record. We also considered
beds where depositional events could not be identified, and the minimum sample
size (50 specimens) was difficult to achieve, looking for their ecostratigraphic sig-
nificance as evidence of ecostratigraphic events. We usually worked on averaged
faunal spectra which were the result of both the repeated sampling for a single
faunal spectrum in an especially fossil-rich bed, or the combination of single
faunal spectra belonging to a biochronostratigraphically-identifiable set of beds
without sedimentary discontinuities and/or significant changes in facies. As might
be expected, the level of required ecostratigraphic interpretation determined the
relationship among the number of faunal components (from basic type-faunas to
selected taxonomic ranks) to be differentiated in the analyzed faunal spectrum and
its size, i.e. around the minimum size sample of 50 specimens, we could consider
only two or three faunal components, and thus reach only general ecostratigraphic
interpretations. On the other hand, more precise interpretations needed the aver-
aged faunal spectra, in which the number of specimens far exceeded 50 (commonly
up to 300), allowing the recognition of more diversified faunal components. With
regard to ecostratigraphic interpretations, we considered fluctuations of specific
faunal components which are below 10% to be sampling noise; moreover, we did
not consider faunal components of below 2-3% to be important in our ecostrati-
graphic interpretations. Thus, the faunal content of under 10% is regarded as being
the unstable component in the recorded fossil assemblage.
Special attention should be paid to taphonomic information. Given that we are
working with fossil megainvertebrate assemblages which are usually very similar
in basic composition, it is important to recognize two clearly different depositional
environments on basis of field observations. Thus, in epicontinental mid- to outer-
shelf deposits, fossil remains are affected mainly by physical processes (crushing,
fragmentation and disarticulation) and buried comparatively rapidly. Conversely,
condensed deposits in epioceanic distal pelagic swells have comparatively longer-
lasting exposure offaunal remains, mostly under low-energy conditions, and there-
fore chemical processes (corrosion) predominate. Mostly, inner-cast preservation
dominates, allowing the recognition of uncovering when epizoa and/or bioerosion
traces are preserved on these inner casts. Epigenized (ca1citized) shells have been
also found. The recognition of post mortem transport is especially significant for
the appropiate evaluation of ammonite assemblages; thus, we paid special attention
to: I) the ammonites/aptychi ratio (ammonite shells vs. aptychi other than 1:2);
2) the record of "epizoaires intrathalames" and "extrathalames" (Tintant, 1980,
1983), which can be identified on inner casts and epigenized shells respectively;
and 3) the presence/absence of shell sorting used to identify traces of post mortem
transport in standard populations.
Papers by Savrda (1991), Pemberton (1992) and Taylor (1992) demonstrate the
use of ichnology in sequence stratigraphy. Since we are not working with nearshore
deposits, and gi ven that emergence and/or well-developed hardgrounds are not very
common, trace fossils are generally unspecialized. Chondrites and Planolites pre-
dominate in shelf deposits, but Diplocraterion, Arenicolites and Thalassinoides
have also been found on firmgrounds and omission surfaces. In the ammonitico
64 F. 016riz et al.
rosso facies, Chondrites and Thalassinoides are the most frequent trace fossils.
Thus, we found ichnology to be of limited use in our Upper Jurassic sections with
cephalopods in Iberia. Nevertheless, we recognize the value of trace-fossil analysis
in ecostratigraphic and sequence stratigraphic interpretations of lithologic succes-
sions, especially when skeletal remains of megainvertebrates are scarce. In cases
of a scarce ammonite record, 0l6riz and Rodrfguez-Tovar (1993) demonstrated the
relationship between benthos-poor records and dysaerobic substrates during the
Lower Kimmeridgian in epicontinental areas of southern Spain (Prebetic Zone).
3. Ecostratigraphy and Sequence Stratigraphy
Ecostratigraphic interpretations of faunal assemblages should be accompanied by

an understanding of the depositional dynamics, with particular attention paid to
the stratigraphic aspects. Depositional conditions are known to differ between
sedimentary environments, and due to local factors, within a given depositional
environment. Different conditions of deposition represent one of the possible ways
in which the accommodation space (Jervey, 1988), or the accommodation enve-
lope (Posamentier and others, 1988), is filled, that is, changing accommodation
influences deposition and thus facies within a sedimentary environment. This is
the basic tenent of sequence stratigraphy when fluctuations in relative sea-level are
the controlling factor. Therefore, relative sea-level fluctuations, changing accom-
modation and shifting facies, produce the cause and effect logic for interpreting
stratal patterns and depositional environments within systems tracts and deposi-
tional sequences.
The stratal patterns are associated with the accommodation history, but they
are not the only evidence of the occupied space between the sea floor and the
sea surface during a given time interval. However, accommodation can be better
thought of as a complex and multidimensional ecological volume (ecospace) which
could be widely, but often unevenly, occupied from the sea surface to the lower
boundary by the infauna (Fig. 2). Extreme situations could be the existence of
ecospace with zero accommodation (Jiving opportunities for endobenthos in final
phases of basin filling), or potential accommodation being beyond the limits of
conditions for living systems (local extreme anoxia and poisonous events).
Fluctuations in relative sea level cause changes in accommodation, with or
without significant shifting of facies, and affect changes in the ecospace volume.
These fluctuations can be easily interpreted from significant lithofacies and biofa-
cies changes. However, the analysis of fossil assemblages is necessary to identify
changes in ecospace, and thus the accommodation, when no changes, or only
subtle changes, in lithofacies occur in response to relative sea-level changes. If
so, relative sea-level fluctuations, even with minor changes in accommodation,
could be interpreted in ecologic terms through the analysis of fossil assemblages
by using an ecostratigraphic approach, whenever the duration of ecologic factors
determining the composition of faunal assemblages is longer than that involved in
relative sea-level fluctuations.
On this basis, shifting ecospaces must also be considered within the systems-
tracts framework and, consequently, within the framework of depositional se-
sea level
Fig. 2. Schematic relationship between accommodation and ecospace. Note: a) two extreme cases of
maximum (black arrow), and minimum (black triangle) accommodation, b) three ecospace volumes
taking into account substrate occupation by endobenthos (longer white arrow corresponding with
larger ellipse), unfavourable substrate for endobenthos occupation (medium size white arrow and
smaller ellipse), and extremely reduced ecospace which could correspond with zero accommodation
(see text).
quences (Figs. 3 and 4). In such a situation ecostratigraphic trends deduced from
the analysis of fossil assemblages could be far more sensitive than traditional
biostratigraphy alone, or at least a useful complement for correlation. In cases
where only inadequate biostratigraphy is available, sequence stratigraphic corre-
lation could be performed by means of ecostratigraphic trends. These trends can
be studied by the analysis of benthos (Baum in Ross and Ross, 1986; Ross and
Ross 1986; Breyer, 1992, Garcia and Laurin, 1992; Sztejn, 1992; Hirsch, 1993,
among others), plankton (Brinkhuis, 1992 and Wilpshaar, 1992), and combined
data of benthic and/or pelagic and planktic fauna (Ol6riz and others, 1991, 1993;
Clavel and others, 1992; Stein and others, 1992, among others). Other examples
centered on cephalopods are found in Donovan (1985), Gygi (1986), Atrops and
Ferry (1989), Hantzpergue (1991), McGhee and others (1991), EI Hariri and others
(1992), Hoedemaeker (1992), Reboulet and others (1992) and Ol6riz and others
(1993). More general approaches to relationship between paleontologic records and
relative sea-level changes interpreted according to sequence stratigraphy can be
found in Kopaevich (1992), Holland (1993) and Patzkowsky and Holland (1993).
In addition to the control of ecostratig raphic trends as tracers of sequence evolution
and/or sequence differences, the identification of ecostratigraphic events (Kauff-
man, 1986, 1988) could potentially provide information about sequence boundaries
in the traditional context of HIRES, especially for Type 1 unconformities.
4. A Theoretical Case as Deduced from the Iberian Upper Jurassic
A theoretical case demonstrating the application of ecostratigraphic analysis and its

relationship to sequence stratigraphic interpretations is shown in Fig. 5. Proximal
(left), mid- to outer-shelf (center) and distal epioceanic (right) columns represent
potential sedimentary records in Iberian paleomargins during the Upper Jurassic.
Lithologic differences are expressed by the siliciclastic/carbonate ratio as well as
by detailed lithofacies. Reference horizons are condensed levels at the middle of
66 F. 016riz et al.
short tenn
EUSTATIC
CURVE ACCOMMODATION
+ + o
SIGMOIDAL SEQUENCE STRATIGRAPHY MODEL
HS
TR
SMW
lllGHSTAND SYSTEM TRACT

+ + 0
1 ~
() ~
---.:v
......:::::
I I
~ ~+
TRANSGRESSIVE SYSTEM TRACT
, ~
+ + 0
~ ~
-- v ....
oco::::::::::::
I
::::= ~
I
SHELF-MARGIN SYSTEM TRACT
a
+ + 0
~
f)
~
I I
"-- ~
Fig. 3. Idealization of expected relationship between eustatic fluctuations, accommodation and
stratal patterns according to the sequence stratigraphic model for a type-2 depositional sequence of
the third order. Note: a) averaged faunal spectra and ecostratigraphic trends belonging to different
systems tracts, and b) differentiated fossil assemblages in relation with epicontinental environments
(left) and epioceanic swell areas (right). Pie-diagrams: Megainvertebrates (other than ammonites)
(black), phylloceratids and lytoceratids (vertical ruling), ammonites (white in left), ammonitina if
other ammonite group are represented (white in right). Systems tracts abbreviations follow normal
convention of Sequence Stratigraphy.
DISTURBED TRANSGRESSIVE SYSTEM TRACT

short \em
EUSTATIC ACCOMMODATION
CURVE ACCOMMODATION
+ _ + 0 + 0
\ ()
1
(~
?1~
I
c_:!:t~
Fig. 4. Idealization of expected relationship between relative sea level and accommodation during
Transgressive Systems Tract development with tectonic interactions (uplift affecting epicontinental
shelves: black arrow). Note: a) differential accommodation (horizontal ruling) and then ecospace on
epicontinental environments (left) and epioceanic swells (right), b) deviation of the averaged faunal
spectra characterizing epicontinental shelves when compared with the equivalent expected standard
in Fig. 3, and c) unchanged situation in unaffected epioceanic swells.
the sections, the mass-mortality horizon in the epicontinental sites (left and center)
and a bio-intraclastic bed sets at the top of the three sections. These different records
are in agreement with the paleogeography.
The time interval represented is assumed to be geochronologically identifiable
between times tJ and t2, when a more accurate geochronological resolution is un-
available. Bio- and chronostratigraphic correlationship are possible by the record
of taxa a (ammonites), band c (planktic and benthic foraminifera, respectively)
and, as might be expected, differences in the biostratigraphic distribution are re-
lated to paleogeography and ecosedimentary dynamics (see comments below).
Therefore, homotaxial differences make intra-biozone a correlationship difficult.
Only correlationship at the biozone a level are reliable using both biostratig-
raphy and geochronology. Bed-by-bed sampling provides an accurate image of
the evolution of fossil assemblages when significant taphonomical effects are
not present. It is worth mentioning the recurrent composition of the ratio am-
monites/megainvertebrates (other than ammonites) (black and soft-grey columns),
as well as peaks of these which can clearly be correlated with sudden lithological
changes unevenly recorded in the sections. The evolution in averaged ammonite
assemblages (pie-diagrams) are in accordance with general trends in faunal com-
position.
On the basis of a bed-by-bed study, we pay special attention to taphonomic
and ichnologic features in order to characterize maximum flooding surfaces. As
a whole, in the proximal section (left) the maximum flooding surface is an event
of short duration determining soft to firmground conditions identifiable by trace
fossils assignable to distal Cruziana ichnofacies (Chondrites, Planolites and Tha-
lassinoides). The same surface in the mid- to outer-shelf site (center) is a firm
RELATIVF
SEA "
LEVEL
..
... v~~~ ~sM.:~ v~_~~ 0\
00
"""
"Vf~~~~rw.~~ V:-:~I.
EPICONTINENTAL SHELVES EPIOCEAN IC ENVIRONMENTS

~
1- -
w~
c=-__-_-----,
~ ~ ~
" " " " "
,2
.anunolllk::s:
o-",n-IDr"",gro...o \ () - b.roekground cnc:'l:Y
RUrl ""
_=__ IuncoID II -fum- Ii;) hiNgound ........ mcgainvc.rtd'll'alcs
_ _ t- limc.Iwnclmul alLcmlition 0. - U:me-....cn~J.ng
. - turdgroun<:! with biocroilion ..,.. olher than ammonila
-
.,_~. ~
i~--
fi..: -
=~ =~nc
I\lJ<h.. t.r h~I.iPO(:
luK,lcOITOJlOn tIlIiCCII -A
.;:: - Id\ column ph)'lloc"CnliLLll~..... unmonilina
_ l-:tCIe capOOn
dcpo"",
,
- (;Q1\c.r column &. l)'lOL~nlilJ..t
o -_ condenKd
un.tabk. S(:I bouorru
- - - ma.rly ammoniUc.'O rouo
--=-~ - ca~tcOW ammonilico ",.so
- right column
,mm- a.mmom\cl:
o megaan... c.l1cbnlCl ~~ .~sodimenta.ry sliding
- and racd.unml.allOn
.:..~..~ bl-bioclutic layer Dmota- other I.ha.n ammonites ~ -Oulocrc~moNlI.c platl~m\S
_ cb .. c-oquiN bed . .. " - ammombCo rotlO r.CICI
_ 2-soc caotion
ABOVE BELOW
Fig. 5 Idealized Southern Iberian paleomargin during the Upper Jurassic with synthetic sections. Below: Idealized Upper ~
Jurassic Southern Iberian paleomargin with main physiography, environmental conditions and stratigraphic features. Differen- Q
tial possibilities to recognize stratal patterns in outcrops (black-I). Siliciclastics and/or turbidites with or without radiolarites CA
in epioceanic areas (black-2). Above: Synthetic sections characterizing proximal (left) and mid-outer (middle) epicontinental ~.
depocenters, and distal epioceanic swells (right) belonging to a geochronologic interval (tl - tz), biostratigraphy, fossil (b
.....
assemblages, selected abiotic traces, and relative sea-level curve. Note differential thickness (scales: x I to X 10-100 from
right to left) and features of mid sections condensed surface (wavy line) with differential trace fossil suites. See text for
~
explanations.
to hardground surface which contains Diplocraterion, Arenicolites and Thalassi-

no ides (Glossifungites suite). This surface in distal epioceanic swells (right) is
marked by extreme condensation and mineralization with dense Thalassinoides,
bioerosion and corrosion phenomena making it hard to recognize other trace fos-
sils. Differences in taphonomy are related mainly to exposure time of fossil remains
increasing seawards as revealed by increasing time averaging in epioceanic envi-
ronments where low-energy conditions prevailed.
Since useful sedimentary structures are usually absent in the Upper Jurassic
deposits studied, the abiotic signals used are clay minerals and geochemical data.
Peaks in detrital clay minerals are recorded in relation to the Sr/Mn ratio and 8 18 0
incursions.
The interpretation of the ecosedimentary evolution in our theoretical case
(Fig. 5) is based on megascopic aspects (lithologic evolution and paleontological
data) together with the proposed record of mineralogical and geochemical signals.
Overall, lithology enables the differentiation of epicontinental (left and center)
from epioceanic (right) sites.
In contrast to nodular marly limestones characterizing the epioceanic column,
marl-limestone rhythmites characterize epicontinental sites which hardly differ
from each other in thickness below the easily identifiable condensed horizon.
Below this condensed horizon increasing condensation is recognizable in all the
sections through stratal patterns. However, it is particularly important to note the
consistent increase in the ratio of ammonites to megainvertebrates (other than
ammonites), which reflects the increasing record of mainly epioceanic dwellers
(ammonites: phylloceratids and lytoceratids, vertical ruling). The maximum values
of these trends are recorded in coincidence with the time-averaged horizons (wavy
lines) which evidence maximum flooding features. Peaks of the Sr/Mn ratio also
fit this interpretation.
Sedimentation above the condensed deposits is well differentiated between
sections. In the epicontinental sections, the proximal section (left) is made up of
comparatively expanded marly deposits with well-developed bioclastic interca-
lations upwards. The lithologic succession in the center is slightly thinner, with
less developed upper bioclastic layers. The coquina bed, which is more apparent
in proximal sites of the epicontinental shelf system, is consistent with incursions
in low oxygenation (negative peak in 8 18 0) and peaks of vermiculite + kaolinite
(Vrm+Kln). These record siliciclastic influx causing mass-mortality. On the other
hand, the epioceanic section (right) clearly differs, with dominant ammonitico
rosso (condensed facies) grading into more extensive bioclastic limestones up-
wards. The proposed record for faunal assemblages shows a consistent decreasing
trend in the ratio of ammonites to megainvertebrates (other than ammmonites)
across localities, with punctuated reversals related to short-term deterioration of
vital conditions for benthos when high-energy sedimentation occurred (bioclastic
layers). The high frequency of these layers correlates well with a more proximal
setting. The record of mainly epioceanic dwellers (ammonites: phylloceratids and
lytoceratids) is assumed to relate to the distance from shore and sea level.
As stated previously, biochronostratigraphic correlationship are difficult within
biozone a, due to homotaxial differences (Fig. 5). In this regard, we note that the:
70 F. 016riz et al.
1) the last appearance datum (LAD) for taxon c occurs at the first record of high-
energy deposits in the middle and right columns, but below that in the proximal
column (left); 2) the first appearance datum (FAD) for taxon c, which is registered
below the hiatal surface (wavy line) only in the mid section; 3) the coincidence of
the LAD for taxon b with the hiatal surface in all the sections; and 4) the concurrent
FADs for taxa a and b in all but the proximal section on the left. These differences
require an explanation.
Interpretation of these biostratigraphic data is possible in ecologic and de-
positional terms. The delayed FAD of taxon b in proximal sites (left section)
without lithological differences when compared to relatively distal depocenters
(mid section), points to ecology as the controlling factor. The fact that the record
of taxon b in the proximal section is condensed when compared to more distal
sections, suggests that the range of taxon b is controlled by the transgressive ef-
fect of a relative sea-level rise. The related maximum flooding could be easily
identifiable in the hiatal surfaces at the top of condensed deposits in all the sec-
tions. The sharp changeover of taxa b to c in proximal (left) and epioceanic (right)
sections contrasts with their overlap in the middle section. Decreasing conden-
sation in outer-shelf sites (middle section) and wider hiatuses under low-energy
conditions on epioceanic swells (right section) could explain these differences in
biostratigraphic records in the middle and right columns, and ecologic limitations
(delayed increase of ecospace) and/or erosion in proximal sites could cause the
observed record in the left column. The recorded differences in LADs of taxon c
are caused by ecosedimentary conditions. In proximal sites (left) a mass-mortality
event locally extinguished taxon c, whereas in relatively distal sites, shifting fa-
cies (high-energy conditions) progressively reduced or eliminated the ecospace for
taxon c basin wards in time.
Taking into account the proposed changes in both lithology and fossil assem-
blages in the sections, and the assumed time content for the tl - t2 interval as
deduced from mean values of ammonite biozones (7 x 105 _106 years), we find
the homotaxial differences to be consistent with relative sea-level fluctuations, as
expected from developing transgressive and highstand systems tracts belonging to
a third-order depositional sequence.
5. Applications to Upper Jurassic Outcrops in the Western Tethys
The case studies discussed here refer to comparative analyses of fossil assem-
blages of megainvertebrates: a) at the stage-substage level (example 1) and at
the ammonite biochronozone level (example 2) from epioceanic and epicontinen-
tal environments; b) from distant sites belonging to epicontinental shelf systems
surrounding Iberia (example 3); c) controlling trends in mineralogical and geo-
chemical variables (example 4); d) revealing the influence of tectono-eustatic
interactions on the fossil record and their interpretation (example 5).
Ecostratigraphic interpretations, involving time-controlled processes, must be
based on both a faunal-reference composition deduced from fossil assemblages
and their paleogeographic setting, especially when combined with the sequence-
stratigraphic approach.
./ vAULTS. WRe:-:CII. TIIRUST

1IS1IALLOW M;\RI:>n. MAINLY
~ CIWBO~.\T[s
Fig. 6. Paleogeography and main tectonic framework during the Upper Jurassic (from Ziegler, 1988)
with location of studied outcrops: 1 (Algarve Basin), 2 (Iberian Cordillera), 3 (Mallorca Island), 4
(Prebetic Zone), 5 (Subbetic Zone), AKB (Alboran-Kabylian Block), AM (Armorican MassiO, AZFZ
(Azores Fracture Zone), BB (Bay of Biscay Rift), IBM (Iberian Meseta).
As shown in Fig. 6, the South and East Iberian margins were near the re-
gion of relative movements between Africa and Iberia on the western extreme
of the Tethys during the Upper Jurassic (Seyfried, 1980; Garcia-Duenas and Co-
mas, 1983; Canerot, 1985; Comas and others, 1988; Manupella and others, 1988;
Ziegler, 1988; Vera, 1988; Aurell, 1990). Basically, epicontinental shelves and
distal pelagic swells belonging to the epioceanic fringes provide a traditionally
accepted image of the paleomargin structure and ecologic environments (Ol6riz
and others, 1991). Rhythmic successions of limestones (mudstones/wackestones)
and marls characterize epicontinental deposits in contrast to condensed nodular
72 F. 016riz et al.
limestones (ammonitico rosso) which develop on epioceanic swells. At times,

thicknesses in epicontinental successions reached 100 times those recorded in epi-
oceanic swells (Marques and others, 1991; Ol6riz and others, 1991). Outcrops with
ammonites show the dominance of relatively low-energy deposits with some sig-
nificant intercalations of clastic inflows, revealing short-time depositional events as
well as synsedimentary sliding of mud-flows and breccias, mainly on slopes (Co-
mas and others, 1981; Ol6riz and others, 1993). Major paleomargin evolutionary
features during the Late Jurassic are: 1) the geological event that caused the basal
unconformity of the Upper Jurassic, which is recognizable in both the epicontinen-
tal and the epioceanic environments; 2) the notable change in subsidence rate near
the Oxfordian-Kimmeridgian boundary; 3) the general shallowing affecting epi-
continental shelves from the Middle Kimmeridgian upwards; and 4) emergence of
inner and mid-epicontinental environments and the near-contemporaneous phys-
iographic reorganization of epioceanic fringes during the Berriasian. All of these
events caused significant changes in ecosedimentary conditions in relation to in-
teractions between tectonics and eustasy (see Marques and others, 1991 for an
extended treatment).
As previously stated, two major ecological environments persisted during the
Late Jurassic related to the main physiographic features of the paleomargins.
Outside the buildups and carbonate platforms, ammonites dominated the vagile
faunas in the epicontinental and epioceanic environments.
The details of ammonite ecology are controversial, but a consensus exists: 1) the
mainly open-ocean character for phylloceratids and lytoceratids which generally
have smooth shells, variable coiling, and high values of the siphuncular-strength
index; 2) the eurytopy of the phylloceratid and shell-constricted Sowerbyceras
within sUbtropical latitudes; and 3) the relatively shallow environments for va-
grant epibenthic or neckthic perisphinctids, in both epicontinental and epioceanic
waters. In the case studies, "other ammonitina" refers especially to aspidoceratids
and haploceratids, which have a comparatively less clear ecology. Megainver-
tebrates (other than ammonites) include belemnites, which only ocasionally are
significant, and benthic faunas among which bivalves and brachiopods are the
most significant, fluctuating in accordance with distance from shore and local
ecology near the substrate. Persistent differences in faunal composition and stabil-
ity between epicontinental and epioceanic records seem to indicate major ecologic
differences associated mainly with the ecospace volume, nutrient availability, and
environmental stability. On epicontinental shelves ecospace volume and environ-
mental stability were comparatively low but nutrients levels relatively high. Local
conditions could make oxygenation to be a significant ecologic factor within epi-
continental environments. Other abiotic factors (e.g. salinity) controlling general
ecology in the cases presented here are rarely identified.
Fluctuations in relative sea level would potentially affect ecosedimentary con-
ditions in very different ways in epicontinental and epioceanic environments.
Shifting facies and ecospaces are assumed to occur widely on the shelf system,
which was far more sensitive to relative sea-level changes as indicated by the
known evolution of facies and fossil assemblages. Transgressive effects favored
incursions of open-sea inhabitants landwards and increased the ratio of ammonites
A 8 C l) E F G H
1.\.1
UPPER 135
T
I 1.'6 I> xv
T
MLOWER S2
N. Alb.
138
139 UXII
Hyb.
K
I 142 I>XI
M
M. 143 dX
- - --
KII
I> IX
bV111
SI
d VII
OX III
- - -
I> VI
dV
OX II
I>IV
(Jill
OX I
Fig. 7. Depositional sequences in the Southern Iberian Margin relative to the cycle chart and
sea-level curve proposed by (Haq and others, 1988). Adapted from Marques and others (1991)
and Ol6riz and others (1993). Middle-Upper Oxfordian coastal onlap curve slightly modified. A)
geochronology, biostratigraphy in the Tethys and sourrounding areas, B) Third-Order cycles, C)
coastal onlap curve, D) systems tracts, E) eustatic sea-level curve, F) time in million years, G)
discontinuities, H) depositional and tectono-eustatic sequences, I) supersequences.
to megainvertebrates (other than ammonites) for a given shelf site. In such a sit-
uation a trend towards decreasing siliciclastics and increasing condensation was
normal, and opposite trends correlated with regressive phases. On the other hand,
ecosedimentary conditions were comparatively constant on epioceanic swells, and
relative sea-level fluctuations were recorded by subtle lithologic changes and com-
paratively minor changes in the general composition of megainvertebrate faunas.
As might be expected, local conditions in epicontinental as well as in epioceanic
environments could change the general trends discussed above.
On this basis we shall discuss some cases in the Upper Jurassic outcrop sur-
rounding Iberia with reference to the cycle chart (Fig. 7) ofHaq and others (1988),
slightly modified by Marques and others (1991) and 0l6riz and others (1993).
Recent proposals by Ponsot and Vail (1991) and Ponsot-Jacquin (1992) are not
applicable.
5.1. EXAMPLE 1: ECOSEDIMENTARY EVOLUTION AT THE STAGE-SUBSTAGE LEVEL
The analysis offossil assemblages of megainvertebrates and their relationship to the

sedimentary record during the Middle-Upper Oxfordian and Lower Kimmeridgian
in Southern Iberia has been introduced by 016riz and others (1991). This example
(Fig. 8) refers to the comparison of averaged megainvertebrate spectra from the
74 F. 016riz et al.
EUSTATIC
AL long teml SB CURVE
+I _ short term
ClI:
+ 1-
Z L1l

.....
p.
p.
::J
~
0 L1l
..J
~ Q
>< ....
0
Q
~
I long tern)
.............
--==-
~
Fig. 8. Averaged megainvertebrate spectra and ecostratigraphic trends at the substage level. AI-
garvian Basin (AL), Subbetic Zone (SB). Concentric pie-diagrams are used to show the data from
the Middle and the Upper Oxfordian. The inner pie-diagram represents the general composition
expressed as ammonites (white) versus megainvertebrates (other than ammonites) (black). In the
outer, enviromentally significant ammonite groups are specified. Note the large amount of megain-
vertebrates (other than ammonites) in epicontinental records (left), decreasing perisphinctids (broken
lines in outer pie-diagrams, AL) and increasing phylloceratids and lytoceratids (SB) during rising
long term sea level. Pie-diagrams as in Fig. 3.
epioceanic Subbetic Zone (Spain) with comparatively homogeneous ammonitico

rosso facies and the epicontinental East Algarve Basin (Portugal), where a locally
variable rhythmite made of limestones and marls was deposited. The interval
analyzed, the Middle-Upper Oxfordian, corresponds to a rising long-term eustatic
curve (Haq and others, 1988), but does not show any direct relationship with
complete depositional sequences (i.e. incomplete cycles 4.3 and 4.4 of LZA-4
Supercycle, ibid.).
More than 600 specimens have been obtained in the Algarve and 2000 in the
Subbetic Zone. Fig. 8 represents the location of the study areas in a synthetic
profile of the Southern Iberian paleomargin. Major differences were the large
amount of megainvertebrates (other than ammonites) in the epicontinental areas
and the increasing record of phylloceratids and lytoceratids in the epioceanic
environment during the rising long-term sea level from the Middle to the Late
Oxfordian. Note also the slight decrease of perisphinctids in the epicontinental
areas. The trends observed are coherent with those expected during expanding
ecospace conditions. However, we considered relatively minor fluctuations to be
probable artifacts caused by lengthly time-averaging of data introduced by the
wide-range compilation of fossil assemblages at the substage level; i.e., the wider
the time range analyzed the more stable the faunal composition due to the averaging
effect. Thus, this level of observation is valid only for general ecostratigraphic
characterizations, and caution should be taken when using averaged data which
PB MA EUSTATIC
CURVE
~ i
~
Q.
---
@_HS ~
-~
~
o~ .
::>
Q.
TR
i
u. '"'
~ !Xl ~ - - - ':) -----@ , ," ~~ ;'
1 I
---- ---
Fig. 9. Averaged megainvertebrate spectra and ecostratigraphic trends at the ammonite biochrono-
zone level. Prebetic Zone (PB), Mallorca (MA). Note the correlation of ammonite chronozones with
short term sea-level fluctuations and systems tracts and changes in benthos correlating with systems
tracts and expected ecospaces evolution in epicontinental environments (PB). Eurytopic phyllocer-
atids (Sowerbyceras) and belemnites are responsible for deviations of expected records in epioceanic
swell (MA). See text. Pie-diagrams as in Figs. 3 and 8.
represent a combination of significantly different segments along the sea-level

curves (e.g., rising vs. falling sea level).
5.2. EXAMPLE 2: ECOSTRATIGRAPHIC INTERPRETATIONS AT THE AMMONITE BIO-

CHRONOZONE LEVEL
The ammonite biochronozone duration approximates that of systems tracts and

occasionally they may share limits, in which case, ammonite biochronozones su-
cessions could be used to analyze the developments within depositional sequences
and then to characterize them according to local conditions. Deviations from the
expected patterns in faunal trends can be explained as having been caused by
local factors distorting eustatic influences on the ecospace evolution. In this case,
the selected example was that of the Bimammatum and Planula Zones (Upper
Oxfordian), and the Platynota Zone (lowermost Kimmeridgian). We considered
the Bimammatum Zone to represent the shelf margin wedge (SMW), the Planula
Zone the transgressive systems tract (TST), and the Platynota Zone the highstand
systems tract (HST) of the 4.4 third-order cycle of Supercycle LZA-4 of Haq and
others (1988).
Fig. 9 represents the general location of the records compared in a synthetic pro-
file combining the main physiographic subdivisions of the southern epicontinental
shelves (Prebetic Zone) and eastern epioceanic swells (island of Mallorca) along
the Iberian paleomargins. This example was investigated using a total amount of
76 F. 016riz et al.
1200 megainvertebrates in the epioceanic (Central Sierra Norte, Mallorca Island)

and more than 1500 in the epicontinental (Central Prebetic Zone) environments. On
the epicontinental shelves the recorded trend in the ratio of ammonites to megain-
vertebrates (other than ammonites) accurately indicated the expected changes in
the ecospace from SMW to HST conditions. The record of perisphinctids was
also consistent with an increasing-decreasing ecospace during the evolution of the
depositional sequence. In epioceanic swells, there was low proportion of benthos
5%). Belemnites ranged from 90% to 50% within the megainvertebrates (other
than ammonites), in contrast to their scarce record in epicontinental areas 5%).
Specialized phylloceratids and lytoceratids evolved consistently with the expected
ecospace evolution, although the ubiquitous Sowerbyceras increased largely under
HST conditions. This unexpected record of Sowerbyceras could exemplify a de-
viant record induced by the evolution of the taxon, but migration from diminishing
platforms cannot be ruled out, especially in the case of a vagile eurytopic taxon.
5.3. EXAMPLE 3: COMPARISON OF FAUNAL ASSEMBLAGES FROM DISTANT EPI-

CONTINENTAL AREAS
Significant information can be obtained by comparing chronostratigraphically-

correlatable averaged spectra of megainvertebrates from distant areas belonging
to the same ecological environment. In this case we tried to analyze differen-
tial ecological responses relative to local environmental conditions. The selected
example was from epicontinental deposits belonging to the Platynota Zone (low-
ermost Kimmeridgian) surrounding Iberia. We shall consider differential ecospace
occupation during the HST conditions in response to a variable connection with
open-sea waters. For an extended explanation the reader is referred to Ol6riz and
others (1988, 1991).
We selected three widely separated sectors which represent paleogeographi-
cally well-differentiated areas (Fig. 10). A greater disconnection with the open sea
was assumed to affect the sections studied in eastern Iberia (A1corisa-Berge sec-
tor) compared with those in the East Algarve Basin (LouIe sector) and the Prebetic
Zone (Central Prebetic). The total number of recorded megainvertebrates were
800, 500, and 850, respectively. The lowest ammonites/megainvertebrates (other
than ammonites) ratio, together with the over-dominance of specialized epiconti-
nental perisphinctids (Ataxioceratinae) in the A1corisa-Berge sector (East Iberia),
clearly indicated the relative proximity to inland-sea conditions. This interpretation
is consistent with the near-absence of eurytopic phylloceratids (Sowerbyceras).
Conversely, more distal settings could be assumed for the Louie sector (Algarve
Basin) and the Central Prebetic Zone along the Southern Iberian margin. Along
this margin, the proximity of the epioceanic branch of the Tethys (see Fig. 6)
is revealed by comparable numbers of phylloceratids, but the higher numbers of
specialized epicontinental perisphinctids (Ataxioceratinae) in the Algarve Basin
could be interpreted as increasing disconnection with epioceanic waters and/or
development of ecologically-favourable environments for these ammonites.
200Km
'-------'
Fig. 10. Averaged faunal spectra from distant epicontinental areas. Algarvian Basin (AL), Prebetic
Zone (PB), Iberian Cordillera (JB). Note inland sea affect in (JB). See text. Pie-diagrams as in Figs. 3
and 8.
5.4. EXAMPLE 4: ECOSTRATIGRAPHY AND TRENDS IN ABIOTIC COMPONENTS
The combination of ecostratigraphic data and evidence for abiotic trends in grain
size, geochemistry and mineralogy can allow the interpretation of the ecosedimen-
tary evolution. Changes in the ecospace and accommodation are to be approached
together, including possibilities of active local tectonism.
To illustrate this we shall discuss the epicontinental Prebetic Zone (Segura de
la Sierra section) in Southern Spain (see Rodrfguez-Tovar, 1993). The succession
is comprised of 70 m of rhythmic alternations of marly limestones and limestones,
with marly intercalations in the lowermost and uppermost parts of the section.
Significant marly intercalations in the lower section, as well as slumps in the middle
to upper section indicate source-area erosion and bottom instability. This section
belongs to the Platynota Zone (lowermost Kimmeridgian), which we considered
to correspond with the HST in the tectonically disturbed 4.4 third-order Cycle of
the Supercycle LZA-4 of Haq and others (1988), as modified by Marques and
others (1991).
After the tectonic pulse recorded at the base of the section by significantly
lower marls, a reduction in accommodation and ecospace was expected during
progressive HST conditions; we interpreted marly intercalations of up to 4m thick
at the top of the section as being related to depositional conditions in the late high-
stand. Thus Fig. 11 shows a decreasing ratio of ammonites to megainvertebrates
(other than ammonites) (inner pie-diagrams) with an increasing relative domi-
78 F. 016riz et al.
PB GRAIN SIZE Cm+Qtz . F.USTATIC

(Segura de la Sierra) (M~n Om.) Cal+Ool Kaolinite A1+K+Fe+TI CURVE
)01"1""1 ..............- J--'---i I I +
o I'D 10 o 3 0 30" 0 ppm. 150
I 01 " I " 01 I r-----r---I t---r--I I , , I

o )'iJI 10 0 3 0 30" 0 ppm. ISO
Fig. 11. Recorded trends in averaged faunal spectra and selected abiotic components in epicontinental
environments in the Prebetic Zone (PS). See text. Pie-diagrams as in Figs. 3 and 8.
nance of specialized perisphinctids (Ataxioceratinae) upwards (broken lines in the

pie-diagrams). In depositional conditions of a near-constant siliciclastic/carbonate
ratio (very slight increase in the upper section) the increased grain size, signifi-
cant detrital elements (AI, K, Fe, and Ti) and kaolinite, are all consistent with the
expected evolution of the local accommodation and of progradation during the de-
velopment of the HST. Except for megascopic features (lower marls and slumps),
local tectonics seems to have had no special effect on combined ecostratigraphic
and sequence stratigraphic patterns.
5.5. EXAMPLE 5: INFLUENCE OF TECTONO-EUSTATIC INTERACTIONS ON TRENDS

IN THE COMPOSITION OF FOSSIL ASSEMBLAGES
If fossil assemblages are interpreted in terms of ecostratigraphic trends, relative

to the expected ecological conditions in relation to relative sea-level fluctuations,
a key point in the combined ecostratigraphic and sequence stratigraphic model

should be the possibility to recognize deviations in the composition of fossil
assemblages. These deviations most likely are caused by local factors and their
interaction with eustasy, as examined in the example below.
This case deals with the interpretation of significant differences in fossil assem-
blages of megainvertebrates recorded in three neighboring sections (GA6, GA7,
AH4) belonging to the proximal epioceanic fringe of the Southern Iberian margin
(Sierra de Gaena). The AH4 section is located about 7 km West of the GA6 and
GA7 sections. The distance between the latter two is less than 2 km. The Lower
Tithonian (AlbertinumiDarwini Zone) is made up of 140 cm (A~), 133 cm (GA7)
and 103 cm (GA6), mainly of calcareous ammonitico rosso facies (broad sense).
The standard lithological column is represented by more calcareous and slightly
nodular beds at the bottom, overlain by an intercalation of red nodular marly lime-
stones, and again red nodular limestones. This is recorded in sections AH4 and
GA7, which differ only in higher condensation, including hiatuses towards the top
in GA7. In the thinner GA6 section, pinching-out horizons are present and the
marly nodular ammonitico rosso intercalation is recorded only at the top.
In these sections, 750 megainvertebrates were stratigraphically sampled
through the Lower Tithonian AlbertinumiDarwini Zone. The standard ecostrati-
graphic trend was registered in the AH4 section, which showed coherent fluctua-
tions in faunal composition fitting the assumed sea-level curve with significantly
counterbalanced records of phylloceratids and perisphinctids (see Fig. 12). On the
other hand, the GA7 and GA6 sections provided an opposite and easily identi-
fiable trend in the composition of megainvertebrate assemblages which adapted
well to shallowing and upward deepening conditions, respectively (tilting blocks).
Significant in this case were the records of phylloceratids and perisphinctids. In
addition, taphonomical bias (differential post mortem transport) was thought to
have occurred as shells floated from highs (GA7) to nearby lows (GA6).
6. Conclusions
Ecostratigraphic trends can be inferred from the analysis of paleontologic data with
special attention to depositional and stratigraphic frameworks. Ecostratigraphy
complements sequence stratigraphy, especially in the analysis of outcrops, where
an iterative interactive approach is most appropriate.
Accommodation, stratal patterns, systems tracts and sequence boundaries can
be related with ecospace, averaged faunal spectra, ecostratigraphic trends and
ecostratigraphic events. Even so, the complicated relationships between shifting
ecospaces and facies, as well as between sequence boundaries and ecostratigraphic
events is recognized, since such relationships can be found at any moment in a
developing depositional sequence. We also recognize the existence of presumed
sequence boundaries in middle to outer platforms and epioceanic swells without
significant ecostratigraphic or even lithologic traces. The key point here is the
variable relationship between the time span involved in these hiatal surfaces and
that required for ecological responses in the faunas involved. As might be expected,
80 F. 016riz et aJ.
ACCOMMODATION
EUSTATIC CURVE AH. GA,
+ GA...... ~ 1
~ 0 G~
TR
HSulSMW
TRfHSI
Fig. 12. Tectono-eustatic interactions and ecostratigraphic trends in epioceanic swells (Subbetic
Zone, SB). Standard records of megainvertebrate assemblages on the left. Note similar faunal com-
position of lower pie-diagrams and the persistent increasing-decreasing record of phylloceratids
and lytoceratids. and perisphinctids on mid and right. Arrows indicating tilting which determines
accommodation and ecospace decrease on right and the reversed on mid sections. This is assumed
to induce ecologic and taphonomic signal on the recorded averaged megainvertebrates spectra. See
text. Pie-diagrams as in Figs. 3 and 8.
depositional sequences and parasequences would correspond with ecostratigraphic

trends of different orders.
The application of our approach requires adequate ecostratigraphic sampling
programs in which biotic and abiotic signals must be taken into account. Tapho-
nomic and ichnologic analyses playa significant role complementing the faunal
analysis.
The case studies examples show that the ecostratigraphic evolution in a given
basin, whether eustatically or tectono-eustatically induced, can be favourably ap-
proached by the combined application of ecostratigraphy and sequence stratig-
raphy. This is especially true when one is dealing with relatively monotonous
lithofacies.
Acknowledgements
This paper benefited from projects PB85-0406 (CAICYT), PB91-0733 (DGI-

CYT) and the EMMI Group (Junta de Andaluda). We thank the participants
of the SEPM'93 Meeting "Stratigraphic Record of Global Changes" Penn State
University (Pennsylvania), especially B.U. Haq (Washington, USA), E. Heydari
(Louisiana State University, USA), E.G. Kauffman (University of Colorado, USA),
M.R. Rampino (New York University, USA), and A. Strasser (Geol. Inst. Fri-
bourg, Switzerland) for comments benefiting this contribution. We acknowledge
A.J. Boucot (Oregon State University, USA) for improving the text and sugges-
tions. We acknowledge stimulating observations and comments by two anonymous
reviewers.
References
ATROPS, E and FERRY, S., 1989, Sequence stratigraphy and changes in the ammonite
fauna (Upper Jurassic, S-E France): 2ieme Congres Fran6aise de Sedimentologie, Meso-
zoic Eustasy Record on Western Tethyan margins, Lyon, p. 7-9.
AURELL, M., 1990, EI Junisico Superior de la Cordillera Iberica central (Provincias
de Zaragoza y Teruel). AnOlisis de cuenca: Tesis Doctoral, Universidad de Zaragoza,
Spain, 389 p.
BREYER, R., 1992, Rudists and sequences in marginal environments: an example from
the Coniacian of SE-France ("Massif de la Ceze"; Gard): Sequence Stratigraphy of
European Basins, CNRS-IFP, Dijon, Abstract Volume, p. 268-269.
BRINKHUIS, H., 1992, Dinoflagellate and sequence stratigraphy of the type-Priabonian;
A Key to its chronostratigraphical position: Sequence Stratigraphy of European Basins,
CNRS-IFP, Dijon, Abstract Volume, p. 270.
CANEROT, J., 1985, La Chaine Alpine des Iberides (Espagne): Un example d'evolution
sedimentaire et tectonique intraplaque: Archives des Sciences, Societe de physique et
d'histoire naturelle de Genove, v. 38, I, p. 37-62.
CLAVEL, B., BUSNARDO, R., CHAROLLAIS, 1. and SCHROEDER, R., 1992, Se-
quence stratigraphy: a paleontological approach for the lower Cretaceous: Sequence
Stratigraphy of European Basins, CNRS-IFP, Dijon, Abstract Volume, p. 478-479.
COMAS, M., OLORIZ, E and TAVERA, J.M., 1981, The red nodular limestones (Am-
monitico Rosso) and associated facies: A key for setting slopes or swell areas in the
Subbetic upper Jurassic submarine topography, in Farinacci, A., and Elmi, S., eds.,
Rosso Ammonitico Symposium Proceedings, Roma, p. 113-136.
COMAS, M., GARCIA-DUENAS, V. AND BALANYA, J.C., 1988, El dominio sudiberico
como margen continental mesozoico: Symposium on the Geology of the Pyrenees and
Betics, Barcelona, Abstract Volume, p. 11.
DONOVAN, D.T., 1985, Ammonite shelf form and transgression in the British Lower
Jurassic, in Bayer, U., and Seilacher, A., cds, Sedimentary and Evolutionary Cycles.
Lecture Notes in Earth Sciences: Springer-Verlag, Berlin, p. 48-58.
EL HARIRI, K., SADKI, D., DOMMERGUES, J.L. and MOUTERDE, R, 1992, Niveaux
condenses, fluctuations eustatiques? Implications sur les faunes d' ammonites juras-
siques dans Ie Haut-Atlas (Maroc): Comptes Rendus de I' Academie des Sciences, Paris,
v. 314, II, p. 507-510.
FURSICH, ET., 1977, Corallian (Upper Jurassic) marine benthic associations from Eng-
land and Normandy: Palaeontology, v. 20, 2, p. 337-385.
82 R 016riz et al.
FURSICH, ET. and SYKES, R.M., 1977, Paleobiogeography of the European Bo-
real Realm during Oxfordian (Upper Jurassic) times: a quantitative approach: Neues
Jahrbuch fUr Geologie und PalQontologie, Abhandlungen, v. 155,2, p. 137-161.
FURSICH, ET. and WERNER, w., 1986, Benthic associations and their environmental
significance in the Luisianan Basin (Upper Jurassic, Portugal): Neues Jahrbuch fUr
Geologie und PalQontologie, Abhandlungen, v. 172,3, p. 271-329.
GALBRUM, B., BASSOULLET, J.-P., BAUDIN, E, DEPECHE, E, LACHKAR, G.,
RENARD, M., GAB ILLY, J., HANTZPERGUE, P., MANIVIT, H. and RUGET, Ch.,
1991, Stratigraphie integn!e du Toarcien stratotype (Coupes de Thovars et Airvault,
Deux-Sevres, France): 3rd International Symposium on Jurassic Stratigraphy, Poitiers,
Abstract volume, p. 50.
GARCIA, J.P. and LAURIN, B., 1992, Brachiopods associations: a biostratigraphical tool
for identifying parasequences and systems tracts in the middle Jurassic of the Paris
Basin: Sequence Stratigraphy of European Basins, CNRS-IFP, Dijon, Abstract Volume,
p.272-273.
GARCIA-DUENAS, V. AND COMAS, M., 1983, Paleogeografia mesozoica de las Zonas
Externas Beticas como borde de la Placa Iberica entre el Athintico y la Mesogea, in
Obrador, A., ed., X Congreso Nacional de Sedimentologia, Menorca, p. 5.26-5.28.
GYGI, R.A., 1986, Eustatic sea-level changes of the Oxfordian (Late Jurassic) and their
effect documented in sediments and fossil assemblages of an epicontinental sea: Eclogae
Geologicae Helvetiae, v. 79, 2, p. 455-491.
HAQ, B.U., HARDENBOL, J. and VAIL, P.R., 1987, Chronology of fluctuating sea levels
since the Triassic: Science, v. 235, p. 1156-1167.
HAQ, B.U., HARDENBOL, J. and VAIL, P.R., 1988, Mesozoic and Cenozoic Chronos-
tratigraphy and cycles of sea-level changes, in Wilgus e.K., Hastings, B.S., Kendall,
e.G.SLe., Posamentier, H.W., Ross, e.A., and Van Wagoner, J.C., eds., Sea-level
changes: An integrated approach: Society of Economic Paleontologists and Miner-
alogists, Special Publication, no. 42, p. 71-108.
HALLAM, A., 1969, Faunal realms and facies in the Jurassic: Paleontology, v. 12, p. 1-18.
HALLAM, A., 1975, Jurassic environments. Cambridge University Press, Cambridge, 269
p.
HALLAM, A., 1985, Jurassic Molluscan migration and Evolution in Relation to Sea-level
Changes, in Bayer, U., and Seilacher, A., eds, Sedimentary and Evolutionary Cycles.
Lecture Notes in Earth Sciences: Springer-Verlag, Berlin, p. 4-6.
HANTZPERGUE, P., 1991, Biogeographie des ammonites et variations du niveau marin:
apport de la stratigraphie scquentielle dans L' analyse des peuplements du Kimmeridgien
Nord-Aquitaine: Gcobios, v. 24, 1, p. 59-64.
HIRSCH, E., 1993, Triassic time, Eustasy and faunal cycles: Society of Economic Pale-
ontologists and Mineralogists 1993 Meeting, Stratigraphic Record of Global Changes,
Penn State University, Pennsylvania, p. 53.
HOEDEMAEKER, Ph.J., 1992, Lower Cretaceous ammonite biostratigraphy and Se-
quence Stratigraphy: Sequence Stratigraphy of European Basins, CNRS-IFP, Dijon,
Abstract Volume, p. 278-279.
HOLLAND, S.M., 1993, Stochastic modelling of the stratigraphic distribution of fossils in
depositional sequences: Society of Economic Paleontologists and Mineralogists 1993
Meeting, Stratigraphic Record of Global Changes, Penn State University, Pennsylvania,
p.42.
JERVEY, M.T., 1988, Quantitative geological modelling of siliciclastic rock sequences and
their seismic expression, in Wilgus e.K., Hastings, B.S., Kendall, C.G.SLC., Posamen-
tier, H.W., Ross, e.A., and Van Wagoner, J.e., eds., Sea-level changes: An integrated
approach: Society of Economic Paleontologists and Mineralogists, Special Publication,

no. 42, p. 47-71.
KAUFFMAN, E.G., 1986, High resolution event stratigraphy: regional and global Creta-
ceous Bio-events, in Walliser, O.H., ed., Global Bio-events: Springer-Verlag, Berlin, p.
279-335.
KAUFFMAN, E.G., 1988, Concepts and methods of high resolution event stratigraphy:
Annual Review Earth Planetary Science Letters, v. 16, p. 605-654.
KAUFFMAN, E.G., ELDER, w.P. and SAGEMAN, B.B., 1991, High-resolution corre-
lation: a new tool in chronostratigraphy, in Einsele, G., Ricken, W., and Seilacher, A.,
eds., Cycles and Events in stratigraphy: Springer-Verlag, Berlin, p. 795-819.
KOPAEVICH, L.F., 1992, Sequence Stratigraphy and Biological Signals in Upper Creta-
ceous of Mangyshlak: Sequence Stratigraphy of European Basins, CNRS-IFP, Dijon,
Abstract olume, p. 282-283.
MANU PELLA, G., MARQUES, B. and ROCHA, R., 1988, Evolution tectono-
sedimentaire du Bassin de l' Algarve pendant Ie Jurassique: 2nd International Sym-
posium on Jurassic Stratigraphy, Lisboa, p. 1031-1046.
MARQUES, B., OLORIZ, F. and RODRIGUEZ-TOVAR, FJ., 1991, Interactions between
tectonics and eustasy during the Upper Jurassic and lowermost Cretaceous. Examples
from the South of Iberia: Bulletin de la Societe geologique de France, v. 162, 6, p.
1109-1124.
McGHEE, Jr.G.R., BAYER, U. and SEILACHER, A., 1991, Biological and Responses to
Transgressive-Regressive Cycles, in Einsele, G., Ricken, w., and Seilacher, A., eds.,
Cycles and Events in stratigraphy: Springer-Verlag, Berlin, p. 698-709.
McKINNEY, M.C., 1986, Estimating Volumetric Fossil Abundance from Cross-Sections:
A Stereological Approach: Palaios, v. I, I, p. 79-84.
OLORIZ, F., MARQUES, B. and MOLlNER, L., 1988, The platform effect: an example
from Iberian shelf areas in the lowermost Kimmeridgian: 2nd International Symposium
on Jurassic Stratigraphy, Lisboa, p. 543-562.
OLORIZ, F., MARQUES, B. and RODRIGUEZ-TOVAR, F., 1991, Eustatism and faunal
associations. Examples from the Southern Iberian Margin during the Late Jurassic
(Oxfordian-Kimmeridgian): Eclogae geologicae Helvetiae, v. 84, 1, p. 83-106.
OLORIZ, F., RODRIGUEZ-TOVAR, F., MARQUES, B. and CARACUEL, I.E., 1993a,
Ecostratigraphy and sequence stratigraphy in the high frequency sea-level fluctuations:
Example from Jurassic macroinvertebrates assemblages: Paleogeography, Paleoclima-
tology, Paleoecology, v. 101, p. 131-145.
OLORIZ, F., RODRIGUEZ-TOVAR, F., 1993, Icnitas del Kimmeridgiense inferior y medio
en el perfil de Puerto Lorente (Sierra de Cazorla, Prebetico Externo). Contribuci6n a la
interpretaci6n de la evoluci6n ecol6gica del substrato: Revista de la Sociedad Geol6gica
de Espana, v. 7, 112.
OLORIZ, F., RODRIGUEZ-TOVAR, F. and CARACUEL, J.E., 1994a, Faunal Assem-
blages, Ecostratigraphy and High Resolution Sequence Stratigraphy: High Resolution
Sequence Stratigraphy: Innovations and Applications, University of Liverpool, Liver-
pool, Abstract Volume, p. 198-203.
PATZKOWSKY, M.E. and HOLLAND, S.M., 1993, Sea-level changes and faunal turnover
in the Paleozoic: Society of Economic Paleontologists and Mineralogists 1993 Meeting,
Stratigraphic Record of Global Changes, Penn State University, Pennsylvania, p. 52.
PEMBERTON, S.G., ed., 1992, Application of ichnology to Petroleum Explaration: A core
Workshop: Society of Economic Paleontologists Mineralogists Core Workshop, no. 17,
429 p.
84 F. 016riz et al.
PONSOT-JACQUIN, C., 1992, Sequence Stratigraphy in the Jurassic of the Paris Basin:
Sequence Stratigraphy of European Basins, CNRS-IFP, Dijon, Abstract Volume, p.
72-73.
PONSOT, C.M. and VAIL, P.R., 1991. Sequence stratigraphy of the Jurassic: New data from
the Paris-London Basin: 6th Meeting of European Union of Geosciences, Strasbourg,
p.308.
POSAMENTIER, H.W., JERVEY, M.T. and VAIL, P.R., 1988, Eustatic controls on clastic
deposition I, in Wilgus c.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier, H.W.,
Ross, c.A., and Van Wagoner, J.c., eds., Sea-level changes: An integrated approach:
Society of Economic Paleontologists and Mineralogists, Special Publication, no. 42, p.
109-125.
RASPLUS, L. and FOURCADE, E. (Coords.), 1987, Stratigraphie integree du sillon
Citrabetique (Sierra de Fontcalcn, Province d' Alicante, Espagne): Geobios, v. 20, p.
337-387.
REBOULET, St., ATROPS, E, FERRY, S. AND SCHAAF, 1992, Renouvellement des
ammonites en fossa vocontienne a la limite Valanginien Hauterivien: Geobios, v. 25,4,
p.469-476.
REY, J., 1992, Sequence stratigraphy and biological signal: Sequence Stratigraphy of
European Basins, CNRS-IFP, Dijon, Abstract Volume, p. 78-79.
RODRIGUEZ-TOVAR, E, 1993, Evoluci6n sedimentaria y ecostratigrOfica en platafor-
mas epicontinentales del margen sudiberico durante el Kimmeridgiense inferior: Tesis
Doctoral, Universidad de Granada, Spain, 374 p.
ROSS, J.R.P. and ROSS, C.A., 1986, How does Sea Level Rise and Fall?: Palaios, News
& Comments, v. I, 3, p. 341-342.
SAVRDA, C.E., 1991, Ichnology in Sequence Stratigraphic studies: An example from the
Lower Paleocene of Alabama: Palaios, v. 6, p. 39-53.
SEYFRIED, H., 1980, Uber die Bildungsbereiche mediterraner Jurasedimente am Beispiel
der Betischen Kordillera (SUdost-Spanien): Geologische Rundschau, v. 69, p. 149-176.
STEIN, J.A., GAMBER, J.H., KREBS, W.H. and LA COE, M.K., 1992, A composite
standard approach to biostratigraphic evolution of the North Sea Paleogene: Sequence
Stratigraphy of European Basins, CNRS-IFP, Dijon, Abstract Volume, p. 292-293.
SZTEJN, J., 1992, Stratigraphy of the Cretaceous deposits of the Polish Lowlands on the
base of the benthic "small" foraminifers except the Upper Albian: Sequence Stratigraphy
of European Basins, CNRS-IFP, Dijon, Abstract Volume, p. 294.
TAYLOR, A.M., 1992, The Application of Trace Fossil fabric (ichnofabric) analysis to Se-
quence Stratigraphic interpretation: Sequence Stratigraphy of European Basins, CNRS-
IFP, Dijon, Abstract Volume, p. 296-297.
TINTANT, H., 1980, Les epizoaires intrathalames, indice d'allochtonie des coquilles de
Cephalopodes: 8eme Reunion Annuel des Sciences de la Terre, Marseille, p. 384.
TINTANT, H., 1983, Autochtonie ou allochtonie chez les cephalopodes: Memoires
Geologiques de I'Universite de Dijon, v. 7, p. 257-268.
TIPPER, J.c., 1976, A method for the quantitative estimation of the faunal content of well
cemented fossiliferous rocks: Journal of Paleontology, v. 50, p. 175-179.
VAIL, P.R., COLIN, J.P., CHENE Jr.R., KUCHLY, J., MEDIAVILLA, E and TRIFILIEFF,
v., 1987, La Stratigraphie scquentielle et son application aux correlations chronostrati-
graphiques dans Ie Jurassique du basin de Paris: Bulletin de la Societe geologique de
France, v. 8, 7, p. 1301-1321.
VERA, J.A., 1988, Evoluci6n de los sistemas de dep6sito en el margen Iberico de la
Cordillera Betica: Revista de la Sociedad Geol6gica de Espala, v. 1, 3-4, p. 373-391.
WILGUS, C.K., HSATING, B.S., KENDALL, Ch.G.St.C., POSAMENTIER, H.W.,
ROSS, c.A. and VAN WAGONER, J.c., eds., 1988, Sea-level changes: An integrated
approach: Society of Economic Paleontologists and Mineralogists, Special Publication,

no. 42, 407 p.
WILPSHAAR, M., 1992, Dinoflagellate cyst associations as a reflection of early Cretaceous
sea-level changes in the Vocontian Basin: Sequence Stratigraphy of European Basins,
CNRS-IFP, Dijon, Abstract Volume, p. 298-299.
ZIEGLER, P.A., 1988, Post -Hercynian Plate Reorganization in the Tethys and Arctic-
North Atlantic Domains, in Manspeizer, W., ed., Triassic-Jurassic Rifting: Elsevier,
Amsterdam, p. 711-749.
CHAPTER 4
Sea-Level Changes, Carbonate

Production and Platform Architecture:
The Llucmajor Platform, Mallorca,
Spain
Luis Pomar and William C. Ward
ABSTRACT: The stratal architecture of the Upper Miocene coral-reef platform of south-
western Mallorca, Spain, is controlled by high-frequency changes in accommodation and
sediment supply (carbonate production), in the absence of significant compaction and
subsidence during progradation. In this example, carbonate production and accommoda-
tion changes are not independent factors and both are, in tum, controlled by the changes
of sea-level and morphology of the depositional profile of the basin floor.
The basic unit of accretion is the sigmoid which stacks in ever larger accretional units:
sets, cosets, and mega sets of sigmoids. All of these accretional units have the same
characteristics in terms of stratal geometries, facies architecture and bounding surfaces,
and may be viewed as depositional sequences reflecting different hierarchical orders
of sea-level fluctuations. The stratal and facies architecture in sigmoids, sets, cosets,
and megasets, reflect higher production of carbonate during sea-level rises and lower
production during sea-level stillstands and sea-level falls. Their stacking patterns allow
definition of four reef-platform systems tracts: low-stillstand, aggrading, high-stillstand
and offtapping.
On larger scale, progradation of carbonate reef complex is extensive (up to 20 km) toward
the south, where the basin was shallow, but progradation is much less (less than 2 km)
toward the west, along the margin of the relatively deeper Palma Basin. This results
from the steepness and overall morphology of the depositional profile within the context
of fluctuating sea level that controls carbonate production. Progradation of the reefal
systems is more significant during sea-level falls on a gentle depositional profile. The
subsequent sea-level rise creates a wide lagoon which enhances carbonate production
and downslope shedding of sediment. A steeper topographic gradient allows only minor
reefprogradation during sea-level falls and, subsequently, a small lagoonal area is created
during flooding of the platform, leading to proportionally small carbonate production and
downslope shedding. This example illustrates how a reefal carbonate platform responds
to high-frequency sea-level changes and how it differs from siliciclastic systems.
87
B. U. Haq (ed.),
88 L. Pomar, w.e. Ward
1. Introduction
Although sequence stratigraphy has been demonstrated to be a successful tool

in sedimentary geology, providing the basis for correlation and prediction, some
aspects of its application to carbonates still remain controversial. Of particular
interest is the question of how changes in accommodation are expressed in the
stratal arhitecture of carbonate depositional sequences as compared to the basic
Exxon models for siliciclastic sequences. Another set of questions centers around
the relative importance of accommodation versus sediment supply in creating and
shaping carbonate depositional sequences.
To what extent can the cause-and-effect concepts of standard sequence stratig-
raphy be directly applied to the carbonate system? There are a number of major
differences between genetic factors that control the development of depositional
sequences in carbonates and clastics. (Sarg, 1988; Haq, 1991; Schlager, 1991,
1992, 1993; Handford and Loucks, 1993; Hunt and Tucker, 1993). Carbonates are
often biological systems that are built up in-situ. Carbonate production is closely
related to the sea level, changes in which affect both accommodation and produc-
tion. This contrasts markedly with clastics that are governed by lateral input of
physically eroded sediment from the continent into the basin. Facies architecture
in carbonates depositional sequences, therefore, is a sensitive reflection of relative
sea-level fluctuations.
The issue has become even more complicated by the recent recognition of
high-frequency depositional sequences from seismic and well-log data, as well
as outcrop studies. The different magnitude sequences are proving to be scale
and time independent (Mitchum and Van Wagoner, 1991; Posamentier and others,
1992; Pomar, 1991; Pomar and Ward, 1994; Sonnenfeld, 1993; Sonnenfeld and
Cross, 1993; Abbott and Carter, 1994). The absence of high-resolution chronos-
tratigraphic criteria, however, makes it difficult to ascertain the temporal scale of
the high-frequency depositional sequences, and consequently the use of cyclicity
as a correlation tool.
The Upper Miocene progradational coral-reef platform cropping out at the
southern sea-cliffs of the Spanish island of Mallorca provides new data which can
address some of the questions about carbonate sequence stratigraphy. This paper
will show how the LIucmajor Carbonate Platform responded to Late Miocene
changes in accommodation, to what extent the depositional sequences differ from
the standard models, and what general applications of carbonate sequence stratig-
raphy may be extracted from the Mallorca example.
2. The Llucmajor Platform
Mallorca Island (Fig. I) has a basin-range configuration, that resulted from Late
Miocene to Early Pleistocene extensional faulting. The mountain ranges are horsts
blocks of the Alpine foldbelt characterized by NE-trending thrust sheets, which
resulted from northwestward thrusting during Early-Middle Miocene (Fontbote
and others, 1983; Ramos-Guerrero and others, 1989). Upper Miocene, Pliocene
and Pleistocene deposits onlap the folded Mesozoic to Middle Miocene rocks,
The Llucmajor Platform, Mallorca, Spain 89
Mallorca
39 30' N
Cap Blanc
I Q~' Campos
50 km
~ Reef Complex platforms

,..".? ; ...'
i tJ #
_.... Reef platform margin
rI::IJ Highlands (Alpine thrust belts)
Fig. 1. Simplified geologic map of Mallorca showing paleogeography of Upper Miocene (Upper
Tortonian-Lower Messinian) reefal platforms. These platforms prograded around islands resulting
from Alpine (Middle Miocene) orogeny. Palma to Alcudia Basins resulted from Late Miocene
to Pleistocene faulting, whereas the Campos Basin down-dropped during early Pleistocene times.
(Modified from Pomar and Ward, 1994).
constructing near-horizontal platforms around the ranges (Santanyf, Alcudia and

Llucmajor Platforms) and filling down-dropped areas (Palma to Alcudia Basins).
The Campos Basin seems to have subsided only during Pleistocene times.
The "Reef Complex" is one of the three Upper Miocene units and crops out
along 20 km of the high sea cliffs of southern Mallorca in the Cap Blanc area (Bar6n
and Pomar, 1985; Pomar and others, 1985; Pomar, 1991). These sea cliffs display
in exquisite detail the facies architecture of part of the reef-rimmed carbonate
shelf (Pomar, 1991, 1993; Pomar and Ward, 1994) that prograded about 20 km
across the Llucmajor Platform. There is still a lack of refined chronostratigraphic
data on the Llucmajor Platform. Nevertheless, diverse regional considerations
allow attribution of this prograding platform to the Late Tortonian-Early Messinian
(Pomar and others, 1985; Pomar, 1991). This is consistent with Bizon and others
(1973) and Alvaro and others (1984), who attributed the platform to the N17 Blow
zone (Late Tortonian). Despite the chronostratigraphic uncertainty, Pomar (1991)
assigned the platform to a third-order cycle (Late Tortonian TB 3.2 cycle of Haq
and others, 1987).
Paleobalhymetry
~o ------4---------:-~_:__------ sea level - -

-10
-20 L agoon Inner
-30
coral framework skeletal

with skeletal grainstone! mudstone!
fine grained coarse skeletal grainstone! packestone wackestone
packestones packestone with corals
poorly bedded sigmoidal horizontal beds
(bioturbated) bedding bounded by erosion surfaces
Fig. 2. Facies model for the reef-rimmed Llucmajor Platform. Zonation of the coral-colony mor-
phologies in the reef-core lithofacies with respect to the paleobathymetry is an important tool for
high-resolution sequence stratigraphic analysis.
The Reef Complex unit, conformably overlies the calcisiltite with Heteroste-
gina unit, a unit assigned to the N16 Blow zone (Early Tortonian). On the Palma
Basin, the Reef Complex is overlain by grey marls, mudstone, and finely laminated
dolostones, attributed to the Messinian "Terminal Complex" (Pomar, 1991), and
by the Pliocene calcisiltites with Ammussium. Pleistocene eolianites and red soils,
composed of sandy skeletal grainstones, onlap the Pliocene deposits on the Palma
Basin and the Reef Complex on the Llucmajor Platform.
The Llucmajor Platform has been buried only a few tens of meters at most, and
large primary and secondary pores remain intact. The platform is mainly flat-lying
with only slight tilting associated with basin subsidence and strike-slip faulting
during the Pliocene and early Pleistocene. The possible influence of gentle uplift
on the depositional patterns can not be ruled out, but the lack of terrigenous influx
onto this flat-lying platform suggests that this was a time of tectonic stability on
Mallorca during the deposition of this platform. Loading subsidence probably also
was an insignificant factor because of the low density of these rocks (high porosity)
and the short time (on the order of 2 my) in which this sheet-like unit prograded
across the shallow platform (Pomar and Ward, 1994). Subsidence on the basins as
well as strike-slip faulting affect the Reef Complex unit and are related to Pliocene
and Lower Pleistocene tectonic flexure.
2.1. REEF PLATFORM LITHOFACIES
Complete exposure of the Upper Miocene Llucmajor progradational reef platform

in vertical sea-cliffs at Cap Blanc allows the establishment of a depositional model
(Fig. 2) and a high-resolution architectural facies model (Pomar, 1991, 1993; Po mar
and Ward, 1994). Four main lithofacies can be distinguished on the Llucmajor
Platform:
1) Lagoon (back-reef) lithofacies are characterized by horizontal beds, bounded
by erosional surfaces. Inner-lagoon deposits are composed of mudstones and
wackestones with vertical root molds, fecal pellets, benthic foraminifera, bi-
valves, and locally thin layers of monospecific gastropods and discontinuous
caliche crusts. Outer-lagoon deposits are mainly composed of coarse skeletal
grainstone/packstone, with abundant mollusk debris, rhodoliths, echinoids, benthic
foraminifera, and coral fragments. Locally, coral patch reefs are also present.
2) Reef-core lithofacies have a characteristic sigmoidal bedding and are com-
posed of skeletal grainstone/packstone within coral framework. Coral-colony mor-
phologies in the framework range from dish-coral in the lower part, to branching-
corals in the middle, and to massive-corals in the upper part of the reef-core. This
coral-morphology zonation with respect to the paleobathymetry is an important
tool for high-resolution sequence stratigraphic analysis.
3) Clinobeds of reef-slope (fore-reef) lithofacies range from coarse skeletal
packstones with abundant mollusks, red-algae fragments, rhodoliths, coral debris
and Halimeda in proximal settings to fine-grained packstones on the distal-slope
to open-shelf settings. These clinobeds may be hundreds of meters long, dipping
seaward up to 30. Distal-reef-slope deposits are extensively burrowed and gently
inclined (less than 10).
4) Open-shelf( shallow basin) lithofacies are composed of poorly bedded (bio-
turbated) fine-grained packstones with planktonic foraminifera; deeper-water oys-
ters and echinoids may also be present. These beds may be interbedded with
biostromes of coarse-grained red-algae packstone with densely-stacked rhodoliths
and laminar corals.
2.2. THE BASIC BUILDING BLOCK
The Mallorca outcrops have provided a high-resolution sequence stratigraphic

analysis, allowing us to determine the key to the architectural relationships and
relative sea-level fluctuations (Pomar, 1993; Pomar and Ward, 1994). The building
block of the L1ucmajor Platform is a sigmoidal depositional unit ("SIGMOID"). It is
composed (Fig. 3) of a horizontal lagoonal bed passing basinward to a sigmoidal
reef-core, then to a reef-slope clinoform and to a horizontal open-shelf bed. This
accretional unit is bounded by erosional surfaces in the landward upper part (lagoon
and reef-core lithofacies) and by their correlative conformities in the basinward
lower part (lower portion of reef-core, slope and open-shelf lithofacies). Some of
these basic accretional units are wedge-shaped as a result of non-depositional or
erosional truncation of the upper part of this accretional unit (lagoon and upper
portion of reef-core lithofacies), or both.
Completely preserved sigmoids often reveal the internal arrangement of the
distribution of lithofacies. In the upper portion of the reef-core facies, massive
(shallow-water) corals encrusting the lower erosional surface pass upward into
branching forms. In the lower portion, dish-shaped (deeper-water) coral colonies
sharply and conformably overlie the proximal reef-slope deposits of the previous
massive corals
SIGMOIDS Lagoon erosional
branching corals
surfaces
dish corals
WEDGE
Open shelf
Fig. 3. The "sigmoid" is the basic building block of the Llucmajor Platform. It is a sigmoidal
depositional unit, composed of a horizontal lagoonal bed passing basinward to a sigmoidal reef core,
then to a reef-slope cIinoform and a horizontal open-shelf bed. The sigmoid is bounded by erosional
surfaces which pass basinward into their correlative conformities. A wedge-shaped unit may result
from non-depositional and/or erosional truncation of its upper part. (Modified from Pomar, 1991;
Pomar and Ward, 1994).
unit, and pass upward into coarse-grained reef-slope deposits. The upper erosional
surface truncates the branching corals of the reef-core facies and correlates bas-
inward with the conformity. In some sigmoids, coral-morphology is characterized
by a shallowing-upwards zonation.
In the lagoonal facies thin laminites or gastropod-rich wackestone (restricted
facies) resting on the erosional surface are overlain by packstone, wackestone, and
grainstone with red algae, echinoids, mollusks, and benthic foraminifera (open la-
goon). In outer lagoonal facies, the basal laminites are overlain by both, coral-patch
reefs and coarse skeletal grainstone (interpatch sediments). The upper erosional
surface truncates everything, including the patch-reefs corals and the grainstone
sediments of the outer-lagoon facies. In the more distal reef-slope and open-shelf
settings, intense bioturbation can destroy the internal arrangement of the lithofacies
and obscure the conformable nature of its boundary.
The vertical sequences within the reef-core reflect aggradation during sea-level
rise (Fig. 4) , with the shallowing- or deepening-upward trends depending on the
ratios between sea-level rise and carbonate production/sedimentation rates (accom-
modation vs. production). In the lagoonal facies, basal laminites record the flooding
of the platform top, and the overlying coral patches record the submergence of the
platform to the optimum production conditions.
The bounding erosional surfaces are most likely related to falls in sea level.
In the reef-core facies, the degree of shift of the coral-morphology zones from
relatively deeper to relatively shallower facies across the boundaries reflects the
amount of sea-level fall. In the lagoonal facies, the erosional truncation on top of
Accretion events
Facies architecture - t i me - -
massive coral zone
SIGMOID branching coral zone Pin n i ng poi nts

dish coral zone
Accretion events
aggradation
during
ac ies ar~~::,~C:U:~___'_'.__~'~~:"
submergence
F
. ..~-+
~---:"~rMo-,.r-r.-:-:-o-~- - ~.
- t i me --
Fig. 4. Internal arrangement of lithofacies within sigmoids (A) and in the lagoonal facies (B)
are interpreted as resulting from the highest-frequency sea-level fluctuations. Paleobathymetric
coral-morphology zonation within the reef-core framework allows accurate determination of the
amplitude of the sea-level cycle (A). See text. The reef-crest curve (pomar. 1991) is defined by the
successive positions of the reef-crest. measured or inferred from the coral-morphology zonation.
the coral patches indicates that the upper shallowing-up part of the cycle is missing,
and the physical correlation of the erosional surface from the lagoon to the reef
core indicates that it also is due to fall of sea level. The boundaries between
consecutive sigmoids which does not show clear shift of the coral-morphology
zones, may also be explained as result of other processes like submarine erosion.
Lack of subaerial-exposure features at most of the sigmoid boundaries lead us to

interpret them as due to a sea-level fall which lowered wave base on a platfonn that
remained submerged or due to submarine erosion during the subsequent flooding.
2.3. LARGER-SCALE ACCRETIONAL UNITS
In the sea-cliff outcrops of the Cap Blanc area, the basic accretional units (sig-
moids) are stacked forming different magnitudes of larger-scale accretional units
(Pomar, 1991, 1993; Pomar and Ward, 1994). Sigmoids stack in "sets ofsigmoids",
and sigmoid sets stack in "cosets" (Fig. 5). Core data from water wells allows in-
terpretation of larger accretional units: the "megasets" ("sets of cosets" in fonner
papers), resulting from the stacking of the cosets of sigmoids.
All of these accretional units have similar characteristics in tenns of facies
distribution, boundaries and internal stacking of the higher-order units. Facies
distribution in a set or in a coset of sigmoids is (in a basinward direction): horizontal
lagoonal beds, sigmoid-bedded reef-core with wavy configuration (upward and
downward shifts of the general progradation), and gently inclined reef-slope to
open-shelf lithofacies. Boundaries are major erosional surfaces in the landward
part, often marked by caliche or microkarst, and their correlative confonnities in
the basin ward part.
The vertical distribution of coral morphologies within the reef facies, as well
as subaerial surfaces within the lagoonal facies, show that most depositional cycles
resulted from sea-level fluctuations rather than sedimentologic (autocyclic) influ-
ences. Thicknesses of the depositional cycles were essentially unaffected by either
mechanical or chemical compaction. For these reasons, the upward and downward
shifts of the reef-core facies and the vertical shifts of coral morphologies within the
reef-core facies make it possible to gauge the amplitudes of sea-level fluctuations.
Thus, the reef-crest curve (pomar, 1991, 1993) is defined by the successive posi-
tions of the reef-crest (measured or inferred from the coral-zonation). This curve
reflects the amount of relative sea-level fluctuation related to progradation, which
is a function of time.
The megasets show an amplitude of fluctuation on the order of 100s of meters
in 3 to 6 km of progradation, the co sets of sigmoids show an amplitude of 50-
70 meters in 1 to 3 km of reef progradation, and the sets of sigmoids show an
amplitude of 20-30 m in a reef progradation of 100s of meters.
Pomar (1993) showed that the approach developed in Mallorca can be applied
to prograding carbonates elsewhere and at different scales. Seismic examples
from the Straits of Andros in Bahamas (Eberli and Ginsburg, 1987, 1988, 1989)
and from the Bali-Flores Sea (Tyrrell and Davis, 1989) show seismic facies and
reflection patterns which can be compared to the facies architecture, bedding
geometries and scale of progradation of the Mallorca example (Pomar, 1993).
The remarkable similarities between these three examples, which occur in quite
different geographic settings, reinforces the general applicability of the Mallorca
approach for refined seismic stratigraphic analysis and for prediction of lithologies
and sequence-stratigraphic packages in prograding carbonate complexes.
lagoonal facies
reef-core facies __- - - - - - - - -
Blanc
~I "t'-~-~__~, ~..~'.'~~.'. ~.i"ii~i,i~lto!ce!ne~~~\"

Cap curve
100m
\ SETS OF SIGMOIDS
1Cap B - I a n c - - - - - - - - -
I Pleistocene
E
8
COSET OF SIGMOIDS
reef-crest MEGASETS
Fig. 5. Reef-platform facies and stratal architecture. The basic accretional units (sigmoids) are
stacked in different magnitudes of increasingly larger scale accretional units: sets, cosets and megasets
of sigmoids. All of these accretional units have similar characteristics in terms of facies distribution,
boundaries and internal stacking of the higher-order units and may be considered as depositional
sequences. The vertical distribution of coral morphologies within the reef-core facies, as well as
subaerial erosional surfaces within the lagoonal facies, show that most of these units resulted from
sea-level fluctuations. The reef-crest curve reflects the amount of relati ve sea-level fluctuation related
to progradation. which is a function of time. (Modified from Po mar and Ward, 1994).
2.4. LLUCMAJOR PLATFORM 3-D ARCHITECTURE
Recent integration of data from 70 water-well core logs, and a number of measured
sections on the sea cliffs allows us to establish the distribution patterns of the reef
complex and its paleogeography on the Llucmajor Platform, the Marratxf Platform
and the eastern part of the Palma Basin (see Fig. 1). It also allows us to construct
a three-dimensional model of the architecture of the Llucmajor Reefal Platform.
The diagnostic criteria for the interpretation of well data and for its correlation
across the Llucmajor Platform (Fig. 6) originate from the sea-cliffs architectural
model (see Fig. 5). These include interpreting aggradation, progradation, or offtap
patterns from the elevation of the facies tops, the reef-core thickness, the upward or
downward shifts of reef-cores, and when possible, the reef-core internal succession
and the thicknesses of the existing coral-morphology zones (see Fig. 2).
Reef-tract progradation lines through the platform (Fig. 7A) have been traced
by correlation of the well data. This correlation is based on two assumptions: 1)
from the margin toward the interior of the platform, the reef-tract lines can be
projected from the directions of the successive reef tracts cropping out on the sea
cliffs and 2) from the interior toward the exterior of the platform, the reef tracts par-
allel the basement contours (Fig. 7B). The reef-tract lines represent the successive
position of the reefs prograding over the Llucmajor Platform. This interpretation
visualizes important differences in the progradation rates, depending on the de-
positional gradients. Progradation is more important towards the south, where the
basin was shallower (Fig. 7C), and it is dramatically reduced toward the west, in
the margin of the relatively deeper Palma Basin (Fig. 70). These differences in the
amount of progradation as controlled by the platform configuration resulted in the
clockwise rotation of the successive reef tracts towards the Palma Basin (Fig. 7A).
3. Sequence Stratigraphic Implications
Sequence stratigraphic analysis carried out on the high-frequency accretional units

within the Upper Miocene reef complex of the Cap Blanc area shows several
differences in the stratigraphic architecture (pomar and Ward, 1994) from that
represented in the "standard" Exxon siliciclastic sequence-stratigraphic models
(Vail and others, 1977, 1991; Posamentier and others, 1988; Van Wagoner and
others, 1988). The Mallorca model, however, is compatible with recent variable
carbonate sequence stratigraphy models (Handford and Loucks, 1993; Hunt and
Tucker, 1993).
3.1. HIGH-FREQUENCY REEFAL PLATFORM DEPOSITIONAL SEQUENCES
The basic reefal accretional unit, the "sigmoid", is not a parasequence in the strict
sense, but has characteristics of a small depositional sequence. The arrangement of
facies within a sigmoid and across the erosional boundaries is interpreted to reflect
changes in accommodation that results from the higher-order sea-level fluctuations
and can be equated to the systems tracts.
~
(b
r
"-
c::
(')
S
c.
~
::E
t::>
~
w~
Shallow eorals preserved I Thin reef; only deep-w. coral. nun reel: deep w. corals Thiele reef
& OIIerla.n by outerlagoon overlain by thick euler lagoon overlain by inner lagoon ove~ain by inner lagoon ~
Very thin proximal slope "-
"-
ThICk distalslope/open sheH thin slope~hic:l< open sheH Thiele proximal slope o~~~n 1~:i~n~r.H o
(-$
TYPE wt::>
SEQTIONS
sb 8f. S . sb 0', $.
sb or. s. fine grained ~
AST sb or. s. fine gr. fine grained 1i ne gralfl8d AST t::>
AST AST packlmd
sb
U er.s.
pack/md
sb aI, s.
packlmd
sb
packlmd
.r, s. coarse 10 fine
sb If. S.
fine grained
s
coarse graIned AST grained pacl\Imd AST packlmd
coarse 9,arns.tone sb sb tH'. s,
AST wnh oorals AST or. s.
.==rmal~~ dish eoral.
& molluscs HST dis massIve
sb 81.5. corals
dish corals
sb [J Of. s. massive OST dis (clinobeds) massl'WI8
mullusc AST "",01.
LST bfatlChin,9:
dish cor $ (mls) rich layel, coat'$G grai ned
dis AST packas'lona brancl'ling corals
dish corals
fine grained (,hodoIiths)
AST fine graIned ... r. algae & molluscs
p.ac:kestone pact< tono dis
AST 1' 1
.. CO
CO AST fine grained
fin. grained packeslona
flna graln&d
...."1
AST packaslona
fine grained pact<estcne
AST .':':' AST
pact<e.tone
CO
~M CO
fine grained
CS ::: co fine grained AST
AST ...... AST AST pact<e.ton. 1-: -:'1 pact< ton.
Fig. 6. The sea-cliff architectural model provides the diagnostic criteria for the interpretation of well data and for its
correlation across the L1ucmajor Platform. Aggradation, progradation, or offlap patterns are interpreted from the elevation of
the facies tops, the reef-core thickness, the upward or downward shifts of reef-cores, and the reef-core internal succession and \0
-...l
the thicknesses of the existing coral-morphology zones.
...
A
DISTRIBUTION OF THE
Marratxi REEF COMPLEX
platform'OO
UIIk~
OOJ ? -
4
A
WIIIt..-us
mtuurtd MCtlone
.-..flrads
OOl~~UUJ
00 T"
. , No"
rrj(ll~
H ghlands
Palma <I-,,"1'l.
basin rOil-- ?
T"'q. .f./-N o 13
-<>-'JJI:JJ ""' ...

MooI!.lI ll1i1
... .to
Fig. 7. Three-dimensional platform architecture interpreted from integration of well data and
measured sections on the sea cliffs. A) Lines of reef-tract progradation and location of cross-sections.
B) Structure map on top of sub-"ReefComplex" basement rock. C) and D) (see Fig. 7 continued) Cross
sections showing stratal architecture across the L1ucmajor Platform. Section A-A' is in direction of
maximum progradation and section B-B' is in direction of minor progradation on flank of Palma
Basin. The basinal deposits of the Reef Complex on the Palma Basin, are overlain by the Messinian
("Terminal Complex") grey marls and by the Pliocene calcisiltites with Ammussium. Pleistocene
eolianites and red soils, composed of skeletal grainstones, onlap the Pliocene deposits in the Palma
Basin and the Reef Complex on the L1ucmajor Platform.
C C'
200
100
mO
100
200
8-37- p,oJtc:ttd ",.11.,
D D'
200
100
mO
100
'"
B
.,,,
110... . "'--t---., . -
...
Fig. 7. COlltillued.
BUILDING
BLOCK
-,
AGGRADING
SYSTEMS TRACT
HIGH STILLSTAND
SYSTEMS TRACT
OFFLAPPING
SYSTEMS TRACT
PROGRADING
PLATFORM
sb: sequence boundary

cl: condensed section
ts: transgressive surlace
mls : maximum flooding surface
dis: downl.p surface
Fig. 8. High-frequency reefal platform depositional sequences and their stacking within the systems
tracts. (Modified from Pomar and Ward, 1994). The entire prograding platform is built up mainly by
stacking of aggrading systems tracts which formed during sea-level rises at all orders of magnitude.
The high-stillstand, offtapping, and low-stillstand systems tracts correspond to the distally condensed
intervals of the open-shelf facies and to the erosion surfaces within the lagoon facies. The erosional
surface and the downlap surface basinward are the only two relevant surfaces bounding systems tracts,
both of which merge into the distally condensed interval. The sequence boundary is an erosional
surface landward, whereas basinward it is represented by the top of the distally condensed interval.
Larger-scale accretional units (sets, cosets and megasets of sigmoids) resulting

from hierarchical stacking of sigmoids also show characteristics of depositional
sequences but, with a number of differences with respect to the standard sequence
stratigraphic model (Fig. 8). All of them are I) sigmoidal in shape; 2) bounded
landward by erosional surfaces which pass basin ward to correlative conformities;
3) and composed of an inner belt of horizontal lagoonal beds, a middle belt of
sigmoid-shaped reef-core lithofacies, and an outer belt of reef-slope and open-
shelf lithofacies. Changes in stacking patterns allow (pomar and Ward, 1994)
definition of four systems tracts in all hierarchical units above the sigmoids: "low
stillstand", "aggrading", "high stillstand", and "offtapping", and the vertical shifts
of the reef-core facies within these units allow the establishment of an accurate
relationship between systems tracts and each part of the sea-level cycle.
1) The "low stillstand" systems tract, which forms during the initial sea-level
rise after the lowest point of the sea-level cycle, is composed of a relatively thin
prograding reef-core facies and poorly developed or absent lagoonal beds. The
forereef-slope and open-shelf facies thin basinwards.
2) The "aggrading" systems tract is characterized by aggradation in all de-
positional systems (from the lagoon to the open shelf) and is formed during the
rise of sea level. It is volumetrically the most important. Lagoonal facies overlies
the LST and onlaps onto the erosion surface. Thick reef-core facies shows mainly
aggradation without backstepping. The fore reef-slope and the open-shelf facies
are also thick and aggradational. This AST differs from transgressive systems tract
by the absence of a condensed section above it and the absence of backstepping;
only the lagoonal facies shows landward onlapping.
3) The "high stillstand" systems tract, which forms during the high part of
sea-level cycle, is composed of thin progradational reef-core facies, forereef-slope
facies wedging out basinward, and volumetrically condensed open-shelf facies.
Commonly, lagoonal beds are absent (if deposited, they may be eroded during the
subsequent sea-level fall).
4) The "offlapping" systems tract, which forms during the falling sea level,
is composed of thin prograding and downstepping reef-core facies, which down-
laps on to the open-shelf facies (of the previous systems tract), without significant
forereef-slope facies. There is no lagoonal facies, and the open-shelf facies are vol-
umetrically condensed. This OST was originally described as "offlapping reefs"
(pomar, 1991) by using the Swain's (1949) definition of offlap (in Bates and Jack-
son, 1987) as the progressive offshore regression of the updip terminations of the
sedimentary units within a conformable sequence of rocks, in which each succes-
sively younger unit leaves exposed a portion of the older unit on which it lies. This
OST differs from the standard models, which do not envision deposition during
sea-level fall. Recently, however, similar units have been designated "stranded
parasequences" on the "forced regressive wedge systems tract" (Hunt and Tucker,
1992), "forced regression" deposits of a lowstand prograding wedge (Posamentier
and others, 1992), or "stranded parasequences" (Van Wagoner and others, 1990).
This OST is bounded at the top by the erosional surface and at its base by the
downlap surface and correlates basin ward with an interval in which sediments are
volumetrically condensed.
The Mallorca sequences appear to be simpler than the Exxon-type depositional

sequences (Jervey, 1988, Posamentier and others, 1988), because two systems
tracts are missing. Lowstand fans are more or less replaced by tracts of down-
stepping reefs (OST). In addition, the backstepping transgressive systems tract is
absent because of high carbonate production on the margin of Llucmajor Platform
during sea-level rise. The platform evolved from lowstand progradational into
aggradational geometries as the sea level rose.
3.2. SYSTEMS TRACT BOUNDARIES
Pomar and Ward (1994) recognized only two relevant surfaces bounding systems
tracts: the erosional surface and the downlap surface basin ward (Fig. 8), both of
which merge into the distally condensed interval. These two surfaces were formed
at the same time and represent the period of sea-level fall. The transgressive surface
and the maximum-flooding surface are not represented by significant changes in
lithology, but they may be identified from the systems-tract stacking patterns.
The reef-slope and open-shelf as well as the reef-core to lagoon facies aggraded
during sea-level rise without backstepping. There is, therefore, no deepening-
upward sequence followed by a shallowing-upward sequence that would allow the
characterization of the maximum flooding surface. The MFS merges basin ward
with the bottom of the distally condensed interval, and the TS merges with the
top of this condensed interval. The TS and MFS merge with the erosional surface
landward. The sequence boundary, which represents the lowest point of sea level
(see Kolla and others, 1995; and Hunt and Tucker, 1995; for more discussion),
is the erosion surface landward and the top of the distally condensed interval
basinward. Without high-resolution data, only two accretional packages may be
identified: an aggradational package related to sea-level rises and an offlapping
(progradational) package related to the stillstands (highs and lows) and falls of sea
level. These two surfaces (erosion surface and downlap surface) are in a "slant-tau
configuration" that can be recognized in seismic lines (Po mar, 1993).
At the outcrop scale, there are no lithologic criteria for defining the boundaries
of the systems tracts nor the level of hierarchy of the discontinuities and erosion
surfaces. Only from stacking patterns (Fig. 9) in the dip direction can the sequence-
stratigraphic framework be adequately defined. The entire prograding platform is
built up mainly by stacking of aggrading systems tracts which formed during
sea-level rises of various orders of magnitude. The high-stillstand, offlapping, and
low-stillstand systems tracts are mainly represented on the reef-core facies, and
correspond to the distally condensed intervals of the open-shelf facies and to the
erosional surfaces of the lagoonal facies.
3.3. ACCOMMODATION VS. CARBONATE PRODUCTION
One of the fundamental concepts derived from the architectural model ofthe Lluc-
major progradational reef platform is the relationship between sea-level fluctuation
and carbonate production. Sea-level change not only determines the accommoda-
Cross-section
30
20
10
grainstone (lagoon)
~ .ha~owwater corals (massive)
'II/ Intermedlatewater ooraJ. (branching)
::::; deep-waler co<a1s (dish)
reefcrest Pinning points

cor,l-morphology;zon
30
3
20
10
Sea-level cycles
7th order 7th, 6th & 5th order sb

30 cycle
20
10 6th order sb
5th order ts
Om 1 - - - - - - - -
Interpretation
30 6th order sb
20 ~~$~~~~~~~:::=~~~5th order ts
10
7th, 6th & 5th order sb
Fig.9. Magnitude and hierarchy of sea-level cycles and sequence stratigraphic framework interpreted
from the vertical shifts of coral morphologies within and accross sigmoids. The low stillstand of a
coset of sigmoids (5th-order sequence) is composed of two sets of sigmoids (6th-order sequences),
which may be composed of several sigmoids (7th-order sequences). The systems tracts boundaries are
not represented by characteristic changes in lithology, but they may be identified by the sigmoidal
stacking patterns. A single surface may represent the boundary of distinct systems tracts in the
different order of units, depending on its hierarchical stacking.
104 L. PomM, w.e. Ward
tion space, it also determines the efficiency of the carbonate factory, and thus, the
amount of carbonate production and sedimentation (Fig. 8).
Major amounts of carbonate sediment are produced when the shallow platform
is flooded during the rise of sea level (Fig. 8). Intermittent sea-level rises following
brief falls cause re-establishment of carbonate production in lagoons that have been
filled by sedimentation during slowing or cessation of sea-level rise. The rate of
carbonate production keeps up with the rate of creation of accommodation space,
and, consequently, instead of transgressive retrogradation, aggradation takes place
in all depositional systems (lagoon, reef, and slope/open-shelf).
During sea-level stillstand (low- or high-stand), the shallow platform is oblit-
erated or filled, causing reduction in the amount of carbonate sediment production
(Fig. 8). The lack of increase in accommodation space, in the absence of significant
subsidence, results in progradation of the reef system onto thin slope deposits.
During sea-level fall, both the absence of productive shallow platforms and
the loss of accommodation space, results in the offlapping of the reef system onto
the open shelf without significant reef-slope deposits (Fig. 8) and with erosional
truncation on the upper part of the previous high-stand reef and lagoonal deposits.
Lagoonal deposits, as well as open-shelf deposits, correspond to stacked aggrada-
tional systems tracts (Fig. 8), and only from the progradational patterns of the reef
may the platform architecture be interpreted.
Another important characteristic of the stratal architecture is the relationship
between accommodation space and the length and angle of the slope. In the
larger accretional units (megasets), the low stillstand units display short and gently
inclined clinobeds. By contrast, the c1inobeds of aggradational and high-stillstand
in the megasets are hundreds of meters long and steep (with up to 30 slopes).
Large slope beds dipping up to 35-45 have been described from Quaternary
foreslopes as well as in ancient examples (see Kenter, 1990; Grammer, 1991; for
more discussion).
This relationship between carbonate production and relative sea-level change is
the key to the reefal platform architecture. This is similar to the relationship between
platform flooding and carbonate shedding documented in Quaternary carbonates
(e.g., Grammer, 1991; Grammer and Ginsburg, 1992; Glaser and Droxler, 1991;
Bosscher, 1992; Schlager, 1992), as well as during Late Miocene to Recent on Great
Bahama Bank (Eberli and others, 1994). Increase of pelagic carbonates during
Late Quaternary warm periods have also been reported on Western Mediterranean
sea (Vazquez and others, 1991), although changes in productivity of calcareous
plankton is difficult to ascertain (Dean and Gardner, 1986). In the Quaternary,
platform shedding takes place when the rimmed platform is flooded, which occurs
during the highstand of sea level (Schlager and others, 1994). Driscoll and others
(1991) also document prograding clinoforms of carbonate sediments during long-
term relative sea-level rise. The Cap Blanc model indicates that much of the
carbonate shedding occurred while the sea level was rising and the platform top
was continuously flooded, rather than during the highstands.
3.4. BASIN FLOOR TOPOGRAPHY VS. CARBONATE PRODUCTION
The three-dimensional interpretation of the Llucmajor Platform (Fig. 7) illustrates

the importance of another factor that controls the architecture of the reefal prograd-
ing platform, i.e. pre-existing topography (depth and steepness) and overall mor-
phology of the depositional profile. This factor not only determines the available
accommodation space, but also controls the extent and efficiency of the carbonate
factory. With a low topographic gradient of the depositional profile (Fig. 7C),
sea-level fall leads to significant progradation of the offlapping reefs. When sub-
sequently the sea level rises, it floods the platform top, creating a wide lagoon,
that enhances carbonate production and downslope shedding of sediments. On the
other hand, when the topographic gradient is steep, the amount of progradation of
the offlapping reefs is minor. Subsequent flooding of the platform creates only a
small lagoonal area where carbonate production is not optimal and the downslope
shedding is proportionally small.
Thus, differing depositional gradients produced dramatic differences in progra-
dation of the platform. The reefs prograded faster toward the southern margin on
the shallow Llucmajor open shelf than toward the west along the margin of the
Palma Basin. The reef tracts of the successive progradational events outline a
clockwise rotation towards the Palma Basin (Fig. 10), as a result of the greater
production of carbonate on the southeastern part, where the lagoons were wider.
This example illustrates the importance of sediment supply (carbonate produc-
tion) in controlling the progradational architecture. The efficiency ofthe carbonate
factory is controlled by the fluctuations of the sea-level plus the bathymetry and
the slope of the depositional profile.
4. Conclusions
The Llucmajor Platform, a 20-km wide Late Miocene reefal platform shows how a
carbonate progradational platform responds to changes driven by high-frequency
eustatic fluctuations and how it differs from the siliciclastic system. The sea-cliff
outcrops allow high-resolution sequence analysis in two-dimensions of the stratal
and facies architecture, and borehole (water wells) data allow interpretation of the
progradational architecture of the platform in three dimensions.
The basic building block of this carbonate complex (the "sigmoid") stacks
in different magnitudes of larger-scale accretional units ("sets", "cosets" and
"megasets" of sigmoids). All these units have the same characteristics in terms of
stratal geometries, facies architecture and bounding surfaces, and may be consid-
ered as depositional sequences. The hierarchy on stacking of different magnitudes
(thickness and extent) of accretional units reflects a hierarchy of the magnitudes
(amplitude and period) of sea-level fluctuations. This is due to 1) the close rela-
tionship between coral-reef growth and the sea surface, 2) the lack of significant
tectonic or loading subsidence, and 3) the absence of mechanical or chemical
compaction on this never-buried Upper Miocene platform.
The distribution of lithofacies within all accretional units and across their
boundaries, the stages of aggradation and the stacking patterns of the higher-order
106 L. Pomar, w:e. Ward
4370 r\-111~~~~=~1
Fig. 10. Extent of reef tracts and lagoons associated with the 10 cosets of sigmoids (4th-order
sequences) interpreted on the L1ucmajor Platform, during their maximum aggradational phases. Re-
construction of reef tracts is prior to the postulated strike-slip faulting, which must have occurred after
the deposition of the reefal prograding platform and before the sedimentation of Lower Pleistocene
eolian deposits.
sequences, reflect how the reefal system reacts to changes in accommodation.

All accretional units are composed of four systems tracts: 1) lowstand systems
tract composed of progradational reef facies and thin forereef to open-shelf facies;
2) aggrading systems tract composed of thick aggradational lagoon, reef-core,
reef-slope and open-shelf deposits; the lagoonal deposits overly the low-stillstand
systems tract and onlap the erosional surface; the reef core does not backstep; 3)
Fig. 10. Continued.
highstand systems tract composed of progradational reef facies, wedging-out reef-

slope and condensed open-shelf facies; and 4) offtapping systems tract composed
of thin down lapping and down stepping reefal facies, without significant reef-slope
and open-shelf facies, which are distally condensed.
The erosional surface (which corresponds to the sequence boundary) and the
downlap surface are the only two relevant surfaces bounding systems tracts,
and both merge basinwards into the distally condensed interval. Without high-
resolution data, only two conspicuous packages can be defined by these surfaces:
a thick aggradational package and a thin progradational (offtapping) package, the
latter passing basin ward to the distally condensed interval.
This depositional-sequence framework (stratal and facies architecture) dif-
fers from standard sequence stratigraphic models mainly in the lack of a typical
transgressive systems tract and by the presence of distally condensed intervals

corresponding to the high-stillstand, offlapping and low-stillstand systems tracts.
These differences result from the way in which carbonate production reacts to
changes in sea level.
Sea-level change not only determines the accommodation space, it also de-
termines the amount of carbonate production and sedimentation (efficiency of the
carbonate factory). Production is highest during the rising sea-level when the plat-
form top is submerged to an optimum depth; intermittent sea-level rises following
brief falls force re-establishment of carbonate production in lagoons which were
filled by sedimentation during slowing or cessation of sea-level rise. Obliteration
of the shallow platform during sea-level stillstand (low or highstand) causes re-
duction in carbonate production. This results in progradation of the reef system
onto thin slope deposits due to the lack of accommodation space (in absence of
significant subsidence). Absence of shaIlow platforms during sea-level falls leads
to significantly reduced carbonate production, which results in offlapping of coral
reefs onto the open shelf.
Another important factor controIling the platform architecture is the steepness
and overaIl morphology of the depositional profile within the context of fluctuating
sea level. With a gentle depositional profile, sea-level fall leads to significant
progradation of the reefal systems. Flooding of the platform during the subsequent
sea-level rise creates a wide lagoon, which enhances carbonate production and
downslope shedding of sediment. On the contrary, a steeper topographic gradient
allows minor reef progradation, and subsequent flooding of the platform creates
only a small lagoonal area where carbonate production and downslope shedding
is proportionally minor.
The Llucmajor Platform illustrates the importance of sediment supply as a
controlling factor in sequence architecture. It also exemplifies the differences
between the reefal carbonate system and a siliciclastic system. As shown in the
Mallorca example, reefs can step down on the slope during sea-level falls (forced
regression). In contrast, siliciclastic sedimentation will generally bypass the upper
slope when sea level faIls below the shelf break. In addition, the AST of the
MaIl orca complex shows that carbonate production can increase to keep pace
with the rising sea levI. In siliciclastic systems, on the other hand, transgression
generaIly means shoreface erosion and backstepping.
In addition to tectonics and subsidence, stratal and facies architecture also
depends on the interaction between sediment supply and accommodation. Changes
in accommodation result from changes in the morphology of the depositional
profile and from base-level changes. In siliciclastic systems the sediment supply
depends on the conditions outside the sedimentary basin (tectonics, climate and
drainage system on the emergent continent) which may be relatively independent
of base-level changes. On the contrary, in the Mallorca example, sediment supply
(carbonate production) and accommodation changes are not independent, because
both are controlled by the same basic factors: changes in sea level (high-frequency
fluctuations) and depositional profile of the basin floor.
Acknowledgments
We gratefully acknowledge A. Bar6n of the Junta d' Aigiies (Hydrological Survey

of the Balearics) for the water-well data of the Llucmajor Platform. Funding for
this study has been provided by Shell Philippines and Shell Sarawak through Shell
Research (KSEPL), and by a NATO collaborative research grant (SAS-2-0S; CRG
931472). We appreciate the constructive comments of F.V. Abbots, A.-J. Everts
and L. Cooper and U. Singh. A. Gallego helped with drafting. We especially thank
B.U. Haq for his encouragement, useful comments and a review of this paper.
References
ABBOTT, S. T. and CARTER, R. M., 1994, The sequence architecture of mid-Pleistocene

(c.l.I-OA Ma) cyclothems from New Zealand: facies development during a period of
orbital control on sea-level cyclicity, in de Boer, P. L., and Smith, D. G., ed., Orbital
forcing and cyclic sequences: International Association of Sedimentologists Special
Publication No. 19, p. 367-394.
ALVARO, M., BARNOLAS, A., DEL OLMO, P., RAMIREZ DEL POZO, J., AND SIM6,
A., 1984, EI Ne6geno de Mallorca: Caracterizaci6n sedimentol6gica y bioestratignifica:
Boletfn Geol6gico y Minero, v. 95, p. 3-25.
BAR6N, A. and POMAR, L., 1985, Stratigraphic Correlation Tables: area 2c Balearic
Depression, in Steininger, F. F., Senes, J., Kleemann, K., and Regl, F., ed., Neogene
of the Mediterranean, Tethys and Paratethys: Institute of Paleontology, University of
Vienna, p. 17.
BATES, R. L. and JACKSON, J. A., 1987, Glossary of Geology: Alexandria, Virginia,
American Geological Institute, 788 p.
BIZON, G., BIZON, 1. J., BOURROUILH, R. and MASSA, D., 1973, Prusunce aux iles
Baluares (Mud. Occ.) de sediments "messiniens" dupossus dans une mer ouverte, U
salinitu normale: Comt. Rend. Acad. Sci. Paris, v. 277 (12), p. 985-988.
BOSSCHER, H., 1992, Growth potential of coral reefs and carbonate platforms: Utrecht,
Vrije Universiteit, 160 p.
DEAN, W. E. and GARDNER, 1. V., 1986, Milankovitch cycles in Neogene deep-sea
sediment: Paleoceanography, v. I, p. 539-553.
DRISCOLL, N. w., WEISSEL, J. K., KARNER, G. D. and MOUNTAIN, G. S., 1991,
Stratigraphic response of a carbonate platform to relative sea level changes: Broken
Ridge, southeast Indian Ocean: American Association of Petroleum Geologists Bulletin,
v. 75, p. 808-831.
EBERLI, G. P. and GINSBURG, R. N., 1987, Segmentation and coalescence of Cenozoic
carbonate platforms, northwestern Great Bahama Bank: Geology, v. 15, p. 75-79.
EBERLI, G. P. and GINSBURG, R. N., 1988, Aggrading and prograding Cenozoic sea-
ways, northwest Great Bahama Bank, in Bally, A. w., ed., Atlas of Seismic Stratigraphy:
Tulsa, OK, American Association of Petroleum Geologists Studies in Geology No. 27,
p.97-103.
EBERLI, G. P. and GINSBURG, R. N., 1989, Cenozoic progradation of northwestern
Great Bahama Bank, a record of lateral platform growth and sea-level fluctuations, in
Crevello, P. D., Wilson, J. L., Sarg, J. F., and Read, J. F., ed., Controls on Carbonate
Platform and Basin Development: Tulsa, OK, Society of Economic Paleontologists and
Mineralogists Special Publication, No. 44, p. 339-352.
EBERLI, G. P., MELIM, L. A. and SWART, P. K., 1994, Distribution of sedimentary

and diagenetic facies in prograding seismic sequences (Great Bahama Bank, Bahamas):
implications for petrophysical properties and fluid flow, in American Association of
Petroleum Geologists Hedberg Research Conference. Application of sequence stratig-
raphy to oil field development: Paris, France.
FONTBOTE, J. M., OBRADOR, A. and POMAR, L., 1983, Islas Baleares, in Geologia
de Espala, T. II (Libro Jubilar J.M. Rios): Madrid, Comisi6n Nacional de Geologia e
Instituto Geologico y Minero de Espala, p. 343-391.
GLASER, K. S. and DROXLER, A. w., 1991, High Production and Highstand Shed-
ding from Deeply Submerged Carbonate Banks, Northern Nicaragua Rise: Journal of
Sedimentary Petrology, v. 61 (1), p. 128-142.
GRAMMER, G. M., 1991, Formation and evolution of Quaternary carbonate foreslopes,
Tongue of the Ocean, Bahamas: University of Miami, 314 p.
GRAMMER, G. M. and GINSBURG, R. N., 1992, Highstand versus lowstand deposition
on carbonate platform margins: insight from Quaternary foreslopes in the Bahamas:
Marine geology, v. 103, p. 125-136.
HANDFORD, C. R. and LOUCKS, R. G., 1993, Carbonate depositional sequences and
systems tracts - Responses of carbonate platforms to relative sea-level changes, in
Loucks, and Sarg, R., ed., Recent Advances and Applications of Carbonate Sequence
Stratigraphy: Tulsa, American Association of Petroleum Geologists Memoir No. 57, p.
3-41.
HAQ, B. U., HARDENBOL, J. and VAIL, P. R., 1987, Chronology of fluctuating sea levels
since the Triassic: Science, v. 235, p. 1156-1167.
HUNT, D. and TUCKER, M. E., 1992, Stranded parasequences and the forced regressive
wedge systems tract: deposition during base-level fall: Sedimentary Geology, v. 81, p.
1-9.
HUNT, D. and TUCKER, M. E., 1993, Sequence stratigraphy of carbonate shelves with
an example from mid-Cretaceous (Urgonian) of southeast France, in Posamentier, H.
W., Summerhayes, C. P., Haq, B. U., and Allen, G. P., ed., Sequence stratigraphy and
Facies Associations: International Association of Sedimento1ogists Special Publication
No. 18, p. 307-341.
HUNT, D. and TUCKER, M. E., 1995, Stranded parasequences and the forced regressive
wedge systems tract: deposition during base-level fall - reply: Sedimentary Geology, v.
95, p. 147-160.
JERVEY, M. T., 1988, Quantitative geological modelling of siliciclastic rock sequences
and their seismic expression, in Wilgus, C. K., Hastings, B. S., Kendall, C. G. S.,
Posamentier, H. W., Ross, C. A., and Van Wagoner, J. C., ed., Sea-Level Changes: an
Integrated Approach: Tulsa, Society of Economic Paleontologists and Mineralogists
Special Publication, No. 42, p. 47-69.
KENTER, J. A. M., 1990, Carbonate platform flanks: slope angle and sediment fabric:
Sedimentology, v. 37, p. 777-794.
KOLLA, v., POSAMENTIER, H. W. and EICHEN SEER, H., 1995, Stranded parase-
quences and the forced regressive wedge systems tract: deposition during base-level fall
- a discussion: Sedimentary Geology, v. 95, p. 139-145.
MITCHUM, R. M. and VAN WAGONER, J. c., 1991, High-frequency sequences and their
stacking patterns: sequence stratigraphic evidence of high-frequency eustatic cycles:
Sedimentary Geology, v. 70, p. 131-160.
POMAR, L., 1991, Reef geometries, erosion surfaces and high-frequency sea-level
changes, upper Miocene reef complex, Mallorca, Spain: Sedimentology, v. 38, p. 243-
270.
POMAR, L., 1993, High-resolution sequence stratigraphy in prograding carbonates: appli-
cation to seismic interpretation, in Loucks, and Sarg, R., ed., Recent Advances and Appli-
cations of Carbonate Sequence Stratigraphy: Tulsa, American Association of Petroleum

Geologists Memoir No. 57, p. 389-407.
POMAR, L., FORNOS, J. J. and RODRIGUEZ-PEREA, A., 1985, Reef and Shallow
Carbonate facies of the Upper Miocene of MaJIorca, in Mild, M. D., and Rosell, J., ed.,
6th. European Regional Meeting, Excursion Guidebook: Lleida, Spain, International
Association of Sedimentologists, p. 493-518.
POMAR, L. and WARD, W. C., 1994, Response of a Late Miocene Mediterranean reef
platform to high-frequency eustasy: Geology, v. 22, p. 131-134.
POSAMENTIER, H. w., ALLEN, G. P. and JAMES, D. P., 1992, High Resolution Se-
quence Stratigraphy - the East Coulee Delta, Alberta: Journal of Sedimentary Petrology,
v. 62 (2), p. 310-317.
POSAMENTIER, H. w., ALLEN, G. P., JAMES, D. P. and TESSON, M., 1992, Forced
regressions in a sequence stratigraphic framework: concepts, examples, and exploration
significance: American Association of Petroleum Geologists Bulletin, v. 76, p. 1687-
1709.
POSAMENTIER, H. w., JERVEY, M. T. and SARG, J. E, 1988, Eustatic controls on
clastic deposition II - sequence and systems tract models, in Wilgus, C. K., Hastings, B.
S., KendaJI, C. G. S., Posamentier, H. W., Ross, C. A., and Van Wagoner, J. C., ed., Sea-
Level Changes: an Integrated Approach: Tulsa, Society of Economic Paleontologists
and Mineralogists Special Publication, No. 42, p. 125-154.
POSAMENTIER, H. w., JERVEY, M. T. and VAIL, P. R., 1988, Eustatic controls on clastic
deposition I - conceptual framework, in Wilgus, C. K., Hastings, B. S., Kendall, C. G.
S., Posamentier, H. W., Ross, C. A., and Van Wagoner, J. c., ed., Sea-Level Changes:
an Integrated Approach: Tulsa, Society of Economic Paleontologists and Mineralogists
Special Publication, No. 42, p. 109- I 24.
RAMOS-GUERRERO, E., RODRIGUEZ-PEREA, A., SABAT, E and SERRA-KIEL, J.,
1989, Cenozoic tectosedimentary evolution of Mallorca island: Geodinamica Acta, v. 3
(1), p. 53-72.
SARG, J. E, 1988, Carbonate Sequence Stratigraphy, in Wilgus, C. K., Hastings, B. S.,
KendaJI, C. G. S., Posamentier, H. W., Ross, C. A., and Van Wagoner, J. C., ed., Sea-
level Changes: An Integrated Approach: Tulsa, Society of Economic Paleontologists
and Mineralogists Special Publication, No. 42, p. 155-182.
SCHLAGER, W., 1991, Depositional bias and environmental change - important factors
in sequence stratigraphy: Sedimentary Geology, v. 70, p. 109-130.
SCHLAGER, w., 1992, Sedimentology and Sequence Stratigraphy of Reefs and Carbonate
Platforms: Tulsa, American Association of Petroleum Geologists Continuing Education
Course Note Series No 34, 71 p.
SCHLAGER, W., 1993, Accommodation and supply - a dual control on stratigraphic
sequences: Sedimentary Geology, v. 86, p. 111-136.
SCHLAGER, w., REIJMER, 1. G. and DROXLER, A., 1994, Highstand shedding of
carbonate platforms: Journal of Sedimentary Research, v. B64 (3), p. 270-281.
SONNENFELD, M. D., 1993, Anatomy ofOfftap: Upper San Andres Formation (Permian,
Guadalupian), Last Chance Canyon, Guadalupe Mountains, New Mexico, in Adams, J.
W., Austin, G., Barker, J., Hawley, J. w., and Love, D. W., ed., Carlsbad Region, New
Mexico and West Texas: New Mexico Geological Society, 44th Field Conference: p.
195-203.
SONNENFELD, M. D. and CROSS, T. A., 1993, Volumetric partitioning and facies dif-
ferentiation within the Permian Upper San Andres Formation of Last Chance Canyon,
Guadalupe Mountains, New Mexico, in Loucks, R. G., and Sarg, E J., ed., Carbon-
ate Sequence Stratigraphy - Recent Developments and Applications: Tulsa, American
Association of Petroleum Geologists Memoir No. 57, p. 435-474.
TYRRELL, W. W. and DAVIS, R. G., 1989, Miocene Carbonate Shelf Margin, Bali-Flores
Sea, Indonesia, in Bally, A. w., ed., Atlas of Seismic Stratigraphy: Tulsa, OK, American
Association of Petroleum Geologists Studies in Geology No. 27, p. 174-179.
VAIL, P. R., AUDEMARD, F., BOWMAN, S. A., EISNER, P. N. and PEREZ-CRUZ, C.,
1991, The stratigraphic signatures of tectonics, eustasy and sedimentation - an overview,
in Einsele, G., Ricken, w., and Seilacher, A., ed., Cycles and Events in Stratigraphy:
Springer-Verlag, p. 617-659.
VAIL, P. R., MITCHUM, J., R.M., TODD, R. G., WILDMIER, J. M., THOMPSON, S. I.,
SANGREE, J. B., BUBB, J. N. and HATFIELD, W. G., 1977, Seismic stratigraphy and
global changes of sea level, in Payton, C. E., ed., Seismic Stratigraphy - Applications to
Hydrocarbon Exploration: American Association of Petroleum Geologists Memoir No.
26, p. 49-212.
VAN WAGONER, J. C., MITCHUM, R. M. J., CAMPION, K. M. and RAHMANIAN,
V. D., 1990, Siliciclastic sequence stratigraphy in well logs, cores and outcrops, in :
American Association of Petroleum Geologists Methods in Exploration Series, No.7,
p. 45 pp.
VAN WAGONER, 1. c., POSAMENTIER, H. w., MITCHUM, R. M. J., VAIL, P. R.,
SARG, F. J., LOUTIT, T. T. and HARDENBOL, J., 1988, An overview of the funda-
mentals of sequence stratigraphy and key definitions, in Wilgus, C. K., Hastings, B. S.,
Kendall, C. G. S. c., Posamentier, H. w., Ross, C. A., and Van Wagoner, J. C., ed.,
Sea-Level Changes: an Integrated Approach: Society of Economic Paleontologists and
Mineralogists Special Publication, No 42, p. 39-45.
VAZQUEZ, A., ZAMARRW, I., REYES, E. and LINARES, J., 1991, Late Quaternary
climatic changes on the southwestern Balearic slope (Western Mediterranean): isotopic,
faunal, and mineralogical relationships: Palaeogeography, Palaeoclimatology, Palaeoe-
cology, v. 81, p. 215-227.
CHAPTER 5
Geometric Responses in Neogene

Sediments of Offshore New Zealand:
Simulated as Products of Changes in
Depositional Base Level Driven by
Eustasy and/or Tectonics
Christopher G.St.c. Kendall, Gregory L. Whittle, Craig S. Fulthorpe, Phil Moore,

T. Don Hickey, Robert Cannon and Douglas Hellmann
ABSTRACT: The stratigraphic sequences that occur throughout the sedimentary record
are the products of independent variations in eustasy, tectonic movement and rates of
sedimentation. The computer program Sedpak, which graphically simulates the sequence
stratigraphic fill of basins by changing sea-level position, tectonic movement and sedi-
mentation rate as independent variables, was used to test interpretations of the evolution
of Neogene sequences in the offshore Canterbury Basin, eastern South Island, New
Zealand. The simulation results for the Canterbury Basin suggest that sequence bound-
aries were created by changes in depositional base level (Le. sea level), which in turn
were driven by variations in the rate of eustatic and/or tectonic change. Varying the rate of
sediment supply in the simulations did not produce sequence-bounding unconformities,
but greatly influenced sequence stacking patterns.
1. Introduction
The purpose of this paper is to illustrate, using the stratigraphic simulation Sedpak,
the importance of changes in depositional base level (i.e. sea level in marine
environments), for creating sequence-bounding unconformities. These base level
changes may be driven either by changes in the rate of tectonic movement, eustasy
or both. Varying sediment supply alone, did not produce unconformities in these
Sedpak simulations. We demonstrate this effect with an example from Neogene of
the Canterbury Basin, offshore New Zealand.
Pitman (1978), Steckler (1981), Watts (1982), Burton and others (1987),
Reynolds and others (1991) and Kendall and others (1982) have demonstrated
113
B. U. Haq (ed.),
114 e.G.St.e. Kendall et al.
how eustasy, sedimentation, and subsidence act together to produce the sequence
architecture we see in the sedimentary record. Many sequence stratigraphers,
however, assume that the ideal Exxon model can be used to interpret most of the
sedimentary stratigraphic record, and that most high frequency changes in this
record are responses to changes in base level caused by eustasy alone (Posamen-
tier and others, 1988; Donovan and others, 1988; Baum and Vail 1988; Loutit and
others, 1988; Greenlee and Moore 1988; Handford and Loucks, 1993; and Eberli
and Ginsburg 1989; to cite few). Thus, the Exxon model assumes that a specific
eustatic curve, coupled to a constant low frequency tectonic subsidence, produces
a particular pattern of sedimentation (Vail and others, 1977; Vail and Todd, 1981;
Van Wagoner and others, 1987; 1988, 1990; Jervey 1988; and Posamentier and
others, 1988). While Sedpak simulations indeed confirm the importance of base
level changes in generating sequence-bounding unconformities, the simulations
also show that attributing all sequences to eustasy remains an assumption. Vertical
tectonism can produce identical base-level changes. Furthermore, these Sedpak
simulations show that changes in sediment unconformities can strongly influence
sequence stacking patterns.
2. Sedpak
Sedpak is one of many sedimentary simulations being used to visualize and analyze
different sequence stratigraphic geometries expressed by sediments deposited in
response to eustasy, tectonic events or climatic change (Aigner and others, 1987
and 1989; Frohlich and Matthews, 1991; Helland Hansen and others, 1988; 1989;
Kendall and others, 1986; 1989a; 1989b; 1991 a; 1991 b; 1992; Lawrence and others,
1989; 1990; Nakayama and Kendall, 1989; Scaturo and others, 1989; Strobel and
others, 1987; 1989a; 1989b; Tang and others, 1989; Bowman and others, 1993;
Eberli and others, 1994; Armentrout and others, 1994; Rankey and others, 1994;
and Ross and others, 1994).
The basin modeling software, Sedpak, uses an empirical approach for strati-
graphic prediction via graphic simulation (Helland-Hansen and others, 1988; and
Strobel and others, 1989a). The geometric output of Sedpak is non-unique because
we assume for Sedpak, as we assume is true for nature, that sediment accommo-
dation is controlled by changing rates of sedimentary fill, tectonic movement and
sea-level position. The major assumptions used in Sedpak are that:
1) Sea-level position, tectonic movement and sedimentation all vary indepen-
dently.
2) Tectonic movement, as modeled by the simulation, is vertical only.
3) Clastic sediments first fill the accommodation on the shelf landward of
the depositional shoreline break (or clinoform breakpoint) and sediment then is
deposited downslope from this break to a prescribed distance, on surfaces that are
not inclined above a prescribed angle.
4) Following deposition and tectonic movement, erosion of sediment takes
place where the sediment surface has slopes above prescribed angles or excessive
relief.
Neogene Sediments of New Zealand Simulated 115
To paraphrase Helland-Hansen and others (1988) and Strobel and others

(1989a), the "area" of clastic sediment deposited for each time step of the sim-
ulation is prescribed by the user. Sediment is first subtracted from this "area" to
fill the accommodation on the shelf up to the angle of the alluvial plane and then
the remainder is deposited downslope from the depositional shoreline break over
a user-defined distance. The latter wedge of sediment may bypass this slope, if
its angle is above a user-defined gradient. Following deposition, compaction and
tectonic movement, if the angle of the sediment surface, or its relief, is too great,
erosion of the sediment takes place. Carbonates accumulate as a function of water
depth and position with respect to "reefs" or breaks in slope closest to deep water.
In the version of Sedpak used for the simulation experiments described in this pa-
per, the time varying rate of tectonic movement (uplift or subsidence) is defined by
the user for different locations across the basin. The program linearly interpolates
between these defined values. This user defined "tectonic" movement substitutes
for the combined effects of crustal cooling and the isostatic response to sediment
loading but does not model them directly. It does not incorporate subsidence due
to compaction, which is handled separately.
Sedimentary simulations like Sedpak can be viewed as mediums for testing
hypotheses that attempt to explain the geometries seen in stratigraphic sequences
resulting from sea-level variation, sediment accumulation, and tectonic movement.
The user has the opportunity to evaluate whether one particular hypothesis is more
reasonable than another. The Sedpak simulation incorporates the rules of modem-
day sequence stratigraphy as they are applied to the interpretation of high quality,
multi-channel seismic data. The output of the simulation is visually matched to
an interpreted stratigraphic section by iteratively changing the input parameters,
including the rate of sediment accumulation, crustal dynamics, and eustasy. These
inputs control the geometry of the sedimentary fill.
The simulation reproduces the interfingering geometry of clastic and carbonate
sediments as they fill basins. It identifies stratal elements with respect to time,
tracking the evolution of sedimentary geometry (be it carbonate, sandstone, shale,
or a mixture of the three) as a function of rate of sediment accumulation, subsidence
behavior, and eustatic sea-level variation. Sediment geometries are plotted as they
are computed, so the results can be viewed immediately.
In this paper we have primarily used the simulation as a mentor which guided
our stratigraphic interpretations. The simulation was initiated by determining the
most reasonable initial shape for the basin, then the subsidence history of the
basin was determined by tracking the behavior of the lowest surface of the basin
through time. The changing rates of subsidence or uplift were then prescribed by
the user for specific locations and times. Next the gross sedimentary geometries
from the seismic lines were matched by eye with simulation outputs, varying such
parameters as rate of sediment input, distance of sediment transport into the basin,
and depth to wave base.
The resulting Sedpak simulations were used to enhance understanding of the
geometric relationships of sedimentary facies and to predict the sedimentary rela-
tionships within the seismic section, while honoring the age picks on the section.
In addition, the magnitudes of the processes responsible for the geometries seen
116 c.G.St.c. Kendall et al.
the interpreted seismic lines were deciphered as the simulation progressed and the
geometries were recreated. We found that for Sedpak to produce realistic geome-
tries, upper bounds had to be placed on tectonic movement, eustatic fluctuations,
and rates and amount of sedimentation.
3. Examples of Sedpak Simulations
We have used the sedimentary simulation Sedpak in conjunction with the Haq
and others (1987) curve to successively model the sequence stratigraphy of the
Tertiary of offshore Korea (Cheong and others, 1993), the Gulf of Mexico Mesozoic
(Whittle and others, 1993a; and Sullillvan and others, 1993) and the carbonate
Great Bahama Bank (Eberli and others, 1994). In each of these cases we were
able to obtain a good visual match between the stratigraphic cross-section and
the sedimentary simulation output. To produce these matches we assumed the
eustatic sea-level curve of Haq and others (1987). We then modified the tectonic
and sedimentary inputs to the simulation until we produced the best visual match
between the simulations and the stratigraphic sections. For instance, in the Gulf
of Mexico (Whittle and others, 1993b), we modeled the development of a low
ramp in the Jurassic, which progressively aggraded and prograded during the time
of deposition of the Smackover Formation. Later in the Cretaceous, we modeled
the amount of clastics entering the basin as increasing enormously, bypassing the
Smackover margin, causing aggradation and progradation of the shelf. In the late
Tertiary and Quaternary, salt tectonics in the Gulf of Mexico caused the movement
of local diapirs which, in conjunction with the rate of sedimentation and variations
in eustasy, controlled the accommodation space that was then filled by sediment.
We were able to obtain a relatively good match between seismic profiles and
the simulation output using the Haq and others (1987) curve, but as we show
later with the example from Canterbury Basin, these were probably non-unique
solutions. As noted at the beginning of the paper, stratigraphers should recognize
that the stratigraphic record is not only the product of global sea-level change but
also that of tectonic movement, sedimentation rate and compaction, and that none
of these variables need be in phase with each other. For instance, in many cases
tectonic movement does not maintain a constant rate. There are many examples of
high-frequency tectonic movement, as in Southern Italy, where a most spectacular
example of irregular rapid tectonic movement can be seen at the temple of Apollo
in Naples. Here, the temple first subsided into the sea, and was then uplifted
to its present level. The mollusk-encrusted pillars record these events (personal
observation). Similar events have occurred in Indonesia ( Chappell, 1974; Ward
and Chappell, 1975; Chappell and others; and Veeh, 1978; and Chappell and
Shackleton, 1986). Similarly, the Tertiary coastlines of many Caribbean Islands
have raised shorelines, with Barbados representing a prime example (Mesollella
and others, 1969 and 1970; Steinen and others, 1973; Matthews, 1986; Fairbanks
and Matthews, 1978).
Sedpak File: Voilelo s t .db Timestep: 6 MY, - 17 .600
lilhOI04jlie Rolio ~ol
Column 100
Fig. 1. Output of the simulation Sedpak while testing the Exxon model. Note the falling sea-level and
the prograding and aggrading sedimentary geometries. The black dots mark the most seaward position
of alluvial sediments, which often coincides with the depositional shoreline break or clinoform
breakpoint. The graph is the right of the simulation tracks eustasy, and sediment supply. The size of
the excursions on the sediment supply are scaled to fit the area of the graph and the maximum value
for the area of sediment displayed is listed at the top of the graph.
Sedpak File: VoilCla s t. db Timestep: 11 MY: - , 5.600
Litholoqlc R(]tio Plot
0.0 Distonce (km) Deplh.:Widlh- l:80o.0 .00.0
Fig. 2. Output of the simulation Sedpak for the Exxon model. Note stillstand in sea-level and the
prograding and aggrading sediment geometries.
Sedpak File: VailClast.db Tirnestep: 16 MY: -13.600
Lithol09ic Ratio Plol
100
Sond
- 350.0
Fig. 3. Output of the simulation Sedpak for the Exxon model. Note the rising sea-level and onlapping
sediment geometries.
Sedpak File: VailClast.db Timestep' 21 MY: -11.600
Lithologic Ratio Plol M~~~=O

t Column , 00 L:ttt 1~ Oe~r:hl
~O~~~~~~~~~mm~~~~nnnn~~~mmmmmm~""nn~ ...
ft
nd
0.0 Distonce (km) Depth W'idU.. - 1;800.0 <00.0 - 350.0
Fig. 4. Output of the simulation Sedpak for the Exxon model. Note the stillstand in the sea-level
curve and the prograding character of the sedimentary geometries.
4. Testing the Exxon Model
Using Sedpak, we have been able to reproduce much of the geometric and sedi-
mentary character described in the Exxon sequence model. For instance (Figs. 1
and 2), we were able to confirm the Exxon model result, that during a eustatic
fall and the subsequent sea-level low, provided the accommodation seaward of
the depositional shoreline break was shallow enough, sediments which bypass the
shelf were deposited to produce aggrading and prograding geometries. During the
following rise, sediments began to onlap landward, across the underlying lowstand
systems tract and shelf, producing a transgressive systems tract which aggraded as
it onlapped (Fig. 3). In the following highstand, the sediments overfilled the space
available on the shelf and began to prograde (Fig. 4).
The geometries in Figs. 1 through 4 match the essence of the Exxon model for
the sedimentary fill at a basin margin. However, the sedimentary patterns expressed
in these diagrams need not have been the products of eustasy alone, but could also
have been the products of tectonic movement (keeping the amount of sediment
entering the depositional system the same in each case), as demonstrated later
in this paper. Indeed, many of the sedimentary sequence patterns of the geologic
record may be independent of eustasy, rather than dependent, as the Exxon model
suggests. Instead, great portions of the sedimentary section may be dependent on
tectonic setting, as for instance can be seen in the geologic record in the vicinity
of island arcs. Here compression may have uplifted the terrain and driven its
erosion (Mesolella and others, 1969 and 1970; Steinen and others, 1973; Bloom
and others, 1974; Chappell, 1974; Ward and Chappell, 1975; Chappell and Veeh,
1978; Chappell and Shackleton, 1986; and Matthews, 1986). Similarly, climatic
variations may have produced sedimentary sequence variations which, in terms
of the Exxon model, were out of phase with sea-level. For instance, this can be
seen in the Quaternary and Upper Tertiary sediments of the Mississippi Delta.
Here Kolla and Perlmutter (1993) have been able to show that the response of
the Mississippi Delta to the melting of ice during the Pleistocene was a maximum
outflow of sediments into the delta area. This occurred during a sea-level rise rather
than during a fall or lowstand.
s. Isolating the Effects of Sediment Supply
To investigate the sensitivity of sediment geometries to fluctuations in rate of

sediment supply, we have performed a series of experiments with Sedpak. Initially
we input sediment in phase with eustatic highstands, and produced a particular
pattern of sediment geometries at the shelf margin (Fig. 5). In this case, a thick
wedge of sediments overlies and envelopes the margin, because the sediment,
instead of filling and prograding into the basin, tended to aggrade at the basin
margin, producing a stacked pattern of sediment layers. In contrast, if the sediment
was programmed to enter the basin at the sea-level lows (Fig. 6), the resulting
wedge tended to be spread out across the basin and the sediments, instead of
aggrading close to the basin margin, prograded out in the basin. The cause of
the difference between these two margin geometries was that, in the first case,
Sedpok Ti"" topt 50
0.0 -281..0
Fig. 5. Output of the simulation Sedpak for testing effects of varying sediment supply and sea level
position. The sediment is programmed to enter the basin at sea-level highs. This produces a thick
wedge of sediment that envelops the basin margin
Sedpok MY, 0 .000
Sand/Sholl!!
litholOQ Ie Ratio Plot Moximum ~o
LJt r 1~l~R~M
0.0 Oistance (ktn) Depth<Widlh -1-80o.0 <00.0
position. The sediment is programmed to enter the basin at sea-level lows and so is caused to build
out into the basin in a series of prograding wedges.
when the sediments were entering the basin at sea-level highs, accommodation
space was available on the shallow shelf to be filled before the sediments bypassed
over the depositional shoreline break into the deeper basin as a series of thin
clinoforms that clung to the adjacent basin margin. Thus, the sediments tended to
form aggradational beds on the shelf, but because so much accommodation space
was available beyond the shelf edge, in the basin, they did not prograde far into
the basin. In contrast, when the sediments entered during the sea-level lows, there
was no accommodation space in which to aggrade on the shelf and, because the
sea-level fall had also reduced the accommodation space available offshore, the
sediments tended to prograde across the shallowed basin, spreading over its floor.
We used Sedpak to perform other experiments to examine further responses.
For instance, we determined how increasing sediment input through time at suc-
cessive sea-levellowstands could cause the sediments initially to aggrade. Then,
as the accommodation space was diminished by the sedimentary fill, the sediments
began to prograde, bypassing the new depositional shoreline break created by the
lowstand systems tracts (Fig. 7). When we programmed a high rate of sediment
input during the first eustatic lowstand, but caused this lowstand sediment input
to decline with each subsequent sea-level low (Fig. 8), the shelf initially built up,
aggrading and prograding. However, as the sediment supply declined, the sediment
volume tended to be absorbed by the accommodation on the top of the shelf rather
than at the shelf margin or the depositional shoreline. The result was a prograding
geometry at the base of the margin, with thicker and thicker aggradational layers
accumulating on the upper portion of this lowstand shelf.
6. Sedpak Simulation of the Canterbury Basin
Using sedimentary simulations that are able to vary sea-level, tectonic behavior
and sedimentation independently, we can produce geometries that are generated
independent of eustasy. This effect is intrinsic to the program but we also believe
it to be true of nature. We have tested this effect with an example from the offshore
Neogene section of the Canterbury Basin in New Zealand. In this particular case
we could not match our simulation output to the stratigraphic cross-sections when
we used the Haq and others (1987) sea-level curve, but were able to obtain a better
match when we superimposed a high-frequency signal on the Haq and others
(1987) curve. Further, in another experiment we varied the tectonics alone, but
kept sea-level constant, and were able to get the same match.
The area of study for this simulation was the offshore Canterbury Basin on
the eastern margin of the South Island of New Zealand. We simulated a seismic
line which is located on the shelf and has been interpreted to show a series of 4th
order sequences in sediments which were inferred to be of Miocene age (Fulthorpe
and Carter, 1989 and 1991 and Fulthorpe, 1991). The middle to upper Miocene
sequences were estimated to have developed between 11.5 and 7 Ma, though
these ages were poorly constrained by the available well biostratigraphic data.
The portion of the section we chose to simulate consisted of several fourth-order
sequences inferred to have been deposited over a two m.y. period between 11.5
to 9.5 Ma, (Fig. 9). Fulthorpe (1991) felt, on the basis of the local geology, that
Sedpok Tin"lle:stopr SO "'y,oooo
Sand/Shale
Lilhol09~ RotJo Plol ~(]ximum Ar~
CoIumf't 100 LfJJft r 199~ {8Ri:hl
II
0.0 Oistol'14;C (t.;m) Daplh,Wid th 1:800_0 400,0
position. The sediments are programmed to both enter the basin at sea-level lows and increase in
amount at each sea-level low. As the accommodation space is diminished by sediment fill, so the
lowstand systems tract is caused to prograde further into the basin than at the preceding sea-level
Sedpok MY, 0.000
Sand/Shale
L..tholog"= Ratio Plot Mo)Cimum Area
1 Column 100 LlttJ" l~leR~hl

0.0 """'~"""'-nTn-np~nTn".,...~rrm'"'"TTlTnTnmT1~"""""'"'""""""""',","m-n""",,
Sond
0.0 Distance (ktn) Oepth,Wi(m, . 1.800.0 - 281.0
position. The sediments are programmed to enter the basin at sea-level lows and decrease in amount
of each sea-level low. As the accommodation space now progressively increases (since there is less
sediment available), so the low stand systems tract progrades shorter and shorter distances into the
basin at each successive sea-level low.
LINE CB- 82- 54

NW SE
2 km
l...-_....L..-_--1.
Fig. 9. Seismic section from the Canterbury Basin, eastern South Island, New Zealand. The simulated
section includes sequence boundaries one (11.5 Ma) through eight (9.5 Ma) (i.e. seven sequences).
the main control on the evolution of the sequences was probably eustatic but
that their geometries were strongly influenced by sediment supply, subsidence
and current activity. While eustatic fluctuations occurred at several frequencies,
the periodicities of sequences resolved on seismic profiles were locally controlled
functions of sediment supply and subsidence (Fulthorpe, 1991).
When we simulated this using the Haq and others (1987) curve with its third-
order variation (Fig. 10), we found the closest visual match we could obtain
between the simulation and seismic was by increasing the accommodation through
changing the rates of sedimentation and tectonic subsidence history between runs
(Fig. 11). To obtain this match we had to assume high rates of sedimentation at the
beginning of the simulation, low constant rates toward the middle of simulation
and a high constant rate at the end of simulation.
When the third-order sea-level curve was used, only one cycle of sea-level
change occurred during the time period simulated (Fig. 11). The Exxon model for
this period, therefore, implies that only one sequence should have formed. How-
ever, to create the accommodation necessary to re-create the appropriate thickness
of sediment, as observed on the seismic section, the rate of tectonic subsidence
had to be varied as a function of time, so as to compensate for the Haq and others
(1987) sea-level positions. Sediment supply was then varied as necessary, so as to
produce the progradational geometries that matched the sequence stacking pattern
of the section. Even so, only two sequences were produced. Variations in sedi-
ment supply alone, within a simulation run, did not create the sequence bounding
unconformities.
We made two simulations of the Canterbury Basin which used sea-level curves
of different frequencies. In the first simulation (described above) we employed
the third-order Haq and others (1987) curve over the two million year period
simulated (11.5-9.5 Ma). In the second simulation we created a sinusoidal sea-
-11.5000
-240
-260
-280
-5. -300
--...
+
I
::5
0- -320
=
Q)
-340
-360
-380
-12 -11.5 -11 -10.5 -10 -9.5 -9
Time - (Mo)
Fig. 10. The portion of the Haq and others (1987) curve used in the first simulation.
level curve with a 25 meter amplitude and a frequency of 300,000 years (Fig. 12).
We superimposed this on the third-order Haq and others (1987) curve to produce
the high-frequency variations in sea-level seen on Fig. 13. We chose this high-
frequency signal on the basis of the number of events identified on seismic profiles
by Fulthorpe and Carter (1991) and their frequency. In both simulations the time
period modeled was 11.5-9.5 Ma, the number of time steps was 110. In addition,
the initial basin surface was the same, tectonic subsidence curves across the basin
angle of repose were the same, and depositional angle was the same. In addition, the
distance of sediment transport into the basin was also identical for each simulation,
so that rate of sediment accumulation could be constrained.
The high-frequency sinusoidal sea-level curve for the second model produced
a very good visual match between the seismic and the simulation output (Figs. 9,14
and 15). The required rate of sediment supply was more uniform, being relatively
low at the beginning of the simulation and relatively high at the end. Seven
fourth-order sequences were interpreted during this interval (Fulthorpe, 1991 and
Fulthorpe and Carter, 1991). For both models sequence boundary one (11.5 Ma)
was used as the initial basin surface and the simulations ended at sequence boundary
eight (Fig. 10).
Sedpok F ile t CQntNZHoq.l0.3.db T imestep. 1 10 MY, - 9 .500
Sm:juenctl Plot Sand/Shole

MaxImum An!.o
L:f r 1~J'tlR~hl
I SeorC'rl'el (m)
High low
0 .0 Distonce (kin) Of!'ptMY-l dth ",34.0
- 69.0 -334.0
Fig. 11. Final output from the Canterbury Basin simulation run with the Haq and others (1987)
sea-level curve. Note that only one third-order sea-level fluctuation occurs over the simulated interval.
To determine if accommodation could be modeled as being produced by tec-

tonic movement only, we performed a further experiment in which we assumed
sea-level was unvarying. We differentiated the high-frequency sea-level curve of
Fig. 13 and created subsidence rate curves from this (Fig. 16) which were 114 cycle
out of phase with the original eustatic signal for input to the simulation. We then
ran the simulation in which the time period modeled (11.5-9.5 Ma), number of
time steps, initial basin surface, angle of repose, depositional angle, and rates of
clastic accumulation were identical to the simulation illustrated in Figs. 11 and
14. The visual match between the simulation (Figs. 17 and 18) and the seismic
cross-section was as good as with Figs. 14 and 15. This result confirms that the
origins of the accommodation for the Canterbury Basin sequences have a non-
unique solution and that tectonics might play an important role in the creation of
the accommodation. However, it is important to note that generation of sequence-
bounding unconformities in the simulations require base-level changes, whether
driven by either eustasy or tectonics (Figs. 14, 15, 17, 18). Changes in the rate
of sediment supply alone did not produce unconformities (Fig. 11). This supports
the finding of Christie-Blick (1991) that eustatics or tectonic base-level change is
essential for producing type 1 unconformities.
There are two reasons why the sediment accumulation rates between the two
simulations in Figs. 11, 14 and 15 differ: 1) differences in accommodation between
126 e.G.St.e. Kendall et aI.
1.0000
30
20
10
E
I
......... 0
+
-=
0..
='"
-10
-20
-30+-----------~--------~----------~--------__.
-11.5 -11 -10.5 -10 -9.5
Time - (Mo)
Fig. 12. High-frequency changes in sea-level which were added to the Haq and others (1987) curve
of Fig. 10.
the models occurred because of their different sea-level curves and 2) the clastic
rates in Sedpak are rates of supply, not accumulation, so some of the sediment may
bypass out of the plane of section. The latter implies that once a critical sediment
supply rate was reached, any increase in supply resulted in bypass of the sediment
beyond the modeled section. Locating this critical rate was simply a matter of trial
and error.
The problems created by the difference in accommodation between the two
simulations was not as easily resolved. The value for the clastic accumulation at any
particular time in these simulations was primarily a function of how far across the
basin the units of that sequence had prograded. The downslope clastic accumulation
rate in the simulation with the third-order sea-level curve (Fig. 11) was significantly
less than the simulation with the fourth-order curve (Figs. 14 and 15) from 11.5 Ma
to 10.368 Ma, although both models prograde sediment across the basin the same
distance. The differences in shelf geometry between the two simulations of Figs. 11
and 14 were a result of the differing sea-level curves: the third-order simulation
showed downlapping geometries associated with regression prior to 10.36 Ma; the
fourth-order simulation showed both the downlapping associated with regression
and the onlapping associated with transgression for each sequence during this
-200
-250
]: -300
I
----
+
:5
CL
= -350
Q)
-400
-450+-----------,-----------,-----------,-----------,
-11.5 -11 -10.5 -10 -9.5
Time - (Mo)
Fig. 13. Haq and others (1987) with added high-frequency sea-level curve.
interval. Both the simulations of Figs. 11 and 14 captured the thin sequence
associated with the mid-Miocene sea-level low (l 0.368 Ma). Subsequent sediment
supply rates were identical in both simulations though the resulting geometries
were slightly different: the third-order simulation showed onlapping associated
with transgression, while the fourth-order simulation continued to show both onlap
and downlap (compare Figs. 11 and 14).
The overall stacking patterns of the sequences of both simulations were very
similar, suggesting that the sedimentation rate could account for the gross ge-
ometries seen in the seismic, regardless of the eustatic signal. The differences in
sequence geometries, however, suggested that relative sea-level behavior is critical
to the development of individual sequence boundaries, which were created by base
level changes driven by eustasy, tectonics or both.
7. Conclusions
We conclude, as others have before us, that there is no determinable unique cause
for stratigraphic sequences. Tectonic movement, variation in sedimentation rate,
and eustasy may all in their respective ways be responsible. If a model assumes
the size of two of these variables, the third variable can be predicted and the ex-
128 c.G.St.c. Kendall et aI.
Sedpok Filal ContNZH+C. l00_db Tlma~l.p' 1 10 ~YI -9.500

Sand/Shote
Sequonce! Plot lIo)Cimum Area
Column 200 ler:t r 1~~cIe~r:h'
c
o
o
o
iil
0.0 ~l(lnce (~m) OcplhoW'IdIn.. 1134.0
Fig. 14. Final output from the Canterbury Basin simulation run with a sine wave sea-level curve
superimposed over the third-order Haq and others (1987) sea-level curve. Note that seven fourth-order
sea-level fluctuation occur over the simulated interval (one for each interpreted sequence).
pected stratigraphic geometries can be produced by means of computer simulation.

Questions raised by the fact that the sequence stratigraphic simulations have non-
unique solutions suggest that model assumptions should be stated by the sequence
stratigraphers interpreting geologic sections or running simulations. The question
is "do the modelers want the truth or do they want beauty?" Most of us opt for
beauty, particularly since a control of varying sea-level alone is easier to accept
because it is "a la mode" and is beautiful! But is it the truth?
As stratigraphers we must accept that in some stratigraphic settings, fluctua-
tions in the rate of sedimentation and variations in subsidence/uplift may be the
driving forces behind sedimentary geometries. This is not to say that sea-level is
not an important factor but that changes in base level are primary factors in form-
ing sequence boundaries and it is essential to have the correct curve of base level
change (eustatic or tectonic) in order to produce the same number of simulated
sequences that are observed on seismic profiles. The size and frequency of the
eustatic and/or tectonic changes can only be inferred from the local and regional
geology and cannot be determined from the simulation. Similarly many geologists
assume sediment supply is rather tightly constrained, but where sediment supply is
believed to fluctuate significantly, the sequence geometries and stacking patterns
in that area are highly dependent upon on those changing supply rates.
- 9.500
Sedpok File r ContN%H+C.l CO.db Timestepl 1 10 ~YI
Sand/Shale
Facie'S F'tol MO;t;imum Ar'ea
Column lftrt l~(r2::h'
High lOw
0.0 Distance (tim) O~p th .W-Kl th 11.34,0 34.0 -.7.0 -376.0
Fig. 15. Same output as Fig. 14 but now showing the depth of water in meters in which sequences
formed. This diagram gi yes a good sense of the character of the sequence boundaries in the simulation
run.
Finally, when interpreting seismic sections, the observations made from the
seismic data are not of the processes that formed the rocks involved, but are the
products of these processes. From these observations, we make empirical inter-
pretations as to how the geometries we see were created. Empirical sedimentary
simulations are constrained by the same geological rules we use to interpret the
sequence stratigraphy of seismic lines. These geological rules include reasonable
but independently varying rates of change of sediment accumulation, eustasy and
tectonic movement. Within these sedimentary simulations, sediment erosion and
deposition are also constrained by slope angles. The result is that these simulations,
obeying the geological rules (incorporated into the software), produce the geome-
tries and lithofacies relationships we expect of them. Though non-unique in origin,
they provide a valuable insight into which geometric responses one can expect
from specific changes in eustasy, tectonic behavior and rates of sedimentation.
There is no doubt in our minds that we would not have been able to confirm the
importance of base level in generating the sequence boundaries of the Canterbury
Basin without the use of Sedpak or a similar sedimentary simulation.
34.0000
0.5
~
0
-=
----
.5-
1 0
----
+
-2
0
=
-0.5
-1+-----------,-----------,-----------,----------,
-11.5 -11 -10.5 -10 -9.5
Time - (ka)
Fig. 16. Differentiated combined Haq and others (1987) and fourth order sea-level curves - expressed
as subsidence and used in the simulations shown in Figs. 17 and 18.
Acknowledgments
We would like to express our appreciation to Lincoln Pratson, Phillip Levine and
an anonymous reviewer whose editorial advice has much improved this paper. We
also would like to thank AGIP, Amoco, Conoeo, Mobil, the Japanese National Oil
Corporation, Statoil, and Texaco for their financial support in the development of
the program Sedpak.
References
AIGNER, T., DOYLE, M. and LAWRENCE, D. T., 1987, Isostatic Controls on Carbonate
Platform Development: American Association of Petroleum Geologists Bulletin, v. 71,
p.524.
AIGNER, T., DOYLE, M. and LAWRENCE, D. T., EPTING, M., VAN VLIET, A., 1989,
Quantitative Modeling of Carbonate Platforms: Some Examples; in P. D. Crevello, J. L.
Wilson, J.F. Sarg, and 1. F. Read, eds., Controls on carbonate platform and basin devel-
opment, Society of Economic Paleontologists and Mineralogists Special Publication, v.
44, p. 27-37.
Sedpok File, ContNZH CNewOS200.db Timt!lstep' 1 10 MY, -9500
Sequer'lee Plot
Sond/Shole
llAaximum kea
COlumn 200 LJ(I" l~Jdt!R::1i1
0.0 Oisl(lnce (km) Oeplh Width-l..:}4.o
Fig. 17. Final output for the Canterbury Basin simulation run with a constant sea-level curve and
a high-frequency curve (Fig. 16) superimposed on the subsidence curve. This result suggests that
sequence stratigraphic geometry alone cannot be used to determine the cause of the accommodation
that produces it.
ARMENTROUT, JOHN M. and LINDA S. ROUCH, 1994, Computer Simulation of

Seismically-Defined Depositional Sequences; Pliocene and Pleistocene Geometry's and
Rates, Offshore Texas: Program of the Annual Meeting of the American Association of
Petroleum Geologists, v. 3, p. 95.
BAUM, G. R. and VAIL P. R., 1988, Sequence stratigraphic concepts applied to Paleocene
outcrops, Gulf and Atlantic Basins, in Wilgus, C. K., Hastings, B. S., Kendall, C. G. St.
C., Posamentier, H. W., Ross, C. A., and Van Wagoner, J. C., eds., Sea-Level Changes-
an Integrated Approach, Society of Economic Paleontologists and Mineralogists Special
publication, v. 42, p. 309-328.
BLOOM, A. L., BROECKER, W. S., CHAPPELL, J. M. A., MATTHEWS, R. K. AND
MESOLELLA, K. J., 1974, Quaternary sea level fluctuations on a tectonic coast: new
231Th1234U dates from the Huon Peninsula, New Guinea: Quaternary Research, v. 4,
p. 185-205.
BOWMAN, SCOT A., 1994, Carbonate Sedimentation Processes in PHIL: Program of the
Annual Meeting of the American Association of Petroleum Geologists, p. 78.
CHAPPELL, J. and SHACKLETON, N. 1., 1986, Oxygen isotopes and sea level: Nature,
v.324,p.137-140.
CHAPPELL, J. and VEEH, H. H., 1978, Late Quaternary tectonic movements and sea
level changes at Timor and at Atauro Island: Geological Society of America Bulletin, v.
89, p. 356-368.
132 e.G.St.e. Kendall et aI.
.... y, - 9.500
Sedpok F11e , COf'l tNZH+ CNlI!wDS200.db Timestep' 1 10
Sand/Shole
lo1o)Cimum Areo
Column 200 le~' 18%50'&2f;M
..
c
o
~
'"
0-0 ~stance (km) Oepth'Widlh 1.,)4.0 .>\.0 - '95.0
Fig. 18. Same output as in Fig. 17 but now showing depth of water in meters in which sequences
accumulated.
CHAPPELL, J., 1974, Geology of coral terraces, Huon Peninsula, New Guinea: a study of
Quaternary tectonic movements and sea level changes: Geological Society of America
Bulletin, v. 85, p. 553-570.
CHEONG D. K., WHIITLE, O. L., KENDALL, C. O. ST. C., CANNON, R. L., MOORE,
P. D. and KIM S. H., 1993, Simulation of the Tertiary sedimentary section of the
Southern East Sea (Sea of Japan) and its application to oil exploration: Program of
Annual Meeting of American Association of Petroleum Geologists, p. 84.
CHRISTIE - BLICK, N., 1991, Onlap, offlap, and the origin of unconformity-bounded
depositional sequences, Marine Geology, v. 97, p. 35-36.
DONOVAN, A. D., BAUM, O. R., BLECHSCHMIDT, O. L., LOUTIT, T. S., PFLUM, C.
G. and VAIL, R. R., Sequence stratigraphic setting of the Cretaceous Tertiary Boundary
in Central Alabama: in Wilgus, C. K., Hastings, B. S., Kendall, C. G. St. C., Posamentier,
H. W., Ross, C. A., and Van Wagoner, J. C., eds., Sea-Level Changes - an Integrated
Approach, Society of Economic Paleontologists and Mineralogists Special publication,
42, p. 299-308.
EBERLI, GREGOR P., WHIITLE, G. L., KENDALL, C. G. ST. C., CANNON, R. L. and
MOORE, P. D., 1994, Testing a seismic interpretation of the great Bahama bank with a
computer simulation: American Association of Petroleum Geologists Bulletin, v. 78, p.
981-1004.
FAIRBANKS, R. O. and MAITHEWS, R. K., 1978, The Marine Oxygen Isotope Record
in Pleistocene Corals, Barbados, West Indies: Quaternary Research, v. 10, p. 181-196.
FROHLICH, C. and MATIHEWS, R. K., 1991, STRATA-VARIOUS: a flexible Fortran

program for dynamic foreward modeling of stratigraphic: in Franseen, E. K., Watney, W.
L., Kendall, C. G. St. c., Ross, w., eds., Sedimentary Modeling: Computer Simulations
and Methods for Improved Parameter Definition, Kansas Geological Survey Bull. 233,
p.449-461.
FULTHORPE, C. S., 1991, Geological Controls on Seismic Sequence Resolution: Geology,
v. 19, p. 61-65.
FULTHORPE, C. S. and CARTER, R. M., 1989, Test of Seismic Sequence Methodology
on a Southern Hemisphere Passive Margin: the Canterbury Basin, New Zealand: Marine
and Petroleum Geology, v. 6, p. 348-359.
FULTHORPE, C. S. and CARTER, R. M., 1991, Continental-Shelf Progradation by
Sediment-Drift Accretion: Geological Society of America Bulletin, v. 103, p. 300-309.
GREENLEE, S. M. and MOORE T. c., 1988, Recognition and late pretation of depositional
sequences and calculation of sea-level changes from stratigraphic data - offshore New
Jersey and Alabama: in Wilgus, C. K., Hastings, B. S., Kendall, C. G. St. C., Posamentier,
H. W., Ross, C. A., and Van Wagoner, J. C., eds., Sea-Level Changes - an Integrated
Approach, Society of Economic Paleontologists and Mineralogists Special publication,
42, p. 329-356.
HANDFORD, C. R. and LOUCKS, R. G., 1994, Carbonate depositional sequences and
systems tracts - responses of carbonate platforms to relative sea-level changes: in R.G.
Loucks and 1.F.R. Sarg, eds., Carbonate Sequence Stratigraphy, American Association
of Petroleum Geologists Memoir 57, p. 3-42.
HAQ, B., HARDENBOL, J. and VAIL, P. R., 1987, Chronology of fluctuating sea levels
since the Triassic (250 million years to present), Science, no. 235, p. 1156-1167.
HELLAND-HANSEN, w., KENDALL, c., LERCHE, I. and NAKAYAMA, K., 1988, A
Simulation of Continental Basin Margin Sedimentation in Response to Crustal Move-
ments, Eustatic Sea-level Change and Sediment Accumulation Rates: Journal of Math
Geology, v. 20, p. 777-802.
HELLAND-HANSEN, w., STEEL, R., NAKAYAMA, K. and KENDALL, C. G. ST. C.,
1989, Review and Computer Modeling of the Brent Group Stratigraphy, in Whately, M.
K. G. and Pickering, K. T., eds., Deltas: Sites and Traps for Fossil Fuels, Geological
Society Special Publication no. 41, p. 237-252.
JERVEY, M. T., 1988, Quantitative Geological Modeling of Siliciclastic Rock Sequences
and their Seismic Expression: in Wilgus, C. K., Hastings, B. S., Kendall, C. G. St. c.,
Posamentier, H. W., Ross, C. A., and Van Wagoner, J. C., eds., Sea-Level Changes - an
Integrated Approach, Society of Economic Paleontologists and Mineralogists Special
publication, 42, p. 47-70.
KENDALL, C. G. ST. C., HARRIS, P. M., STROBEL, J., CANNON, R., MOORE, P.,
BEZDEK, J. and BISWAS, G., 1989b, Simulation of the West Texas, New Mexican
Permian Guadalupian Basin Margin - A Response to Eustatic Change, An Example
of Sedpak: Abstracts of the 28th International Geological Congress, Washington D.C.,
U.S.A., v.2, p. 2-174.
KENDALL, C. G. ST. c., LERCHE, I. and NAKAYAMA, K., 1986, Simulation of Conti-
nental Margin Sedimentation: American Association of Petroleum Geologists Bulletin,
v. 70, p. 606.
KENDALL, C. G. ST. c., MOORE, P., STROBEL, J. S., CANNON, R. L., PERLMUT-
TER, M., BEZDEK, 1. and BISWAS, G., 1991b, Simulation of the Sedimentary Fill
of Basins: in Franseen, E. K., Watney, W. L., Kendall, C. G. St. C., Ross, W., eds.,
Sedimentary Modeling: Computer Simulations and Methods for Improved Parameter
Definition, Kansas Geological Survey Bull. 233, p. 9-30.
134 CG.St.C Kendall et al.
KENDALL, C. G. ST. C., STROBEL, J. S., CANNON, R. L., BEZDEK, J. and BISWAS,
G., 1991 a, The Simulation of the Sedimentary Fill of Basins: Journal of Geophysical
Research, v. 96, p. 6911-6929.
KENDALL, C. G. ST. c., STROBEL, J., HE, T., MOORE, P., CANNON, R., BEZDEK,
J. and BISWAS, G., 1989a, Simulation of the west Texas, New Mexican Permian
Guadalupian Basin Margin - A Response to Eustatic Change, An Example of Sedpak:
in Harris, P. M. and Grover, G. A., eds., Subsurface and Outcrop Examination of the
Capitan Shelf Margin, northern Delaware Basin Society of Economic Paleontologists
and Mineralogists Core Workshop No. 13, p. 423-426.
KENDALL, CHRISTOPHER G. ST. C., MOORE, P., WHITTLE, G. L. and CANNON,
R. L., 1992, A Challenge: Is it possible to determine eustasy and does it matter?: in
Dott, R. H. ed., Eustasy: The Historical Ups and Downs of a Major Concept: Boulder,
Colorado, Geological Society of America Memoir 180, p.93-107.
KOLLA, V. and PERLMUTTER, M. A., 1993, Timining of sedimentation on the Mis-
sissippi Fan: American Association of Petroleum Geologists Bulletin, v. 77, n. 7, p.
1129-1141.
LAWERENCE, D. T., DOYLE, M. and AIGNER, T., 1989, Calibration of Stratigraphic
Models in Exploration Settings: American Association of Petroleum Geologists Bulletin,
v. 73, n. 3, p. 379-380.
LAWRENCE, D. T., DOYLE, M. and AIGNER, T., 1990, Stratigraphic Simulation of
Sedimentary Basins: Concepts and Calibrations: American Association of Petroleum
Geologists Bulletin, v. 74, p. 273-295.
LOUTIT T. S., HARDENBOL, J., VAIL, P. R. and BAUM, G. R., 1988, Condensed Sec-
tions: The Key to Age Determination and Correlation of Continental Margin Sequences:
in Wilgus, C. K., Hastings, B. S., Kendall, C. G. St. c., Posamentier, H. w., Ross, C.
A., and Van Wagoner, J. c., eds., Sea-Level Changes - an Integrated Approach, Society
of Economic Paleontologists and Mineralogists Special publication 42, p. 183- 216.
MATTHEWS, R. K., 1986, Carbonate diagenesis-equilibrium of sedimentary mineralogy
to the subaerial environment, Coral Cap of Barbados, West Indies: Journal of Sedimen-
tary Petrology, v. 38, p. 110-119.
MESOLELLA, K. J., MATTHEWS, R. K., BROECKER, W. S. and THURBER, D. L.,
1969. The astronomical theory of climatic change: Barbados data: Journal of Geology,
77: 250-274.
MESOLELLA, K. J., SEALY, H. A. and MATTHEWS, R. K., 1970. Facies geometries
within Pleistocene reefs of Barbados, West Indies: American Association of Petroleum
Geology Bulletin 54: 1899-1917.
NAKAYAMA, K. and KENDALL, C. G. ST. C., 1989, A Simulation of Basin Margin
Sedimentation to Infer Geometry and Lithofacies- A Carbonate Example: in Taira, A.
and Masuda, F., eds., Sedimentary Facies in the Active Plate Margin, Terra Scientific
Publishing Company (TERRAPUB), Tokyo, p. 17-31.
POSAMENTIER, H. W. and VAIL, P. R., 1988, Eustatic Controls on Clastic Deposition II
- Sequence and Systems Tract Models: in Sea- Level Changes, an integrated approach:
An Integrated Approach, Wilgus, C. K., Hastings, B., Kendall, C. G. St. C., Posamen-
tier, H., Ross, C., Van Wagoner, 1. C., eds., Society of Economic Paleontologists and
Mineralogists, Special Publication 42, p. 125-154.
POSAMENTIER, H. w., JERVEY, M. T. and VAIL, P. R., 1988, Eustatic Controls on
Clastic Deposition 1- Conceptual Framework: in Sea- Level Changes: An Integrated
Approach, Wilgus, C. K., Hastings, B., Kendall, C. G. St. C., Posamentier, H., Ross,
C., Van Wagoner, 1. C., eds., Society of Economic Paleontologists and Mineralogists,
Special Publication 42, p. 109-124.
RANKEY, EUGENE C., 1994, Data Sim: A Relational Database for Analysis of Strati-
graphic/Sedimentologic Computer Models, Program ofthe Annual Meeting of the Amer-
ican Association of Petroleum Geologists, v. 3, p. 241.
ROSS, W. C., B. A. HALLIWELL, J. A. MAY AND D. E. WATTS, 1994, Integrated Se-
quence Stratigraphic Analysis: An Example from the Mesozoic/Cenozoic of Australia,
Program of the Annual Meeting of the American Association of Petroleum Geologists,
v. 3, p. 246.
SARG, J. E, 1988, Carbonate sequence stratigraphy: in Wilgus, C. K., Hastings, B. S.,
Kendall, C. G. St. c., Posamentier, H. W., Ross, C. A., and Van Wagoner, J. C., eds.,
Sea-Level Changes - an Integrated Approach, Society of Economic Paleontologists and
Mineralogists Special publication, 42, p. 155-182.
SCATURO, D. M., STROBEL, J. S., KENDALL, C. G. ST. C., WENDTE J. C., BISWAS,
G., BEZDEK, J. and CANNON, R., 1989, Judy Creek: A Case Study of a Two-
Dimensional Sediment Deposition Simulation: in Wilson, J. L., Crevello, P., and Read, J.
E" eds., Controls on Carbonate Platform and Basin Development, Society of Economic
Paleontologists and Mineralogists Special publication 44, p. 64-76.
STEINEN, R. P., HARRIS, R. S. and MATTHEWS, R. H., 1973. Eustatic low stand of
sea level between 125,000 and 105,000 B.P. Evidence from the subsurface of Barbados,
West Indies: Geological Society of America Bulletin, 84: 63-70.
STROBEL, J., CANNON, R, KENDALL, C. G. ST. c., BISWAS, G. and BEZDEK, J.,
1989a, Interactive (Sedpak) Simulation Of Clastic And Carbonate Sediments In Shelf
To Basin Settings: Computers and Geoscience v. 15, p. 1279-1290.
STROBEL, J., KENDALL, C. G. ST. c., BISWAS, G., BEZDEK, J. C., AND CANNON,
R., 1987, Preliminary Description of the Program SEDFIL with Carbonate Module
Added: in D. C. Peters and S. A. Krajewski, eds., Proceedings of Denver GeoTech
Computer-Aided Methods and Modeling in Geology and Engineering, p. 341-349.
STROBEL, J., SOEWITO, E, KENDALL, C. G. ST. c., BISWAS, G., BEZDEK, J.
and CANNON, R., 1989b, Interactive Simulation (Sedpak) of Clastic and Carbonate
Sediments In Shelf To Basin Settings: in T. A. Cross (ed.), Quantitative Dynamic
Stratigraphy, p. 433-444.
SULLIVAN N. M., WHITTLE, G. L., KENDALL, C. G. ST. C., LOWRIE, A., CANNON,
R. L., MOORE, P. D., HELLMANN, D. R. and SPAINHOUR, K.,1993, Variations
explained between the sequence stratigraphy of east and west offshore Louisiana using
a sedimentary simulation: Program of Annual Meeting of American Association of
Petroleum Geologists, p. 187.
TANG, J., KENDALL, C. G. ST. c., STROBEL, J., CANNON, R., MOORE, P., BEZDEK,
J. and BISWAS, G., 1989, Simulation of the Sedimentary Fill of the Sichuan Basin by
Upper Permian Coals and Carbonate Sequences- An Example of the use of Sedpak:
American Association of Petroleum Geologists Bulletin, v. 73, n. 3, p. 418.
VAIL, P. R. AND TODD, R. G., 1981, Northern North Sea Jurassic Unconformities,
Chronostratigraphy and Sea Level Changes from Seismic Stratigraphy: in L. V. IIIing
and G. D. Hobson (Eds.), Proceeding of the Petroleum Geology of the Continental Shelf
ofNW Europe Conf., March 4-6,1980, London, Heydon and Son Ltd., p. 216-235.
VAIL, P. R, MITCHUM JR., R. M., TODD, R. G., WIDMIER, J. M., THOMPSON III, S.,
SANGREE, J. B., BUBB, J. N. and HATLELID, W. G., 1977, Seismic Stratigraphy and
Global Changes of Sea-level: in C. E. Payton (ed.), Seismic Stratigraphy - Applications
to Hydrocarbon Exploration, American Association of Petroleum Geologists Memoir
26; p. 49-212.
VANWAGONER, J. c., MITCHUM, R. M., CAMPION, K. M. and RAHMANIAN, V. D.,
1990, Siliciclastic Sequence Stratigraphy in Well Logs, Cores and Outcrops: American
Association of Petroleum Geologists Methods in Exploration Series, No.7, American

Association of Petroleum Geologists, 55 p.
VAN WAGONER, 1. c., MITCHUM, R. M., POSAMENTIER, H. W. and VAIL, P. R.,
1987 Key Definitions of Seismic Stratigraphy: American Association of Petroleum
Geologists Atlas of Seismic Stratigraphy, v. 1, p. 11-14.
VAN WAGONER, 1. C., POSAMENTIER, H. w., MITCHUM, R. M., VAIL, P. R., SARG,
J. F. and HARDENBOL, J., 1988, An Overview of the Fundamentals of Sequence
Stratigraphy and Key Definitions: in Wilgus, C. K., Hastings, B. S., Kendall, C. G. St.
C., Posamentier, H. W., Ross, C. A., and Van Wagoner, J. C., eds., Sea-Level Changes-
publication, 42, p. 39-45.
WARD, W. T. and CHAPPELL, J., 1975. Geology of coral terraces, Huon Peninsula, New
Guinea: A study of Quaternary tectonic movements and sea level changes: discussion
and reply. Geological Society of America Bulletin, 86: 1482-1486.
WHITTLE, G. L., KENDALL, C. G. ST. C., FOGARTY, M. A. and LOWRIE, A., 1992,
Paleotectonic restoration and simulation modeling applicable to the Louisiana offshore:
Gulf Coast Association of Geological Society Transactions, v. XLII, p. 429-444.
WHITTLE, G. L., KENDALL, C. G. ST. C., SCOTT R. w., MOORE, P. D. and CANNON,
R. L., 1993a, Simulation of the evolution of a Lower Cretaceous ramp to rim carbonate
margin in Texas and Louisiana: Program of Annual Meeting of American Association
of Petroleum Geologists, p. 200.
WHITTLE, G. L., KENDALL, C. G. ST. c., SULLIVAN N. M., FOGARTY, M. A.,
KROTZER C., LOWRIE, A., CANNON, R. L., MOORE, P. D., HELLMANN, D. R.
and SPAINHOUR, K., 1993b, Stratigraphic modeling of two contrasting Gulf Coast
Basins: Program of Annual Meeting of American Association Of Petroleum Geologists,
p.200.
CHAPTER 6
Controls on Long-Term Global Rates of

Coal Deposition, and the Link between
Eustasy and Global Geochemistry
L. Bruce Railsback
ABSTRACT: Global rates of coal deposition correlate positively with long-term lowering
of global sea level, probably because long-term first-order regression leaves sedimentary
shelves exposed as coastal plains for development of peat swamps during higher-order
short-term transgressions. Global coal depositional rates are weakly correlative with
orogenic activity, probably because development of foreland basins favors peat deposi-
tion. Global coal depositional rates are not correlative with paleogeographic parameters,
although previous work has shown that paleogeography is a major control on location
of coal deposits. Global coal depositional rates are correlative with neither abundance of
plant species nor changes in abundance of plant species.
The paleoclimatologic significance of these results is that, although coal is clearly a cli-
matically sensitive sediment, its paleolatitudinal sensitivity may diminish during global
regression. The geochemical significance is that coal deposition, like sulfate deposition,
is a parameter important in governing the atmosphere's 02 and C02 contents. If, as
recent work suggests, both coal deposition and sulfate deposition are functions of sea
level change, eustasy may have been a major control on the changing composition ofthe
earth's atmosphere in the Phanerozoic.
1. Introduction
Coal is a sedimentary rock of major economic significance as an energy resource,

of considerable paleogeographic utility as a paleoclimatic indicator, and of geo-
chemical importance as a reservoir of organic carbon. Despite this significance in
several fields of study, our understanding of the controls on rates of coal deposition
is relatively limited. Several efforts (e.g., Parrish and others, 1982) have considered
the paleogeographic distribution of coal and its implications for past climates and
plate motions. However, relatively little attention has focused on changing rates,
rather than locations, of coal deposition at the global scale, despite the greater
importance of accumulation rates from geochemical and economic perspectives.
137
B. U. Haq (ed.).
Sequence Stratigraphy and Depositional Response to Eustatic. Tectonic and Climatic Forcing. 137-159.
138 L.B. Railsback
This is partly because it is not easy to reconstruct the record of coal deposition
from existing data, and perhaps because it is usually assumed that rates of coal
deposition have been a function of paleogeography and climate.
This paper examines the controls on global rates of coal deposition. It examines
the hypotheses that paleogeographic positions of the continents, orogenic activity,
plant evolution, and changes in eustatic sea level have controlled rates of coal
deposition. The first hypothesis is implicit in the work of paleogeographers such
as Parrish and others (1982), in that paleogeographic configurations that have
provided more coal-depositing localities should also have allowed volumetrically
greater coal deposition. The second comes from the association of tectonic activity,
and especially foreland basin development, with coal deposition (McCabe, 1991).
The third follows logically from obvious linkage of plants and coal accumulation.
The last hypothesis is implied by statements like that of Holser and others (1988)
that high sea level minimizes development of coal swamps, and by Tissot's (1979)
observation that major economic accumulations of coal formed during periods
following shifts in the onlap curve of Vail and others (1977). The evaluation of
these hypotheses leads to a model suggesting that regression has been a major
control on global rates of coal deposition since the Devonian.
2. The Record of Global Coal Depositional Rates
Coal or coal-like deposits are known from the Proterozoic (Tyler and others, 1957)
and Devonian coals are found in Russia (Vishemirsky, 1978) and the Appalachian
Basin (Englund and Thomas, 1990), but significant coal deposition began only in
the Early Carboniferous. Two methods can be used to generate a record of coal
depositional rates since the Devonian. One approach uses published estimates of
the distribution of coal resources and reserves with respect to geologic age. The
main problem with this method is that it uses amounts of coal that are economically
exploitable resources, rather than actual amounts of coal. Thus amounts of coal
reported for a given period typically exclude thinly bedded coals that cannot be
easily mined, and they thus exclude most coal. The reported resources also exclude
coal buried below depths reasonable for economic recovery, and they exclude any
coal in Antarctica.
In addition to these problems, data on coal resources. also contain biases ac-
cording to nation, which affect the distribution of coal with respect to time. For
example, Bestougeff (1980) noted that the 1977 World Energy Conference reached
estimates of Soviet coal reserves lower than most Soviet estimates, whereas esti-
mates of U.S. reserves were considered a relatively "conservative estimate", and
reserves of the Peoples' Republic of China include "hypothetical" deposits not ac-
cepted by the 1977 WEe. Because Bestougeff's (1980) and Vishemirsky's (1978)
data show that some nations contain disproportionate shares of coal resources of
a particular age, this means that time distributions from these resource data may
be further biased. All these problems suggest that the validity of estimates of coal
depositional rates from resource data are highly debatable. Resource data are also
commonly divided into such large time divisions, often by geologic period, that
Controls on Global Rates of Coal Deposition 139
-0-
7 Vishemirsky's (1978) dala wilh
60x "real-la-resource" ralio
(ri 6
iil
CI:>.S -6:-
Slrakhov's (1969) dala wtth
<aE 50x "real-la-resource" ralio
.Q~ 4
.... 0 -0-
~() 3 8eslougeff's (1980) dala wilh
Cl.0l
GlOO 80X "real-la-resource" ralio.
0'0 2
-----
<a"-
o Ronov's (1976,1983) dala assuming
()
coal basin sedimenl density~2.5
and 2% coal in coal basin sedimenls
o 100 200 300 400

MYBP
Fig. 1. Global rates of coal deposition through geologic time. Filled squares indicate rates calculated
from volumes of coal-basin sediments; open symbols indicate rates calculated from estimates of
coal resources and reserves, multiplied to account for coal not included in economically exploitable
seams. All show high rates in Late Paleozoic, low rates in Triassic, and moderate or fluctuating rates
in post-Triassic. Resource-based rates plunge in the last 20 m.y. because of immaturity of young
organic sediments. Note that the plotted values are apparent rates derived from present resources and
rock masses, and these rates do not take into account modification of the record by erosion.
they are of little use when compared to parameters like sea level that change over
shorter time scales.
A second method of reconstructing global coal depositional rates uses Ronov's
data regarding the global abundance of various sedimentary rock types through the
Phanerozoic. Ronov (1982) provided data on the volume of "coal basin sediments"
as one of seven sediment types. Given (1) the density of these sediments and (2) the
proportion of these sediments that is actually coal, one can calculate the mass of
coal deposited in each time period, and division by the duration of the period yields
an average depositional rate for the period. This method offers several advantages
over estimates based on reserves. Ronov's data do not exclude coals buried beyond
the range of economic exploitation, and they include thinly-bedded coals that are
not economically exploitable. They are also probably less susceptible to the biases
between nations noted in coal reserves. In addition, Ronov's data are more finely
divided with respect to time than are most resource data, allowing better temporal
resolution in modeling. Thus the use of Ronov's data on sediment abundance
provides a better record of coal depositional rates, and it is the data set used in
most recent work, such as Berner and Canfield's (1989) development of a global
model for atmospheric oxygen.
Fig. 1 shows the global rates of coal deposition calculated from these records.
Ronov's data were converted to depositional rates assuming an average density of
2.5 gm/cm 3 for coal basin sediments and assuming that coal constitutes only 2%
of coal basin sediments. The latter value is well within the range of 0.5 to 5 %
discussed by Berner and Canfield (1989) but less than their "best guess" of 2.5%.
140 L.B. Railsback
This conversion yields coal depositional rates of 1 to 4 X 10 18 glm.y. except in

the Triassic, when rates dip to 2 to 5 X 1017 glm.y. Rates calculated from Ronov's
data are generally 50 to 80 times as great as those derived from resource data, so
the only constraint that resource data place on the proportion of coal in coal basin
sediments is that the latter must be at least 0.04%.
The first-order trends in depositional rates are the same for Ronov's data
and for the resource data, which are taken from Strakhov (1969), Vishemirsky
(1978), and Bestougeff (1980). In all four cases, depositional rates increase from
nearly zero in the Devonian to relative highs in the late Paleozoic, and then
decrease precipitously to Triassic lows before returning to variably intermediate
to high levels in the post-Triassic. The only serious discrepancies between the
lithologically-derived and resource-derived rates are that the resource-derived rates
are greater in the Permian and, to a lesser extent, in the Early Jurassic. However,
71 % of Permian reserves and 75% of Jurassic reserves are in the former Soviet
Union (Bestougeff, 1980), and Bestougeff pointed out that Soviet reserves may
have been overestimated. Similarly, large Jurassic reserves in China (e.g., Fen,
1984) may be among the "hypothetical" Chinese reserves discussed by Bestougeff.
In addition, rates deri ved from the resources of all nations plummet in the Neogene
because of the thermal immaturity of most young organic sediments. Because of
these problems with resource-derived rates of coal deposition, this paper will use
the approach of Berner and Canfield (1989), which was supported by the results
of Berner (1991), and thus will concentrate on the lithologically-derived rates in
attempting to discern the controls on global coal depositional rates.
3. Controls on Coal Deposition
3.1. CLIMATE AND GEOGRAPHY
Most discussions of controls on the deposition of peat and thus coal center on two
factors, precipitation and temperature, because standing water inhibits oxidation
of fallen plant matter and low temperatures slow the processes of decay. Many
workers (e.g., Schopf, 1973) have focused on the importance of precipitation in
peat deposition, and the occurrence of most peats in regions of high rainfall (Figs. 2
and 3) supports that view. However, most modern peats form at high latitudes rather
than in the rainy tropics (Fig. 2), and McCabe(l984) pointed out that Canadian
and Siberian peats accumulate under relatively little precipitation, and that swamps
presently occur in dry regions, such as Sudan and Chad.
Thus there are no clear rules regarding the location of peat swamps, and peat de-
position is commonly attributed to both tropical and temperate climates. However,
abundant precipitation and cool temperatures clearly favor peat accumulation. This
can be seen in Ireland's thick, not just widespread, peat deposits that have formed
in a cool climate with abundant precipitation provided by a warm ocean current.
Similarly, Strakhov (1969, vol. 2., p. 359) stated that modern peat thicknesses
in Russia vary with temperature: "To the south [of Russia's boreal forests] the
thickness of peat deposits is again reduced, since the increase in summer and mean
annual temperature with relatively dry air brings about decomposition of organic
Fig. 2. Map showing location of modern peat swamps (taken from Ziegler and others, 1987) and
generalized ocean currents. Peat swamps are concentrated at high temperate latitudes but occur at all
latitudes in locations where warm currents bring precipitation.
Fig. 3. Map showing land area where precipitation is greater than 0.5 m/yr. Comparison with
Fig. 2 shows overlap of precipitation with peat deposition. Adapted from Fig. 4 of Ziegler and others
(1987), which was reprojected from an unpublished map by T.T. Fujita. Arrows indicate warm western
boundary currents warm promoting precipitation on eastern sides of continents, and on western sides
of continents at 45 to 60 latitude. For a more detailed treatment of global precipitation patterns,
see Lottes and Ziegler (1994).
matter ... ". Strakhov likewise argued that high temperatures in tropical regions
limit the modem accumulation of peat, but he conceded that most Pennsylvanian
coals accumulated in tropical regions.
This uncertainty regarding the contribution of tropical regions to coal depo-
sition through time may result from the temporal variability of monsoonal circu-
lation. Parrish and others (1982) pointed out that modem monsoonal circulation
in the Indian Ocean allows little rainfall in tropical East Africa in the winter, so
that this major portion of the earth's tropical land area has an unexpectedly dry
climate in some seasons that limits peat deposition. More recently, Lottes and
Ziegler (1994) have demonstrated that seasonality of rainfall is a critical control
on the distribution of modem peats. Thus, our expectation of high tropical rain-
142 L.B. Railsback
fall and peat accumulation across a large part of the modem earth is invalidated
during times of monsoonal circulation, and Parrish and others (1982) argued that
monsoonal circulation may have limited tropical areas of peat accumulation in the
Triassic as well as from the Miocene to the present.
This emphasis on atmospheric and oceanic circulation is also instructive in
examining the map of modem peat occurrences compiled by Ziegler and others
(1987) (Fig. 2). The map shows large areas of peat accumulation in the temperate
zones, but also shows peats scattered across nearly all latitudes less than 70. This
is especially true in the western hemisphere, where peats occur along the eastern
margin of North and South America, even in the Horse Latitudes. Such occurrences
are enigmatic from a strictly zonal perspective, but they are readily understood if
one considers that warm currents travel poleward on the west sides of ocean basins
to generate precipitation on the eastern margins of continents (Fig. 3) . Those
western boundary currents then produce abundant precipitation in the temperate
regions, accounting for the widespread peats across Canada and western Eurasia.
However, central Eurasia at the same temperate latitude is so far removed from
oceanic effects that its limited rainfall allows little peat deposition. Cool boundary
currents, on the other hand, allow little precipitation in tropical regions on the west
sides of continents and thus cause the scarcity of peat occurrences on the western
sides of North and South America and in tropical western Africa (Fig. 2). Thus
an understanding of modern oceanic and atmospheric circulation, including the
monsoonal effect in East Africa pointed out by Parrish and others (1982), explains
the modem distribution of peats better than a strictly zonal model, and should be
more useful in predicting ancient distributions.
3.2. TECTONICS
Strakhov (1969, p. 361) observed that coal deposition has often occurred in tecton-
ically active regions, and he concluded (p. 378) that major coal accumulation in the
Late Paleozoic occurred during the folding of geosynclines. McCabe (1991) sim-
ilarly noted that many large coal deposits are in foreland basins. McCabe argued
that subsidence rates in foreland basins approximate typical rates of peat accumu-
lation, so that peat swamps in foreland basins are neither flooded and drowned
(as in strike-slip basins) nor exposed and oxidized (as might happen on mature
passive margins). McCabe also pointed out that recent models of foreland basin
deposition (e.g., Heller and others, 1988) suggest that siliciclastic sediments may
be trapped near their uplifted sources long enough to allow significant peat accu-
mulation before swamps are flooded with sediment. McCabe (1991) thus proposed
an explanation for the link between tectonic activity and coal accumulation, but
he did not test the hypothesis against the stratigraphic record of coal accumulation
rates.
3.3. CHANGES IN SEA LEVEL
The relationship of coal deposition to long-term first-order sea level change is

nearly as confused as its relationship to geography, climate, and tectonics, but the
role of sea level has been explored less thoroughly. Holser and others (1988) stated
that development of peat swamps is minimal when sea level is high, presumably
because continental area is reduced, and perhaps because the remaining continental
area is intensely weathered. However, Fig. 1 shows that coal deposition was also
minimal in the Triassic, a time of low sea level. The Triassic example is debatable,
however, because arid climate in the Pangaean continental interior and Tethyan
monsoonal circulation may have limited the extent of low-latitude Triassic peat
swamps.
The role of sea level becomes clearer if one considers the geologic setting of
modern peat swamps and ancient coal deposits. Ziegler and others (1987) observed
that the huge region of peat bogs along the southwest side of Hudson Bay (roughly
300,000 km 2 ) is the result of poor drainage on a nearly flat "marine sedimen-
tary platform" exposed by isostatic rebound after glacial retreat. The Okefenokee
Swamp of southeastern Georgia (Cohen, 1974) similarly lies on a low coastal plain
of marine sediments exposed by regression. In considering ancient deposits, Ward
(1984, p. 161) concluded that "the best parallels [for environments that formed
major coal seams in the past] are probably to be found in modern coastal areas,
on river deltas or near barrier island coasts", rather than in smaller bogs at higher
elevations that cannot have generated large volumes of coal. Cecil and Englund
(1985) pointed out the need for "large, flat areas" for major accumulations of
coal, Haszeldine (1989) suggested that "abandoned coastal plains", rather than
deltas, were major sites of coal deposition, and Strakhov (1969, vol. 2., p. 361)
similarly concluded that ancient coal deposition commonly took place on "a plain,
inclined toward the sea". Cross (1988) emphasized the importance of sea-level
rise in generating accommodation space over coastal plains for coal deposition,
but that suggests prior longer-term regression to expose sedimentary shelves as
coastal plains.
This emphasis on recently drained coastal plains suggests that regression is
important in exposing sedimentary platforms or shelves to generate broad coastal
plains where peat swamps are likely to form (Fig. 4). Major transgression, on the
other hand, floods previously incised coastal valleys and so generates only small
areas of marsh or swamp. Still-stand, as well as minor transgression flooding a
previously generated coastal plain, would not necessarily inhibit peat deposition,
so long as it followed a regression that generated coastal plains. Tissot (1979)
speculated that major economic accumulations of coal were associated with re-
gressions, although his use of an onlap curve, rather than a sea level curve, left
uncertainty as to the actual temporal relationship of coal deposition to sea level
change. While one might question the preservation potential of any non-marine
sediment deposited during regression, the impressive resistance of peat to erosion
(McCabe, 1984) suggests that preservation of peat during regression is not an
issue.
3.4. EVOLUTION OF LAND PLANTS
Because peat production requires land plants, and particularly land plants adapted
to swampy environments, one might expect that land plant evolution could be an
144 L.B. Railsback
Long-term Sea Level Fall

------ -----SL 1
Inc ised prior -----------SL2 ~
coastal plain SL3
Coastal plains: exposed
t
Time I----~~""")
but not yet Incised shen
Long-term Sea Level Rise

+
---------------------- ---~~~:~~ 1
~~~~~-=-------------------------SL3
Drowned valleys over

Incised prior ooastal plains
Flooded shell unavailable
tor coal depos~ion
Fig. 4. Cartoon illustrating the hypothesized relationship between long-term sea level change
and global rates of coal accumulation. Long-term transgression floods incised continental margins,
generating few large swamps for coal deposition as coastlines move through valleys and onto uplands.
During long-term transgression. most depositionally flattened terrain is below sea level and is thus
beyond the reach of coal deposition. Long-term regression instead exposes broad flat sedimentary
shelves where large swamps can form for major coal deposition. Epoch-by-epoch correlation supports
this hypothesis (Table I). Note that more detailed sequence stratigraphy suggests that most coal
deposition probably occurs in shorter-term. higher-order transgressions within overall long-term
regression (asterisks on oscillating curve) as exposed shelf is episodically re-flooded (see text).
700
T K J T P P M D
600
CJ)
(l)
u 500
(l)
a..
en
......
400
c
ctl
-Q;
0... 300
0
200
.0
E
~ 100
Z
0
0 50 100 150 200 250 300 350 400
MYBP
Fig. 5. Plot of number of species in major groups of land plants from Silurian to Miocene,
adapted from Figs. 1. 2 and 3 of Niklas and others (1985). "Lycopods" includes arborescent and
rhizomatous lycopods; lycopod diversity shown for Cretaceous to Miocene is author's extrapolation
of pre-Cretaceous trend shown by Niklas and others (1985).
important control on rates of coal deposition. For example, land plant diversity at
the species level increased greatly in the Mississippian (Fig. 5), the earliest period
of significant coal accumulation. Increased plant diversity with the radiation of
angiosperms in the Cretaceous and Tertiary (Fig. 5) likewise coincides with an
increase in coal accumulation rates (Fig. 1). Conversely, land plant diversity dipped
at the Permian-Triassic boundary, when coal accumulation rates plummeted. This
suggests some relationship between land plant evolution and coal accumulation.
3.5. HYPOTHESES
Consideration of the above controls on peat deposition suggests four general

hypotheses. These hypotheses are that global rates of coal deposition are controlled
by (1) the area of the continents with zones of high rainfall; (2) the extent of
orogenic activity generating basins suitable for coal deposition; (3) changes in
global sea level; and (4) plant evolution and diversity. These hypotheses are tested
below.
4. Methods and Results
4.1. PARAMETERS
To test the hypotheses above, four general parameters A, T, R, and D were

considered. A is the proportion of total continental area within zones of high
rainfall and thus likely to have peat swamps, determined for each of the geologic
epochs for which Ronov (1982) provides data on coal accumulation. One zonal
variant of this parameter is A z , the proportion of total continental area in latitudinal
zones from 0 to 15 and 45 to 80. Az thus measures the movement of the continents
through latitudinal zones of high rainfall as the result of plate tectonics. A second
variant, A c , is the proportion of total continental area in zones of high rainfall
suggested by paleocirculation patterns. Ac was determined with Fig. 2 as a guide,
so that areas selected included regions from 50 to 70 latitude (especially on the
west sides of continents and excluding centers of continents), equatorial regions on
the east side of continents, smaller equatorial areas on the west sides of continents,
and eastern margins of continents regardless of latitude (see Fig. 6 for an example).
A third variant, A p , is the proportion of total continental area likely to enjoy high
rainfall as suggested both by paleocirculation patterns and by orographic or rain-
shadow effects. Ap is essentially a subset of Ac, because it takes into account the
monsoonal modification of tropical rainfall in the Permo-Triassic and Miocene to
present suggested by Parrish and others (1982), Kutzbach and Gallimore (1989),
and Dubiel and others (1991), and because it excludes areas in rain shadows.
A (for all variants) was measured for each of Ronov's (1982) epochs using
paleogeographic maps generated by the Terra Mobilis software of Denham and
Scotese (1988). Patterns of atmospheric as well as oceanic paleocirculation were
qualitatively generated using the paleogeographic maps of Scotese and others
(1979) and Ziegler and others (1983), which include mountain ranges to estimate
rain shadow effects. Temporal variation in Ac is shown in Fig. 7.
146 L.B. Railsback
Fig. 6. Map showing hypothetical continents and areas for parameters A .. Ac and Ap. Az is the
proportion of total continental area in high-rainfall latitudinal zones from 0 to 15 and 45 to 80. Ac
is the proportion of total continental area in zones of high rainfall as suggested by paleocirculation
patterns. Ap is the proportion of total continental area likely to enjoy high rainfall as suggested both
by paleocirculation patterns and by orographic or rain-shadow effects and by monsoonal effects.
1.0 1cj:Q:::::O-r- 0.4

1....-_ _ _.;;........1
0.8
~o
Ac
"0
c:: 0.6
III 0.3
C\j
a: 0 .4
0 .2
0 .0 -+-----,.--.,.---.---.---ot-r-----.---r-L- 0.2
o 100 200 300
MYBP
Fig. 7. Temporal variation in R2, the proportion of time in each epoch in which sea level did not rise
(open squares), A c , the proportion of total continental area likely to host peat swamps as suggested
by paleogeography and patterns of atmospheric and oceanic circulation (small closed squares), and
To-20, the proportion of time in each epoch in which an orogeny took place (circles).
T, the second parameter, is the proportion of time during which orogenic

activity took place during each of Ronov's epochs, and thus presumably the time
in which foreland basin development would have favored peat deposition. Ts is
based on the global record of orogenies by Stille (1936), and uses Stille's first-
rank and second-rank orogenies (of four ranks) and assumes that each lasted 20
million years. Tc is based on the timing of orogenies tabulated by the lUGS
Alpine 0
Era <Il
c-
o ""0 0
E
Variscan! ""0
:;
j'" Laramide
0
~
:::l
Hercynian CIl
c- Er~ c- c- E,@ r-
C.ille,9.-0nianc-Era c-
c c c
c c '" c '5i'"
.~ '"
'55 c lij
c
'" c c
'" 'c"
:c0 <Il
<Il 0'"
~.-E
<Ill:; E
E .!l! 0 lij I"~ c
E "iii
N ~ ' :c .9
c'" c
c
0
0
0
'5 E'"
'" iii I iii 0
~ <Il I,:::l
'" 0 it <Il 0
'" ~
~
.0 0
.f:
CIl
w
~
III ~ '"
I- CIl al
a
~
:::l
CIl
....I
'"
W
I I I I I I
o 100 200 300 400 500 600
MYBP
Fig. 8. Chronology of orogenies compiled by International Union of Geological Sciences Commis-

sion on Geochronology (1968), with anglicized terms from Stille (1936). Width of bars equals 20
million years on horizontal axis.
Commission on Geochronology (1968), and sub-variants Tc -5, Tc - IO , Tc -20 and

Tc-30 assume orogenies lasting 5 million, 10 million, 20 million, and 30 million
years, respectively (Fig. 8). Temporal variation in T c- 20 is shown in Fig. 7.
R, the third parameter, is the proportion of time during which sea level fell
during each of Ronov's epochs, and thus it is presumably the time in which ex-
posed coastal plains favored development of peat swamps. Variants of R include
RI, the time in which sea level decreased, and R2, the time in which sea level
did not increase (and thus includes stillstand as well as regression) (Fig. 9). Hal-
lam's (1984) sea level curve was digitized at one million year intervals for this
purpose, and in each of Ronov's epochs the proportion of time of sea level fall
was calculated. Sahagian and Holland (1991) and Algeo and others (1992) have
argued that Hallam's sea level curve overestimates absolute values of ancient sea
level because global hypsography in the past differed from that of today, but the
trends in Hallam's curve, which are the issue is this application, remain essentially
unchallenged. Temporal variation in R2 is shown in Fig. 7.
D, the fourth parameter, is the diversity of land plants as expressed by the
number of species known from each of Ronov's epochs, using the results ofNiklas
and others (1985) (Fig. 5). LlD is the change in diversity from that in the previous
epoch.
4.2. RESULTS
Table 1 shows the correlation of each of these parameters with global rates of coal
deposition. None of the paleogeographic parameters correlate significantly with
coal depositional rates, and in fact the most sophisticated parameter, A p , produces
148 L.B. Railsback
R2 = Tl+T3
250 (150-120) T 1+T2+ T3+ T 4
T3~ +- T2~ +-- Tl
225
<II 200
CD
'lD 175
~
150
1254----r--~--~~--~--~--~
..;50
130 MYBP 140
........
~ 400 ....
.' .-
.5!!
t1l
~ 300 ....
C
~ 200
~
Co
~ 100
o
..c
t1l 0
l!!
CD
'lD -1 OO+--.,..,..,'T"I'..,..,..'"T"Ir-I-r-.+,....,..,r-r-r..........,......,...............T"'I""......."'T"I
~ 0 50 100 150 200 250 300 350
MYBP
Fig. 9. Hallam's (1984) sea level curve for Devonian and later (lower portion of diagram) and
explanation of calculation of R2 (upper portion), where R2 is the proportion of time during which
sea did not rise in any interval. In the example shown, R2 (as well as Rl) would be about 0.60 for
the interval from 150 to 120 MYBP.
the weakest correlation. Neither overall land plant diversity, D, nor diversity of
gymnosperms, lycopods, nor gymnosperms plus lycopods, nor change in plant
diversity, D.D, correlates significantly with coal depositional rates. Ofthe tectonic
parameters, only Tc -20 correlates significantly with coal depositional rates, and
that is only if one accepts a 10% chance of faulty correlation. On the other hand,
the parameters related to sea level change show significant correlations with global
rates of coal deposition (Table 1), suggesting that regression has indeed been a
major control on rates of coal deposition.
4.3. NUMERICAL MODEL
One can generate a simple numerical model for coal depositional rates through
time using A, R and an average global rate constant k for coal deposition. The rate
constant k can be approximated in two ways. First, if we take the world's present
coal and lignite reserves, which total 10.1 x 10 18 gm, and divide by half the 350
million years in which those reserves were deposited (i.e. assuming discontinuous
coal deposition over only half the total time elapsed) and by 30% of the 135.8 Mm2
TABLE 1
Correlation of model parameters with global rates of coal deposition
Parameter Description Correlation Statistical

(R2) Significance
Az Paleolatitude 0.017 None (p > 0.10)

Ac Circulation 0.019 None (p > 0.10)
Ap Circulation and paleogeography 0.003 None (p > 0.10)
T. Tectonic activity 0.018 None (p > 0.10)
Tc-s Tectonic activity - 5 m.y. orogenies 0.087 None (p > 0.10)
Tc-1O Tectonic activity - 10 m.y. orogenies 0.114 None (p > 0.10)
Tc-2o Tectonic activity - 20 m.y. orogenies 0.172 p =
0.10
Tc-3o Tectonic activity - 30 m.y. orogenies 0.098 None (p > 0.10)
D Plant Species 0.140 None (p > 0.10)
t:..D Change in plant Species 0.000 None (p > 0.10)
Rl Regression 0.235 p = 0.05
R2 Regression and stillstand 0.339 p = 0.02
R2 x Ac Combination of above 0.415 p < om
R2 x Ac x Tc Combination of above 0.458 p < 0.01
.
R2 x Ac x Tc x Dn Combination of above
P = probability that no correlation exists
0.545 p < 0.01
of continental area under consideration (Le. assuming coal deposition across only
parts of the continents, as suggested by Ac, and ignoring Antarctica), we obtain
k = 1.4 x 10 15 gmIMm 2 m.y. This accounts, however, for only the present reserves,
not for other presently non-economic coal deposits. The "real-to-resource" ratios
on Fig. 1 suggest that k must thus be increased by a factor of about 60, giving a k
of about 0.85 x 10 17 gmIMm2 m.y. A second way to calculate k is to begin with the
modern depositional rate of carbon in peat swamps of 1.6 x 10 14 gm/year, which
corresponds with an average accumulation rate of 0.5 mm/year of peat (Bramryd,
1980). Modern deposition occurs across about 350 x 106 hectares of peat swamps,
which yields an areal rate constant of 4.6 xl 0 19 gmlMm 2 m.y. That area is, however,
only about 3.9% of Ac for the modern world, so the modified value of k would be
only 1.8 x 10 18 gmIMm2 m.y., and if we assume 10% preservation of these peats
in the geological record, k is then reduced to about 1.8 x 10 17 gmlMm2 m.y., a
value very close to that derived by the first method.
If we average the values above and use k = 1 X 10 17 gmIMm2 m.y., the rate
of coal deposition D for any epoch e is thus
De = kRA c 135 Mm 2 , (1)

150 L.B. Railsback
5 "T""-----I ----- Observed 1--_ _---.

--0- Model A
Q)
a; 4
a:.
-::>:'-
~ E 3
0:::'
. - (1j
~o
~()
arc? 2
O~
a
(tj..-
o
()
O+--~--r--r-~-D~-_,--~
o 100 200 300
MYBP
5 , . - - - - - - 1 ----- Observedl-_ _--.
- - 0 - Model B
$
4
a:.
(1j
-::>:'-
(1j E IJ
6:::'
. - (1j
3
~o
~()
arc? 2
o~
a
(tj..-
o
()
O+--~--r-~~~~~r-_,-~~
o 100 200 300
MYBP
Fig. 10. Results of numerical models for global rates of coal deposition. A) Results of Equation 2
using sea level change (R2) and paleogeography (Ac) as controls. B) Results of Eq. (3) using sea
level change (R2), tectonics (Tc), plant diversity (D) and paleogeography (Ac) as controls. Solid
symbol here is same as solid symbol in Fig. 1.
where the final quantity is the area of the continents minus that of Antarctica.
Table 1 indicates that the most effective version ofEq. (1) is
(2)
The results of Eq. (2) are shown in Fig. lOa. As Table 1 suggests, the agreement
between the model results and the observed record is strong. As Eq. (2) shows,
the model assumes a constant continental area. If one assumed an increasing
continental area from the Carboniferous to the present, the fit of the model results
with the data would be even better.
Another modification of Eq. (1) is to include the parameters Tc-20 and D.
Because T c -20 is less than 1.0 for some periods (Fig. 7), the value of k must be
increased by about half. A normalized value of D, Do, can be obtained by dividing

by the mean value of D (about 300 plant species), to give a modified Eq. (2):
(3)
The results of Eq. (3) are shown in Fig. lOb. Eq. (3) yields a better fit to the
observed data in the Late Paleozoic but a worse fit in the Jurassic. The Jurassic
is, however, a period for which coal reserves may have been over-reported, as
noted above, so model results for the Jurassic lower than Ronov's data may not be
cause for great concern. As Table 1 indicates, the De calculated by Eq. (3) yields
a considerably better correlation with the observed record than that produced by
Eq. (2).
5. Discussion
5.1. PALEOGEOGRAPHIC IMPLICATIONS
The results indicate that falling sea level has been a major control on global rates
of coal deposition, whereas paleogeographic considerations have been of lesser
significance. The relative success of A c , which involved oceanic and atmospheric
circulation, compared to the purely zonal A z , is to be expected from the work of
Parrish and others (1982), who similarly found that circulation models were in
general more effective than zonal approaches in predicting locations (rather than
rates) of coal deposition.
The failure of A p , the paleographic parameter including orographic and mon-
soonal effects, may result from overestimation of the importance of rain shadows,
and the fact that plate tectonics generates numerous mountain ranges at the edges
of continents to potentially, but not necessarily, block transport of vapor to con-
tinental interiors (as in the Americas today). For example, the paleogeography of
North America in the Pennsylvanian, when an eastern equatorial mountain range
might have blocked moisture-laden trade winds, would not seem conducive to the
vast accumulation of Appalachian coal. Appalachian coal deposits are instead an
example of coal accumulation in a major foreland basin, as would be expected
from the observations made by McCabe (1991).
Despite their failure here, paleogeographic and paleocirculation patterns no
doubt have controlled locations of coal deposition, as shown by Parrish and others
(1982) and McCabe and Parrish (1992). The results above instead suggest that
global rates of coal accumulation are controlled by sea level changes. Thus, coal
must still be considered a paleocIimatically sensitive sediment, but it may be a
less effective paleolatitudinal indicator in times of falling sea level, when peat
deposition may be more widespread. This caveat regarding times of regression is
supported both by the model results for the Phanerozoic and by the presence of peat
swamps at 30 latitude on the modem eastern margins of North and South America,
which have been made available for peat swamps by Cenozoic regression.
152 L.B. Railsback
5.2. COAL ACCUMULATION AND PLANT EVOLUTION
The poor correlation of plant di versity with coal accumulation rates may have more
to do with the fossil record than with actual controls on coal accumulation. Any
geologic or paleontologic record contains a bias toward the recent, because more
material is available for study. The plant record is further complicated by species
and genera defined by particular plant organs, as described by Stewart (1983, p.
24-26), and although Niklas and others (1985) strove to eliminate form genera and
to correct for synonymies caused by organ genera, problems no doubt remain in
the record used here. Furthermore, comparison with overall plant diversity may
not provide the best comparison with coal accumulation rates. For example, the
minor decrease in overall diversity in the Late Permian and Triassic masks a major
extinction of cordaites and arborescent lycopods, major tree-forming groups in the
Carboniferous and Permian (Taylor, 1981, p. 121 and 425) whose loss may have
contributed to the dearth of Triassic coal. Thus, while the record of plant diversity
is easily compared with other parameters like coal accumulation, that record may
be the least meaningful of the four examined here.
A different perspective on evolutionary controls on coal accumulation is that
evolution of degraders of plant matter may control organic preservation and thus
coal accumulation (Tom Algeo, pers. comm., 1993). This is a difficult hypothesis
to test, because we have no fossil record of the micro-organisms responsible for
degradation of plant matter. The poor correlation of change in diversity with coal
accumulation does not support the hypothesis, because one might expect that
preservation of plant matter from newly-evolved species might be favored until
later evolution of degraders. However, some support for the hypothesis comes from
Strakhov's (1969) compilation of the changing loci of coal deposition. Strakhov
(1969) argued that Paleozoic coals were largely deposited in tropical to subtropical
settings, whereas Late Carboniferous to Cenozoic coals have increasingly been
deposited in moist temperate settings like those in which most peats form today
(Fig. 11). One might conclude that Paleozoic low-latitude coals accumulated prior
to extensive evolution of microbial degraders, whereas later coal accumulation
was favored at high latitudes after the evolution of temperature-sensitive micro-
organisms whose activity is diminished in colder climates. However, this linkage
between evolution and coal accumulation is at present only speculative.
5.3. STRATIGRAPHIC CONSIDERATIONS
The correlation of high coal accumulation rates with falling sea level may seem
anomalous, because sequence stratigraphers often find coals in lowstand systems
tracts that were deposited during early transgression (e.g., Van Wagoner and others,
1990, 1991; Kirschbaum and McCabe, 1992). Three observations resolve this
anomaly, however. First, Hallam's (1984) sea level curve used above is a first-
order sea level curve at time scales too large to identify higher-order sea level
changes. Thus minor higher-order transgressions undoubtedly occurred during
the periods of major first-order regression identified by the model. The model
results in no way contradict the argument that most coals are deposited during
short-term, higher-order transgressions. Secondly, deposition of coal during such
.-..
~140~--------------------------~
~
III
c:
.8120
III
c: 100
~
iIi
-; 80
tii
a:
lii 60
c:
,g
'iii
0 40
a.
<I>
0
lii 20
0
()
100 300 400
Fig. 11. Plot of Strakhov's (1969, p. 356-359) compilation of coal deposition in different humid
zones. Note general shift from low-latitude coals to high-latitude coals through time.
minor and/or early transgressions is expected in the depositional scheme described

above. Small transgressions across recently drained shelves would cause flooding
of coastal plains to create extensive coal swamps, but such swamps would have a
lower probability during major transgression extending across eroded and incised
shelves or uplands. Thirdly, the preservation of coals despite regression is probably
explained by the common deposition of coals in foreland basins (McCabe, 1991;
McCabe and Parrish, 1992). The relatively rapid subsidence of those basins may
explain how coals are preserved despite the seemingly high likelihood of erosion
during a long-term global drop in sea level.
Short-term transgressions generating coal swamps during long-term regres-
sions that expose coastal plains to non-marine environments thus probably explain
the seeming anomaly of coals in lowstand systems tracts. For example, Pennsyl-
vanian coals in Illinois and the U.S. Midcontinent were deposited when lowlands
were flooded during the transgressive phases of third or fourth order cycles (e.g.,
Heckel, 1984) within the first-order regression of Hallam's (1984) sea level curve.
Similarly, deposition of Maastrichtian coals in Alberta (Saunders and Pemberton,
1988) at the beginning of transgressive cycles took place during the longer-term
Maastrichtian regression. Haszeldine's (1989) generalization that "coal beds un-
conformably transgress" underlying deltaic deposits likewise suggests small-scale
transgressions independent of larger-scale or longer-term sea level change.
In short, although peat deposition may require short-term transgression across
a pre-existing coastal plain, construction of that coastal plain probably requires de-
position of a sedimentary surface and then exposure of that surface, which implies
a longer-term regression (Fig. 4). Short-term transgression during longer-term
regression thus floods exposed coastal plains, whereas short-term transgression
154 L.B. Railsback
10~-------------------------r5
--->-
4~ E
0:-
- 0
CIl_
3C 0
B E
iii!'
0"-
20. 0
0:::
CD..-
(iiQ5
-
10 "0
o 0
~
O+-~---r--~~~~--T-~--~O
o 100 200 300 400
MYBP
Fig. 12. Results of numerical model for global rates of coal deposition (solid line) and rates of
burial of organic carbon generated by numerical models by Berner (1987) (short dashes) and Lasaga
(1989) (long dashes).
during longer-term transgression floods incised uplands that have little potential
for development of widespread peat swamps.
5.4. EUSTASY AND GLOBAL GEOCHEMICAL CYCLES
The dependence of coal deposition on sea level is significant to geochemistry

because coal is one the sedimentary reservoirs in which organic carbon resides.
Although shales are a larger reservoir for organic carbon than coal (De gens, 1965,
p. 203), the depositional flux of organic carbon to coal may have at times rivaled
the corresponding flux to shales. For example, Berner (1987) speculated that 60%
to 90% of the deposition of organic matter during the Carboniferous and Permian
took place in fresh water, and most of that would presumably have been as coal.
The model results are roughly correlative with estimates of burial of organic carbon
generated by isotopically-driven numerical models by Berner (1987) and Lasaga
(1989) (Fig. 12), further suggesting that major events in coal deposition help
determine major events in the deposition of organic carbon.
Deposition of organic carbon in turn plays a major role in controlling the con-
centration of 02 and CO2 in the earth's atmosphere. For example, Garrels and
Perry (1974) developed a global mass balance equation
4 FeS2 + 8 CaC03 + 7 MgC03 + 7 Si02 + 31 H 20 =
8CaS04H20 + 2 Fe203 + 7 MgSi03 + 15CH20, (4)

linking the storage of carbon in its reduced form as organic matter to the oxidation
of sulfur to sulfate. Garrels and Lerman (1981, 1984) subsequently constructed
numerical models based on this equation, and showed that its linkage of the carbon
and sulfur cycles could account for the general negative correlation of al3C of
sedimentary carbonates and 834 S of marine sulfates through the Phanerozoic.
Berner (1987, 1991) and Berner and Canfield (1989) built on the work of Garrels
and Lerman to construct models for atmospheric 02 and C02 in the Phanerozoic,
and burial of organic carbon was a major control in those models. The atmospheric
C02 concentrations predicted by those models have since been partly confirmed
by independent investigations (e.g., Mora and others, 1991).
The significance of this paper's results with regard to the broader questions of
atmospheric chemistry is that one of the major depositional fluxes of organic carbon
(coal deposition) is controlled by change in sea level. Railsback (1992) showed
that the depositional flux of marine sulfates in the Paleozoic, and perhaps over
longer time periods, was also a function of sea level. If fluxes of two of the major
species in Eq. (4) are to some extent controlled by sea level, it in tum follows
that changes in the O2 and CO 2 contents of the atmosphere in the Phanerozoic
have also been affected by sea level change. Geochemists have long used the
isotopic records of carbonates and sulfates to estimate some of the fluxes implied
by Eq. (4), but a better understanding of the linkage of eustasy and deposition
of these geochemically important sediments may allow us to discern the ultimate
controls on the global geochemical system. Furthermore, sea level change may
be responsible for other changes in atmospheric chemistry, as Haq (1993) has
suggested. Our growing understanding of eustasy and global geochemistry may in
tum help us understand the enigmatic correlation of sea level change and extinction
(Newell, 1962, 1967: Jablonski, 1985).
6. Conclusions
1. Global rates of coal deposition calculated from various sources of data show suf-
ficient agreement to indicate that the long-term Phanerozoic record is meaningful
(Fig. 1).
2. Neither paleogeographic parameters nor measures of land-plant evolution,
as determined herein, are correlative with global rates of coal deposition (Table 1).
3. Global rates of coal deposition are weakly correlative with orogenic activity,
probably because development of foreland basins favors accumulation of peat
(Table 1).
4. Global rates of coal deposition are strongly correlative with long-term first-
order sea level fall, probably because regression exposes sedimentary shelves that
provide coastal plains for formation of peat swamps (Table 1; Fig. 4). However, se-
quence stratigraphic studies (Cross, 1988; Saunders and Pemberton, 1988; Haszel-
dine, 1989; Van Wagoner and others, 1990, 1991; Kirschbaum and McCabe, 1992)
show that most coal deposition probably occurs during shorter-term higher-order
transgressions within long-term first-order regression.
156 L.B. Railsback
5. A numerical model using parameters for eustasy, tectonics, paleogeography,

and land-plant diversity accounts for over 50% of the variance in global coal
depositional rates through time (Table 1; Fig. 10).
6. Control of coal deposition by sea level change, combined with control of
evaporite deposition by sea level, suggests that changes in the 02 and C02 contents
of the atmosphere may also be controlled by sea level change.
Acknowledgments
Steven Holland of the University of Georgia kindly shared his knowledge of

sequence stratigraphy and eustasy, and he and R.S. Haszeldine of the University of
Glasgow are thanked for careful reviews of an early version of the manuscript. Fred
Ziegler (University of Chicago), Anne Raymond (Texas A&M University), and
Tom Algeo (University of Cincinnati) provided valuable comments on an earlier
version of this work presented at the 1993 SEPM Meeting on the Stratigraphic
Record of Global Change. John S. Compton (University of South Florida) and an
anonymous reviewer provided thorough and helpful reviews of the manuscript.
References
ALGEO, TJ., SESLAVINSKY, K.B. and WILKINSON, B.H., 1992, The Paleozoic world:
Continental flooding, hypsometry, and sea level: Geological Society of America Ab-
stracts with Programs v. 24. p. AI07.
BERNER. R.A.. 1987. Models for carbon and sulfur cycles and atmospheric oxygen:
Application to Paleozoic geologic history: American Journal of Science., v. 287. p.
177-196.
BERNER, R.A., 1991, A Model for Atmospheric CO 2 over Phanerozoic Time: American
Journal of Science, v. 291, p. 339-376.
BERNER, R.A. and CANFIELD. D.E., 1989, A new model for atmospheric oxygen over
Phanerozoic time: American Journal of Science, v. 289, p. 333-361.
BESTOUGEFF, M.A., 1980, Summary of mondial coal resources and reserves. Geographic
and geologic repartition: 26th International Geological Congress Colloque C2 Energy
Resources, p. 353-366.
BRAMRYD, T., 1980, The role of peatlands for the global carbon dioxide balance: Pro-
ceedings of the Sixth International Peat Congress (Duluth, MN), p. 9-11).
CECIL, C.B. and ENGLUND. KJ., 1985, Geologic controls on sedimentation and peat
formation in the Carboniferous of the Appalachian Basin, in Englund, KJ., Gille-
spie, W.H., Cecil, C.B., Windolph, J.F., Jr., and Crawford, T.J., Characteristics of the
Mississippian-Pennsylvanian boundary and associated coal-bearing rocks in the south-
ern Appalachians: U.S. Geological Survey Open-File Report 85-577, p. 27-33.
COHEN, A.D., 1974, Petrography and paleoecology of Holocene peats from the Oke-
fenofee swamp-marsh complex of Georgia: Journal of Sedimentary Petrology, v. 44, p.
716-726.
CROSS, T.A., 1988, Controls on coal distribution in transgressive-regressive cycles, Up-
per Cretaceous, Western Interior, U.S.A., in Wilgus, C.K., Hastings, B.S., Kendall,
C.G.St.C., Posamentier, H.W., Ross. c.A., and Van Wagoner, J.C., eds., Sea-Level
Changes: An Integrated Approach: SEPM Special Publication 42, p. 371-380.
DEGENS, E.T., 1965, Geochemistry of Sediments: Englewood Cliffs, NJ, Prentice-Hall,

342p.
DENHAM, C.R. and SCOTESE, c.R., 1988, Terra MobilisTM: Houston, Earth in Motion
Technologies.
DUBIEL, R.E, PARRISH, J.T., PARRISH, J.M. and GOOD, S.C., 1991, The Pangaean
Megamonsoon - Evidence from the Upper Triassic chinle formation, Colorado Plateau:
Palaios, v. 6, p. 347-370.
ENGLUND, K.J. and THOMAS, R.E., 1990, Late Paleozoic depositional trends in the
central Appalachian Basin: U.S. Geological Survey Bulletin 1839f, p. FI-F9.
FEN, M., 1984, The tectonic control for sedimentation of coal-bearing sequences in East
China, in Rahmani, R.A., and Rores, R.M., eds., Sedimentology of Coal and Coal-
bearing Sequences: International Association of Sedimentologists Special Publication
7, p. 375-385.
GARRELS, R.M. and LERMAN, 1981, Phanerozoic cycles of sedimentary carbon and
sulfur: National Academy of Science Proceedings, v. 78, p.4652-4656.
GARRELS, R.M. and LERMAN, A., 1984, Coupling of the sedimentary sulfur and carbon
cycles - an improved model: American Journal of Science, v. 284, p. 989-1007.
GARRELS, R.M. and PERRY, E.A., 1974, Cycling of carbon, sulfur, and oxygen through
geologic time, in Goldberg, E.D., ed., The Sea, Volume 5: New York, John Wiley &
Sons, p.303-336.
HALLAM, A., 1984, Pre-Quaternary sea level changes: Annual Reviews Earth Planetary
Science, v. 12, p. 205-243.
HAQ, B. U., 1993, Deep-sea response to eustatic change and significance of gas hydrates for
continental margin stratigraphy: International Association of Sedimentologists Special
Publication 18, p. 93-106.
HASZELDINE, R.S., 1989, Coal reviewed: depositional controls, modern analogues, and
ancient climates, in Whateley, M.K.G., and Pickering, K.T., eds., Deltas: Sites and Traps
for Fossils Fuels: Geological Society of London Special Publication 41, p. 289-308.
HECKEL, P.H., 1984, Changing concepts of Midcontinent Pennsylvanian Cyclothems,
North America, in Belt, E.S., and Macqueen, R.W., eds., Sedimentology and Geochem-
istry: Neuviome Congros International de Stratigraphie et de Geologie du CarbonifOre,
Compte Rendu, v. 3, p. 535-553.
HELLER, P.L., ANGEVINE, c.L., WINSLOW, N.S. and PAOLA, c., 1988, Two-phase
stratigraphic model of foreland-basin sequences: Geology, v. 16, p. 501-504.
HOLSER, W.T., SCHIDLOWSKI, M., MACKENZIE, ET. and MAYNARD, J.B., 1988,
Geochemical cycles of carbon and sulfur, in Gregor, C.B., Garrels, R.M., Mackenzie,
ET., and Maynard, J.B., eds., Chemical Cycles in the Evolution of the Earth: New York,
John Wiley & Sons, p. 105-173.
INTERNATIONAL UNION OF GEOLOGICAL SCIENCES COMMISSION ON
GEOCHRONOLOGY, 1968, Tableau comparatif des echelles geochronologique
recrement publiees pour les temps phanerozoiques: Notice explicative: Annales de
la Societe Geologique de Belgique, v. 90, bulletin 10, p. 825-828.
JABLONSKI, D., 1985, Marine regressions and mass extinctions: a test using the modern
biota, in Valentine, J.W., ed., Phanerozoic Diversity Patterns: Princeton, N.J., Princeton
University Press, p. 325-354.
KIRSCHBAUM, M.A. and MCCABE, PJ., 1992, Controls on the accumulation of coal and
on the development of anastomosed fluvial systems in the Cretaceous Dakota Formation
of southern Utah: Sedimentology, v. 39, p. 581-598.
KUTZBACH, J.E. and GALLIMORE, R.G., 1989, Pangaean climates: Megamonsoons of
the megacontinent: Journal of Geophysical Research, v. 94, p. D3341-3357.
158 L.B. Railsback
LASAGA, A.C., 1989, A new approach to isotopic modeling of the variation of atmospheric
oxygen through the Phanerozoic: American Journal of Science, v. 289, p. 411- 435.
LOTTES, A.L. and ZIEGLER, A.M., 1994, World peat occurrence and the seasonality of
climate and vegetation: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 106, p.
23-37.
MCCABE, PJ., 1984, Depositional environments of coal and coal-bearing strata, in Rah-
mani, R.A., and Flores, R.M., eds., Sedimentology of Coal and Coal-bearing Sequences:
International Association of Sedimentologists Special Publication 7, p. 13-42.
MCCABE, PJ., 1991, Tectonic controls on coal accumulation: Bulletin de la Societe
Geologique de la France, v. 162, p. 277-282.
MCCABE, P.l. and PARRISH, J.T., 1992, Tectonic and climatic controls on the distribution
and quality of Cretaceous coals, in McCabe, PJ., and Parrish, J.T., eds., Controls on the
Distribution and Quality of Cretaceous Coals: Geological Society of America Special
Paper 267, p. 1-15.
MORA, c.1., DRIESE, S.G. and SEAGER, P.G., 1991, Carbon dioxide in the Paleozoic
atmosphere: Evidence from carbon-isotope compositions of pedogenic carbonate: Ge-
ology, v. 19, p. 1017-1020.
NEWELL, N.D., 1962, Paleontological gaps and geochronology: Journal of Paleontology,
v. 36, p. 592-610.
NEWELL, N.D., 1967, Revolutions in the history of life, in Albritton, C.C., Jr., ed.,
Uniformity and simplicity: a symposium on the principle of the uniformity of nature:
Geological Society of America Special Paper 89, p. 63-91.
NIKLAS, KJ., TIFFNEY, B.H. and KNOLL, A.H., 1985, Patterns in vascular land plant
diversification: an analysis at the species level, in Valentine, J.W., ed., Phanerozoic
Diversity Patterns: Princeton, NJ., Princeton University Press, p. 97-128.
PARRISH, J.T., ZIEGLER, A.M. and SCOTESE, C.R., 1982, Rainfall patterns and the
distribution of coals and evaporites in the Mesozoic and Cenozoic: Paleogeography,
Paleoclimatology, Paleoecology, v. 40, p. 67-101.
RAILSBACK, L.B., 1992, A geological numerical model for Paleozoic global evaporite
deposition: Journal of Geology, v. 100, p. 261-277.
RONOV, A.B., 1982, The Earth's sedimentary shell: quantitative patterns of its struc-
ture, compositions, and evolution: International Geology Review, v. 24, p. 1313-1388
(Reprinted by American Geological Institute, Reprint Series V, 1983).
SAHAGIAN, D.L. and HOLLAND, S.M., 1991, Eustatic sea-level curve based on a stable
frame of reference: Preliminary results: Geology, v. 19, p. 1209-1212.
SAUNDERS, T. and PEMBERTON, S.G., 1988, Trace fossils and sedimentology of a
Late Cretaceous progradational barrier island sequence: Bearpaw-Horseshoe Canyon
Formation transition, Dorothy, Alberta: Canadian Society of Petroleum Geologists Field
Trip Guide, 166 p.
SCHOPF, J.M., 1973, Coal, climate, and global tectonics, in Tarling, D.H., and Runcorn,
S.K., eds., Implications of Continental Drift to the Earth Sciences 1. (London, Academic
Press) p. 609-622.
SCOTESE, c.R., BAMBACH, R.K., BARTON, c., VAN DER VOO, R. and ZIEGLER,
A.M., 1979, PALEOZOIC BASE MAPS: JOURNAL OF GEOLOGY, V. 87, P. 217-277.
STEWART, W.H., 1983, Paleobotany and the Evolution of Plants: Cambridge, Cambridge
University Press, 405 p.
STILLE, H., 1936, The present tectonic state of the earth: American Association of
Petroleum Geologists Bulletin, v. 20, p. 849-880.
STRAKHOV, N.M., 1969, Principles of Lithogenesis, trans. by 1. Paul Fitzsimmons: New
York, Consultants Bureau, 3 vols.
TAYLOR, T.N., 1981, Paleobotany: New York, McGraw-Hill Book Company, 589 p.
TISSOT, B., 1979, Effects on prolific petroleum sources rocks and major coal deposits
caused by sea-level changes: Nature, v. 277, p. 463-465.
TYLER, S.A., BARGHOORN, E.S. and BARRETT, L.P., 1957, Anthracitic coal from
Precambrian Upper Huronian black shale of the Iron River District, northern Michigan:
Bulletin Geological Society of America, v. 68, 1293-1304.
VAIL, P.R., MITCHUM, R.M., JR. and THOMPSON, S., 1977, Seismic stratigraphy and
global changes of sea level, part 4: Global cycles of relative changes in sea level,
in Payton, C.E., ed., Seismic Stratigraphy-Applications to Hydrocarbon Exploration:
Amer. Assoc. Petroleum Geologists Mem. 26, p. 83-97.
VAN WAGONER, J.e., JONES, C.R., TAYLOR, D.R., NUMMEDAL, D., JENETTE,
D.e. and RILEY, G.W., 1991, Sequence stratigraphy applications to shelf sandstone
reservoirs: American Association of Petroleum Geologists Field Conference Guidebook.
VAN WAGONER, J.C., MITCHUM, R.M., CAMPION, K.M. and RAHMANIAN, V.D.,
1990, Siliciclastic sequence stratigraphy in well logs, cores, and outcrops: Concepts
for high-resolution correlation of time and facies: American Association of petroleum
Geologists Methods in Exploration Series No.7, 55 p.
VISHEMIRSKY, V.S., 1978, Stratigraphic distribution of fuel mineral products: Geology
and Geophysics, v. 6, p. 3-9. (In Russian).
WARD, e.R., 1984, Coal Geology and Technology: Melbourne, Blackwell Scientific Pub-
lications, 345 pp.
ZIEGLER, A.M., SCOTESE, e.R. and BARRETT, S.F., 1983, Mesozoic and Cenozoic
paleogeographic maps, in Brosche, P., and Syndermann, J., eds., Tidal Friction and the
Earth's Rotation II: Berlin, Springer-Verlag, p. 240-252.
ZIEGLER, A.M., RAYMOND, A.L., GIERLOWSKI, T.C., HORRELL, M.A., ROWLEY,
D.B. and LOTTES, A.L., 1987, Coal, climate, and terrestrial productivity: the present
and early Cretaceous compared, in Scott, A.e., ed., Coal and Coal-bearing Strata: Recent
Advances: Geological Society of London Special Publication No. 32, p. 25-49.
CHAPTER 7
Sequence Stratigraphy and Sea-Level

Changes in the Early to Middle Triassic
of the Alps: A Global Comparison
Thomas RUffer and Rainer ZUhlke
ABSTRACT: During Early and early Late Triassic times, the Northern Calcareous Alps
(Austria, southern Germany) and the Dolomites (northern Italy) were situated at the
margin of the western Tethys. In the Scythian, widespread clastic-carbonate deposition
on the shelf prevailed. Carbonate ramps revived in the earliest Anisian. From the late
Anisian to the early Ladinian, carbonate ramps evolved to rimmed carbonate platforms.
The Dolomites comprise five Scythian sequences, controlled by low amplitude sea-level
changes and progressively increasing tectonic subsidence rates. During the Anisian to
Ladinian, the sea-level fluctuations increased in amplitude. Five Anisian, three Ladinian
and two early Carnian depositional sequences developed. Tectonic subsidence rates
changed significantly over intervals of 2-5 Ma in the northwestern Dolomites, but de-
veloped steadily in the northeastern Dolomites. The Northern Calcareous Alps comprise
two Scythian, five Anisian, four Ladinian and two early Carnian depositional sequences.
The completely marine succession were only weakly affected by early tectonics. A dis-
tinct increase in subsidence occurred in the late Ladinian, leading to the change from
distally steepened ramps to rimmed platforms. Only during this time interval, a rapid
tectonic subsidence signal overprinted the sea-level signal.
Depositional sequences in the Early to early Late Triassic of the Northern Calcareous
Alps and the Dolomites can be correlated, supported by biostratigraphic data. Local
controls, for instance varying subsidence rates, were either subdued or can be accounted
for by comparing different sections within one study area or both study areas as a whole.
Deposition in the northwestern Tethys realm was strongly controlled by basinwide sea-
level fluctuations. However, this need not imply eustatic control. In order to assess global
sea-level changes, data from the northwestern Tethys have been compared to sea-level
data from other Pangean margins. Although biostratigraphic resolution in other basins is
limited, depositional sequences of other basins in the northwestern and eastern Tethys,
epicontinental seas and the Arctic Sea can be correlated during much of the Scythian to
early Carnian: the late Scythian to latest Anisian/earliest Ladinian, the late Ladinian to
earliest Carnian and the late early Carnian. Correlative sequences in these basins suggest,
that eustatic sea-level changes had a major influence on the development of depositional
systems during Scythian to early Carnian times.
161
B. U. Haq (ed.),
162 Th. Ruffer, R. Zuhlke
Fig. 1. Locations of the study areas in the Alps of central Europe. The Northern Calcareous Alps
are indicated by the upper, the Dolomites by the lower rectangle.
1. Introduction
During Early and Middle Triassic times, the western Tethys and adjacent epi-
continental basins extended over a large part of Europe. The two study areas,
the Northern Calcareous Alps (Austria, southern Germany) and the Dolomites
(northern Italy) were situated at the margin of the western Tethys. Important sedi-
mentological and paleogeographical changes took place in these regions in the late
Scythian to early Carnian time. In the Scythian, widespread clastic-carbonate sedi-
ments were deposited on the shelves. Carbonate ramps, that had ceased to function
due to the global sea-level fall near the Permiantrriassic boundary, were revived in
the earliest Anisian. In the late Anisian to early Ladinian, carbonate ramps evolved
into rimmed carbonate platforms. In the early/middle Carnian, the platforms were
affected by strong input of terrigenous clastics and partly experienced subaerial
exposure. The Scythian to early Carnian interval spans approximately 15-20 Ma,
depending on the radiometric time scale applied (see for example, Forster and
Warrington, 1985; Haq and others, 1987, 1988).
The first objective of this paper is to present an integrated litho- and sequence
stratigraphy of the Early and Middle Triassic of the Northern Calcareous Alps and
the Dolomites (Figs. 1 and 2). Both areas have been studied in detail over the
last decade and a dense grid of measured sections and lateral outcrop observations
exist. The second objective is to analyze Tethys-wide sea-level changes by COffi-
Early and Middle Triassic Sea-Level Changes 163
Geologic-Tectonic Units in the Alps Vie,.,.

M<ric:h
o 50 100
"----'---'
km
IZiZ!:!l fielvelic Domain

_ AIpne l'ilrusives
Fig. 2. Major geologic-tectonic units of the Alps.
paring data from other Tethyan areas and adjacent epicontinental basins. Finally, a
comparison with data from other basins, situated in different parts of the Pangea,
lead to an assessment of possibly global sea-level changes in the Early and Middle
Triassic.
The first estimates of global sea-level changes were presented by Vail and
others (1977). The Early and Middle Triassic interval was not resolved to the level
of third-order sea-level changes. The revised sea-level chart published by Haq and
others (1987, 1988) estimated global sea-level changes for the Early to early Late
Triassic based on data from the Dolomites (Anisian to Carnian), the Salt Range
(Pakistan, Scythian), and from Bear Island and Spitsbergen (Scythian to Carnian).
Hirsch (1992) compared eastern Mediterranean data with depositional sequences
from the northwestern Tethys and the Canadian Arctic.
Brandner (1984) correlated relative sea-level changes in the northwestern
Tethys, especially from the Northern Calcareous Alps, the Dolomites and sections
from Iran. In general, he followed the earlier study of Schlager and Sch611nberger
(1974), who subdivided the northern alpine Triassic into four depositional units,
bounded by three events in the late Scythian, late Anisian and early Carnian.
These northern alpine "depositional units" essentially reflect second-order peak
transgressions and maximum regressions (Ruffer and BechsUidt, in preparation).
Bechstadt and others (1993a, 1993b) and Ruffer and others (1994) assessed the
basinwide and possibly eustatic controls on sedimentation in the Early to early
Late Triassic.
2. Geological Setting
The paleogeographic setting of the Middle Triassic western Tethys realm and
adjacent areas has been compiled, among others, by Ziegler (1982, 1988) and
Dercourt and others (1993). Both, the Dolomites (Italo-Dinarid domain) and the
Northern Calcareous Alps (Austro-Alpine domain) form a part of the Triassic
European continental margin overlying continental crust. This wide carbonate and
partly evaporitic shelf formed the northwestern margin of the Tethys ocean. The
exact spatial relationship between the Italo-Dinarid domain, the Austro-Alpine
and other Tethyan domains at the southern margin of the European continent is
still considered unresolved.
Today, the Italo-Dinarid and Austro-Alpine domains are separated by the
major Periadriatic lineament, which extends over more than 1000 km in east-west
direction (Fig. 2). In the Late Cretaceous to Paleogene, oblique collision between
Africa and Europe resulted in a left-lateral offset of several hundred kilometers
along this lineament.
The oceanic Tethys was situated in considerable distance further to the south-
east. Deep marine entrants, bounded by normal faults and mostly underlain by
continental crust existed in the Hellenides and Dinarides (Greece, Bosnia, Croa-
tia). These features indicate a tensional stress field at the eastern margin of the
Pangean supercontinent ("Tethys Rift" sensu Ziegler, 1988). Aborted rifting ac-
tivity (Bechstadt and others, 1978) has been made responsible for the facies evo-
lution of the Middle Triassic in the Dolomites and other parts of the Southern
Alps (Italo-Dinarid block sensu Ziegler, 1988). However, geochemical studies
on early/middle Longobardian plutonic and volcanic rocks in the Southern Alps
(Dolomites) have not yielded unambiguous data for the paleogeographic recon-
struction of the Triassic western Tethys area. An evolutionary trend from early
calc-alkaline rhyolites/andesites to late shoshonitic basalts (Castellarin and oth-
ers, 1988), simultaneous transpressive and transcurrent tectonics (Doglioni, 1984,
1987; Blendinger, 1985) and the arcuate distribution of volcanics seem to be in-
compatible with rifting. According to the model of Marinelli and others (1980)
and Castellarin and others (1988) the Triassic central Southern Alps represent a
wide back-arc basin.
Further to the north, several epicontinental basins extended well onto the Eu-
ropean plate (Germany, southern North Sea, southern France, eastern Spain). They
were connected to the western Tethys by a few shallow marine and periodically
open seaways (e.g. Burgundy seaway, Silesian seaway). In contrast, the Triassic
Tethys was not connected to the Arctic Sea north of the Laurentian-Greenland and
Fennosarmatian cratons.
3. Methods
The Northern Calcareous Alps and the Dolomites are quite well exposed. About
100 sections have been measured in the field yielding interpretations of lithofa-
cies (specific particles, sedimentary structures, energy levels, paleoecology) and
depositional environments. Relative changes of in-situ carbonate production have
been assessed based on the presence of bioconstructions, dysaerobic to anoxic
conditions, and input of fine-grained terrigenous clastics or distal volcaniclastics.
The composition and frequency of carbonate turbidites in slope or middle to lower
ramp deposits indicate relative changes in sediment input (detrital carbonates).
Subaerial unconformities (fluvial deposition, erosion, karstification), drowning

unconformities (boring surfaces, abrupt shifts to deep marine and/or anoxic con-
ditions), correlative conformities and stratal patterns have also been taken into
account. Large-scale stratal patterns (e.g. on lap, downlap) are rarely visible in the
late Scythian to late Anisian time strata. This is mainly due to relatively small
depositional gradients (carbonate ramps) and to the restricted lateral extension
of many of the outcrops. In addition, some stratal patterns have been destroyed
during burial diagenesis and dolomitization. Large-scale lateral outcrops with well
preserved geometries are usually restricted to the Ladinian. In the Dolomites, they
have been analyzed in a number of published studies (Bosellini, 1984; Yose, 1991;
Biddle and others, 1991; Doglioni and Bosellini, 1989; Doglioni and others, 1990).
Based on analyses of lithofacies and depositional environments, measured
sections have been subdivided into regressive and transgressive cycles. They rep-
resent paleobathymetric changes, which may differ from relative sea-level changes
in the sense of Posamentier and others (1988). Changes in water depth do not nec-
essarily represent changes in accommodation ("accommodation envelope" sensu
Posamentier and others 1988) due to varying sediment supply (cf. Schlager, 1993).
However, small-scale transgressive/regressive cycles can usually be correlated
throughout each study area. Large-scale transgressive/regressive trends can be
correlated between the Northern Calcareous Alps and the Dolomites, even though
they represent two tectonically and paleogeographically different realms. These
facts indicate that transgressive/regressive cycles in both study areas represent
relative sea-level changes. In some stratigraphic intervals, e.g. the early Anisian,
cycle stacking patterns in carbonate-clastic ramp deposits allow a direct analysis
of accommodation changes.
Main parts of the sequence-stratigraphic interpretations given in this paper
are based on our own field studies in the Dolomites and the western part of
the Northern Calcareous Alps. Reinterpreted sections from other authors served
as additional database, especially for the late Ladinian to early Carnian of the
Southern Alps and the late Scythian to early Anisian of the Northern Calcareous
Alps. Biostratigraphic data based predominantly on ammonites and conodonts
provide good chronostratigraphic control on depositional sequences.
4. Dolomites
The Dolomites are situated in northernmost part of Italy and represent the central
portion of the Southern Alps, which extend for 500 km in east/west direction
(Fig. 2). The study area covers an area of about 120 to 80 km (Figs. 3 and 4), that
was little affected by Alpine tectonics compared to the more internal or northern
parts of the Alps. The sedimentary record comprises middle Permian to early
Tertiary deposits.
4.1. PREVIOUS STUDIES
Broglio Loriga and others (1983, 1990) and Neri and Posenato (1985) studied the
lithofacies and biostratigraphy of the Scythian Werfen Formation. Bosellini (1968),
Bechstadt and Brandner (1970), Pisa and others (1978), Masetti and Neri (1980),
Blendinger (1983) and Farabegoli and de Zanche (1984) provided stratigraphic
control and lithofacies information on the Anisian of the northern, southern and
central Dolomites. The lithofacies and paleogeography in the northern Dolomites
were reevaluated by Zi.ihlke in Senowbari-Daryan and others (1993). Studies on
the whole Triassic succession by Assereto and others (1977), Gaetani and others
(1981), Casati and others (1982), de Zanche (1990) and de Zanche and Farabegoli
(1988) also include data on Anisian lithofacies and paleogeography. Sequence-
stratigraphic investigations in Anisian deposits were performed by de Zanche and
others (1992, 1993) and Zi.ihlke in RUffer and others (1994). The Ladinian and
Carnian sequences have been extensively studied, for instance by Bosellini and
Rossi (1974), Viel (1979a, 1979b), Fois and Gaetani (1982), Blendinger (1986),
Goldhammer and Harris (1989), Bosellini and Neri (1991), Bosellini and Stefani
(1991), and Masetti and others (1991). The biostratigraphy of the late Scythian
to early/middle Carnian in the Dolomites and adjacent areas was established by a
number of workers, especially Brack and Rieber (1986, 1993, 1994), Mietto and
Manfrini in de Zanche and others (1993), Broglio Loriga and others (1983, 1990),
and Pisa and others (1980).
4.2. REGIONAL GEOLOGY
4.2.1. Permian
The post-Hercynian sedimentary history of the Dolomites started in the middle
Permian with terrigenous clastics of an alluvial plain (Groden Formation). Cyclic
sabkha and subtidal open-marine ramp deposits of the Bellerophon Formation were
deposited in the late Permian (see also Fig. 13). At the PermianlTriassic boundary,
the coastline of the Tethys was situated immediately west of the Dolomites along
the Etsch and Eisack valleys (Fig. 3).
4.2.2. Scythian
A renewed long-term sea-level rise shifted the coastline further west. The initially
carbonate and subsequently clastic-carbonate ramp of the lower and middle Wer-
fen Formation came into existence. The lithofacies subdivision is based on the
intercalations of periodic intertidal to supratidal fiats, that interrupted the prevail-
ing subtidal deposition above storm wave base. Intervals of regional tectonic uplift
south of the Dolomites (Recoaro area and Po Plain subsurface; de Zanche and
Farabegoli, 1981) resulted in a strong input of terrigenous clastics (e.g. Campil-
Member).
4.2.3. Anisian
Carbonate-clastic and carbonate ramps first developed in the Aegean and Bithynian
(Lusnizza, Lower SarI and Gracilis Formations). Around the BithynianlPelsonian
boundary, ramps were locally emergent and subject to erosion and karstification
(northern, southern Dolomites). After incised valley, lower shoreface and subtidal
Dolomites / Southern Alps (Italy) : schematic map

Prags Toblach
N
1
Sennes
Puez M.
Odles Popera
Fanes
Garde-
Cavalese

Predazzo
o~
10
(km)
Fig. 3. Schematic map of the Dolomites (Southern Alps, Northern Italy).
deposition (Voltago, Agordo and Recoaro Formations) distally steepened ramps

with higher depositional gradients evolved in middle/late Pelsonian times (Upper
Sari Formation). Other ramp areas experienced a complete drowning and sub-
sequent deepening upward since the late Bithynian, shown by the presence of
deposition in narrow basins and marine inlets (Dont, Ambata, Bivera, and Moena
Formations). At the Pelsonian/Illyrian boundary, ramps in the eastern, northern and
southern Dolomites again became subaerially exposed. In the western Dolomites,
Anisian deposition only started after a major depositional and erosional hiatus. In
the Illyrian, alluvial, lower shoreface and subtidal shelf environments (Richthofen
and Morbiac Formations) prevailed once again and evolved into distally steepened
ramps and platforms (Contrin Formation). Contrin platforms usually experienced
complete drowning. Pelagic anoxic deposition covered wide parts of the basin
in the late IlIyrian (Buchenstein Formation, ita!. Livinallongo Formation). Few
sections in the northeastern and southwestern Dolomites show subaerial exposure
and minor erosion at the top of the Contrin platform.
4.2.4. Ladinian
Carbonate platforms (Schlern Formation, ita!. Sciliar Formation) initially devel-
oped in areas of subaerial exposure of the Anisian platforms. From these areas,
three platform stages aggraded and prograded into adjacent basins (Buchenstein,
Zoppe, Aquatona and Fernazza Formations). Topographic relief in the late Fassa-
nian to early Longobardian reached up to 800-1000 m (Bosellini and Rossi, 1974;
Bosellini, 1984; upper Buchenstein Formation). Platform lagoons and basins show
small-scale shallowing upward cyclicity (5th, 4th order). Discontinuous reef mar-
gins with mainly encrusting and baffling reef organisms (algae, microproblematica)
bordered the platforms. Prograding slope deposits, dominated by megabreccias,
Dolomites I Southern Alps (Italy) : locations of sections
N
~,,: ...... .
r.==c;;-N~
o section, inoorporated
from the lITerature
Fig. 4. Locations of sections in the Dolomites (Southern Alps, Northern Italy). Sections measured
in this study are indicated as small black dots. Large open circles represent sections mentioned in the
text. Small open circles indicate sections incorporated from the literature (Bechst!1dt and Brandner
1970, ZUhlke in Senowbari-Daryan and others, 1993; Pisa and others, 1978; B1endinger, 1983).
reached gradients of up to 35. In middle Longobardian times, subaerial volcanic

activity in the southern Dolomites abruptly interrupted further carbonate platform
buildup. Platforms experienced subaerial exposure and/or were suffocated by vol-
caniclastics. Basins received large quantities of clastics (Wengen Group) when a
subaerial volcanic high (Predazza, Monzani) in the Southern Alps was gradually
eroded. Carbonate production on isolated platforms or platforms that were attached
to volcanic islands, returned in the late Longobardian.
4.2.5. Carnian
Major platform development only returned in the early Carnian (Cassian Dolomite).
The strong progradation of early Carnian platforms completely filled the basins of
the San Cassiano Formation in the western Dolomites. In the middlellate Carnian,
the vast carbonate platform domains of the Cassian and Dilrrenstein Dolomite
(cyclic interior lagoon) were subaerially exposed, karstified and covered by ter-
rigenous clastics (alluvial plain, shoreface), evaporites and marginal marine to
shallow subtidal carbonates (Raibl or Schlernplateau Formation).
4.3. FACIES AND SEQUENCE STRATIGRAPHY OF SELECTED SECTIONS
The Early and Middle Triassic facies and sequence stratigraphy of three key
sections called Selvapiana, Elferschartenspitze and Pufels will be discussed in
greater detail.
~:::~~m~i~~Part) =~ BH~W;::;OR c.p~

~-0m I ~
~l;i~~ ~~ Jl j: ll~
...j. ~
....: .. I I ISf;:.?,
... I
...
~~
..i
.. "HOT

I
"
1--'"I----'
t- I I~
~
it t
l.---- .1
1---
_.1
1........-
I ~
.-k---.1
I
= p- j
.--
.- ~
I .- ~ J
Fig. 5. The Selvapiana section in the northeastern Dolomites (cf. Figs. 6 and 7). Total thickness is
850 m. The section is split into two columns. The right column depicts the lower, the left column the
upper part of the section; for legend see Fig. 20.
Fig. 6. The Selvapiana section in the northeastern Dolomites (cf. Figs. 5 and 7; see text).
Fig. 7. The Selvapiana section in the northeastern Dolomites (cf. Figs. 5 and 6; see text). The
photograph depicts the late Bithynian to middle Pelsonian part of the section. The Lower Sari
ramp was incipiently drowned. After deep subtidal deposition (Dont Formation), carbonate ramp
conditions returned with the Upper Sari Formation.
4.3.1. The Selvapiana Section

This section (Figs. 5, 6 and 7) covers a 900 m thick late Scythian to late Anisian
ramp/platform succession in the northeastern Dolomites. The lower part of the San
Lucano Member indicates a middle ramp to shoreface depositional environment.
Bindstones with microbial mats mark a shift to intertidal flat deposition, that is
interpreted as type 2 sequence boundary (Sc 5/An 1). The uppermost San Lucano
Member and the lower/middle Lusnizza Formation document shallow subtidal to
peritidal deposition with evaporitic conditions and minor subaerial exposure re-
sulting in solution and collapse breccias (An 1, shelf margin wedge). Overlying
middle ramp carbonates were deposited below storm wave base when accom-
modation increased (An 1 transgressive systems tract). The uppermost Lusnizza
and basal Lower SarI Formations represent a renewed regressive trend with pe-
riodic flux of storm deposits, input of shallow-marine particles, and decreasing
input of terrigenous clastics (An 1, highstand). The An 2 shelf margin wedge
and transgressive systems tracts show a progressive shift to lower ramp environ-
ments, with some distal carbonate turbidites from shallow marine areas. Strong
input of fine grained terrigenous clastics and two deepening-upward macrocycles
(4th order?) mark the maximum flooding surface. Mud- and wackestones from
lower and middle ramp areas with sparse distal storm deposits dominate the re-
maining Bithynian. Aggradation prevailed during the An 2 highstand. Around
the BithynianlPelsonian boundary, the Lower Sari ramp was incipiently drowned
(Fig. 7). Middle ramp deposits evolve gradually but rapidly to hemipelagic deep
subtidal deposits (Dont Formation). Turbidites with echinoderm debris and distal
volcaniclastics were intercalated. Radiolarians, pelagic bivalves and a maximum
of shale deposition characterized the An 3 maximum flooding surface. Pelsonian
highstand deposits are represented by the Upper Sari ramp, that aggraded from
subtidal to shallow subtidal depths. Some progradation towards adjacent narrow
basins occurred during the late highstand. Sequence boundary An 3/An 4 is marked
by subaerial exposure and karstification. Subtidal terrigenous clastics of the An 4
transgressive systems tract directly overlie the An 3 highstand deposits of the
Upper Sari Formation. Aggradation once again became dominant in the middle
and upper Morbiac Formation, that had already aggraded to sea-level (microbial
bindstones). The An 4/An 5 sequence boundary (not illustrated in Fig. 5) is marked
by minor subaerial exposure of the uppermost Contrin Formation.
Fig. 8 shows accommodation plots ("Fischer plots", cf. Sadler and others,
1993) with a litho- and cyclostratigraphic correlation of 5th order shallowing
upward-microcycles in the Werfen, Lusnizza and Lower SarI Formations of three
sections (Selvapiana, Cervera, Innergsell, compare Fig. 4). Although the number
of cycles varies, individual systems tracts of the sequences Sc 5 to An 2 can be
well defined and correlated.
4.3.2. The ElJerschartenspitze Section

The Elferschartenspitze section (Fig. 9) covers a 200 m thick early Anisian to early
Ladinian succession in the northeastern Dolomites. In contrast to the Selvapiana
section, the late Anisian interval comprises lower slope and basinal sediments.
Selvaplana section
NEDoIomiles /71 cycles
Innergsell seclion
NEDoIomlles /80 cycles
Cervera section
E-Dolomiles / SO cycles
Lithostratigraphic correlation I
Werten Formation Lusnlzza Formation Lower Sa~ Formation
~
Selvapiana section
NEDoIomlle.171 cycles
---- ~
~ ---
~ Innergsell section
NE DoIomlles /80 cycles
Cervera section
EDotomites /50 cycles
I Accommodation trends I correlation I

HST SMW TST HSl SMW 1ST
Sc5 An 1 An2
Fig. 8. Accommodation plots ("Fischer plots") covering the uppermost Werfen Formation to Lower
SarI Formation (Dolomites). The same accommodation is plotted twice. The upper sketch illustrates
a lithostratigraphic correlation. The lower sketch illustrates a correlation based on characteristic
long-term trends, which represent 3rd-order changes in accommodation. Each triangle represents a
single 5th-order cycle, plotted with the assumption of steady total subsidence in the time interval
depicted (cf. Fig. 13)
During An 2 highstand times, middle and lower ramp environments prevailed.

The top of the Lower SarI Formation shows a drowning unconformity (sequence
boundary An 2/An 3) with a small-scale submarine relief. Transgressive deposits
of sequence An 2 are condensed and much thinner than on the ramps (cf. Selva-
piana section, Fig. 5). The maximum flooding surface is located in the basal Dont
Formation. The middle and upper Dont Formation shows two parasequences (Ps)
with progressive shallowing upward of the basin (early highstand An 3). Above the
flooding surface of parasequence 2, debris flows and turbidites parasequences doc-
ument a stepwise progradation of the Upper Sari ramp (late highstand An 3). The
overlying Dont Formation documents a renewed progressive deepening upward to
pelagic, partly anoxic, basinal conditions with increasing amounts of shale (An 4,
transgressive systems tract). Open-marine basinal deposition with an increase in
carbonate turbidites shed from shallow marine platform areas to the east (Contrin
platform, Selvapiana section) marks the highstand of sequence An 4 (Ambata For-
mation). Sequences An 4 and An 5 are separated by a correlative conformity. Due
to the basinal position of the section, no shelf margin wedge deposits can be de-
fined. The condensed transgressive systems tract deposits (An 5) comprise basinal
mud-/wackestones of the B uchenstein Formation. Thick grain/rud- to wackestones
with graded bedding document a rapid infill of the late Anisian basin during the
An 5 highstand, when adjacent Schlern platforms were prograding (Selvapiana).
4.3.3. The Pufels Section

The Pufels section (Fig. 10, cf. Figs. II and 12) covers a 200 m thick middle
Scythian to early Ladinian succession in the western Dolomites. The base of
the section shows the middle Scythian (Smithian) Campil Member of the Wer-
fen Formation. It is bounded by a low angle (approx. 5) subaerial unconformity
(sequence boundary Sc 3/An 4). Around the PelsonianlIllyrian boundary time,
incised valleys (cf. Fig. 12) in the erosion surface were filled with alluvial con-
glomerates and breccias (An 4 lowstand systems tract). The transgressive surface
is located in the overlying sand-/siltstones, that were deposited in alluvial to lower
shoreface environments (lower/middle Richthofen Formation). Shoreface deposits
to subtidal bioturbated mud- and wackestones constitute the An 4 transgressive
systems tract. The maximum flooding surface occurred near the top of bitumen-
rich ramp wackestones deposited in subtidal anoxic conditions of the top Morbiac
Formation. During An 4-highstand times subtidal to shallow subtidal middle ramp
conditions prevailed. Finally, the Contrin Formation experienced complete drown-
ing. Anoxic, hemipelagic shales and carbonates of the Moena Formation abruptly
cover the Contrin ramp (drowning unconformity "I"). Debris flows and detrital
carbonates characterize the middle and upper Moena Formation. The An 4/An 5
sequence boundary was produced by a rapid, two-step increase in accommodation
(drowning unconformity "II"). Thin carbonate and terrigenous clastics (Ambata
Formation) and radiolarian wackestones with nodular chert and distal volcani-
clastic input (Buchenstein Formation, "Nodular Limestone" Member) indicate a
progressive deepening of the depositional environment to open marine, basinal
conditions (An 5 transgressive systems tract). Adjacent sections in the western
Dolomites show turbidites and finally slope clinoforms of the Schlern Formation
prograding into the Buchenstein basin from late Fassanian times onward (early to
late An 5 highstand, cf. Fig. II).
EHerschMenspitze section
Iia il (-- (--
I
PArtICles DIp. II"lVlronmenl
I----'-no:.c,:..::rth.:.;:9.::;8s:..:;t9:.;,m::oDo;:..::;,lom=it.::;es= ----1
~i ill i j
..
E
""
e.,

<
>- .,<

.. ""-
~ E
0. ~
~ c
E
E
""
Fig. 9. The Elferschartenspitze section in the northeastern Dolomites (see text). Total thickness is
200m.
Fig. 10. The Pufcls section in the northeastern Dolomites (see text, cf. Figs. 11 and 12). Total
thickness is 200 m.
Fig. 11. Western flank of the Rosengarten (Ita!. Catinaccio, Dolomites).The Rosengarten section is
comparable to the Pufels section as far as the Scythian to early Fassanian sequences are concerned
(cf. Fig. 10). In contrast to the model of Bosellini and Stefani (1991), no subaerial exposure or
karstification is documented at the top of the Contrin Formation. Thin, anoxic, basinal carbonates
abruptly overlie the Contrin Formation (drowning unconformity "I", cf. Pufels section, Fig. 10).
Breccias at the top of the Contrin Formation are related to the collapse of the carbonate ramp. They
are again overlain by anoxic basinal carbonates ("Platy Limestone Member", drowning unconformity
"n", cf. Pufels section, Fig. 10) and basin to basin margin deposits ("Nodular Limestone" Member)
ofthe Buchenstein Formation. Adjacent to the area shown in the photograph, the Contrin Formation
shows ramp-interior basins, with large-scale slump structures (several tens of meters).
4.4. CONTROLS ON SEDIMENTATION IN THE DOLOMITES
Fig. 13 illustrates the tectonic and total subsidence, the paleobathymetry, sea-level
history and 3rd-order sequences for two representative sections covering the late
Permian to late Triassic time interval in the northeastern and northwestern Dolomi-
tes (see also Figs. 13,22). Only the Scythian to early Carnian interval, which is the
topic of this paper, will be discussed. The reconstructions are based on incremen-
tally decompacted sediment thicknesses and on paleobathymetric trends inferred
from lithofacies changes and stratal patterns. Regional sea-level fluctuations have
been interpreted from accommodation analyses, subaerial unconformities with sub-
sequent fluvial or terrigenous deposition and from lithostratigraphic correlations
to other parts of the Southern Alps (de Zanche and others, 1993; Gaetani, 1993).
Amplitudes were estimated based on the facies and thickness of regressive units
and the interpreted paleobathymetric shift between maximum flooding and max-
imum regression. Subsidence histories for the Triassic of the Dolomites have not
Fig. 12. The Sass Rigais section in the northwestern Dolomites is comparable to the lower part
of the Pufels section (see text. Fig. 10). An incised valleys of sequence An 4 cuts deeply into the
underlying highstand deposits of the Gastropodenoolith Member (Werfen Formation). The opposite
side of the valley is covered under scree in the left foreground. Incised valleys were filled during the
lowstand. Peritidal flat to carbonate-clastic subtidal deposits (Morbiac Formation) are overlain by
the carbonate ramp of the Contrin Formation.
been modeled before (cf. estimates in Bosellini, 1984; Doglioni and Goldhammer,
1987).
4.4.1. Scythian to Ladinian in the Northeastern Dolomites

The early Scythian sequence Sc 1 is characterized by tectonic and total subsidence
rates of 30/40 b (b=Bubnoff units, mmlka). Deposition in the middle and late
Scythian sequences Sc 2 to Sc 5 showed a gradual increase in tectonic and total
subsidence rates to 60-100 and 100-200 b, respectively. Small amplitude 3rd-order
sea-level changes (10-20 m) produced four type 2 sequences which are correlat-
able over wide parts of the Southern Alps. Accommodation steadily increased
from the late Scythian to the early/late Anisian boundary. Tectonic subsidence
rates ranged between 100-150 b. The underlying Permian to Scythian deposits
experienced strong compaction, leading to high total subsidence rates (max. 300 b,
compaction-induced subsidence). An abrupt increase in tectonic subsidence to
200-250 B in the late Bithynian triggered the drowning of the An 2-highstand
ramp. Amplitudes of sea-level changes during the early Anisian were probably
higher (20-40 m?) than those during the Scythian. In the late Anisian, tectonic
and total subsidence rates were reduced (-10 b to 30 b, 0-70 b). Accommoda-
tion space was predominantly controlled by two 3rd-order sea-level changes with
relatively larger amplitudes (approx. 50 m) than during the Scythian and early
Anisian. No major breaks in subsidence occurred at the AnisianlLadinian bound-
ary or during the Ladinian. Tectonic and total subsidence gradually increased to
50 b, 100 b respectively. Carbonate production could keep up and periodically
outpaced accommodation during the Pelsonian to late Longobardian. Aggrading
and prograding carbonate platforms prevailed. Marine reentrants, documented in
the Elferschartenspitze section (Fig. 9), were local features with narrow widths
(1-3 km) and fault-controlled boundaries. They show a separate subsidence de-
velopment not illustrated in Fig. 13.
4.4.2. Scythian to Ladinian in the Northwestern Dolomites

Scythian subsidence trends and rates in the northwestern Dolomites were fairly
similar to those in the northeastern Dolomites. In the early Anisian (Aegean), an
abrupt reversal in basin subsidence took place. It was probably triggered by the
activation of several strike-slip faults in the northern and central Dolomites (trans-
pression/transtension, cf. B1endinger, 1983, 1985; Doglioni, 1984, 1987). Uplift
with rates of -50 b to -10 b resulted in the development of a regional horst block
("dorsale badioto-gardenese", Bosellini, 1968). On this high, the erosional surface
was continuously above the fluctuating sea level from Aegean to late Pelsonian
times. In the northern Dolomites, where tectonic uplift was less pronounced, sub-
aerial exposure with stream rejuvenation in incised valleys or shoreface deposition
occurred at the base of the Piz da Peres Conglomerate and the Voltago Formation.
Sequence boundaries An 1/An 2 and An 2/An 3 may be interpreted in terms of
sea-level falls (10-20 m, 20-30 m), that were superimposed on tectonic uplift, or
on reduced tectonic subsidence. Deposition on the high in the northwestern Dolo-
mites resumed, when tectonic uplift changed to subsidence in the early II1yrian
(0-20 b) and was superimposed by sea-level fluctuations with higher amplitudes.
Two major breaks in subsidence took place in the late Anisian and the Ladinian. In
the late IIIyrian (ReitziiSecedensis Zone), an abrupt increase in tectonic subsidence
to 150-200 b occurred. The associated increase in accommodation was enhanced
by a marked 2nd-order sea-level rise (cf. regional sea-level curve in Fig. 13) .
Only in few areas could carbonate production keep up with rising sea-level in the
latest Anisian (e.g. Latemar, northern M. Popera Range, northern Rosengarten, cf.
Fig. 3), so that the Contrin ramp is directly overlain by the early Schlern platform.
Contrin ramp areas, that could not match the late II1yrian increase in accommo-
dation, were incipiently or completely drowned for the duration of the Ladinian
and covered by basinal deposits of the Buchenstein Formation (cf. Pufels sec-
tion, Fig. 10). Tectonic subsidence gradually decreased until the late Fassanian.
Basins experienced a deepening-upward trend (cf. Bosellini, 1984). Subsidence
rates continued to slow down in the early Longobardian due to the development
of a volcanic high in the southwestern Dolomites (Predazzo; see Viel, 1979 a,b;
Yose, 1991; Gaetani and others, 1981). Tectonic uplift reached up to -100 b in
the late Longobardian, resulting in subaerial exposure and karstification of plat-
forms. However, the correlation of sequence boundaries between the Dolomites
and western parts of the Southern Alps (e.g. Lombardy) indicates, that even in
200
400
rtectoniCSUbsidence 600
1000
1200
O Subsidence history
NW . Dolomites
L-------l
/-tot&ISUbSidence
1400
1600
1800
c----
~~ H
Cassian
Bellerophon Formation Werlen Formation ~~ Schlem FormatiOn
~~ ~iI'
Ii :;
Dolomite ~" Z
II:
Low. Sari Fm.
~:
T"i,l, ~~
~
~.r!:; ~~
"
"Trans/barwl Badiota Sail I
Member"
Campi!
Mill. Anchz a.lrap. V. Sa .' ~~ I III
"~"" !!.
iiI~~g iJg ~~f ~~g Ii .. filS; (~ i.g ii~i j;~~~
~a!"OB
t~t~1 f ii
~.
.~
"~~l
it i. 0
i&:ep- a "U[
~! .~=
3~~ ~f it~l I~a !~i i!.
~!h i!~ ~.
." .
""i!5'iilD.
!h~~
j~
~H~ ~crl'i ~ ~.~ ~J i!a~ ~~i jI&
'---
255 250 245 240 235 230 22S 220 215
-100
regional sea-lel/81 Time (maJ

~balhymetry aflerGradslein 200
burial depth 400

1m]
600
.00
nedata 1000
(present-day
erosional 1200
surface)
1400
1600
~::
1800
2000
50b
r lotal subsidence
2SOb
100b 2200
~ Subsidence history 2400

approximate
subsidence
NE Dolomite. rates 2600
1.-------'
1 b[Bubnolf}
",1rnm11t8 >SOO
Zuhlke (in prep.) 3000
Fig. 13. Subsidence histories for the top of the Hercynian basement in the northwestern (A) and north-
eastern Dolomites (D). Data come from various sections in the Sass da Putia and Puez Ranges, the M.
Popera Range respectively (see Fig. 3). The Pufels and Selvapiana sections (Figs. 10,5) illustrate the
late Scythian to early Ladinian sequences in both regions. The modelling of decompacted subsidence
histories is based on a Iithophysical parameter set comprising 15 lithologies, sequence-stratigraphic
and/or lithostratigraphic units, decompacted thicknesses and paleodepth interpretations. The top
of the Hercynian basement ("Drixener Quarzphyllit") has been taken as lithosphere with definite
strength. Schematic columns show the lithostratigraphy and the main depositional environments
in the late Permian to Norian of the northwestern and northeastern Dolomites (cf. Fig. 22). Only
long-term subaerial or submarine hiata are indicated. For the discussion of subsidence histories see
text.
the Longobardian, sea-level fluctuations at least partly controlled deposition in the

Dolomites.
Sediment supply increased from 10 mmlka at the Permiantrriassic boundary to
a maximum of 250 mmlka at the Scythian/Anisian boundary (Ta 2ISc 1 sequence
boundary). This was due to the significant input of terrigenous clastics in the
Smithian (Sc 3) and increasing carbonate production in the Spathian after input
of terrigenous clastics had ceased (Sc 4 to An 1 shelf margin wedge). Rates of
erosion in the Aegean to Pelsonian reached up to 100 mmlka. In the Late Anisian,
deposition returned with an abrupt increase from 10 to 250 mmlka (An 4). After
the drowning of ramps in the Reitzi Zone, low production rates of 10-50 mmlka
prevailed in carbonate basin and basin margin settings (An 5 to La 2 transgressive
systems tract). Longobardian to early Carnian (La 2 highstand systems tract to
Ca 2) slopes and platforms (La 2 highstand systems tract to Ca 2) show maximum
carbonate production rates of up to 400 mmlka. The long-term increase in carbonate
production on ramps and platforms in the early to early late Triassic was connected
to the progressive recovery of reef-building organisms since the PermianlTriassic
reef crisis.
4.4.3. Summary (controls on deposition)

Early to early Late Triassic deposition in the Dolomites was strongly controlled
by fifteen 3rd-order sea-level changes (Sc 1 to Ca 2), superimposed on shifts in
subsidence: (a) increasing tectonic and total subsidence from the early Scythian
to the Aegean or to the late Bithynian (7 to 11 Ma); (b) slow tectonic uplift and
minimized total subsidence from the Aegean or early Pelsonian to the middle 11-
lyrian (3 to 6 Ma); (c) strong, gradually decreasing, tectonic and total subsidence
from the late IIlyrian to the late Fassanian or into the Carnian (4 to >8 Ma) and,
(d) regional tectonic uplift in the Longobardian (3 Ma), giving way to renewed
tectonic subsidence in the Carnian (>5 Ma). Deposition in areas with decompacted
total subsidence rates of more than 150 b and less than 0 b was primarily con-
trolled by varying subsidence/uplift rates and associated processes, e.g. input of
terrigenous clastics from emerged topographic highs. Areas and time intervals with
decompacted subsidence rates between 150 band 0 b show sequences that are cor-
relatable within the Dolomites and with other basins in the NW-Tethys (cf. Fig 23).
Third-order sea-level changes were the main controlling factor of deposition.
5. Northern Calcareous Alps
The Northern Calcareous Alps are situated between the Swiss-Austrian border in
the west and Vienna in the east. They cover an area of about 500 by 60 km and form
the main part of the Upper Austro-Alpine nappe complex (Figs. 2 and 14), which
comprises a sedimentary record from the Paleozoic (Grauwacken Zone) through
the Permo-Mesozoic to the syn- and late Alpidic Gosau sediments and Tertiary
successions.
An abundance of detailed local data has been published on Middle Triassic car-
bonates of the Northern Calcareous Alps (treated extensively by Tollmann, 1976
Middle 10 Early Upper Triassic

--
Kaiser-
gebirge
(Reichenhall. Virgloria. Steinalm . Reilling.

WeUerstein and PMnach Formations)
o 10 20
ISect Ions I km
Fig. 14. Location map for the western part of the Northern Calcareous Alps.
and 1985). In contrast, concise stratigraphic reviews and syntheses are lacking.
The main exception is the important study by Schlager and SchOllnberger (1974),
subdividing the Mesozoic succession into depositional units. With reference to the
Early to Middle Triassic, the depositional units were bounded by events ("strati-
graphic turning points") in late Scythian, late Anisian, and early Carnian times. A
more recent study (RUffer and BechsUidt, in preparation) of the Triassic sequence
stratigraphy in the western Northern Calcareous Alps gives a concise review on
lithologies, depositional environments and biostratigraphic data.
5.1. LATE SCYTHIAN TO ANISIAN DEPOSITIONAL ENVIRONMENTS
The early to middle Scythian comprises coarse terrigenous clastics (Lower Alpine
Buntsandstein, Fig. 15) in the western part and marine clastic-carbonate sedi-
ments (Werfen Formation) in the eastern part of the Northern Calcareous Alps.
The analysis of sedimentary structures, lithofacies and paleocurrent directions of
the Scythian Lower Alpine Buntsandstein indicate a braided river system with
the main source areas of terrigenous clastics to the north and probably to the
west (Stingl, 1987). In the western part of the Northern Calcareous Alps a first
marine ingression occurred in the uppermost Lower Alpine Buntsandstein. After
the deposition of the fluvial Upper Alpine Buntsandstein, transgression caused
the change to fluvial dominated estuaries, followed by shallow marine to tidal
flat clastic-carbonate sediments (Werfen Formation). The late Scythian (? to early
Anisian) Reichenhall Formation was formed in a restricted marine depositional
environment with episodic evaporitic conditions (Spotl and Burns, 1991).
Within the Anisian, a shallow marine subtidal to peritidal shelf and homoclinal
ramp (Virgloria and Steinalm Formation, Fig. 15) formed. Carbonate successions
were deposited in an area of little relief, gently dipping to the east. Therefore, more
open marine deposition took place in eastern parts of the Northern Calcareous
Alps (Gutenstein and Hallstatt Formations), a trend that was persistent throughout
Stratigraphic Depositional
column environment
(main I~hology) (lithostratigraphic units)
Carn ian:
mixed siliciclasticcarbonate
shelf margin sedimentation
(Ralbl Group)
latest Anisian Early

Carnian:
distally steepened
ramps (Fassanian) and
rimmed platforms
(WeHerslein Fm.).
carbonate basins
(Reijling Fm.). shaly
basins (Pannach Fm.)
An isian :
homoclinal ramp
(without clastics
Steinalm Fm . wilh
shales Vlrgloria Fm.)
Scythian
(1 to early Anisian):
lIuvlal (Alpine
Buntsandstein) to
marginal marin ramp
deposits (Reichenhall
Fm .. Werlen Fm.)
I0: 1(mainly coarse) clastics

o
~ carbonate ramp deposits
~ shales. partly WIth (basinal) carbonates platform margin and slope
~ carbonate ramp deposits (w,lh shales) g inner platform carbonates
o carbonale basin sediments ~ Inner platform dolomites
Fig. 15. Early to early late Triassic depositional environments and lithological units in the western
part of the Northern Calcareous Alps.
Triassic times. Compared to the Southern Alps, the succession is much more
homogeneous and rather uniform. Clastics in the Anisian are quite fine-grained
and are strongly diluted by carbonates. Again, this is in strong contrast to the
situation in the Southern Alps.
The Steinalm Formation comprises inner to outer ramp deposits, including
shallow intraramp basin sediments. Due to the very slow early diagenesis and the
predominance of micritic facies in all depositional areas, an unstable, muddy sub-
strate evolved. Encrusting organisms prevailed in shallow marine environments.
Organisms, capable of building reefal structures are missing completely during the
Steinalm time. Anisian "reefs" of the Mieming Range, described by Miller (1965)
.-
'200
e:
'E'"
c:
!l!
~
"
-,
'"
(,) ~
:::
~ terngenous :
dasHes .:
.
.
c
.
C
'6
..J
~
0
..J
basin
(condensed sections)
I
h
.. e:
and basin margin
..
IL (t-
~
~ lIIynan
'-' c:
~ ,!l!
'c"'
depression and ramp .nHuanced by terrigenous clastics
Fig. 16. Middle Triassic depositional environments in the western part of the Northern Calcareous
Alps.
and quoted later by Tollmann (1976) and Flugel (1982), are definitely mid- to inner
ramp deposits with few microbial stromatolites. The "reef-like" geometry of the
outcrop (Miller, 1965) resulted from Alpine tectonics (Ruffer, 1994).
5.2. LADINIAN TO EARLY CARNIAN DEPOSITIONAL ENVIRONMENTS
The main feature of the Middle Triassic succession in the Northern Calcareous Alps
is a transgressive pulse in latest Anisian to early Ladinian times, called "Reiftinger
Wende" or Reifling event (Schlager and Schollnberger, 1974). This event caused
the quick alteration from intertidal to shallow subtidal carbonate ramp deposits
of the Steinalm Formation to the widespread Ladinian basin sediments. In the
Northern Calcareous Alps no indications of any synsedimentary tectonics breaking
up Anisian platforms can be recognized. This is in contrast to the Southern Alps (see
above). Instead, a quick but gradual transformation took place, from the Anisian
homoclinal ramps through early Ladinian distally steepened ramps to basins and
rimmed platforms in middle Ladinian to early Carnian times.
All of the mainly shallow water Ladinian to early Carnian deposits belong
to the Wetterstein Formation, carbonate basin to basin margin successions to the
Reifling Formation, and terrigenous sediments filling the basins to the Partnach
Formation (Figs. 15 and 16). The widespread Wetterstein Formation span a deposi-
tional area from the lower slope through a high energy reefal platform margin into
inner lagoon and tidal flats. Difficulties in establishing stratigraphic subdivisions
arise, when Anisian and Ladinian carbonate platform sediments are superimposed
on one another without intercalations of basinal successions. During the Longobar-
dian and early Carnian, Wetterstein platforms aggraded and prograded over basin
areas, resulting in increasingly larger platforms and, due to distinct subsidence, in
often more than 1000 m thick platform deposits. No basinal intercalations separated
the incessant platform development. Even during the late Longobardian to early
Carnian influx of fine terrigenous clastics, forming a few hundred meters of sedi-
ments deposited within small and relatively shallow basins, carbonate production
mostly flourished in the platform and upper slope environments (Fig. 16).
In Middle Carnian times carbonate platforms as well as carbonate and shale
basins were abruptly covered by terrigenous clastics (Raibl Group in the western
and central part, Reingraben, Lunz and Opponitz Formations in the eastern part
of the Northern Calcareous Alps). This general onset of Julian shallow marine
clastics, which has been named "Reingrabener Wende" ("Rein graben event") by
Schlager and Schollnberger (1974), marks the upper boundary of the described
succession (see also Bechstadt and Schweizer, 1991).
5.3. FACIES AND SEQUENCE STRATIGRAPHY OF SELECTED SECTIONS
In contrast to the uniform Anisian depositional environments, remarkable depo-

sitional differences occur from about the AnisianlLadinian boundary onwards. In
the following, three sections are discussed, representing different environments in
late Anisian to Ladinian times. For biostratigraphic data from these sections see
Ruffer (1994) and Ruffer and Bechstadt (in preparation).
5.3.1. The Hochgliickkar Section

This key section (Figs. 17 and 18) is located in the central northern Karwendel and
represents (on top of Scythian/early Anisian deposits) the widespread transition
from a homoclinal ramp in the Anisian through latest Anisian/early Ladinian
basinal sediments to rimmed platform deposits in the late Ladinian and early
Carnian.
The Scythian of this area includes caliche and sabkha deposits consisting
mainly of dolomites and terrigenous clastics (Werfen Formation). A regressive
trend with increasing evaporation during the late Reichenhall Formation (latest
Scythian or early Anisian) interrupted the generally transgressive trend in Scythian
and Anisian times (Szulc, pers. comm., 1994). The early Anisian ramp is docu-
mented by mainly strongly bioturbated mudstones to wackestones with a low but
continuous shale content (Virgloria Formation). Szulc (pers. comm. 1994) ana-
lyzed fourth-order cyclicity using the infauna distribution patterns and concluded
that the transgressive trend in the Anisian resulted in slightly deeper environments
in the late Anisian. Succeeding peritidal inner ramp deposits, consisting mainly
of aggregate and microbial lumps, a shallow subtidal early Pelsonian transgres-
sive succession followed, largely comprised of mud- to wackestones with crinoids
and ostracods (transgression A3). This interval, extremely low in species, reflects
restricted conditions, probably due to input of fine-grained clastics. Wackestones
with maximum diversity of non microbial skeletal grains, and radiolarian mud- to
packstones are characteristic for the interval around the maximum flooding surface
and the early highstand (Fig. 17). The main part of the Pelsonian is built up by
- c
oS
:~
H
.
....
Fig. 17. The HochglUckkar section (Karwendel. Northern Calcareous Alps) with Anisian ramp.
early Ladinian basin and late Ladinian rimmed platform; for legend see Fig. 20.
Fig. 18. The HochglOckkar section in the western part of the Northern Calcareous Alps (Karwendel,
cf. Fig. 17); Anisian homoclinal ramp deposits (Steinalm Formation) and late Ladinian rimmed
platform sediments (Wetterstein Formation) are intercalated by early Ladinian basinal deposits
(Reifling Formation). In the background, the Anisian homoclinal ramp deposits were directly overlain
by a platform succession.
restricted shallow subtidal, low to medium energy pack- to rudstones, consisting

of aggregate lumps, shell debris, coated grains, and microbes. Stromatolitic and
microbial bindstones with dasycladaceans occur especially during the late Pelso-
nian lowstand time (lowstand A4). After crinoid coquinas in the subsequent early
transgressive systems tract and a following decrease in particles, late transgres-
sion A4 and early highstand A4 deposits comprise high diversity nonmicrobial
fauna. In contrast to the preceding Pelsonian highstand, the IIlyrian highstand
is characterized by shallow subtidal medium to high energy deposition, during
which marine cementation increased. The depositional surface remained in subti-
dal environments, even during the following lowstand. Restricted conditions were
present only in a few areas further north (landward) as indicated by stromatolitic
mudstones.
With the transgressive surface AS, a drastic change in geometry and deposi-
tional environment began. Near the AnisianlLadinian boundary the late IIlyrian
carbonate ramp was drowned by transgression, which caused a gradual, but rapid
shift to deeper marine conditions, that persisted over large areas during the Fassa-
nian and lowermost Longobardian. During this time interval more than 100 m thick
basinal and basin marginal sediments of the Reifting Formation were deposited.
The lowermost basinal deposits consist of bituminous bioclastic wackestones (syn-
chronous to the first reefs in nearby areas), followed by nodular limestones with
cherts rich in radiolarians and intercalated volcaniclastics.
Slight increase in marine cementation and the first distal turbidites from the
platform shed into the basin indicate the early Longobardian highstand (high-
stand Ll) and the following lowstand (lowstand L2), whereas the autochthonous
sediments show gradual deepening. A short-term renewed cessation of platform
shedding is accompanied by the occurrence of deeper marine fauna (transgression
L2). The subsequent few hundred meters thick slope and platform margin succes-
sion starts with lower slope deposits, dominated by turbidites that mainly contain
binding organisms. Within the upper slope sediments a sudden increase in early
cements, a decrease in micrite and a change in composition occurs. Lithoclasts of
baffle- and framestones with corals and Tubiphytes obscurus prevail over encrust-
ing organisms. No sequence-stratigraphic subdivision of this part of the succession
has been attempted.
5.3.2. The Berglental Section

This section (Fig. 19) is of special interest due to its restricted intraplatform-basin
sediments ofFassanian age. Up to the Pelsonian (late Anisian) this landward section
in the southeastern Wetterstein Mountains is made up of slightly shaly, intensively
burrowed peloidal mudstones of the Virgloria Formation. The Steinalm Formation
starts with an extremely thin, dolomitic-calcitic lowstand (lowstand A4), made up
of partly reworked stromatolites and microbial mudstones. Showing a sharp basal
contact, interpreted as a ravinement surface, the early transgression A4 is made
up of bioclastic packstones or intraformational breccias. After a short deepening,
indicated by a distinct increase in deeper marine particles, the following highstand
(highstand A4) sediments are rather uniform. Micritic inner ramp carbonates are
mainly composed of crinoids, brachiopods and microbes. An increase in micro-
bial organisms (encrusters, microbial lumps, Tubiphytes) reflects more restricted
conditions prior to the deposition of the latest Anisian to early Ladinian trans-
gressive sediments (transgression AS). Again, this transgression is marked by a
distinct shift to more open marine conditions. Succeeding a keep up phase around
the AnisianlLadinian boundary, (shallow) subtidal sediments (Reifting Formation)
indicate the existence of an intraplatform basin during the Fassanian. In contrast to
other early Ladinian basinal deposits, particles typical of lagoonal inner platform
settings occur. In the mid Fassanian, a shallowing upward trend with increasing
shedding from inner platform areas (highstand AS) leads to backmargin buildups,
composed mainly of (calci)microbes. Situated in the second-order peak transgres-
sion interval (Ruffer and Bechstadt, in preparation), the depositional sequence Ll
cannot be recognized everywhere in the Northern Calcareous Alps (Berglental
section and Igelskopf section; Fig. 19). No distinct lowstand is developed.
5.3.3. The Laliderer Wande Section

In few areas (Figs. 20 and 21) Anisian and Ladinian carbonate platform deposits
are superimposed on one another without intercalation of basin to basin-margin
Fig. 19. Sections with Fassanian restricted intraplatform basin deposits in the western part of
the Northern Calcareous Alps (upper section: Berglental. Wetterstein Mountains; lower section:
Igelskopf. Mieming Range).
successions. This largely continuous platform deposition took place, when car-
bonate production recovered rapidly after incipient drowning. Distally steepened
ramps evolved from the Anisian homoclinal ramps gradually, but quickly, after a
very short initial lag phase. Transgression AS is characterized by pelmicrites dur-
ing early transgression and subsequently by high-energy shoal and reef deposits,
representing the uppermost Illyrian and/or lowermost Fassanian. Generally, most
of the early Ladinian reefal structures developed during rising or high sea level.
S.4. MIDDLE TRIASSIC SEQUENCE STRATIGRAPHY OF THE NORTHERN CALCARE-

OUS ALPS
So far few studies have attempted sequence-stratigraphic applications on the Meso-

zoic succession of the Northern Calcareous Alps. This include Bechstadt and
Schweizer (1991) for the Carnian, Ruffer (1994) for the Middle to lower Up-
per Triassic, and ROffer and BechsUidt (in preparation) for the Triassic. Several
third-order depositional sequences can be identified in the Middle Triassic of the
Northern Calcareous Alps (Figs. 16, see also Fig. 22). This entirely marine succes-
sion was only weakly affected by early (Triassic) tectonics. Rates of subsidence
were moderate to high and no major erosional hiatus is known from the succession.
The area was separated from the stable European northern hinterland due to post-
depositional rifting, strike slip and thrust tectonics and no coastal onlap relations
for the defining sequences can be observed. Thus sequence stratigraphy cannot be
applied in its strict sense. The most prominent, as well as datable, traceable, and
correlatable features are transgressive surfaces and maximum flooding surfaces
(Ruffer and Bechstadt, in preparation).
The most striking features of the north alpine Anisian are third-order trans-
gressive intervals, causing a distinct stacking pattern of ramps (Fig. 21). Due to the
very gentle relief, third-order sea-level fluctuations caused distinct lateral shifts
in facies, but did not change the fundamental mechanism of sediment production,
reworking and transportation. In the long term, accommodation and accumula-
tion balanced each other and only slightly prograding depositional units can be
observed. High rates of sedimentation within the outer ramp during low seastand
compensated low rates during high seastand. In contrast, within the inner ramp,
the lowest rates of sedimentation took place during lowstands with restricted per-
itidal facies. Substantially higher rates were achieved during early highstands.
Within one sequence the overall rates of carbonate production and sedimentation
on the homoc1inal ramp were therefore virtually unrelated to the depositional en-
vironment. These relations account for the fairly slow increase in the depositional
relief.
The early Ladinian transgression (transgression AS, "Rei flinger Wende") is the
most significant event during Middle Triassic time. Basinal deposits (Reifling For-
mation) overlie shallow subtidal to intertidal carbonates of the Steinalm Formation.
In some areas, carbonate production recovered rapidly after incipient drowning
and platform sedimentation continued without intercalations of basinal sediments
from the Anisian to the Ladinian (e.g. "Laliderer Wande"). According to Schlager
and Sch6llnberger (1974) and Bechstadt and others (1978) this event was due to
190 Th. RUffer, R. ZUhlke
Leg end
D-
IB@--.
~nw.5Nt~ ~pIIIy
LJ- IEEE- O-
129-- .....
..
~
I*l--' ~-- ~:::.

~-- 0 .... ~"","""
D-
Fig. 20. The Laliderer Wande section (Northern Calcareous Alps, Karwendel); continuous Middle
Triassic platform succession.
Fig. 21. Late Anisian depositional sequences, the Laliderer Wilnde section (Northern Calcareous
Alps, KarwendeI, cf. Fig. 20).
graben tectonics. However, detailed investigations do not support any indication

of synsedimentary tectonics breaking up the Anisian platforms of the Northern
Calcareous Alps. This contrasts with what is known from the Southern Alps. In
contrast to the widespread assumption, our modeling shows that late Anisian and
Fassanian rates of subsidence have the same order of magnitude (20-30 b, accord-
ing to almost all existing time scales). Detailed facies analysis and biostratigraphy
indicate a distinct increase in subsidence only in late Fassanian to early Longo-
bard ian times (some 50 b) and in the middle Longobardian (about 300 b) using
existing time scales (RUffer, 1994). This strong increase in subsidence is due to
syndepositional extensional tectonics, causing the change from distally steepened
ramps to rimmed platforms. Due to these high rates of subsidence, sequences in
the middle to late Longobardian are difficult to recognize. Distribution patterns of
the terrigenous influx provide the main depositional information, but there is no
well-founded sequence-stratigraphic analysis for the latest Ladinian/earliest Car-
nian succession. Relying on a basinal section in the eastern Northern Calcareous
Alps, a preliminary sequence-stratigraphic analysis was attempted by Ruffer and
Bechstadt (in preparation). A third-order sea-level rise for at least the upper 140 m
(representing a few hundred thousand years) of the Wetterstein Formation can be
assumed from the type of high frequent cycles in lagoonal deposits (Zeeh, 1994),
and leads to the Julian increased input of clastics (Raibl event, see also Bechstadt
and Schweizer, 1991).
6. Depositional Sequences in the Northwestern Tethys
In order to analyze basinwide sea-level changes in the northwestern Tethys, we

have compared depositional sequences from both study areas and from other parts
of the Alps. The sequence-stratigraphic correlation of different domains in the
same basin will support one of the following two observations:
(1) Individual sequences in the study areas can be widely correlated. Deposition
was controlled by basinwide sea-level changes, local controls were subdued.
(2) Individual depositional sequences can not be correlated. Local controls
dominated deposition in each area. Basinwide sea-level changes were not active
or have not been documented in the stratigraphic succession.
Convincing sequence-stratigraphic correlations between different parts of a
large basin require biozonations with resolutions finer than the time intervals
of 3rd-order sea-level changes (0.5-3.0 rna, Duval and others, 1992). Therefore,
biostratigraphic data and their correlation between the study areas was emphasized.
6.1. BIOSTRATIGRAPHY
Different biostratigraphic subdivisions were used for the sequence-stratigraphic

correlation of the two study areas. Correlations are based on (sub-)zones, especially
ammonite zones. Although stages and stage boundaries have been indicated in
Figs. 22 and 23, they did not serve as correlative criteria. Stage boundaries were
assigned following the opinions of the authors of the biostratigraphic schemes. The
AnisianlLadinian boundary is placed at the top of the NevaditeslSecedensis-zone in
the biozonal scheme of Brack and Rieber (1993, 1994) used for the Southern Alps,
but at its base in the scheme of Krystyn (1983) used for the Northern Calcareous
Alps. The Anisian-Ladinian boundary definition by Brack and Rieber (1993, 1994)
is identical with the boundary definition in the American realm (e.g. Silberling and
Nichols, 1982, OccidentalislSubasperum zone).
Biostratigraphic data for the Scythian northwestern Tethys (Fig. 22, central
column) predominantly come from the Dolomites. The biostratigraphy of the
Werfen Formation, established by Broglio Loriga and others (1983, 1990), is
only partly based on standard ammonite zones (Cassianus, Carniolicus zones,
Spathian). Lower and middle Scythian biozonation is founded on characteris-
tic bivalves (Lingula, Claraia, Eumorphotis) and foraminifers (Meandrospira).
While the Permian/Griesbachian, the earlyllate Griesbachian boundary, the Gries-
bachian/Dienerian and the Smithian/Spathian boundaries are defined by biozones,
the Dienerian/Smithian boundary is still a matter of dispute.
The biostratigraphic subdivision for the Anisian to Ladinian indicated in
Figs. 22 and 23 was established by Brack and Rieber (1993, 1994) in differ-
ent parts of the Southern Alps (Dolomites, Lombardy) and has been extensively
documented and analyzed in several studies. Stages indicated in Figs. 22 and 23
have not been defined in these studies, but were added to support the comparison
between the Northern Calcareous Alps and the Southern Alps. For the Ladinian
to Carnian interval of the Northern Calcareous Alps, we have additionally used
the biostratigraphic zonation of Krystyn (1983). It correlates well with the zona-
Western Northern NW-Tethys Dolomites

Chronostratigraphy central Southern Alps Chronostratigraphy
Calcareous Alps (NCA) Depositional Stages and (Sub-)Zones
Stages and (Sub-lZ0419S
Strati- sequences Stratigraphic Biostraligraphyafter
Depositional Depositional
BiostratigraphyafterKrystyn
(1978.1983)
graphic sequences Basinwide sea-level
changes sequences Units ~~~~:r(~~~'~)
Units (schematic) (stages added for C(lIlll8rison)
225 Ca2 225
~~~I~- Austriacum ~/6i~~ C2 Ca2
Durrenstein Fm. Austriacum
~
j
~ ~~~~~~t~~ri~~~~~J
~d~' I J Cassian Julian 3
.r
Aonoides Ca 1 ? --- Dolomite!! E Aonoides
Aon .1----- C1 Ca 1
~~~>------4------r-~
"
Regele-I----
Daxatina J:.: Cassian
------ La3----- La3 ja Raga/edanus
~
? --- Dolomite I
danus Rage/edan. ~~ L4 oil
111111 111111
"- " u
I-- L3 Schlem
La 2 La2 Fm.1II
C L2
230 .!! ~ 230
~ Gr8r!-~. Grad/ari
,i"
Ii"
______ La 1 . ____ _
j f---.--+-_Je_n-l-:::Me:::,"=e:::riJ:::o':::j
~
LI La 1 ------ Schlem Fm.1I
~ JI---r----I "
~ 1-____-jWJlliWlliWillllill di ~
.i Gur;oni; I----- ~>-----I Schlem
Guriani; Fassanian
Je--I---J
Fm.1
~~I------->-----+--I
Curionii
AS An 5 Ans------rn;::!\' ---
Neva- fil Secedensis
diles
_ .+~~ E~~:e (Nevadi/es)
J
.;. Reitzi lIIyrlan
235 3: Contrin Fm. ~ & (RertzVKelineriles) 235
IIlyrlan
Morbiac Fm. a:i
A4 --An4-- An4 Ri hh Fm.1111 Trinodosus
11.p. [~~ Sari Fm.

Pelsonian
An3
An 3------
1 BaJaJon'cu, Pelsonlan
,.
.
c
.!!! e - - - - - - - I
A3 --- ~~m.~~~~____ _
Ismidicum
"ii'iii"
~ Blthynian An 2
Gracilis Lg;~r 1_---1 Bithynlan :::J
-----An 2 ------ Fm.
Fm. Osman!
-
240
~
~ -----?
~ 240
Aegean ~~? A1 --Anl-- Ani _ Lowe,

Sari
~
.~ E
Paracro-
choreJiceras Aegean
Fm. :iiLL
j
~+-----------~~--~---?----
I Reich. Fm. S. Lucano Mb Camiolicus
Spathlan WerfenFm. S2
-----. ScS ------ ScS =.:..:..:. cenMe;ighe ~ ~~ ~r::t~ ~ Spathlan
Up. Alp. Sts
~Sc4= SC4 =.:..:..:. Val Badia Mb.
Cassianus
C. subrolunda
*
Smlthlan ,~ Campil Mb. Smithlsn '"
Sc 3
Sc3
_
~ ~G-.,-'-Io-po-cd-en-_+-E-.h-'n-n'-tid-ea+--- ~ 245
Dlenerlan SI I_----~ IIIIII~ >-,o""O""lith"M",,bc-.-t f.~;pvMo[n)r.$ Dlenerlan :::J
Seis Mb. C. aurils

Sc2 Sc2
- e. claraia
Andraz Mb.
Gries-
Griesbachian e. wang!- bachlan
----- Sc 1 ------ Sc 1 ------ Mazzin Mb. griesbach!
- Tesero Mb.
Lingu/a
Permian Bellerophon Fm. Permian
[[[[[[]J[[I subaerial hiatus - - - sequence boundary - - - - - - - - maximum flooding surface - - - transgressive surface ~ submarine hiatus
Fig. 22. Correlation of depositional sequences in the Northern Calcareous Alps and the Dolomites_
Anisian to Carnian stage names (Dolomites) have been added only to support the comparison of both
study areas, although stages are not formally defined on the basis of biozones (Brack and Rieber,
1993,1994)_ Depositional sequences in the northwestern Tethys (middle column) reflect basinwide
changes in sea level. Note the di ffcrences in the definition of the Anisian/Ladinian (IIlyrianiFassanian)
boundary in the biostratigraphic schcmes_
.....
\0
~
CIWOrlQo
8t01tntlgqlitY 'Ion...nd .ubzonH) TirMIn,t.t\>altOf
tlt.tlgraph, sequenu S.,.llgr,p"1 Poulbk En~tltConlroi
O_many r---;;tants5ea ~Arc:ti(:. hIedr~" y=::;
~. '. ~:~ ~ ~..s::., HW.:.T,::YII =
"""'-'-1:" ......... ' ......... , ~1 .... 1It' ~~ !J!Il~lb";,~==',..:.:..,.:.:.T-,,rl;;;
CO2
Hl
, 1 J ~ iip _.. _, ... --
~ ~~~~~'+---~~h.~------~~~ ~ I---
- lo' '-~l m-13-4. j
7 I---
la, jl T003
L ~ ,,,,
l" TO-' ~
I j~ ~
)1------: f-'-
..
"- +- s...-
- An5 I T,., i r
,-,
"'r""-~I
f.1I
f
...
,II...... "-
t-::-
f-"3 A'II 1111
."
'-+----1-.;:1 1:;-
..." I . . . ..
'-L;r: lui ' ,- T2-5
jf-- fl ~
- A1-I
J 1 1-'-
f~
-- I I I...... T2-4
I .- r-'" Ii T2-3 f
Illf --1 -..-:.::--:- f= Sc5 S2 ~;:: }! !
~se..
F-t ~I1--7-
" ill
Sc3 .!
T:! j}! i .... I!
~
'
2 1~7- W~ , . . ~
!L ,.:.._
,. l~ lIl _fE~~ 1-----1
c-=- Sc,
50 1 ~
i I
f- If I ~ ~
1:'
.-:j~~ ~Sc' ~ _.
If ~
.!"'I
,..,..rniIIn ... i.J I~ ...
;:0
Fig. 23. Global correlation of depositional sequences. Columns on the left-hand side show standard ammonite biozones in
Europe (Tethyan realm) and North America Details of biostratigraphic correlation are explained in the text. The biozonal ~
::r
subdivisions of Skjold and others (1992) in the arctic Barents Sea and Silberling and Nichols (1982) in Nevada are added
for comparison. Columns on the right-hand side indicate globally correlative depositional sequences and time intervals of ~
possible eustatic control.
tion by Brack and Rieber (1986, 1993). The Secedensis and Nevadites zones, the
Curionii, Gredleri, Archelaus and Regoledanus zones of both schemes are correla-
tive. Only the subdivision of the early/middle Longobardian (Gredleri, Archelaus
zones) is preliminary and not well defined (Krystyn, 1983:244; Brack and Rieber,
1994:Fig. 1). The faunal subdivision of the early Carnian of both study areas has
been taken from Krystyn (1983) (cf. Brack and Rieber, 1993:420).
Mietto and Manfrini in de Zanche and others (1993) have published a new
biostratigraphic scheme for the Triassic, which comprises a large number of am-
monite subzones for the Anisian to Carnian stages. The maximum stratigraphic
resolution in the late Anisian to early Longobardian seems to reach up to 180 ka
(An 2 to La 2 sensu de Zanche and others, 1993:8). Zones of other biostratigraphic
schemes have been changed to subzones with significantly shorter ranges (e.g.
Trinodosus zone/subzone). The sequence-stratigraphic zonation of de Zanche and
others (1993) relies strongly on this new biostratigraphic zonation. The database
for this scheme, however, has never been documented in literature.
Some slight differences in sequence-stratigraphic levels, e.g. sequence bound-
ary La IILI in the late Fassanian and La 2 and L2 in the early Longobardian,
might be due to biostratigraphic problems. Biostratigraphic data for the Dolomites
are heavily based on ammonites, while conodonts mainly served for establishing
the sequence stratigraphy in the Northern Calcareous Alps. However, Ladinian
conodont zones can be readily integrated into standard ammonite schemes.
6.2. SEQUENCE-STRATIGRAPHIC CORRELATIONS
6.2.1. Scythian
Sequence-stratigraphic data for the Scythian predominantly come from the Dolo-
mites, where marine deposition prevailed. The time-equivalent Lower Alpine
Buntsandstein of the Northern Calcareous Alps supplied no sequence-stratigraphic
information, as it represents mainly terrigenous deposition which lacks biostrati-
graphic data. Sequence boundaries in the Scythian succession of the Southern Alps
are mainly fixed to the base of thin intertidal units intercalated with (shallow) sub-
tidal deposits. Sequence boundaries are positioned in the Tesero Member and at
the base of the Andraz, Gastropodenoolith, Val Badia and Cencenighe Members.
We have defined a single late Smithian/early Spath ian sequence Sc 4, comprising
the whole Val Badia Member. In contrast, de Zanche and others (1993) regard
the twofold intercalation of peritidal horizons (units a, c sensu Broglio Loriga and
others, 1990) into subtidal units (units b, d sensu Broglio Loriga and others, 1990)
of the Val Badia Member as two sequences. The Sc 4 sequence sensu de Zanche
and others (1993) restricted to the lower Val Badia Member would measure less
than 20 m in thickness without any significant erosion at the base or top. For
this reason, an interpretation as two shallowing-upward parasequences, aggrading
from shallow subtidal to peritidal settings (sabkha) is preferred.
According to Broglio Loriga and others (1990) transgressive/regressive trends
in the Scythian of the Dolomites can be partly correlated to the Transdanubian
Mountain Range in Hungary, 500 km to the east. In the Triassic, this region
covered a more oceanward position as the Northern Calcareous Alps and the
Southern Alps. Although exact biostratigraphic timing is not always possible, the
Permianffriassic boundary transgression (ScI transgressive surface), the regres-
sion in the late Dienerian (sequence boundary Sc 2/Sc 3) and the late Smithian
regression (sequence boundary Sc 3/Sc 4) in the Dolomites are backed by this
correlation. The sequence boundaries Sc lISc 2 and Sc 4/Sc 5 have no counterpart
in Hungary.
First marine ingressions reached the western part of the Northern Calcareous
Alps in the late Scythian (upper part of Lower Alpine Buntsandstein). Biostrati-
graphic data for this transgression (S 1) and the following sequence S2 consisting of
fluvial (Upper Alpine Buntsandstein), sabkha and shallow marine deposits (Werfen
and Reichenhall Formations) do not allow a distinct correlation to the late Scythian
depositional sequences of the Dolomites. For the Werfen Formation in the western
part of Northern Calcareous Alps only a Spathian age has been assigned on the
basis of conodonts and other microfossils (Mostler and Rossner, 1984). The top of
the Reichenhall Formation has been placed by most authors at the Scythian/Anisian
boundary, although an early Anisian age cannot be excluded. Therefore, the first
Anisian sequence boundary in the Dolomites (An 1) and the Northern Calcareous
Alps (A 1) can be correlated only with some reservations.
6.2.2. Anisian
The sequence stratigraphy of the early Anisian is only partly constrained by bios-
tratigraphic data. The Scythian/Anisian boundary may be placed at or slightly
above the lithofacies boundary between the Werfen Formation and the overlying
Anisian dolomites (Neri and Posenato, 1985, southwestern Dolomites). Accommo-
dation analyses show that at least in the eastern Dolomites sequence An 1 starts in
the lowermost Lusnizza-Formation (Cervera section, Fig. 8), that is placed approx-
imately at the Scythian/Anisian boundary. Sequence boundary An lIAn 2 lacks
biostratigraphic constraints ("early Anisian"). Based on considerations of subsi-
dence and sediment production, it is placed near the AegeanlBithynian boundary.
In the Dolomites, the late Bithynian/early Pelsonian sequence boundary
An 2/An 3 is well constrained by the occurrence of Balatonites in basinal settings
of the lower Dont Formation (Casati and others, 1982) and by equivalent data
from sections in the adjacent Carnian Alps (Pisa, 1972; Pisa and others, 1978).
In the western Teyths, first reefs after the Permianffriassic boundary evolved in
the middle to late Pelsonian of the Dolomites (Recoaro and Upper Sari forma-
tions, ZUhlke in Senowbari-Daryan and others, 1993). For the Northern Calcare-
ous Alps biostratigraphic data are available only for the late Anisian. Therefore,
biostratigraphically-controlled comparisons between the study areas are only pos-
sible for the late Anisian (late Pelsonian to late lIIyrian) depositional sequences
A4/An 4. These sequences are definitely not correlatable within the Alps. The
excellent biostratigraphic data, especially for the transgressive interval, indicates a
Pelsonian age (Balatonicus zone) for the Northern Calcareous Alps (transgression
A4) and an lIIyrian age (Trinodosus zone) for the Southern Alps (transgression
An 4). Whereas the south alpine sequence fits well with non-Alpine sequences
(see below), the north alpine sequence seems to be locally controlled. Therefore,
sequence A4 of the Northern Calcareous Alps is interpreted as the result of very

low rates of subsidence during the late Anisian leading to restricted conditions, as
discussed earlier.
Already recognized as the main transgressive event by Schlager and Scholln-
berger (1974), the latest Anisian to early Ladinian transgression is one of the
most striking features throughout the Alps. In the Northern Calcareous Alps this
transgression caused the change from a homoclinal to a distally steepened ramp
and the evolution of the first north alpine reefs after the extinction event at the
Permianrrriassic boundary. In the Dolomites, distally steepened ramps evolved
into rimmed carbonate platforms.
6.2.3. Ladinian to Carnian

Despite some remarkable differences in depositional environments and syndepo-
sitional tectonic and volcanic activity, most Ladinian to early Carnian sequences
of the Southern Alps and the Northern Calcareous Alps correlate very well. Due
to the significant radiation of ammonite and conodont faunas, the biostratigraphic
control is good. The middle Longobardian sequence L3 from the Northern Calcare-
ous Alps resulted from increased rates of subsidence, modifying the depositional
setting of the Wetterstein platform. No correlative sequence has been found in
other alpine areas. Late Ladinian to early Carnian sequences, La 3 to Ca 1 in the
Southern Alps and L4 to Cl in the Northern Calcareous Alps, can be correlated
well within the Alps and throughout the northwestern Tethys area.
The early Carnian sequence C2 in the Northern Calcareous Alps comprises the
upper Wetterstein platform with its local subaerial emersions, the early transgres-
sive uppermost Wetterstein Formation and the late transgressive to early highstand
first Raibl shale (Bechstadt and Schweizer, 1991; Zeeh, 1994). For the sequence-
stratigraphic scheme of the northwestern Tethys we followed the biostratigraphy
of Krystyn (pers. comm.) and the sequence-stratigraphic model established for
the Northern Calcareous Alps (RUffer and Bechstadt, in preparation), which indi-
cates an early Carnian age (Aonoides zone, Aonoides subzone) for the sequence
boundary C2 and the beginning of the Raibl Group.
The biostratigraphic position of Carnian terrigenous and marginal marine de-
posits in the Southern Alps (Raibl, Schlernplateau Formations) is still a matter
of dispute. De Zanche and others (1993) placed the lower boundary of the Raibl
Formation into the early/middle Tuvalian (Dilleri, Subbullatus zone ?). Based on
an old finding of Protrachyceras archelaus in the upper Schlernplateau Formation
Brandner (1991) proposed a completely different scheme for the Schlern area in the
western Dolomites in which the Schlernplateau Formation covers the middle-late
Longobardian interval (La 2, Fig. 22) and after a major subaerial hiatus at the top of
the Schlernplateau Formation, the Raibl Formation (s.s., "Falzarego Formation",
western Tofane area) covers the late Julian to Tuvalian interval. However, this
scheme has not been confirmed by other studies. In contrast, Gaetani and others
(1992) and Gaetani (1993) placed the Breno Formation (equivalent to the Raibl
Formation in the western Southern Alps) in the early Carnian (Julian). Therefore,
their depositional sequence C3 could be correlative to our northwestern Tethys

sequence Ca 2.
6.2.4. Summary: Depositional Sequences in the Northwestern Tethys

Depositional sequences in the Early to Middle Triassic of the Northern Calcareous
Alps and the Dolomites can be readily correlated. Local controls, for instance
varying subsidence rates, were either subdued or can be factored out by compar-
ing different regions within one study area or the study areas with one another.
Deposition in the northwestern Tethys realm was strongly controlled by relative
sea-level changes active over the whole basin.
7. Global Depositional Sequences
Basinwide sea-level changes in the northwestern Tethys documented by correlative

sequences in the study areas do not necessarily represent global fluctuations of sea-
level. Global sea-level changes can only be assessed, when data from tectonically
independent basins are considered.
We have compared sea-level data from the northwestern Tethys (Northern
Calcareous, Southern Alps) to sea-level data from adjacent epicontinental seas
bordering the southern margin of the European plate and the northern margin of
Gondwana. Compared information comes from Spain (Catalan Basin; Cal vet and
others, 1990), southeastern France (Ardeche; Courel and others, 1992), the Paris
Basin (Jacquin and others, in preparation), Germany (Aigner and Bachmann, 1992;
Visscher and others, 1993), southern Poland (Silesia; Szulc, 1993), and the eastern
Mediterranean (Hirsch, 1992). Fig. 23 only depicts the data from southeastern
France, Germany and the eastern Mediterranean. However, these epicontinental
basins may share the same regional controls as the AI ps due to their proximity to the
northwestern Tethys. Sea-level changes in interconnected basins might be related
to the same plate tectonic movements (e.g. thermal and tectonic subsidence).
The only reliable approach to analyze sea-level changes active on a global
scale is to compare data from independent basins that did not experience the same
crustal dynamics. Data from outside the northwestern Tethys come from the Arctic
Sea, especially the well investigated Barents Sea (Skjold and others, 1992) and the
Sverdrup Basin (Embry, 1988, 1991). Additional data have been published from
southwestern and southeastern China (Wang and others, 1994), Tibet (Liu and
Einsele, 1994) and Nepal (von Rad and others, 1994; Ogg and von Rad, 1994). All
the latter domains were situated in the eastern Tethys, although their plate tectonic
position is still a matter of some debate (cf. Dercourt and others, 1993; Ogg and
von Rad, 1994). Fig. 23 depicts data from the Arctic Sea and southern China.
Global sequence-stratigraphic correlations are stilI hampered by insufficient
biozonal subdivision in many basins. Furthermore, the correlation of individual
biozones between different basins is still preliminary based on a limited number
of macrofaunal elements (Brack and Rieber, 1994). Skjold and others (1992)
established their sequence-stratigraphic model for the Triassic Barents Sea based
on the North American (Nevada) biozonal scheme of Silberling and Tozer (1968),
which was revised by Silberling and Nichols in 1982. Our correlation between
Northamerican and European biozones is based on Brack and Rieber (1994) for
the late Anisian to early Ladinian. Additional correlations have been made on the
basis of data by Krystyn (1983) for the late Ladinian (cf. Exxon-chart of Haq and
others, 1988).
The correlation of depositional sequences from different basins is plotted in
Fig. 23. Columns on the left hand side show standard biozonal subdivisions in Eu-
rope (Tethyan realm), North America and the Arctic Sea (boreal realm). Columns
in the center indicate sequence-stratigraphic levels and lithostratigraphic units from
a variety of Early to early Late Triassic basins. We have focused on correlations
to basins outside the western Tethys realm. Additional data from the northwestern
Tethys have been incorporated in the column labeled "NW-Tethys" in Figs. 22
and 23. Finally, columns on the right hand side depict time intervals of possibly
eustatic control based on the correlation of sequence boundaries and maximum
flooding surfaces. Based on the current data, it is does not seem to be possible
to correlate transgressive surfaces on a global scale, because they have not been
documented in a sufficient number of studies.
7.1. SCYTHIAN
The sequence boundary at the Permianrrriassic boundary is well documented in

the Southern Alps (this study; Gaetani and others, 1992; Broglio Loriga et al.,
1990), Iran and Pakistan (Baud, 1993) as well as in southwestern China (Wang and
others, 1994). This corresponds to the maximum regression in the latest Permian
in the coastal onlap curve of Ross and Ross (1988) and the Exxon sea-level chart
(UAA-l.2; Haq and others, 1988).
Early to middle Scythian (Griesbachian, Dienerian) sequences (Sc 1, Sc 2)
are not easily correlatable on a global scale. A sequence boundary around the
Dienerian/Smithian boundary exists in Germany (Aigner and Bachmann, 1992),
the Arctic Barents Sea (Skjold and others, 1992), the eastern Mediterranean and
southwestern China (Wang and others, 1994). Sequence boundary Sc 2/Sc 3 in
the northwestern Tethys (this study) is probably equivalent, although it has been
placed in the late Dienerian. Correlations of Late Scythian sequences are ambigu-
ous. In contrast to the Early Scythian, most basin are characterized by a large
number of sequences with short time intervals and sequence-stratigraphic resolu-
tion is much higher than the available biostratigraphic resolution. The numerous
sequences documented in the epicontinental basin of central Europe (Germany)
and the Barents Sea can not be unequivocally correlated with other basins. At
least one sequence boundary at the SmithianlSpathian boundary is indicated in
the northwestern Tethys (this study; Gaetani and others, 1992), Germany (Aigner
and Bachmann, 1992), the Barents Sea (Skjold and others, 1992), the Eastern
Mediterranean (Hirsch, 1992) and China (Wang and others, 1994). Although an
intra-Spath ian sequence boundary exists in the northwestern Tethys (Sc 5), the
Barents Sea (T2-2) and the Canadian Arctic, eustatic control seems uncertain.
7.2. ANISIAN
In contrast to the Scythian, Anisian sea-level changes are more readily correlated.
Especially the sequence boundary near the Scythian/Anisian boundary occurs in
nearly all basins considered, from western to eastern Tethys including various
epicontinental seas, and in the Arctic Basin at the northern margin of Pangea.
Late Aegean to early Bithynian sequence boundaries in the northwestern Tethys
and the Barents Sea are poorly constrained by biostratigraphic data and have no
counterparts in epicontinental seas and the eastern Tethys. Therefore, sequence
boundary An VAn 2 (this study, see paragraph controls on sedimentation in the
Dolomites) and T2-4rr2-S (Skjold and others, 1992) may reflect regional controls.
The northwestern Tethys, adjacent epicontinental seas (Germany; Paris Basin,
Jacquin, in preparation), the Arctic Sea and southwestern China show an equivalent
sea-level fall in the late Bithynian, indicating eustatic control. The existence of a
global sequence boundary is supported by correlative maximum flooding surfaces
in the sequences An 2 (this study), 4 (Aigner and Bachmann, 1992) and T2-5
(Skjold and others, 1992).
Two IIIyrian sequence boundaries (early Trinodosus, Reitzi zone) occur in
the northwestern Tethys, epicontinental seas (Ardeche, Courel and others, 1992;
Germany; Paris Basin) and the Barents Sea. The younger sequence boundary
(Reitzi zone) is additionally encountered in southeast China and probably in the
eastern Mediterranean (Hirsch, 1992). Consequently, we propose that this boundary
represents an eustatic event in the IIlyrian (including Nevadites zone).
7.3. LADINIAN TO CARNIAN
At the Fassanian/Longobardian boundary, sea-level trends on a global scale seem

to have returned. The La 1 maximum flooding surface of the northwestern Tethys
can be correlated with maximum flooding surfaces in epicontinental seas, the
Arctic Sea and the eastern Mediterranean. This conspicuous global sea-level rise
marks a second-order peak transgression that started in the late Scythian. The sub-
sequent two sequences in the Longobardian (Archelaus, Regoledanus zone) are
documented in the northwestern Tethys, epicontinental seas and the Barents Sea.
Additionally, both maximum flooding surfaces are also correlative. The sequence
boundary at the GredleriiArchelaus transition is affirmed by data from southwest-
ern China and the late Archelaus sequence boundary by data from the Canadian
Arctic.
The correlative sequence boundary at the Ladinian/Carnian boundary is well
constrained by biostratigraphic data and indicates a global control. A global sea-
level fall in the early Julian (Aonoides zone, Ca 2 in the northwestern Tethys)
remains uncertain as sequence boundaries are restricted to the northwestern Tethys
and the Canadian Arctic. Opposing sea-level trends existed in the Barents Sea
and possibly in southeastern China. Finally in the Austriacum zone (late Julian)
correlative sea-level trends returned in the northwestern Tethys, epicontinental
seas and the Barents Sea once again.
8. Conclusions
1. The Early to early Late Triassic successions of two key areas, the Dolomites in
Italy and the Northern Calcareous Alps in Austria, are divided into transgressive-
regressive cycles. The comparison of regional sea-level changes from both areas
results in a sequence-stratigraphic scheme for the northwestern Tethys, reflecting
basinwide sea-level fluctuations. The correlation of depositional sequences be-
tween different basins is based on standard biozonal schemes. In order to assess
global sea-level changes, data from the northwestern Tethys have been compared
to sea-level data from other Pangean margins and epicontinental basins.
2. The Dolomites comprise five Scythian, five Anisian, three Ladinian and
two early Carnian depositional sequences; the Northern Calcareous Alps comprise
two Scythian, five Anisian, four Ladinian and two early Carnian depositional
sequences.
3. Depositional sequences in the Early to early Late Triassic of the Northern
Calcareous Alps and the Dolomites are correlative. Local controls, for instance
varying subsidence rates, are either subdued or can be accounted for by comparing
different regions within one study area or the study areas with each other. Depo-
sition in the northwestern Tethys realm is strongly controlled by relative sea-level
changes active over the whole basin.
4. Global correlations outside the western Tethys are still hampered by re-
stricted biostratigraphic resolution. Nevertheless, several time intervals are char-
acterized by depositional sequences, that can be correlated between a number
of tectonically independent basins. Correlative sequences in Early to early Late
Triassic basins in the northwestern and eastern Tethys, epicontinental seas and
the Arctic Sea suggest, that eustatic sea-level changes control the development of
depositional systems.
Acknowledgments
We thank our advisor Thilo Bechstadt for his support. David K. Jacobs and Bilal
U. Haq reviewed the paper and made helpful comments. Funds for the regional
research projects were partly provided by the German Research Council (DFG).
Leopold Krystyn, Vienna, analyzed and dated conodonts.
References
AIGNER, T. AND BACHMANN, G.H., 1992, Sequence stratigraphic framework of the

German Triassic: Sedimentary Geology, v. 80, p. 115-135.
ASSERETO, R., BRUSCA, C., GAETANI, M., JADOUL, F. AND CASATI, P., 1977, Le
mineralizzazioni Pb-Zn nel Triassico de11e Dolomiti. Quadro geologico e interpretazione
genetica: L' Industria Mineraria, v. 28, p. 367-402.
BAUD, A., 1993, From Murgabian (late Permian) to Olenekian (early Triassic) sequence
stratigraphy; the Tethyan margins record, in Annual Convention of the Canadian Society
of Petroleum Geologists (CSPG) with Global Sedimentary Geology Program (GSGP),
"Carboniferous to Jurassic Pangea", August 15-19, Calgary: Abstract Volume, p. 14.
BECHSTADT, T. AND BRANDNER, R., 1970, Das Anis zwischen St. Vigil und dem
Hohlensteintal (Pragser und Olanger Dolomiten, Sildtirol): Festband des Geologischen
Instituts, 300-Jahr-Feier Universitat Innsbruck, v. 9, p. 9-103.
BECHSTADT, T. AND SCHWEIZER, T., 1991, The carbonate-clastic cycles of the East-
Alpine Raibl group: results of third-order sea-level fluctuations in the Carnian: Sedi-
men tar. Geology, ':: 70, p. 241-27Q. ..
BECHSTADT, T., KOPPEN, A., RUFFER, T., SCHWEIZER, T. AND ZUHLKE, R.,
1993a, Anisian to Carnian (Triassic) sequence stratigraphy, Eastern and Southern Alps,
in Society for Sedimentary Geology, Annual meeting, Stratigraphic record of global
change, Abstract Volume, p. 31-32.
BECHSTADT, T., KOPPEN, A., RUFFER, T., SCHWEIZER, T. AND ZUHLKE, R.,
1993b, Anisian to Carnian rrriassic) sequence stratigraphy, Eastern and Southern Alps,
in Annual Convention of the Canadian Society of Petroleum Geologists (CSPG) with
Global Sedimentary Geology Program (GSGP), "Carboniferous to Jurassic Pangea",
August 15-19, Calgary: Abstract Volume, p. 21.
BECHSTADT, T., BRANDNER, R. AND MOSTLER, H., 1978, Aborted rifting in the Tri-
assic of the Eastern and Southern Alps: Neues Jahrbuch fUr Geologie und Palaontologie,
Abhandlungen, v. 156,2, p. 157-178.
BIDDLE, K.T., SCHLAGER, w., RUDOLPH, K.w. AND BUSH, T.L., 1992, Seismic
Model of a progradational carbonate platform, Picco di Vallandro, the Dolomites, north-
ern Italy: Bulletin of the American Association of Petroleum Geologists, v. 76, I, p.
14-30.
BLENDINGER, W., 1983, Anisian sedimentation and tectonics of the M.Pore-M.Cernera
area (Dolomites): Rivista Italiana di Paleontologia e Stratigrafia, v. 89,2, p. 175-208.
BLENDINGER, w., 1985, Middle Triassic strike-slip tectonics and igneous activity of the
Dolomites (Southern Alps): Tectonophysics, v. 113, p. 105-121.
BLENDINGER, w., 1986, Isolated stationary carbonate platforms in the Middle Triassic
Ladinian of the Marmolada area, Dolomites, Italy: Sedimentology, v. 33, p. 159-183.
BOSELLINI, A. AND STEFANI, M., 1991, The Rosengarten: a platform-to-basin car-
bonate section (middle Triassic, Dolomites, Italy): Dolomieu Conference on Carbonate
Platforms and Dolomitization, September 16-21, St. Ullrich/GrOden, Guidebook Excur-
sion C, 24 p.
BOSELLINI, A., 1968, Paleogeologia pre-anisica delle Dolomiti centro-settentrionali:
Atti della Accademia Nazionale dei Lincei, Memorie, Classe di Scienze Fisiche, Math-
ematiche e Naturali, Serie 8, v. 9, Sezione 2, p. 1-32.
BOSELLINI, A., 1984, Progradation geometries of carbonate platforms: examples from
the Triassic of the Dolomites, northern Italy: Sedimentology, v. 31, p. 1-24.
BOSELLINI, A.S. AND NERI, C., 1991, The Sella platform, upper Triassic, Dolomites,
Italy: Dolomieu Conference on Carbonate Platforms and Dolomitization, September
16-21, St. Ullrich/Groden, Guidebook Excursion B, 30 p.
BOSELLINI, A.S. AND ROSSI, D., 1974, Triassic carbonate buildups of the Dolomites,
northern Italy, in Laporte, L.E, ed., Reefs in time and space, Society of Economic
Paleontologists and Mineralogists, Special Publication, v. 18, p. 209-233.
BRACK, P. AND RIEBER, H., 1986, Stratigraphy and ammonoids of the lower Buchen-
stein beds of the Brescian Prealps and Giudicarie and their significance for the
AnisianlLadinian boundary: Eclogae Geologicae Helveticae, v. 79, I, p. 181-225.
BRACK, P. AND RIEBER, H., 1993, Towards a better definition of the AnisianlLadinian
boundary : new biostratigraphic data and correlations of boundary sections from the
Southern Alps: Eclogae Geologicae Helveticae, v. 86, 2, p. 415-527.
BRACK, P. AND RIEBER, H., 1994, The AnisianlLadinian boundary: retrospective and
new constraints: Albertiana, v. 13, p. 25-36.
BRANDNER, R., 1984, Meeresspiegelschwankungen und Tektonik in der Trias der NW-
Tethys: Jahrbuch der Geologischen Bundesanstalt, v. 126,4, p. 435-475.
BRANDNER, R., 1991, Geological setting and stratigraphy of the Schlern-Rosengarten
buildup and Seiser AIm basin, in
BRANDNER, R., FLUGEL, E., KOCH, R. AND YOSE, L.A., EDS., 1991, The north-
ern margin of the SchlernlSciliar-RosengartenJCatinaccio platform, Dolomieu Confer-
ence on Carbonate Platforms and Dolomitization, September 16-21, St. Ullrich/Groden,
Guidebook Excursion A, p. 1-16.
BROGLIO LORIGA, C., GOCZAN, F., HAAS, L., LENNER, K., NERI, C., SCHEFFER,
A.O., POSENATO, R., SZABO, I. AND MAKK, A.T., 1990, The lower Triassic se-
quences of the Dolomites (Italy) and Transdanubian mid-mountains (Hungary) and their
correlation: Memorie di Scienze Geologiche, v. 42, p. 41-103.
BROGLIO LORIGA, c., MASETTI, C. AND NERI, c., 1983, La Formazione di Werfen
(Scitico) delle Dolomiti occidentali: sedimentologia e biostratigrafia: Rivista Italiana di
Paleontologia e Stratigrafia, v. 88, 4, p. 501-598.
CALVET, F., TUCKER, M. AND HENTON, 1., 1990, Middle Triassic carbonate ramp
systems in the Catalan basin, northeastern Spain: facies, systems tracts, sequences and
controls, in Tucker, M., Wilson, J.L., Crevello, P.D., Sarg, R. and Read, J.F., cds.,
Carbonate platforms - facies, sequences and evolution: International Association of
Sedimentologists, Special Publication, v. 9, p. 79-108.
CASATI, P., JADOUL, F., NICORA, A., MARINELLI, M., FANTINI SESTINI, N. AND
FOIS, E., 1982, Geologia della Valle dell' Anisei e dei gruppi M. Popera-Tre Cime di
Lavaredo: Rivista Italiana di Paleontologia e Stratigrafia, v. 38, I, p. 11-212.
CASTELLARIN, A., ROSSI, P., SELLI, L. AND SIMBOLI, G., 1988, The middle Triassic
magmatic-tectonic arc development in the Southern Alps: Tectonophysics, v. 146, p. 79-
89.
COUREL, L., BOURQUIN, S., BAUD, A., JACQUIN, T. AND VANNIER, F., 1992, Se-
quence stratigraphy of the Triassic series on the western peri-Tethyan margin (France,
Switzerland); role of tectonic subsidence and eustasy on the stratal pattern and lithol-
ogy of depositional sequences and systems tracts, in Centre National de la Recherche
scientifique and Institut Fran6ais du Petrole, eds., Sequence Stratigraphy of European
Basins, May 18-20, Dijon: Abstract Volume, p. 120-121.
DE ZANCHE, V., 1990, A review of Triassic stratigraphy and paleogeography in the
eastern Southern Alps: Bolletino della Societu Geologica Italiana, v. 109, p. 59-71.
DE ZANCHE, V. AND FARABEGOLI, E., 1981, Scythian tectonics in the Southern Alps:
Geologisch-Paliiontologische Mitteilungen Innsbruck, v. 10, p. 239-304.
DE ZANCHE, V. AND FARABEGOLI, E., 1988, Anisian paleogeographic evolution in
the central-western Southern Alps: Memorie di Scienze Geologiche, v. 40, p. 399-411.
DE ZANCHE, V., FRANZIN, A., GIANOLLA, P., MIETTO, P. AND SIORPAES, c.,
1992, The Piz da Peres section (Valdaora-Olang, Pusteria Valley, Italy). A reappraisal
of the Anisian stratigraphy in the Dolomites: Eclogae Geologicae Helveticae, v. 85, I,
p.127-144.
DE ZANCHE, v., GIANOLLA, P., MIETTO, P., SIORPAES, C. AND VAIL, P.R., 1993,
Triassic sequence stratigraphy in the Dolomites: Memorie di Scienze Geologiche, v. 45,
p.I-27.
DERCOURT, J., RICOU, L.E. AND VRIELYNCK, B., 1993, Atlas Tethys Palaeoenvi-
ronmental Maps. Explanatory Notes: Paris, Gauthier-Villars, 307 p.
DOGLIONI, C., 1982, Tettonica triassica della Valle di Livinallongo (Dolomiti centrali):
Annali dell' Universitu di Ferrara, Nuova Serie, Sezione IX: Scienze Geologiche e
Mineralogiche, v. 8, p. 1-20.
DOGLIONI, C., 1984, Tettonica transpressiva nel1e Dolomiti: Giornale di Geologia, v. 46,
p.47-70.
DOGLIONI, C., 1987, Tectonics of the Dolomites (Southern Alps, northern Italy): Journal
of Structural Geology, v. 9, 2, p. 181-193.
DOG LION I, C. AND GOLDHAMMER, R., 1987, Compaction-induced subsidence in the
margin of a carbonate platform: Basin Research, v. I, p. 237-246.
DOGLIONI, C. AND BOSELLINI, A., 1989, Platform break-downlap planes relationship
in prograding carbonate platforms: a tool for the reconstruction of basin evolution:
Bol1etino del1a Societu Geologica Italiana, v. 108, p. 175-182.
DOGLIONI, C., BOSELLINI, A. AND VAIL, P.R., 1990, Stratal Patterns: a proposal of
classification and examples from the Dolomites: Basin Research, v. 2, p. 83-95.
DUVAL, B., CRAMEZ, C. AND VAIL, P.R., 1992, Types and hierarchy of stratigraphic
cycles, in Conseil du Recherche National de Sciences and Institut Franoais du Petrole,
eds., Sequence Stratigraphy of European Basins, Conference, May 18-20, Dijon: Ab-
stract Volume, p. 44-45.
EMBRY, A.F., 1991, Mesozoic history of the Arctic Islands, in Trettin, H.P., ed., Geology
of the Innuitian Orogen and Arctic Platform of Canada and Greenland: Geology of
Canada,v. 3,E,p. 369-434.
EMBRY, A.F., 1988, Triassic sea-level changes: evidence from the Canadian Arctic
archipelago, in Wilgus, c.K., Posamentier, H.W., Ross, C.A. and Kendall, C.G.St.C.,
eds., Sea-level changes: an integrated approach:, Society of Economic Paleontologists
and Mineralogists, Special Publication, v. 42, p. 249-259.
FARABEGOLI, E. AND DE ZANCHE, v., 1984, A revision of the Anisian stratigraphy
in the western Southern Alps: Memorie di Scienze Geologiche, v. 36, p. 391.
FLOGEL, E., 1982, Evolution of Triassic reefs: current concepts and problems: Facies, v.
6, p. 297-328.
FOIS, E. AND GAETANI, M., 1982, The northern margin of the Civetta buildup. Evolution
during the Ladinian and the Carnian: Rivista Italiana di Paleontologia e Stratigrafia, v.
86, 3, p. 469-542.
FORSTER, S.c. AND WARRINGTON, G., 1985, Geochronology of the Carboniferous,
Permian and Triassic, in Snel1ing, N.J., ed., The geochronology of the geological record:
Memoir of the Geological Society, v. 10, p. 99-113.
GAETANI, M., 1993, Sequence stratigraphy in the Triassic of Southern Alps, Italy, in
Annual Convention of the Canadian Society of Petroleum Geologists (CSPG) with
GAETANI, M., FOIS, E., JADOUL, F. AND NICORA, A., 1981, Nature and evolution of
the middle Triassic carbonate buildups in the Dolomites (Italy): Marine Geology, v. 44,
p.25-57.
GAETANI, M., GNACCOLINI, M., JADOUL, F. AND GARZANTI, E., 1992, Multiorder
sequence stratigraphy in the Triassic of the Southern Alps with emphasis to the western
part, in Centre National de la Recherche scientifique and Institut Franoais du Petrole,
eds., Sequence Stratigraphy of European Basins, May 18-20, Dijon: Abstract Volume,
p.52-53.
GOLDHAMMER, R.K. AND HARRIS, M.T., 1989, Eustatic controls on the stratigraphy
and geometry of the Latemar buildup (middle Triassic), the Dolomites of northern Italy,
in Crevello, P.D., Wilson, J.L., Sarg, J.F. and Read, J.F., eds., Controls on carbonate plat-
form and basin development: Society of Economic Paleontologists and Mineralogists,
Special Publication, v. 44, p. 323-338.
HAQ, B.U., HARDENBOL, J. AND VAIL, P.R., 1987, Chronology of fluctuating sea-
levels since the Triassic: Scicnce, v. 235, p. 1156-1167.
HAQ, B.U., HARDENBOL, 1. AND VAIL, P.R., 1988, Mesozoic and Cenozoic chronos-
tratigraphy and cycles of sea-level change, in Wilgus, c.K., Hastings, B.S., Kendall,
C.G.St.c., Posamentier, H.W., Ross, c.A. and van Wagoner, J.C., eds., Sea-Level
changes: an integrated approach: Society of Economic Paleontologists and Mineral-
ogists, Special Publication, v. 42, p. 71-108.
HIRSCH, E, 1992, Circummediterranean Triassic eustatic cycles; Israel Journal of Earth
Sciences, v. 40, p. 29-38.
KRYSTYN, L., 1978, Eine neue Zonengliederung im a)'pin-mediterranen Unterkarn:
Schriftenreihe der Erdwissenschaftlichen Kommission, Osterreichische Akademie der
Wissenschaften, v. 4, p. 37-75.
KRYSTYN, L., 1983, Das Epidaurus-Profil (Griechenland) - ein Beitrag zur Conodonten-
Standardzonierung des tethyalen Ladin und Unterkarn: Schriftenreihe der Erdwis-
senschaftlichen Kommission, Osterreichische Akademie der Wissenschaften, v. 5, p.
231-258.
LIU, G. AND EINSELE, G., 1994, Sedimentary history of the Tethyan basin: Geologische
Rundschau, v. 83, p. 32-61.
MARINELLI, M., VIEL, G. AND FARABEGOLI, E., 1980, II Permo-Trias dell Alpi
Meridionali: evoluzione tardo-ercinica di un bacino marginale di retroarco sialico: L'
Industria Mineraria, v. 6, p. 1-14.
MASETTI, D., NERI, C. AND BOSELLINI, A., 1991, Deep-water asymmetric cycles and
progradation of carbonate platforms governed by high-frequency eustatic oscillations
(Triassic of the Dolomites, Italy): Geology, v. 19, p. 336-339.
MASETTI, M. AND NERI, C., 1980, L' Anisico deIla Val di Fassa (Dolomiti occiden-
tali): sedimentologia e paleogeografia: Annali deIla Universitu di Ferrara, Nuova Serie,
Sezione 9: Scienze Geologiche e Mineralogiche, v. 7, I, p. 1-19.
MILLER, H., 1965, Die Mitteltrias der Mieminger Berge mit Vergleichen zum westlichen
Wettersteingebirge: Verhandlungen der Geologischen BundesanstaIt, v. 1965, p. 187-
212.
MOSTLER, H. AND ROSSNER, R., 1984, Mikrofazies und PalOkologie der hOheren
Werfener Schichten (Untertrias) der Nordlichen Kalkalpen: Facies, v. 10, p. 87-144.
NERI, c., 1991, Sequence stratigraphy of the early Triassic Werfen Formation (Dolomites,
Northern Italy): in Dolomieu Conference on Carbonate Platforms and Dolomitization,
September 16-21, St. Ulrich/Groden: Abstract Volume, p. 194-195.
NERI, C. AND POSENATO, R., 1985, New biostratigraphic data on uppermost Wer-
fen Formation of western Dolomites (Trento, Italy): Geologisch-Palaontologische Mit-
teilungen Innsbruck, v. 14,3, p. 83-107.
OGG, J.G. AND VON RAD, U., 1994, The Triassic of the Thakkhola (Nepal). II: Paleolat-
itudes and comparison with other eastern Tethyan margins of Gondwana: Geologische
Rundschau, v. 83, p. 107-129.
PISA, G., 1972, Stratigraphie und Mikrofazies des Anis und Ladin in den westlichen
Karnischen Alpen, Italien: Mitteilungen der GeseIlschaft fUr Geologie- und Bergbaus-
tudenten in Osterreich, v. 21, p. 193-224.
PISA, G., FARABEGOLI, E. AND OTT, E., 1978, Stratigrafia e paleogeografia dei terreni
anisici della Conca di Agordo e dell' alta Val di Zoldo (Dolomiti sudorientali): Memorie
di Scienze Geologiche, v. 18, p. 63-92.
PISA, G., PERRI, C. AND VENERI, P., 1980, Upper Anisian conodonts from Dont and M.
Bivera Formations, Southern Alps (Italy): Rivista Italiana di Paleontologiae Stratigrafia,
v. 85,4, p. 807-828.
POSAMENTIER, H.W., JERVEY, M.T. AND VAIL, P.R., 1988, Eustatic controls on clastic
deposition I - conceptual framework, in Wilgus, c.K., Hastings, B.S., Kendall, C.G.St.C.,
Posamentier, H.W., Ross, C.A. and van Wagoner, J.C., eds., Sea-level changes: an
integrated approach: Society of Economic Paleontologists and Mineralogists, Special

Publication, v. 42, p. 109-124.
ROSS, e.A. AND ROSS, J.P.R., 1988, Late Paleozoic transgressive-regressive deposition,
in Wilgus, e.K., Hastings, B.S., Kendall, e.G.St.e., Posamentier, H.W., Ross, e.A. and
van Wagoner, J.C., eds., Sea-level changes: an integrated approach: Society of Economic
Paleontologists and Mineralogists, Special Publication, v. 42, p. 227-248.
RUFFER, T., 1994, Entwicklung einer Karbonat-Plattform: Fazies, Kontrollfaktoren und
Sequenzstratigraphie in der Mitteltrias der westlichen Nordlichen Kalkalpen (Tirol,
Bayern): Gaia Heidelbergensis, v. I, p. 1-288.
RUFFER, T., ZUHLKE, R. AND BECHSTADT, T., 1994, Middle to upper Triassic se-
quence stratigraphy in the N-/S-Alps (Austria, Italy) and adjacent basins: implications
for regional and global controls on sedimentation: International Association of Sed-
imentologists, 15th Regional Meeting, April 13-15, IschialItaly: Abstract Volume, p.
362-363.
SADLER, P.M., OSLEGER, D.A. AND MONTANEZ, I.P., 1993, On the labeling,length
and objective basis of Fisher plots: Journal of Sedimentary Petrology, v. 63, 3, p. 360-
368.
SCHLAGER, W. AND SCHOLLNBERGER, w., 1974, Das Prinzip stratigraphischer
Wenden in der Schichtfolge der Nordlichen Kalkalpen: Mitteilungen der Geologischen
Gesellschaft Wien, v. 66/67, p. 165-193.
SCHLAGER, w., 1992, Sedimentology and sequence stratigraphy of reefs and carbon-
ate platforms: American Association of Petroleum Geologists, Continuing Education
Course Notes, Series 34, p. 1-71.
SENOWBARI-DARYAN, B., ZUHLKE, R., BECHSTADT, T. AND FLUGEL, E., 1993,
Anisian (middle Triassic) buildups of the northern Dolomites (Italy): the recovery of
reef communities after the PermianlTriassic crisis: Facies, v. 28, p. 181-257.
SILBERLING, N.J. AND NICHOLS, K.M., 1982, Middle Triassic molluscan fossils of
biostratigraphic significance from the Humboldt Range, northwestern Nevada: Geolog-
ical Survey Professional Paper, v. 1207, p. 1-77.
SILBERLIN G, N.J. AND TOZER, E. T., 1968, Biostratigraphic classification of the marine
Triassic in North America: Geological Society of America, Special Paper, v. 110, p. 1-63.
SKJOLD, L.1., VAN VEEN, P., KRISTENSEN, S.-E. AND RASMUSSEN, A.R., 1992,
Triassic sequence stratigraphy in the Barents Sea, in Centre National de la Recherche
scientifique and Institut Fran6ais du Pctrole, eds., Sequence Stratigraphy of European
Basins, May 18-20, Dijon: Abstract Volume, p. 86-87.
SPOTL, e. AND BURNS, S.1., 1991, Formation of 18-0-depleted dolomite within a
marine evoporitic sequence, Triassic Reichenhall Formation, Austria: Sedimentology,
v. 38, p. 1041-57.
STINGL, v., 1987, Die fazielle Entwicklung des Alpinen Buntsandsteins (Skyth) im West-
abschnitt der Nordlichen Kalkalpen (Tirol/Salzburg, Osterreich): Geologische Rund-
schau., v. 76, p. 647-664.
SZULC, J., 1993, Early alpine tectonics and lithofacies succession in the Silesian Part of
the Muschelkalk. A synopsis, in Hagdorn, H. and Seilacher, A., eds., Muschelkalk -
Schontaler Symposium 1991: Stuttgart/Germany, Goldschneck, p. 19-28.
TOLLMANN, A., 1976, Analyse des klassischen nordalpinen Mesozoikums: Vi-
enna!Austria, Deuticke, 580 p.
TOLLMANN, A., 1985, Geologie von Osterreich, Auf3erzentralalpiner Anteil: Vi-
enna!Austria, Deuticke, 710 p.
VAIL, P.R., 1987, Seismic stratigraphy interpretation procedure, in Bally, A.W., ed., Atlas
of seismic stratigraphy: American Association of Petroleum Geologists, Studies in
Geology, v. 1, p. 1-10.
VAIL, P.R., MITCHUM, R.M. AND THOMPSON, S., 1977, Seismic stratigraphy and
global changes of sea-level, part 3: relative changes of sea-level from coastal onlap,
in Payton, C.E., ed., Seismic stratigraphy - application to hydrocarbon exploration:
American Association of Petroleum Geologists, Memoir, v. 28, p. 63-97.
VIEL, G., 1979a, Lithostratigrafia ladinica: una revisione. Riconstruzione paleogeografica
e paleostrutturale dell'area dolomitica-cadorina (AI pi Meridionali). I. Parte: Rivista
Italiana di Paleontologia, v. 85, I, p. 88-125.
VIEL, G., 1979b, Litostratigrafia ladinica: una revisione. Riconstruzione paleogeografia
e paleostrutturale dell' area dolomitica-cadorina (Alpi Meridionali). II. Parte: Rivista
Italiana di Paleontologia, v. 85, 2, p. 297-352.
VISSCHER, H., BRUGMAN, w'A. AND VAN HOUTE, M., 1993, Chronostratigraph-
ical and sequence stratigraphical interpretation of the palynomorph record from the
Muschelkalk of the Obernsee well, south Germany, in Hagdorn, H. and Seilacher, A.,
eds., Muschelkalk - Schon taler Symposium 1991: Stuttgart/Germany, Goldschneck, p.
145-152.
VON RAD, U., DURR, S.B., OGG, J.G. AND WIEDMANN, J., 1994, The Triassic of
the Thakkhola (Nepal). I: stratigraphy and paleoenvironment of a north-east Gondwana
rifted margin: Geologische Rundschau, v. 83, p. 76-106.
WANG, H., CHU,X., LIU, B., HOU, H. AND MA,L.,1994,Atlasofthepaleogeographyof
China: Institute of Geology, Chinese Academy of Geological Sciences, Wuhan College
of Geology.
YOSE, L.A., 1991, Sequence stratigraphy of mixed carbonate/volcaniclastic slope de-
posits flanking the Sciliar-Schlern-Catinaccio buildup, Dolomites, Italy, in Brandner,
R., FlUgel, E., Koch, R. and Yose L.A., eds., The northern margin of the SchlernlSciliar-
Rosengarten/Catinaccio platform: Dolomieu Conference on Carbonate Platforms and
Dolomitization, September 16-21, St. Ullrich/Groden, Guidebook Excursion F, p. 17-39.
ZEEH, S., 1994, The unusual cyclicity of the Triassic (Carnian) Bleiberg facies of the
Wetterstein Formation (Drau Range, Austria): Geologische Rundschau, v. 83, p. 130-
142.
ZIEGLER, P.A., 1982, Geological atlas of western and central Europe: Shell International
Petroleum Maatschappij B.Y., 130 p.
ZIEGLER, P.A., 1988, Evolution of the Arctic-North Atlantic and the western Tethys:
American Association of Petroleum Geologists, Memoir, v. 43, p. 1-198.
ZUHLKE, R. AND BECHSTA.DT, T., 1993, The Anisian (middle Triassic) of the Dolo-
mites (S-Alps, N-Italy): Eustatic-tectonic control on carbonate-siliciclastic sediments,
in Annual Convention of the Canadian Society of Petroleum Geologists (CSPG) with
CHAPTER 8
Reconstructing Eustatic and

Epeirogenic Trends from Paleozoic
Continental Flooding Records
Thomas J. Algeo and Kirill B. Seslavinsky
ABSTRACT: Eustatic and epeirogenic trends may be independently quantified given

knowledge of flooding patterns for multiple, coexisting. tectonically-independent land-
masses. Herein, we develop a method of analyzing paleo-continental flooding records in
order to reconstruct both global sea-level trends and individual continental epeirogenic
histories. Our method is based on the hypotheses that: 1) co-existing landmasses are
likely to have experienced the same range of eustatic fluctuations, 2) differences in
flooding are thus primarily a function of differences in coastal hypsometry, and 3) dif-
ferences in estimated sea-level elevations between individual landmasses and the world
may reflect continental epeirogeny. We apply the method to the flooding records of 13
Paleozoic landmasses for which detailed paleogeographic reconstructions are available
(Ronov and others, 1984; Khain and Seslavinsky. 1991).
Our analysis indicates that Paleozoic eustatic highstands were probably + 100 to +225 m
above present sea level, which is substantially lower than previous estimates of +300 m
(Vail and others, 1977) to +600 m (Hallam, 1984). Reconstructed epeirogenic histories
suggest that Paleozoic continents experienced 100 m of independent vertical motion
relative to global sea level at a 10-40 m.y. timescale. Most large epeirogenic excursions
coincided with major tectonic events such as rifting, passive-to-active margin transi-
tions, and continental collisions, and may reflect a range of epeirogenic mechanisms for
Paleozoic continents comparable to that documented for modern continents. Close links
between eustasy and continental epeirogeny are suggested by the antithetic pattern of
Gondwanan crustal motions and global sea-level elevations during the mid-Paleozoic.
1. Introduction
Detailed knowledge of secular changes in global sea-level elevations (eustasy)

and broad vertical crustal movements (epeirogeny) is required for thorough under-
standing of underlying mantle and plate tectonic processes. Although eustasy has
been widely regarded as the dominant control on long-term patterns of continental
209
B. U. Haq (ed.),
210 T.I. Algeo, K.B. Seslavinsky
flooding (e.g., Vail and others, 1977; Haq and others, 1987), epeirogeny is now
considered to be a factor of equal importance and to respond to the same underlying
mantle processes (e.g., Gumis 1992a,b). In general, eustatic and epeirogenic fluc-
tuations cannot be separated based on stratigraphic data from a single locale, basin,
or continent. Isolation of these variables at a sub-continental scale is difficult ow-
ing to the flexural rigidity of continental lithosphere and to frequent linkage of the
subsidence histories of adjacent sedimentary basins (e.g., Quinlan and Beaumont,
1984; Cloetingh, 1988).
Independent quantification of eustatic and epeirogenic trends may be possible
given stratigraphic information from multiple, co-existing, tectonically indepen-
dent sources. The goals of this contribution are to: 1) develop a method of analysis
of the flooding records of co-existing paleo-continents that permits reconstruction
of both global sea-level trends and individual continental epeirogenic histories,
2) apply the method to the flooding records of 13 Paleozoic landmasses, and 3)
consider the significance of the results of the analysis in relation to documented
mechanisms of eustatic and epeirogenic motions.
1.1. EUSTASY
Long-term (> 10 m.y.) changes in global sea-level elevation are of considerable

importance as an expression of variations in mantle convection and heat flow
(Turcotte and Burke, 1978; Galer, 1991) and owing to a strong influence on
global climate, geochemistry, and biosystems, including greenhouse-icehouse cy-
cles (Fischer, 1984), atmospheric CO 2 levels (Wilkinson and Given, 1986; Berner,
1994), seawater composition and marine carbonate mineralogy (Wilkinson and
others, 1985), and biodiversity and mass-extinction patterns (Riding, 1984; Wyatt,
1987). Because many aspects of paleo-mantle behavior and paleo-atmospheric
and -oceanic chemistry are difficult or impossible to reconstruct, eustasy has been
widely used as a proxy for other variables in global models (e.g., Gaffin, 1987;
Galer, 1991).
Despite the obvious importance of and need for accurate information regarding
the amplitude of secular eustatic variations, surprisingly little research on long-
term eustasy has been undertaken: to date, the only widely-cited Phanerozoic
sea-level curves are those of Vail and others (1977) and Hallam (1984). Most
extant studies of Paleozoic eustasy are based on sequence stratigraphic or facies
analysis of limited geographic areas and have yielded only qualitative trends
(e.g., Johnson and others, 1985; Heckel, 1986; Johnson and others, 1991; Schenk,
1991). Quantitative analyses of Paleozoic eustatic trends using well-documented
databases are lacking.
1.2. EPEIROGENY
Broad (>100 km) vertical crustal motions are potentially of major significance
as indicators of mantle-crustal thermal interactions, far-field responses to plate-
margin processes, and lithospheric density anomalies associated with melting,
intrusion, or phase changes (McGetchin and others, 1980). The epeirogenic his-
Paleozoic Eustasy and Epeirogeny 211
tories of modern continents from the Cretaceous to the present are moderately
well documented based on a combination of geomorphologic evidence (e.g., King,
1962; Bishop and Brown, 1992), hypsometric analysis of flooding patterns (e.g.,
Bond, 1976, 1978a,b, 1979; Veevers, 1984; Harrison 1988, 1990), and shoreline
elevation records (Sahagian, 1987, 1988). Paleozoic continental epeirogeny has
received little study owing to problems in distinguishing crustal motions from sea-
level changes in the absence of independent eustatic information. Disentangling
these processes is difficult owing to their similar amplitudes and timescales of
operation (e.g., Harrison, 1988, 1990; Gurnis, 1992a). This difficulty may be over-
come through analysis of the flooding records of multiple, co-existing landmasses,
which has the potential to isolate common (i.e., eustatic) components from those
unique to each landmass (i.e., epeirogenic factors).
2. Analysis of Paleozoic Eustasy and Epeirogeny
Paleo-eustatic and -epeirogenic trends may be isolated and separately quanti-

fied given knowledge of flooding patterns for mUltiple, coexisting, tectonically-
independent landmasses. Our method of analysis is based on the assumptions that
such landmasses are likely to have experienced the same range of eustatic fluctua-
tions and that differences in their flooding records are thus primarily a function of
differences in coastal hypsometry. Given secular flooding records for a group of
landmasses, it is possible to resolve flooding variance for each landmass into: 1) a
long-term mean hypsometric component, 2) a eustatic component, and 3) a residual
component, representing a combination of secular changes in coastal hypsome-
try (e.g., of epeirogenic origin) and errors in flooding estimates. The long-term
hypsometric component is unique to each continent but constant for all epochs
(i.e., spatially variant, temporally invariant); the eustatic component is common to
all continents but unique to each epoch (i.e., spatially invariant, temporally vari-
ant); and the residual component is unique to each continent for each epoch (i.e.,
spatially and temporally variant).
Our analytical procedure involves the following steps: 1) reconstruction of rel-
ative (i.e., vertically unscaled) coastal "hypsometric chords" for individual paleo-
continents, 2) scaling of the complete set of paleo-hypsometric chords against a
suitable modern hypsometric analog, 3) conversion of epochal flooding values for
individual continents to sea-level elevations using the scaled coastal hypsometric
chords, 4) calculation of eustatic elevation as an average of individual continental
sea-level elevations for a given epoch, and 5) calculation of elevation residuals as
the difference between eustatic and individual continental sea-level elevations for
each epoch. Below, we develop this analytical method using as an example the
flooding histories of 13 Paleozoic landmasses.
2.1. PALEOZOIC CONTINENTAL FLOODING
2.1.1. Paleogeographic Database

The thirteen Paleozoic landmasses used in this study are Pangea, Gondwana, Lau-
rentia, Laurussia, Baltica, Armorica, Siberia, Kazakhstania, North China, South
China, China (united), Indochina, and Chukotka (Fig. 1). Flooding records for
these paleo-continents were assembled in two stages. In the first stage, lithofacies
data were plotted on 1:25,000,000 polyconic base maps of individual continents
in modem coordinates. Data were compiled for 18 Proterozoic and Paleozoic time
slices for each of six modern continents, totalling 108 maps. In the second stage,
lithofacies data were transferred to the global paleogeographic reconstructions of
Scotese and others (1979), and areas of paleo-flooding were interpreted based
on the distribution of extant marine and continental strata of a given age. The
lithofacies data were published in Ronov and others (1984) as 18 global maps
at a 1:60,000,000 scale and the paleogeographic reconstructions as 7 maps at a
1:120,000,000 scale. Methods of compilation of the original database were dis-
cussed in Ronov (1980) and Ronov and others (1980,1984).
The original lithofacies base maps have been updated since publication using
new sources, including Wang (1985), Paleogeographic Atlas of Australia (1988,
1992), and Ziegler (1989), among others. These data have been replotted on a
new generation of paleogeographic reconstructions (Zonenshain and others, 1987,
1990) with modifications from other sources (Scotese and McKerrow, 1990; Nie
and others, 1990; Robardet and others, 1990; Torsvik and others, 1991). The new
paleogeographic maps were published in Khain and Seslavinsky (1991); see also
Seslavinsky (1987, 1991) for further discussion of compilation methods. Both sets
of maps are under continuous revision as new information becomes available, and
new editions are in preparation for pUblication (Seslavinsky).
2.1.2. Flooding Estimates

Integration of areas of continental flooding over time intervals of unequal duration
introduces the possibility of bias toward greater flooding for longer time slices
(Wise, 1974). We have attempted to minimize this problem by using 15 time slices
of sub-equal duration, ranging from 9 to 40 m.y. and averaging 22 m.y. (timescale
of Harland and others, 1990; Table I). Each Paleozoic system is divided into two
or three time slices, which we informally refer to as "epochs" (although most of
these combine two or more formal stratigraphic epochs).
Total landmass and flooded areas were calculated from digitized equal-area
Mollweide versions of the original Mercator projections (Khain and Seslavinsky,
1991). Three areal parameters were determined for each Paleozoic landmass by
epoch: I) total landmass area, 2) area of extant marine strata, and 3) area of
probable original marine flooding ("most likely" flooding estimate; Fig. 1). Areal
estimates are based on paleo-continental shelf margins rather than present emer-
gent continental outlines. Reconstructed areas of probable original flooding link
extant marine strata of a given age in a manner that is consistent with available
information regarding subsequent regional uplift and erosional patterns (Khain
and Seslavinsky, 1991; Ronov, 1994). For each flooding estimate, the quality of
available paleogeographic data is indicated on a semi-quantitative scale (Fig. 1).
The scale represents the degree of extrapolation of "most likely" flooding estimates
beyond known marine deposits (as a fraction of total landmass area) and averages
3% for excellent, 6% for good, 9% for fair, and 25% for poor flooding data.
;.p
"-
~
0
N
TABLE 1 0
(')
-.
Paleozoic flooding data by continent and epoch*
trl
c::
epoch age a GOND PNG LRNT LRSS BALT SIB ARM CHK KAZ N-CHN CHN S-CHN IND Vl
LPM 250 15.86 35.7 84.0 8'

~
EPM 270 18.7 55.7 88.0 :.,
::J
LCB 305 16.8 34.6 41.8 20.0 68.5 92.0 Q.,
ECB 340 12.0 39.4 44.5 76.6 65.7 27.2 63.6 ~~

LDV 370 19.4 41.7 38.1 82.1 57.1 17.0 59.0 -.
MDV 380 21.1 31.3 45.7 82.1 42.8 12.1 45.4 ~
a
~
::J
EDV 395 27.6 13.4 44.7 75.0 51.5 9.7 50.0 '-<:
LSL 415 23.7 17.0 18.1 37.1 75.0 46.6 46.8 11.4
ESL 430 25.1 28.7 33.7 46.5 84.3 43.3 53.3 28.2
LOR 445 21.7 60.9 29.8 34.6 80.0 40.0 80.0 6.6
MOR 465 25.1 41.9 36.3 50.4 83.3 33.3 76.0 70.0
EOR 490 24.5 41.4 32.4 52.4 76.6 30.0 72.0 73.3
LCM 515 26.9 47.8 18.1 42.5 60.7 36.6 76.0
MCM 525 27.2 31.0 29.8 59.4 71.4 33.3 76.0
ECM 550 24.7 21.0 36.3 72.1 78.5 33.3 68.0
Estimates are based on updated versions of the interpretative paleogeographic maps of Khain and Seslavinsky (1991).
a Ages of epoch midpoints in Ma.
b Fractional flooding of total landmass area in percent for "most-likely" case, F( ij}mean.
N
\.>l
-
214 T.l. Algeo, K.B. Seslavinsky
1~ GONDWANA A .... E ~ 10
: : [l.---4.O--... 0-0-0-0-....... ....... I SIBERIA

80
60
A(ij) I 6
F(r)" PANGEA
~ .-1-
~ .~~n F(IJI mean F(ij) max
~
20~~~ 2
o 0
B F
30 ~STANIA 3
r-- _
LAURUSSIA
LAURENTIA
.
20 ............1--4...........
_ .....
_. ....... " .. .,'
'
:'
:.
",
2
'"~ 10
~
....; --
........... :, '"-""E
-
"
'"o '"0
~ o~~~~~~==~~~~~========~ 0
~
'-... ..... C G
L5 BALTICA
<{
W
CHINA 6 0::
~ 6 <{
ARMORICA D H CHUKOTKA S. CHINA INDOCHINA
:~
3
...., .. 2
~~
AGE (Ma)
Fig. 1. Flooding data for 13 Paleozoic continents of this study. Shown are total landmass area
(A(ij); upper solid line), area of extant marine strata (F(ij)min; dashed line), "most-likely" area of
original marine flooding (F(ij)moan; lower solid line), and maximum area of original marine flooding
(F(ij)max; dotted line). Flooding data quality indicated by symbols at top: excellent (solid squares),
good (solid triangles), fair (open squares), or poor (open triangles). All data plotted versus age
midpoints for Paleozoic 15 epochs: EarlylMiddle/ Late Cambrian, EarlylMiddlelLate Ordovician,
Early/Late Silurian, Early/Middle/Late Devonian, Early/Late Carboniferous, and Early/Late Permian
(Harland and others, 1990). Sources of data cited in text.
Uncertainties regarding the original extent of marine strata on continents result

from loss of sedimentary rocks through erosion, metamorphism, and deep burial.
To allow for uncertainties in flooding estimates, we will bracket our "most likely"
value for continent i at epoch j, F(ij)mean, by minimum and maximum estimates
of original marine flooding, F(ij)min and F(ij)max. As the minimum flooding
value, we will adopt the area on continent i of extant marine strata of epoch
j. Meaningful upper limits on flooding are difficult to establish as these require
greater extrapolation beyond available stratigraphic data. For simplicity, we will
assume that the potential maximum area of lost marine strata is twice that of our
"most-likely" estimate, yielding a maximum flooding value of:
F(ij)max = 2 * F(ij)mean - F(ij)min. (1)

Estimated values of F(ij)min, F(ij)mean, and F(ij)max are shown by continent
in Fig. 1, and values of F( ij)mean are tabulated by continent and epoch in Table 1.
2.2. PALEOZOIC CONTINENTAL HYPSOMETRY
2.2.1. Application of Hypsometry to Sea-Level Analysis

The first step in determining eustatic and epeirogenic trends is to reconstruct
paleo-continental hypsometries. The hypsometry of a topographic surface is its
cumulative areal frequency with respect to elevation. Continental area-elevation
distributions form sigmoidal curves that are convex up at low elevations, concave
up at high elevations, and have an inflection point close to sea level (Fig. 2). The
slope of a hypsometric curve is the rate of change of elevation with respect to
cumulative area for a given elevation or elevation range (Algeo and Wilkinson,
1991). The inflection point of a hypsometric curve represents the elevation at which
the slope of the curve is gentlest and, thus, at which the modal areal frequency
occurs. The fundamental control on the sigmoidal shape of hypsometric curves
is the balance between net continent-interior erosion and net continent-margin
deposition. For landmasses in "hypsometric equilibrium", the inflection point is
located approximately at the updip margin of coastal-plain wedges at an elevation
of several tens of meters above sea level.
Hypsometric curves can be used to convert flooding data to sea-level elevations
and to reconstruct secular sea-level trends from a series of paleogeographic maps
(Fig. 2). Because the elevation range of Phanerozoic eustatic fluctuations has
been within a few hundred meters of present sea level (e.g., Vail and others,
1977; Worsley and others, 1984), only the low-elevation portion of a continental
area-elevation distribution ("coastal hypsometry") is of significance for eustatic
studies. Flooding data provide information about the hypsometric characteristics of
a landmass only within this limited elevation range, and, therefore, reconstruction
of paleo-continental hypsometries is constrained to the same range of elevations.
As a first approximation, we assume that paleo-continental hypsometries within the
elevation range of interest are linear. Although modern continental hypsometries
are strongly non-linear over elevation ranges of several kilometers (Harrison and
others, 1983), deviations from linearity are small within a few hundred meters of
present sea level, and this is likely to have been valid for paleo-continents as well.
A B
t
:>
INFLECTION
POINT
w
...J
W
YI
,-- -----r--"-" ""PRESENT si~ Yo
o X, 50 100
% CUMULATIVE AREA
c o
,EPOCHZ t
:>
w
...J
W
:>
~
w ~S. L.
..,..JRANGE
TIME ---. o 50 100

% CUMULATIVE AREA
Fig. 2. Example of calculation of sea-level elevations using hypsometry and continental flooding
data. (A) Area flooded for a given epoch is measured from an equal-area paleogeographic map relative
to a specified reference level, such as the shelf margin. (B) Fractional area flooded (x 1) is converted to
sea-level elevation (Yl) using a landmass-specific hypsometric curve. (C) Secular sea-level variation
curve is constructed from a series of paleogeographic maps. (D) Given time-invariant continental
coastal hypsometry, a specified flooding range yields a corresponding range of sea-level elevations.
A "hypsometric chord" (H. C.) is a linear representation of part of a continental area-elevation
distribution.
We term linear segments of continental area-elevation distributions "hypsometric

chords" (Fig. 2D).
The principal difficulty in using hypsometry for eustatic analysis is in selecting
an appropriate hypsometric curve with which to transform ancient flooding records
to sea-level elevations. Because paleo-continental hypsometries cannot be recon-
structed on purely theoretical grounds, modem area-elevation distributions must
be utilized as analogs. However, continental hypsometries change through time,
and, consequently, modem area-elevation distributions are likely to yield increas-
ingly less accurate sea-level estimates for progressively older flooding records
(Algeo and Wilkinson, 1991). This limitation has constrained previous hypso-
metric studies of eustasy to the Cretaceous or younger epochs (e.g., Bond, 1976,
1978a,b, 1979; Harrison and others, 1981, 1983; Hallam, 1984; Veevers, 1984;
Harrison, 1988, 1990). Rather than using the area-elevation distributions of mod-
ern continents, we have developed a method of reconstructing mean long-term
hypsometries for individual paleo-continents based on comparative analysis of
their flooding records.
2.2.2. Hypsometric Reconstruction of Paleozoic Continents

Definitions and Assumptions. Calculation of paleo-continental hypsometries and
sea-level elevations relies on a few basic mathematical relationships and underly-
ing assumptions. For each continent i at each epoch j, the original extent of marine
strata is estimated as F( ij)mean (Table 1). Mean Paleozoic flooding for each conti-
nent is calculated as an unweighted average of epochal flooding estimates, and the
range of Paleozoic flooding is calculated as plus or minus one standard deviation
about the mean:
n
JLF(i) =j L: F(ij)jn, (2)
n
CTF(i) = [j L:[F(ij) - JLF(i)f jnjO.5, (3)
I
rF(i) = JLF(i) CTF(i), (4)
where JLF( i) is mean flooding, CTF( i) the standard deviation of flooding, and rF( i)
the range of flooding values of continent i for the Paleozoic as a whole, and n
is the number of epochs during which continent i was an independent entity. As
true range is often a poor measure of the dispersion or variance in a dataset, we
use a "range" defined as one standard deviation about the mean (i.e. 16th-84th
percentiles) for flooding and sea-level estimates. Means and standard deviations of
Paleozoic flooding values for each continent are given in Table 2 based on values
of F(ij)min, F(ij)mean and F(ij)max.
Our analytical method entails several fundamental assumptions. Eustatic ele-
vation is assumed to be uniform at any moment in time, and eustatic changes are
assumed to be globally synchronous. Although departures from this assumption
exist in the modem world in the form of the geoid (e.g., Marner, 1976), the spa-
tial and temporal domains of geoidal variation are largely unknown and will be
assumed to average out globally at epochal timescales. Any deviations of individ-
ual continental sea-level elevations from the global mean as a result of geoidal
variations will appear as elevation residuals (see below) and can be interpreted as
such at that point. The corollary to this assumption is that co-existing continents
experience the same mean, standard deviation, and range of eustatic elevations
during any specified time interval (Fig. 3A).
Despite the assumption of eustatic uniformity, large differences in flooding
can nonetheless be expected among co-existing continents owing to differences in
hypsometry. Given invariant continental hypsometries and a characteristic disti-
bution (e.g., Gaussian) of eustatic elevations, similar (but offset) distributions of
A SEA LEVEL
FREQUENCY
~ HYPSOMETRIC CHORDS ~
.. .. .. ..Z:::l
. . ~.................\Y.f . . . . .
. . . . . . ~. . . . !'r.(:;~
. . . . . . . . Z:::::l ... . . . . . ..Z::::........ -ld
\l,(wl
I ,II
I
r f (1) , r f (2) r f (3)

1'1 I
,
:-+--: :..
I I I I
I _: ~
I
I I
_ I '
I I 'I I
....
b~b
=t "+
bE
"7 ~
b
"+
(!)~
Zz
-w
0:::>
80
...JW
u.ff
a 20 40 60 80 100
AREA FLOODED (%)
8 SEA LEVEL
FREQUENCY
300
200
E
...... 100
z
0
~ r ,(a)
>
~
a I
14 I -
W I
, I
....
:'0
I
: . =t ...
-100 ~ b
I :
I
I
-200
a 20 40 60 80 100
CUMULATIVE AREA (%)
Fig. 3. Reconstruction of paleo-continental hypsometric chords. (A) Relative (non-dimensional)

scaling of hypsomctric chords for three co-existing landmasses exhibiting low (L), medium (M), and
high (H) degrees of average flooding, /IF(i), and variable flooding ranges, /IF(i) + uF(i) (bottom).
As co-existing landmasses experience the same mean and range of eustatic elevations (upper right),
differences in means and ranges of flooding values among continents record differences in their
coastal hypsometries (top). (B) Absolute scaling of hypsometric-chord elevations using a modern
hypsometric analog, for which an "expected" range of flooding is estimated. Actual elevation range
is dependent on choice of analog.
TABLE 2
Flooding ranges for Paleozoic continents
Mean" Min." Max."
Gondwana 22.84.3b 16.33.8 29.36.1

Pangea 17.31.4 14.31.2 20.31.7
Laurentia 36.213.6 32.413.9 40. l 13.5
Laurussia 32. l 10.0 29.21O.O 35.010.1
Baltica 29.36.8 28.06.7 30.87.0
Siberia 46.99.6 42.79.0 51.210.9
Armorica 77.16.2 61.31O.5 92.07.6
Chukotka 37.05.3 16.75.0 57.56.4
Kazakstania 60.416.6 53.516.3 67.417.5
North China 28.424.1 25.824.3 31.024.0
China (united) 53.313.5 50.512.8 56.214.3
South China 54.57.1 44.35.9 64.88.7
Indochina 88.03.2 48.05.7 100.00.0
a Means and standard devIatIOns of floodmg values for
each continent, J.lF(i) O"F(i), based on values of
F(ij)mcan, F(ij)min, and F(ij)max, respectively.
b Fractional flooding of total landmass area in percent.
flooding values should be recorded for a group of paleo-continents. Differences in

the mean, standard deviation, and range of flooding values for each landmass can
be used to partially reconstruct a characteristic long-term mean coastal hypsometry
for each landmass (Fig. 3A). Hypsometric reconstruction proceeds in two steps: 1)
non-dimensional scaling of paleo-continental hypsometric chords, and 2) absolute
scaling of hypsometric chords using a modern continental analog. In the first step,
the means and ranges of flooding values for each paleo-continent are set equal
to the same non-dimensional mean and range of eustatic elevations, 'JLs(w) and
'rs(w) (Fig. 3A). In the second step, dimensional values of the mean and range
of eustatic elevations, JLs(w) and rs(w), are substituted for 'JLs(w) and'rs(w),
permitting calculation of dimensional values for the slopes and y-intercepts of
paleo-continental hypsometric chords:
m(i) = rs(w)/rF(i), (5)

b(i) = JLs(w) - m(i) JLF(i), (6)
where m( i) and b( i) are the slope and y-intercept of the hypsometric chord of
continent i, and JLs (w) and rs (w) are estimates of the mean and range of global
sea-level elevations (Fig. 3B). This relative scaling procedure minimizes variance
in resultant sea-level elevation estimates among continents and assigns as much
of the total variance in flooding values as possible to differences in the long-term
mean coastal hypsometries of continents.
400
300
~ 200
.5
z
0
100
~
>
w
--' ICE-FREE SEA LEVEL
w
0 PRESENT SEA LEVEL
-100
-200 I
0 10 20 30 40 50 60 70 80 90
CUMULATIVE AREA (%)
Fig. 4. Coastal hypsometries of modern continents (excluding Antarctica) with respect to shelf
margins (ca. -200 m). Hypsometric curves are cubic-splined versions of the area-elevation data of
Harrison and others (1983). The modern Indochina curve was calculated from equal-area 1:2,500,000
topographic maps and includes Borneo, Sumatra, Java, the Malaysian peninsula, the Southeast Asian
peninsula (northward to the Song Ma-Red River region in northern Vietnam and westward to
the Sagaing Fault zone in central Burma; Hutchison, 1989), and all intervening shallow-marine
areas (2: -200 m). Inflection points (I, in meters; circles) and hypsometric-slope minima (m,
in meters/%area) for Eurasia: I = 30, m = 5.0, Africa: I = 310, m = 6.7; North America:
I = 15, m = 6.7; South America: I = 90, m = 5.6; Australia: I = -60, +llO, m = 6.1,
5.6; and Indochina: I = -25, m = 1.4. Modern Indochina is included to illustrate the range of
potential hypsometric variation; it may be a better analog for small Paleozoic landmasses than modem
continents.
Choice of Modern Hypsometric Analog. As our analytical method only per-

mits reconstruction of relative coastal hypsometries based on paleo-continental
flooding data, development of an absolute elevation scale requires recourse to a
modem hypsometric analog. Next, we consider the hypsometric characteristics of
modem continents, evaluating their suitability as hypsometric analogs. The best
compilation of modem area-elevation data is that of Harrison and others (1983),
who utilized a Defense Mapping Agency database consisting of global elevation
values for 1 degree squares at elevation intervals of 100 m. We have fitted the
area-elevation data of Harrison and others with cubic splines to produce coastal
hypsometric curves for each modem continent (Fig. 4).
The curves for North America, South America, and Eurasia are rather similar,
having hypsometric slope minima of 5.0 to 6.7 mJ%area and inflection-point el-
evations slightly above present sea level (+ 15 m to +90 m; Fig. 4). For reasons
discussed below, these continents are probably close to their long-term hypsometric
equilibria. Although Africa and Australia have similar hypsometric slope minima
(6.7 and 6.1 mI%area, respectively), their hypsometric curves exhibit anomalous
inflection-point elevations (+310 m and -60 m, respectively). Thus, these conti-
nents have area-elevation distributions that appear to have been offset vertically
in relation to those of other modem continents by about +250 m and -75 m, re-
spectively. The implied large-scale uplift of Africa and subsidence of Australia is
consistent with other evidence of geologically recent crustal motions (e.g., Bond,
1978a; Veevers, 1984; Sahagian, 1988). The apparent hypsometric disequilibria of
these continents render them unsuitable for use as area-elevation analogs, and we
will utilize analogs based on Eurasia and America (the latter an areally-weighted
average of the coastal hypsometries of North and South America, which are similar
in size and tectonic character).
Scaling of Paleo-Continental Hypsometries. The procedure we use in dimen-
sional scaling of paleo-continental hypsometric chords is based on relationships
between landmass area, hypsometry, and flooding. The scaling procedure entails
estimation of two parameters for a Paleozoic landmass equal in area to the chosen
modem analog: 1) an "expected" mean flooding value, J.Lf(a), and 2) an "expected"
range of flooding values, rf( a). Once the mean and range of flooding values have
been estimated, these can be converted to absolute elevations using the hypsome-
try of the modem analog. The entire set of paleo-continental hypsometric chords
is then scaled to the mean and range of elevations determined from the modem
analog (J.Le(a) and re(a), respectively; Fig. 3B).
Among Paleozoic continents, both mean flooding and flooding variability are
related to landmass area. Mean (and maximum) flooding decreases with increas-
ing landmass area: small Paleozoic continents 107 km 2 ) average about 50%
flooding with a maximum> 90%, whereas large continents (> 5 x 107 km2 )
average about 20% flooding with a maximum near 30% (Figs. 5, 6A). Large con-
tinents are generally less "floodable" than small continents owing to a smaller
ratio of low-elevation coastal area to total landmass area (Cogley, 1985; Algeo and
Wilkinson, 1991). Flooding variability (i.e., the standard deviation of observed
flooding values for a given continent) also decreases with increasing landmass
area: large Paleozoic continents exhibit variabilities of less than 5%, mid-sized
continents 5-15%, and small continents 5-25% (Figs. 5, 6B). Because large conti-
nents, as a group, exhibit less variability in flooding range and, hence, hypsometric
character than small continents, the equilibrium hypsometries of large modem
continents (i.e., North America, South America, and Eurasia) are likely to be good
analogs for those of large Paleozoic continents. This assumes a large degree of
uniformitarianism in continental geomorphology during the Phanerozoic.
These relationships permit estimation of means and ranges of flooding val-
ues for the chosen Eurasian and American hypsometric analogs. Given areas of
7.810g km 2 and 7.4 log km 2 for these two continents, "expected" mean flooding
values for area-equivalent Paleozoic landmasses (J.Lf(a) are 24.7% and 34.8%
(Fig. 6A), "expected" values of flooding variability (CTf(a) are 6.7% and 8.8%
(Fig. 6B), and "expected" flooding ranges (rf( a) are 18.0-31.4% and 26.0-43.6%,
respectively. Conversion of "expected" flooding values to sea-level elevations us-
ing the hypsometric curves of the modern analogs yields a mean elevation (J.Le(a)
of +60 m and an elevation range (re(a) of24-102 m based on the Eurasian analog
100r-------~----------------------------------_.
o CHINA (united) o PALEOZOIC
o
i
MODERN (+50 m)
80 o MODERN
o 0
o
fa
o
60 o 0 0 o
o o Cb______
-... 0 0
~
o...J o 0)~ A1~~ 0
u.
40 t/O ~80
1
fGONDWANA
w
a:
ARMORICA
8 00 AU.
Q
0
0
-.............
' f PANGEA
~
o. ......
CHUKOTKA
20 INDOCHINA 0
00 0 fi! NA EA 8-.........:
8
KAZAKHSTANIA
N. CHINA - SA 0 0 0 0
S.CHINA 0 0
o 8AF
O+-______~--------~------~--------,_------~
6.0 7.0 8.0
LANDMASS AREA (log km 2 )
Fig. 5. Flooding versus landmass area for Paleozoic and modem continents. For Paleozoic conti-
nents, mean and maximum flooding decrease with increasing landmass area. For modem continents,
flooding values are shown with respect to both present sea level (circles with dots) and that of an
ice-free modem world (+50 m above present sea level; solid circles). Modem continents: AU =
=
Australia, SA South America, NA = North America, AF = Africa, and EA Eurasia. Note that =
modem continents exhibit generally low degrees of flooding in relation to Paleozoic landmasses of
equivalent area.
(Fig. 7A), and a mean elevation of + 160 m and an elevation range of 96-224 m
based on the American analog (Fig. 7B). Although the hypsometries of Eurasia
and North and South America are similar (Fig. 4), the two analogs yield large dif-
ferences in estimated elevation ranges owing to differences in the degree to which
the actual flooding of modern continents diverges from their "expected" Paleozoic
flooding values: present flooding of Eurasia (23%; Fig. 5) is close to its expected
Paleozoic mean (25%; Fig. 6A), but present flooding of America (22%) is con-
siderably lower than its expected mean (35%). This makes clear the implications
of choosing Eurasia versus America as a hypsometric analog: the former implies
that modern sea-level elevations are rather typical for the Phanerozoic as a whole,
whereas the latter implies that they are unusually low.
2.3. PALEOZOIC SEA-LEVEL AND EPEIROGENIC TRENDS
2.3.1. Calculation of Sea-Level Elevations

Once coastal hypsometric chords have been constructed and scaled for individual
Paleozoic landmasses (Fig. 7), calculation of sea-level elevations for each continent
100~----------------------------------------'
,
, 61ND A
80 ~)'~, ... .L(f(a)
'~~: /
"I::.~
:;
C!J
60 AMERICA
C:~' ,SIB
Z
o KAZ /
9 40
LL
~""'......
1 EURASIA
Z ... - : ) - E'CHK - - - _ _ 6A. . . .....LRS

LNT.
w
(_'v - _y ........ .....
BLT - - ..:::::::::.... ......GND
~
N-CHN - .. O~~ :--.. PNG
20 ----::s. .
-~-
~
o~=====c====~======~=====c====~
25 .N-CHN
B
.. ~,
'~)
cfl. 20
--- ' ..... , .....:;o'f(a} AMERICA
~
:J
15
~KAZ .............. /
'f. . ~Lr:...S
[Q
....
.CHN ............
LNT
a EURASIA
0:
~
C!J 10 (-10') SIB ~ ..... 1
~~ ........ -
Z
o
o ;-~N~-:---.--
9LL 5
ARM v
CHK 6 BLT -- - - - _ ~- ....
-~. -.
61ND GND PN~" -_
o+--------.--------.--------.--------.-------~
6.0 7.0 8.0
LANDMASS AREA (log km 2 )
Fig. 6. Mean flooding and flooding variability for Paleozoic continents. Both (A) mean flooding,
{F(i), and (B) flooding variability, aF(i), decrease and exhibit a narrower range with increasing
landmass area. These relationships are used to estimate "expected" means and ranges of flooding
values for hypsometric analogs. Paleozoic continents equivalent in area to modern Eurasia and
America would exhibit flooding ranges of 25 7% and 35 9%, respectively (open circles).
Uncertainties in estimates of "expected" mean flooding and flooding variability (solid vertical bars)
are based on 1 regression lines (dashed). The low flooding variabilities exhibited by some small
paleo-continents (e.g., Chukotka) reflect time-invariant flooding estimates that result from a lack of
detailed stratigraphic information rather than from genuinely small ranges of flooding values.
224 T.]. Algeo, K.B. Seslavinsky
3OO~----------~-----------------------------~~-------, ,
A /
o'<l: /
/ EURASIAN / /
ANALOG /
,,
rF/
200 ~/ /
/
3/4
/
112
u/m ~1!ij_~~~::.o!JIII!!*,~'-----::":'--4-
1~~~~~~----'-7
,
... J ......... ~.~ ................~.......... .,~.~ ....... ... . -1 cf
PRESENT SEA LEVEL ................... .
,,- ,,--
J / _---
/
/
I l
1,'/
*'
,"
-100
8 /
/
r-_______ /-,1r-~----~--_
/
200
,'---1--1--...,..,.
/
g 100
1----
,
1______----=
z
0
~
>
UJ
--I
UJ
0
-100 3 BALTICA
4 LAURUSSIA 10 KAZAKHSTANIA
5 CHUKOTKA 11 ARMORICA
6 LAURENTIA 12 INDOCHINA
-200+----.----.----,----r----.---.r---.----.----.---~
o 20 40 60 80 100
CUMULATIVE AREA (%)
Fig. 7. Scaled hypsometric chords of Paleozoic continents using Eurasian (A) and American (B)
analogs. The Eurasian model yields a mean elevation of +60 m and an elevation range of 24-102 m,
whereas the American model yields a mean elevation of +160 m and an elevation range of 96-224
m. The area-elevation curves of modern Africa and Indochina bracket the coastal hypsometries of
most Paleozoic continents. Note that choice of an American analog implies steeper paleo-continental
hypsometries and higher sea-level elevations than choice of a Eurasian analog.
at each epoch is possible (Figs. 2, 8). This involves conversion of flooding values
to sea-level elevations using the characteristic hypsometric chord of each paleo-
continent:
S(ij) = m(i) . F(ij) + b(i), (7)
where S( ij) is the sea-level elevation estimate for continent i at epoch j. For
example, Laurentia has flooding values of 21 %, 32%, and 48% for the Lower,
Middle, and Upper Cambrian, respectively (Table I), yielding sea-level elevations
of 20 m, 45 m, and 90 m based on the Eurasian analog, or 80 m, 130 m, and 220 m
based on the American analog (Figs. 7, 8). Sea-level elevation estimates are not
appreciably affected by use of linear hypsometric "chords" rather than continuously
varying curves, because most modern continents have fairly uniform hypsometric
slopes at low elevations (Fig. 4). Significant non-linearity would develop only at
extreme flooding values (i.e., > +2a).
Once continental sea-level elevations have been determined for each epoch,
global sea-level elevations may be calculated by several methods. In theory, each
paleo-continent is an independent recorder of eustasy having equal validity, and,
therefore, global sea-level elevations could be calculated as a simple unweighted
average of individual continental sea-level elevations for each epoch:
m
S(wj) = i L S(ij)jm, (8)
1
where S( wj) is the unweighted estimate of global sea-level elevation, and m is the
number of continents extant during epoch j. However, the variance of individual
continental sea-level estimates is related to flooding data quality: on average,
estimates based on excellent- to good-quality data deviate by 40--45 m from global
mean sea-level elevations, whereas those based on poor-quality data deviate by
62 m. Therefore, we prefer eustatic estimates reflecting data quality:
m m
5.(wj) = i L[S(ij) . Q(ij)]ji L Q(ij), (9)
1 1
where 5.( wj) is the quality-weighted estimate of global sea-level elevation, and
Q( ij) is the quality value of the flooding estimate for continent i at epochj. Quality
values were assigned to individual continental sea-level estimates according to a
scale of 1.0 for excellent-, 0.5 for good-, 0.33 for fair-, and 0.125 for poor-quality
flooding data (quality values based on ratios of flooding extrapolation values; see
above). Global sea-level trends for quality-weighted estimates are shown in Fig. 9.
2.3.2. Calculation of Epeirogenic Trends

Individual continental sea-level estimates diverge to greater or lesser degrees from
the global mean elevation for each epoch, yielding "elevation residuals":
eS(ij) = S(ij) - S(wj), (10)

226 T.l. Algeo, K.B. Sesla vinsky
ISO
GONDWANA A SIBERIA
300
100 , PANGEA "
so " .
S(IJ)mean -',
1\
,'..
"\
'- ~ \, , 100
" . . ,/ "',' '.,t ',,'
0 " S(ijjmln/
.... , ..............................
.~
\
"',' . ,;;/. .......... . ............................. - .. -.. . .................... 0
so '.'
(!)
9 LAURENTIA B F KAZAKHSTAN IA

<{
Z
ISO LAURUSSIA 300 ~
<{ <{
Z
<{ 200 ~
(i) so SJ
<{
c: 100 ffi
:::> ~
w 0
o
<{
so
I I
z c G z
>
BALTICA NORTH CHINA
ISO 300
~~\
i=
<{
W
100
200
~
W
...J ...J
W W
SO
...J 100 ...J
W W
>
W
0 >
W
...J ..................................................... . ... . 0 ...J
<{ so
w
(/) ,,
ISO
,
'. I
,; o ,"
,
INDOCHINA 300
! ,
J: "\:
I
I ,
100
" ARMOR ICA
SO
.. ___1
," "-
I \
\
... ,
I
0 I I
................ ..... -,, ................ ................... .
I ,. , I
; , 'It , 0
SO ,,
em Pm
AGE (Mal
Fig. 8. Sea-level elevations for individual Paleozoic continents. Estimates are based on the
"most-likely" or mean flooded area (F(ij)mcan; solid line) and on minimum and maximum flooding
estimates (F(ij)min and F(ij)ma,. respectively; dashed lines). Sea-level trends are significant where
elevation changes exceed the uncertainty range. Elevation scales are shown for both Eurasian (left)
and American (right) analogs.
~ EURASIAN ANALOG AMERICAN ANALOG .....
o - - S(wil mu
150
.. . -" .
- S(wj)mean
300
- - S(wil m1n
:g ]
Z 100
o z
200 i=
o
~
>
W ~
-l W
..J
W 50 W
-l ..J
W W
>
W 100 Gj
..J ..J
~
C/) 0
~
W
(/)
"'\,:
"1&""
o
-50
500 400 300

AGE (Ma)
Fig. 9. Global and stacked continental sea-level elevations for the Paleozoic. Symbols on continental
sea-level curves (dotted) represent data quality; refer to Fig. 1 for meaning of symbols and to Fig. 8
for identification of paleo-continents. The mean global curve (heavy solid line) is a quality-weighted
average of continental sea-level elevations using weights of 1.0 for excellent (solid squares), 0.5
for good (solid triangles), 0.33 for fair (open squares), and 0.125 for poor data (open triangles).
The uncertainty range for global sea-level elevations (shaded) is based on minimum and maximum
flooding values for all paleo-continents at each epoch (Fig. 1). Elevation scales are shown for both
Eurasian (left) and American (right) analogs.
where eS(ij) is the difference between the sea-level estimate for continent i
and the global mean at epoch j. Generally, high (low) flooding values (Fig. 1)
yield positive (negative) elevation residuals (Fig. 10). The maximum amplitudes
of elevation residuals are about +90 m and + 150 m based on the Eurasian and
American analogs, respectively, which are comparable in magnitude to eustatic
ranges of ca. 100 m and 225 m for the respective hypsometric analogs (Fig. 9).
Deviations of continental sea-level estimates from the global mean are due
to one of two causes: 1) secular changes in paleo-continental hypsometry, or 2)
errors in flooding estimates. Although we have assumed to this point that the
coastal hypsometry of each Paleozoic landmass is time-invariant, secular changes
in the elevation or slope of a hypsometric chord would result in deviations of
continental sea-level elevations from the global mean. The simplest interpretation
of elevation residuals in terms of secular hypsometric variation invokes vertical
displacement of hypsometric chords such that a positive (negative) motion yields
a negative (positive) elevation residual of equal magnitude. This interpretation
Accelerated
ActiveMargin A E Collision with
AldaniBarzugin
200
50 ' .... -.......

Convergence
"..:', I Terrains
100
50
, PANGEA
100
\
\
.~
,I
SIBERIA 200
GONDWANA
,
(!) PassiveMargln B F rlrahan
..:-, I
2OO(!)
9 100 ~~g~7:,~nce M'Clintock
Collision with
Kokchetav &
,?rogen y
9
,. . . . ,I,
Orogeny AntierlCanbou Other Terrains
Z 100 Z
50
'- ,~ I
'\f;orogen y
Z
0
I. ~ - ..
.. , ~.~., ..... .. .. . :~:::::: -- .. -- -.. .. o ~
(ij ()
a: 50 ii:
100 W
::J ::!:
w 100
LAURENTIA LAURUSSIA KAZAKHSTANIA 200
:g
;J, Finnmarkian C G Subduction along
Suolun-Xar & Shanyang- ...J
. ... . . .~.../J;' ~
100
orogen y Caledonian Tongcheng Zones ~~g
::J
Q -50 ,
/
/orogeny , S'bena 100 5
~.. ....... ....... --....... ---
(JJ
W
a: 0 .. .. -.
Z
o 50
i=
i= , 100
:; 100 NORTH
W
BALTICA CHINA CHINA
200 W
>
...J ...J
W W
/Syn.rift Thermal
Uplift D H CHUKOTKA
-200
100
, /,
INDOCHINA
50 ,
\
I -100
o
50 /\ ~ .... ,
I \ I \ ,
/~ \ I \;'
100 / \,.."'" ""', / ARMORICA 200
Cm Pm
500 400 300 500 400 300
AGE (Mal
Fig. 10. Elevation residuals for individual Paleozoic continents. An elevation residual is the deviation
ofa continental sea-level estimate from the global mean at a given epoch (Fig. 9). Residuals represent
a combination of errors in flooding estimates and secular changes in paleo-continental hypsometry,
e.g., as a consequence of epeirogeny. To facilitate interpretation of residuals as epeirogenic motions,
elevation scales are inverted: a negative residual (low relative sea level; Fig. 8) is equivalent to
positive epeirogenic motion (uplift) and, conversely, a positive residual (high relative sea level)
is equivalent to negative epeirogenic motion (subsidence). If viewed as vertical movement about
a fixed, neutral level of buoyancy, these curves yield an "epeirogenic history" for each landmass.
Uncertainty ranges for elevation residuals (shaded) are based on minimum and maximum flooding
values (Fig. I). Elevation scales are shown for both Eurasian (left) and American (right) hypsometric
analogs; values cited in text are for the American analog.
receives empirical support from modern continental hypsometric anomalies (e.g.,

Africa and Australia; Fig. 4), in which the sense and scale of offset of hypsometric
inflection points is consistent with the direction and relative magnitude of recent
vertical crustal motions.
Alternatively, elevation residuals may result in part or in whole from errors in
flooding estimates (Fig. I). The potential effects of such errors may be evaluated by
calculating an uncertainty range for elevation residuals: by substituting F(ij)max
and F( ij)min for F( ij)mcan in Eq. (10), upper (8S(ij)max) and lower (8S(ij)min)
limits on elevation residuals for continent i at epoch j are established (dashed
lines, Fig. 10). When the uncertainty range encompasses the origin of the ordinate
(Le., 0 m), the difference between individual continental and global mean sea-level
estimates is probably not significant (e.g., Cambro-Silurian, Gondwana; Fig. lOA).
When elevation residuals are larger than the enclosing uncertainty envelope, they
are likely to represent secular changes in paleo-continental coastal hypsometries
(e.g., Cambro-Silurian, Laurentia; Fig. lOB).
3. Discussion
3.1. EUSTASY
3.1.1. Paleozoic Eustatic Elevations

First-order Paleozoic eustatic trends are similar in our curve and those of Vail
and others (1977) and Hallam (1984), all of which exhibit a Cambro-Silurian
Caledonian and a Devono-Permian Appalachian-Hercynian cycle (Fig. 11). On the
other hand, these curves exhibit large differences with regard to eustatic amplitudes:
the Vail and Hallam curves have Paleozoic sea-level maxima of +200 m to +300 m
and +300 m to +600 m, respectively, which are substantially higher than those
of this study (+ I 00 m to +225 m). Independent estimates of Paleozoic eustatic
amplitUdes are few. Backstripping methods of subsidence analysis have yielded
"changes in accommodation" (eustasy plus local tectonism) of ca. 100-200 m
in the North American midcontinent area during major Paleozoic transgressions
(e.g., Bond and Kominz, 1991), although such estimates may exceed 300 m for
continental margin sequences (Osleger and Read, 1993).
A comparison of the Paleozoic supercycle with the better documented Meso-
zoic-Cenozoic supercycle is warranted. The Late Cretaceous highstand has been
estimated at + 175 m to +250 m based on hypsometry, mid-ocean ridge volume
analysis, sequence stratigraphy, and subduction rate analysis (e.g., Bond, 1979;
Harrison and others, 1983; Kominz, 1984; Haq and others, 1987; Engebretson and
others, 1992), which is lower than the Paleozoic highstand elevations of Vail and
others (1977) but higher than those of this study. Further, the total length of passive
margins created during supercontinent breakup was greater during the lurassic-
Cretaceous (ca. 35,000 km; Harrison and others, 1981; Heller and Angevine, 1985)
than during the Eocambrian (ca. 18,000 km; Bond and others, 1984). Because
passive-margin lengths and eustatic elevations are both controlled by geotectonic
supercycles (e.g., Heller and Angevine, 1985; Gurnis, 1992a), inferred positive co-
600 (\
\
:g
)
I II \.....--- HALLAM (1984)
1\ _I \
:
z I
I l
I I
I VAIL & OTHERS (1977)
0 400
~
> :/_r~/j r/,/i\
W
-I " ____ . __ , I '- .... , HAC & OTHERS /:'
W I ",_ ,(1987) J :
() 200 I '--- \ \ I,'
i= I \ r\) -

t-
\ /),/
(J) ,- ( -'
:::> " \ _, '/.' . . '~ 1 r',
w
0 .......... J..... ...... ...... ......... ........ ....... .................... ...........................................1/. ..\
.:' THIS STUDY 'I
SUPERCYCLE
CYCLE
PERIOD Cz
600 o
Fig. 11. Phanerozoic eustatic trends according to Vail and others (1977), Hallam (1984), Haq
and others (1987; smoothed version of their long-term curve), and this study. Our Paleozoic curve
(heavy solid line; spliced to Haq curve at Permo-Triassic boundary) is based on mean values for
all model parameters and a hypsometric analog that is an unweighted average of the Eurasian and
American curves; uncertainty range (shaded) from Fig. 9. Note the substantially lower Paleozoic
eustatic elevations of our curve relati ve to those of the two existing Phanerozoic sea-level curves.
variance between these parameters implies that eustatic elevations associated with
the Paleozoic were no greater than those of the Mesozoic-Cenozoic supercycle.
3.1.2. Controls on Long-Term Eustasy

Although many factors influence eustasy (e.g., Donovan and Jones, 1979; Harri-
son, 1988), long-term (Le., > 10 m.y.) sea-level cycles are commonly attributed
to changes in the lengths and spreading rates of mid-ocean ridges (MORs; Pitman,
1978; Kominz, 1984; Larson, 1991) or to interbasinal differences in the age distri-
bution of oceanic lithosphere (Heller and Angevine, 1985). The potential effects
of these factors on global sea-level elevations are substantial: two-fold reduction
of global spreading rates since the Late Cretaceous may have caused a eustatic fall
of ca. 230 115 m (Kominz, 1984), and evolution of ocean-crust age distributions
during growth of an Atlantic-type ocean basin may induce eustatic fluctuations
of ca. 40-100 m (Heller and Angevine, 1985). A related factor linked to geotec-
tonic cycles is ocean basin volume, which decreases during continental rifting and
attenuation of passive margins and increases during continental collision and litho-
spheric thickening in orogens. The potential effects of this factor are substantial:
Heller and Angevine (1985) estimated a 50-90 m eustatic rise owing to continen-
tal margin extension during development of an Atlantic-type ocean, and Harrison

(1988) estimated a 70-m eustatic rise associated with post-Jurassic passive-margin
creation and a 20-m eustatic fall associated with the Eocene-to-Recent collision of
India and Eurasia.
The largest changes in Paleozoic eustatic elevations are associated with the
onset and termination of geotectonic supercycles. A rise of 50-100 m during the
Early-Middle Ordovician (Figs. 9, 11) may reflect a combination of increases in
MOR lengths and spreading rates, formation of new Atlantic-type ocean basins,
and decreases in ocean basin volume associated with attenuated passive margins.
During geotectonic rift phases, all three factors operate to raise global sea levels,
as during the Jurassic-Cretaceous disintegration of Pangea (Heller and Angevine,
1985; Wyatt, 1986). Absence of information regarding lengths and spreading rates
of Paleozoic MORs precludes determination of the relative contribution of each
factor to the Early-Middle Ordovician eustatic highstand (e.g., Vail and others,
1977; Hallam, 1984; this study). However, the timing of the Early Paleozoic
highstand is consistent with peak sea-level elevations some 50-100 m.y. after
supercontinent breakup in response to shifts in the area-age distribution of oceanic
lithosphere (e.g., Heller and Angevine, 1985).
The Caledonian and Appalachian-Hercynian cycles are terminated by eustatic
falls of 40-70 m during the Late Silurian and Late Permian, respectively (Figs. 9,
11), which coincide with major continental collisions, Le., Late Silurian suturing
of Laurentia, Baltica, and Chukotka to form Laurussia, and Late Carboniferous-
Permian suturing of Gondwana, Laurussia, Kazakhstania, and Siberia to form
Pangea. Several mechanisms could account for synchroneity of major Paleozoic
eustatic falls with continental collision events, although some explanations (e.g.,
reduced global spreading rates) lack empirical support. The simplest explanation is
an increase in ocean basin volume resulting from orogenic thickening of continental
lithosphere during the Caledonian and AppalachianlHercynianiUralian orogenies,
similar to but larger than that proposed for the India-Eurasia collision (Harrison,
1988).
3.2. CONTINENTAL EPEIROGENY
Paleozoic continental elevation residuals (Fig. 10) may record epeirogenic motions,
as suggested by large-scale vertical displacements of modern continents (e.g., Fig.
4). Given the limited existing data on continental epeirogenic histories (Bond, 1976,
1978a,b, 1979; Harrison and others, 1981, 1983; Hallam, 1984; Veevers, 1984;
Harrison, 1988, 1990), these residuals represent a potentially valuable source of
information regarding the frequency, magnitude, and origin of continental crustal
motions. However, interpretation of paleo-continental elevation residuals requires
a thorough understanding of epeirogenic processes. Therefore, before proceeding
to consider the significance of Paleozoic elevation residuals, we begin with a
review of the Cretaceous-Cenozoic epeirogenic histories of modern continents
and of proposed underlying mechanisms.
3.2.1. Modern Continental Epeirogeny

Complex patterns of crustal uplift and subsidence have been documented for
most modern continents during the Cretaceous and Cenozoic. In Africa, uplifts
occur as broad plateaus (ca. 106 km 2 ) up to 1 km in height with local domes (ca.
105 km 2 ) greater than 3 km in height (Sahagian, 1988; Westaway, 1993). Domes
exhibit complex Tertiary uplift histories with distinct phases of ca. 2-10 m.y.
duration (King, 1962; Rosendahl, 1987; Sahagian, 1988). In Australia, the most
prominent long-wavelength crustal feature is a north-south depression through the
center of the continent, transecting the Simpson Desert, Euroka Arch, and Gulf
of Carpentaria, and continuing southward at least as far as the Southeast Indian
Ridge, where it is known as the Australian-Antarctic Discordance (AAD; Palmer
and others, 1993).
Elsewhere, broad crustal uplifts are mostly associated with Late Cretaceous
and younger orogenies. In Asia, the Tibetan Plateau has been uplifted 4-5 km
in response to the collision of India with Eurasia (Molnar, 1989; Harrison and
others, 1992). In North America, large areas of the West, including the Basin and
Range, Cordillera, Colorado Plateau, and western Great Plains, have been uplifted
by 1-3 km during the Late Cretaceous-Recent Laramide Orogeny (Bond, 1979;
Sahagian, 1987; Ruddiman and others, 1989). Recent broad uplift has occurred
in central and southern Europe in association with the Alpine-Carpathian and
Pyrennean orogenies (Bond, 1979), as wel1 as in the Andean Altiplano of South
America (Al1mendinger, 1986). On the other hand, broad areas of the Russian
Platform may have subsided (Bond, 1979) or maintained a stable elevation since
the Cenomanian (Sahagian, 1989). Thus, al1 modern continents appear to have
been affected to varying degrees by recent epeirogenic motions.
Hypsometric analysis is capable of identifying major epeirogenic events. How-
ever, because this method integrates elevations over the surface of an entire conti-
nent, hypsometric estimates of elevation changes for a given event are invariably
muted in relation to actual crustal displacements occurring at a subcontinental
scale. Thus, although large areas of eastern Africa and western North America
have experienced uplift of 1-3 km during the Cenozoic, Bond's (I 978b} analysis
identified an average (continent-scale) uplift of 210 m for Africa (Oligocene-
Recent) and an average uplift of 150 m for North America (Paleocene-Eocene).
Thus, it must be recognized that hypsometric estimates of average continent-scale
elevation changes do not represent the actual length scales of epeirogenic events.
3.2.2. Mechanisms of Modern Continental Epeirogeny

Although the phenomenon of epeirogeny is wel1 documented, large uncertainties
exist regarding the operation and relative importance of various causative factors.
Three broad categories of mechanisms have been proposed: thermal isostatic,
non-thermal isostatic, and dynamic (i.e., non-isostatic; Table 3). Each mechanism
exhibits characteristic lateral, vertical, and temporal scales of operation. The lateral
scale of epeirogenic motions is definitional1y delimited to > 102 km (shorter
flexural wavelengths are characteristic oflocal tectonic processes) and is controlled
by the rheology of continental lithosphere and the size and locus (i.e., supra-, intra-,
or subcrustal) of applied forces (Forsyth, 1985). Vertical crustal displacements

range from a few meters to a few kilometers, and characteristic timescales range
from 103 yr to 108 yr (Table 3).
Among the numerous proposed mechanisms of continental epeirogeny (Ta-
ble 3), some are more thoroughly documented than others. In this section, we
consider several epeirogenic mechanisms in greater detail, focusing on those that:
1) have well-documented examples, and 2) operate at length and time scales large
enough to permit recognition through hypsometric analysis, and 3) are linked to
tectonic processes for which independent geologic evidence may be available:
1) Thermal subsidence of rift margins. Conductive and advective cooling of
attenuated passive margins following rifting results in exponentially-declining
subsidence of2-3 km, often amplified substantially owing to sediment loading
(Steckler and Watts, 1978; Royden and others, 1980). Characteristic features
of this process are hinged subsidence of long, linear belts along young passive
margins. Examples include the Paleozoic (Bond and others, 1989; Osleger
and Read, 1993) and Mesozoic-Cenozoic passive margins of North America
(Steckler and Watts, 1978; Heller and others, 1982).
2) Mantle plumes. Thermal buoyancy, in conjunction with dynamic mantle up-
welling, results in uplift of continental lithosphere by 1-3 km over mantle
plumes. Characteristic features of this process are small domical uplifts within
broad plateaus and association with tensional stress regimes, alkaline mag-
matism, and continental fragmentation. Examples include Mesozoic West
Gondwana (White and McKenzie, 1989; Peate and others, 1990) and late
Cenozoic East Africa (Sahagian, 1988; Westaway, 1993).
3) Dynamic topography. ThermaVdensity anomalies of mantle convective origin
result in broad topographic and geoid anomalies (Hager and others, 1985;
Gurnis, 1990b). The largest anomalies are associated with low-degree patterns
of mantle convection: highs over mantle upwelling zones and lows over
mantle downwelling zones. Because continents tend to move away from
the former and toward the latter, long-term cycles of continental flooding
may be linked to drift with respect to dynamic topography (Gurnis, 1988).
Examples include Africa, which is probably located over a mantle upwelling
zone (Hager and others, 1985), and Australia, which may straddle a mantle
downwelling zone, as suggested by seismic, gravimetric, and geochemical
data (Sempere and others, 199 I; Pyle and others, 1992; Kuo, 1993; Palmer
and others, 1993).
4) Variations in intraplate stress fields. Changes in the horizontal forces acting
on a plate cause changes in lithospheric density, leading to uplift under ten-
sional stress regimes and subsidence under compressional ones (Gay, 1980;
Cloetingh, 1988; Cathles and Hallam, 1991). A characteristic feature of this
process may be episodes of linked subsidence in intracratonic basins lacking
an obvious tectonic or eustatic driver. Possible examples include the North
American craton during the Devono-Mississippian (Kominz and Bond, 1991)
and the Australian craton during the Cenozoic (Lam beck, 1983).
TABLE 3 IV
Mechanisms of continental epeirogeny W
Mechanism Scales Reference "'"

Lateral (km) Vertical (m) Temporal (yr)
Thermal Isostasy
Thermal subsidence of rift margins 100-300 0-3000 107 _108 Steckler & Wails, 1978; Royden & others, 1980
Mantle plumes 100-1000 0-3000 106 _10 7 Griffiths & others, 1989; Westaway, 1993
Dynamic topography 500-5000 0-300a 107 _10 8 Hager & others, 1985; Gumis, 1990b, 1992a
Siabless windows 300-1000 0-2000 106 _107 Crough & Thompson, 1977; Smith, 1982
Mantle upwelling owing to
lithospheric delamination 100-1000 0-5000? 106_107 ? Bird,1978
Shear heating along
lithosphere-asthenosphere interface >10,000 0-3000? 107 _1O 8 ? Melosh & Ebe1, 1979
Partial crustal melting 100-1000 0-1000? 106 _1O7 ? McGetchin & others, 1980
Lower crustal magma intrusion 100-1000 0-3000? 106_1O7 ? McKenzie, 1984
Lithospheric phase changes 1000-5000 0-1500? 107 _1O 8 ? Smith,1982
Non-thennallsostasy
Lithospheric flexure/glaciation 100-3000 0-1000 IO J _IO S Peltier, 1980
Lithospheric flexure/tectonic loading 100-300 0-3000 106_107 Jordan, 1981; Quinlan & Beaumont, 1984
Lithospheric thickening/imbricate thrusting 100-1000 0-5000 106_108 Allegre & others, 1984; Molnar, 1989 t-i
~
Subducted slab buoyancy 100-1000 0-2000 106_108 Cross, 1986
106_IOR?"
:>
Variations in intraplate stresses 100-1000 0-100 Gay, 1980; Cloetingh, 1988 QQ
(b
Lithospheric underthrusting 100-1000 0-5000 106_108 Ni & Barazangi, 1983; Powell, 1986187 ~o
Crustal denudation 100-1000 0-20? < lOs Bishop & Brown, 1992
~
Subsurface dissolution 100-1000 0-100 104 _106 Opdyke & others, 1984 ~
Dynamic (Non-Isostatic) til
(b
II>
Viscous dynamic topography 200-1000 0-3000 107 _108 Mitrovica & others, 1989; Gumis, 199Oa, 1992b ~
-S
II>
Vertical offset in relation to center-of-Earth is 0-3000 m; value cited is in relation to hydrost?tic geoid.
b Cathles & Hallam favor a shorter timescale: 104_106 yr. ~
5) Lithospheric thickening owing to imbricate thrusting. Overthrusting and com-

plex deformation of continental margins under strong compressive regimes
may result in lithospheric thickening over wide areas (Allegre and others,
1984). Thickening of the lithosphere underlying the Tibetan Plateau is proba-
bly due to complex internal structural displacements, although limited under-
thrusting of the margins of the Tibetan Plateau (ca. 50-80 km) by the Indian
and Tarim Basin plates is likely (Hirn and others, 1984; Molnar, 1989). The
degree of lithospheric thickening and scale of uplift is primarily a function of
the rate of plate convergence (Molnar, 1989).
6) Viscous dynamic topography. Viscous coupling of subducting oceanic and

overriding continental lithosphere may result in continental subsidence in
back-arc settings (Hager, 1984; Mitrovica and others, 1989; Gurnis, 1992b).
Changes in subduction rate and in the age and angle of penetration of subduct-
ing oceanic lithosphere result in variable crustal motions; generally, increased
(decreased) subduction rates cause subsidence (uplift). This mechanism is of
greatest potential significance during passive-to-active margin transitions,
when onset of subduction may initiate back-arc subsidence (Gurnis, 1992b).
A possible example is the Cretaceous-Tertiary epeirogenic history of western
North America (Cross and Pilger, 1978; Mitrovica and others, 1989).
7) Subducted slab buoyancy. Subduction of young, hot oceanic lithosphere may

result in shallowing subduction angles and physical buoyancy of adjacent
continental margins (McGetchin and others, 1980; Cross, 1986). Progressive
uplift may result when a subduction zone migrates toward a mid-ocean ridge.
Tertiary uplift of the western margin of North America has been linked to
approach of the East Pacific Rise (Cross, 1986).
8) Slabless windows. Collision of a mid-ocean ridge with a trench may produce a

transform margin, along which spreading and formation of oceanic lithosphere
cease and craton ward of which a hole develops in the subducting oceanic
lithosphere. The progressively enlarging hole ("slabless window") permits
advection of hot mantle material, resulting in thermal uplift of overlying
continental lithosphere. High heat flow and uplift of the Basin and Range
Province since the Early Miocene have been attributed to collision of North
America with the East Pacific Rise and to development of a slabless window
east of the San Andreas Fault Zone (Crough and Thompson, 1977).
3.2.3. Paleozoic Continental Epeirogeny

In analysis of paleo-continental epeirogeny, the temporal and spatial character-
istics of the database of choice constrain the range of epeirogenic mechanisms
about which inferences may be drawn. Epeirogenic histories reconstructed from
continental flooding data at epochal intervals (as in this study) have a temporal
resolution of 1-4 x 107 yr and a vertical resolution of 10 to 100 m (the latter
a function of flooding data quality and resultant uncertainty ranges of elevation
residuals; Fig. 10).
Because many epeirogenic mechanisms operate at similar length and time

scales, it is generally not possible to identify the process responsible for a given
crustal motion from these parameters alone. However, most types of epeirogenic
motion have been shown to occur within specific plate tectonic contexts and, there-
fore, analysis of continental epeirogenic histories within a plate tectonic frame-
work should lead to a better understanding of the origins of broad crustal motions.
A complete analysis of this type is beyond the scope of the present paper, but
the following discussion will serve to demonstrate the potential utility of linking
paleo-continental epeirogenic and tectonic histories in order to constrain proba-
ble mechanisms of crustal motion. We will focus on Gondwana and Laurentia,
touching only lightly on other Paleozoic continents:
Gondwana. Gondwanan elevation residuals suggest broad continental subsi-
dence (-130 m) during the Early Silurian-Early Devonian and uplift (+260 m) dur-
ing the Middle Devonian-Early Carboniferous (Fig. lOA). Siluro--Carboniferous
epeirogeny may be related to the tectonic evolution of Gondwanan continental
margins. Late Silurian-Early Devonian flooding of broad areas of South America,
West Antarctica, and eastern Australia was followed by marine regression during
the Middle Devonian-Early Carboniferous (Khain and Seslavinsky, 1991; Scotese
and Golonka, 1993). In South America, where a widespread Late Devonian un-
conformity developed (Barrett and Isaacson, 1988), coeval arc-related magmatism
and terrane accretion occurred along the southern Andean margin (Ramos and
others, 1986). The Australian and Antarctic paleo-Pacific margins underwent a
late Middle Devonian transition from transtensional to convergent active margins
(Veevers, 1984), resulting in development of continental volcanic arc systems
within the Lachlan Fold Belt of eastern Australia (Jell, 1988) and in North Vic-
toria Land and Marie Byrd Land on the eastern Antarctic margin (Bradshaw and
Webers, 1988).
An unusual aspect of the Gondwanan "epeirogenic" record is that major ex-
cursions are antithetic to coeval eustatic trends: 130-m subsidence in the Early
Silurian-Early Devonian correlates with a 70-m sea-level fall, 260-m uplift in
the Middle Devonian-Early Carboniferous with a 50-m sea-level rise, and 11O-m
subsidence in the Late Carboniferous with a 30-m sea-level fall (Figs. 9, 10).
The large magnitude of these excursions and the 150-m.y. duration of the inter-
val of covariance argues against a coincidental relationship and implies a strong
causal connection between mid-Paleozoic eustasy and Gondwanan epeirogeny.
One possibility is that faster MOR spreading rates caused both a global sea-level
rise and uplift of Gondwanan continental margins owing to subduction of young
oceanic lithosphere (e.g., Gurnis, I 992b) or continent-terrane collisions. Deceler-
ation of MOR spreading rates in the mid-Carboniferous may be consistent with
a major plate reorganization following collision of Laurussia and Gondwana, as
evidenced by a mid-Carboniferous cusp in the North American APWP (DiVenere
and Opdyke, 1991).
Laurentia. LaurentialLaurussia exhibits sizable elevation residuals throughout
the Paleozoic, implying large-scale subsidence during the Middle-Late Cambrian
(-170 m), Late Ordovician (- 140 m), and Middle-Late Devonian (-150 m),
and uplift during the Early Ordovician (+ 100 m) and Early Silurian (+ 150 m;
Fig. lOB). These residuals exhibit a close relationship to major events in the tec-
tonic history of Paleozoic North America. Laurentia probably occupied a central
position in a Late Proterozoic supercontinent (Moores, 1991), and development of
long, thermally-subsiding passive margins resulted in transgression of its eastern
and western flanks during the Middle-Late Cambrian (Bond and others, 1989;
Levy and Christie-Blick, 1991; Osleger and Read, 1993). However, this mecha-
nism cannot account for broad transgression of midcontinent Laurentia during the
Middle Cambrian-Early Ordovician (Cook and Bally, 1975) or for the absence
of equivalent submergence of probable conjugate rift partners (i.e., Gondwana,
Siberia, and Baltica; Moores, 1991; Dalziel, 1992), which were already exten-
sively flooded during the Lower-Middle Cambrian (Lindsay and Korsch, 1989;
McKerrow and others, 1992). On the other hand, movement away from a dynamic
topographic high underlying a late Proterozoic supercontinent (e.g., Gurnis, 1992a)
might explain Laurentia's Cambrian flooding history in light of its unique plate
tectonic setting.
Large elevation residuals correlate with a shift from passive to active margins
around Laurentia during the Middle Paleozoic. Late Ordovician-Early Silurian up-
lift and subsidence in Arctic Canada are probably related to the coeval M'Clintock
Orogeny (Trettin, 1989) and to the Early Silurian (Llandovery) collision of north-
eastern Greenland with Barentia (Chukotka; Hurst and others, 1983). Middle-Late
Devonian transgression of western North America and parts of the midcontinent
(Cook and Bally, 1975) coincided with transformation of the western continental
flank to an active margin during the Antler/Caribou Orogeny (Oldow and others,
1989; Dixon and Dietrich, 1990). Subsidence in the Cordilleran region resulted
from back-arc basin formation and probably represents onset of viscous dynamic
coupling (e.g., Gurnis, 1990a), but coeval flooding of the cratonic interior requires
a different mechansim, e.g., increased intraplate stresses (Bond and Kominz, 1991)
or dynamic subsidence associated with drift of Laurentia into a geoid low (e.g.,
Gurnis, 1992b). Uplift of Laurussia during the Early Devonian may have been the
consequence of lithospheric thickening associated with the Caledonian Orogeny,
e.g., through imbricate thrusting (e.g., Molnar, 1989).
Baltica. The Baltic craton exhibits few Paleozoic elevation excursions, and
these appear to be coeval with major tectonic events (Fig. 1OC). Late Cambrian
uplift (+ 70 m) coincided with the Finnmarkian Orogeny, during which the northern
margin of Baltica collided with the southern end of the Chingiz-Tuva volcanic arc
and accreted the Trans-Uralian microcontinent (Zonenshain and others, 1990;
Koenemann, 1993). Late Silurian uplift (+70 m) may reflect collision of Baltica
with Laurentia during the mid- to Late Silurian Scandian phase of the Caledonian
Orogeny (Ziegler, 1989; Koenemann, 1993). The relationship of epeirogenic uplift
to major orogenic episodes is supported by paleomagnetic data showing cusps in the
Baltic APWP during the Middle-Late Cambrian and Late Silurian-Early Devonian
(Torsvik and others, 1992).
Armorica. In this study, Armorica comprises the Armorican Massif and East-
ern Avalonia (southern Britain/northern France; Torsvik and others, 1992). These
microcontinents rifted away from the northern Gondwanan margin in close succes-
sion during the Arenig and Llanvirn (late Early-early Middle Ordovician; Torsvik
and Trench, 1991; Cope and others, 1992). Thus, Middle-Late Cambrian uplift
(+ 140 m) and Early-Late Ordovician subsidence (-170 m) may reflect syn-rift
thermal doming and post-rift subsidence (Fig. lOD).
Siberia. Middle-Late Cambrian uplift (+150 m) may reflect collision and
suturing of the Siberian craton with the Aldan and Barguzin terranes, and Late
Devonian uplift (+60 m) may be related to the coeval Taimyr Orogeny in eastern
Siberia (Fig. 10E; Zonenshain and others, 1990).
Kazakhstania. Owing to progressive assembly through accretion of a series
of arcs and terranes around a microcontinental nucleus, Paleozoic Kazakhstania
has a complex tectonic history (Zonenshain and others, 1990; Sengor and others,
1993). Late Carboniferous uplift (+140 m) was probably related to suturing of
Kazakhstania with Laurussia and Siberia (Fig. lOF). Absence of coeval uplift of
the latter two continents (Fig. 1OB,E) is likely to have been a function of subduction
zone polarity (Khain and Seslavinsky, 1991).
North China, South China, and Indochina. North China, South China, and In-
dochina fissioned sequentially from the Australian margin of Gondwana, probably
during the Early-Middle Ordovician, Late Silurian-Early Devonian, and Siluro-
Permian, respectively (Hutchison, 1989; Nie and others, 1990; Metcalfe, 1991).
Late Ordovician uplift (+ 110m) of the North China craton coincided with onset of
dual-margin subduction along the Suolun-Xar Moron and Shanyang-Tongcheng
tectonic zones (Fig. lOG; Wang, 1985). Early-Late Permian uplift of united China
(+ 120 m) may have been linked to docking of North China with Siberia along the
Suolun-Xar Moron suture (Wang, 1985; Nie and others, 1990).
Chukotka. Possible large-scale subsidence (-280 m) during the Late Ordovi-
cian-Late Silurian may record docking of Chukotka along the northern margin of
North America, which culminated in the Late Silurian-Early Devonian Franklinian
Orogeny (Fig. lOH). Docking occurred through sinistral oblique collision, which
may have depressed the Chukotkan continental margin along a southward-dipping
subduction zone (Trettin, 1989; Klaper, 1992).
This brief survey of the epeirogenic histories of Paleozoic continents suggests
that most large elevation residuals may represent epeirogenic motions connected
with major tectonic events. Although some elevation residual trends are consistent
with known Paleozoic continental tectonic histories, as for Laurentia, others are
enigmatic, such as the large mid-Paleozoic excursions exhibited by Gondwana.
The range of mechanisms inferred for Paleozoic epeirogenic motions is nearly
as broad as that documented for Cenozoic continents. Continued study of paleo-
continental crustal motions within a plate tectonic context should provide new
insights regarding mechanisms of continental epeirogeny.
4. Conclusions
Analysis of the flooding records of multiple, co-existing, tectonically-independent

landmasses allows reconstruction of global sea-level trends and paleo-continental
hypsometries and epeirogenic histories. The fundamental assumption underlying
the method is that co-existing landmasses must have experienced the same range
of eustatic fluctuations, and differences in degree of flooding therefore reflect
differences in continental coastal hypsometry. Patterns of hypsometric variation

(i.e., similar hypsometric slopes but divergent inflection-point elevations) imply
that epeirogenic motions are largely responsible for "hypsometric disequilibrium"
among modern continents. Thus, continents may have stable long-term hypsomet-
ric profiles that are important controls on their flooding patterns, validating the
reconstruction of paleo-continental hypsometric chords.
Scaling of paleo-continental elevations using Eurasian and American analogs
yields mean Paleozoic sea-level estimates of +60 m and + 160 m and sea-level
elevation ranges of 24-102 m and 96-224 m, respectively. Choice of a Eurasian
analog implies that modern eustatic elevations are rather typical for the Phanerozoic
(modern sea level in an ice-free world +50 m), whereas choice of an American ana-
log implies that they are unusualy low. Regardless of choice of analog, highstand
elevations of + 100 to +225 m are substantially lower than previous Paleozoic
estimates of +300 m to +600 m but only slightly lower than estimates of + 175 to
+250 m for the Late Cretaceous highstand.
Differences between individual continental sea-level estimates and the global
mean for each epoch are "elevation residuals" that represent either secular changes
in paleo-continental hypsometry or errors in flooding estimates. Large elevation
residuals are likely to have an epeirogenic origin, and many of these may be
understood within the context of the tectonic history of individual continents.
The most enigmatic continent is Gondwana, which exhibits large excursions in
an antithetic sense to global sea-level trends during the Siluro-Carboniferous.
This pattern requires a global control linking eustasy to epeirogeny, possibly via
increased MOR spreading rates leading to uplift of Gondwanan continental margins
through subduction of young oceanic lithosphere or continent-terrane collisions.
Acknowledgments
We would like to thank Bruce Wilkinson for considerable support and assistance
during the early phases of this project, Robert Berner and Dave Osleger for helpful
reviews, George Klein, Leigh Royden, and Dork Sahagian for commentary on a
paper devoted to a related aspect of this project, and David Nash, Warren Huff,
and Madeleine Briskin for stimulating discussions on this topic. Research support
was provided by a University of Cincinnati Research Council grant and a Project
Development Grant from the NAS National Research Council.
References
ALGEO, T. J., AND WILKINSON, B. H., 1991, Modern and ancient continental hyp-
sometries: Journal of the Geological Society of London, v. 148, p. 643-654.
ALLEGRE, C. J., AND 34 OTHERS, 1984, Structure and evolution of the Himalaya-Tibet
orogenic belt: Nature, v. 307, p. 17-22.
ALLMENDINGER, R. w., 1986, Tectonic development, southeastern border of the Puna
Plateau, northwestern Argentine Andes: Geological Society of America Bulletin, v. 97,
p. 1070-1082.
240 T.!. Algeo, K.B. Seslavinsky
BARRETT, S. F., AND ISAACSON, P. E., 1988, Devonian paleogeography of South

America, in McMillan, N. J., Embry, A. F., and Glass, D. J., eds., Devonian of the
World, Vol. I: Regional Syntheses: Calgary, Canadian Society of Petroleum Geologists,
p.655-667.
BERNER, R. A., 1994, Geocarb II: a revised model of atmospheric C02 over Phanerozoic
time: American Journal of Science, v. 294, p. 56-91.
BIRD, P., 1978, Initiation of intracontinental subduction in the Himalaya: Journal of
Geophysical Research, v. 83, p. 4975-4987.
BISHOP, P. AND BROWN, R., 1992, Denudational isostatic rebound of intraplate high-
lands: The Lachland River valley, Australia: Earth Surface Processes and Landforms, v.
17, p. 345-360.
BOND, G., 1976, Evidence for continental subsidence in North America during the Late
Cretaceous global submergence: Geology, v. 4, p. 557-560.
BOND, G., 1978a, Evidence for Late Tertiary uplift of Africa relative to North America,
South America, Australia and Europe: Journal of Geology, v. 86, p. 47-65.
BOND, G., 1978b, Speculations on real sea-level changes and vertical motions of continents
at selected times in the Cretaceous and Tertiary Periods: Geology, v. 6, p. 247-250.
BOND, G. c., 1979, Evidence for some uplifts oflarge magnitude in continental platforms:
Tectonophysics, v. 61, p. 285-305.
BOND, G. c., AND KOMINZ, M. A., 1991, Disentangling Middle Paleozoic sea level
and tectonic events in cratonic margins and cratonic basins of North America: Journal
of Geophysical Research, v. 96, no. B4, p. 6619-6639.
BOND, G. c., NICKERSON, P. A., AND KOMINZ, M. A., 1984, Breakup of a super-
continent between 625 Ma and 555 Ma: new evidence and implications for continental
histories: Earth and Planetary Science Letters, v. 70, p. 325-345.
BOND, G. C., KOMINZ, M. A., GROTZINGER, 1. P., AND STECKLER, M. S., 1989,
Role of thermal subsidence, flexure and eustasy in the evolution of Early Paleozoic
passive margin carbonate platforms, in Crevello, P., Wilson, J. L., Sarg, J. R., and
Read, J. F., eds., Controls on Carbonate Platform and Basin Development: Society of
Economic Paleontologists and Mineralogists Special Publication 44, p. 39-62.
BRADSHAW, M. A., AND WEBERS, G. F., 1988, The Devonian rocks of Antarctica, in
McMillan, N. J., Embry, A. F., and Glass, D. 1., eds., Devonian of the World, Vol. 1:
Regional Syntheses: Calgary, Canadian Society of Petroleum Geologists, p. 783-795.
CATHLES, L. M., AND HALLAM, A., 1991, Stress-induced changes in plate density,
Vail sequences, epeirogeny, and short-lived global sea level fluctuations: Tectonics, v.
10, p. 659-671.
CLOETINGH, S., 1988, Intraplate stresses: a tectonic cause for third-order cycles in ap-
parent sea level?, in Wilgus, C. K., and others, eds., Sea-Level Changes: An Integrated
Approach: Tulsa, Society of Economic Paleontologists and Mineralogists, Special Pub-
lication 42, p. 19-29.
COGLEY, J. G., 1985, Hypsometry of the continents: Zeitschrift fyr Geomorphologie,
Supplementband, v. 53.
COOK, T. D., AND BALLY, A.W., eds., 1975, Stratigraphic Atlas of North and Central
America: Princeton, Princeton University Press, 272 p.
COPE, J. C. w., INGHAM, 1. K., AND RAWSON, P. F., eds., 1992, Atlas of Palaeo-
geography and Lithofacies: London, Geological Society of London, Memoir 13, 153
p.
CROSS, T. A., AND PILGER, R. H., 1978, Tectonic controls of Late Cretaceous sedimen-
tation, Western Interior, USA: Nature, v. 274, p. 653-657.
CROSS, T. A., 1986, Tectonic controls of foreland basin subsidence and Laramide style
deformation, western United States, in Allen, P. A., and Homewood, P., eds., Foreland
Basins: International Association of Sedimentologists, Special Publication 8, p. 15-40.
CROUGH, S. T., AND THOMPSON, G. A., 1977, Upper mantle origin of the Sierra
Nevada uplift: Geology, v. 5, p. 396-399.
DALZIEL, I. W. D., 1992, Antarctica; a tale of two supercontinents?: Annual Review of
Earth and Planetary Sciences, v. 20, p. 501-526.
DIVENERE, V. J., AND OPDYKE, N. D., 1991, Magnetic polarity stratigraphy and
Carboniferous paleopole positions from the Joggins section, Cumberland structural
basin, Nova Scotia: J. Geophysical Research, v. 96, p. 4051-4064.
DIXON, J., AND DIETRICH, J. R., 1990, Canadian Beaufort Sea and adjacent land areas,
in Grantz, A., Johnson, L., and Sweeny, J. E, eds., The Arctic Ocean Region: Boulder,
Colorado, Geological Society of America, The Geology of North America, vol. L, p.
239-256.
DONOVAN, D. T., AND JONES, E. J. W, 1979, Causes of world-wide changes in sea
level: Journal of the Geological Society of London, v. 136, p. 187-192.
ENGEBRETSON, D. C., KELLEY, K. P., CASHMAN, H. J., AND RICHARDS, M. A.,
1992,180 million years of subduction: GSA Today, v. 2, p. 93-100 (not incl.).
FISCHER, A G., 1984, The two Phanerozoic supercycles, in Berggren, W A., and Van
Couvering, J. A, eds., Catastrophes and Earth History: Princeton, Princeton University
Press, p. 129-150.
FORSYTH, D. W, 1985, Subsurface loading and estimates of the flexural rigidity of
continental lithosphere: Journal of Geophysical Research, v. 90, p. 12,623-12,632.
GAFFIN, S., 1987, Ridge volume dependence of seafloor generation rate and inversion
using long term sealevel change: American Journal of Science, v. 287, p. 596-611.
GALER, S. J. G., 1991, Interrelationships between continental freeboard, tectonics and
mantle temperature: Earth and Planetary Science Letters, v. 105, p. 214-228.
GAY, N. c., 1980, The state of stress in the plates, in Bally, A W, Bender, P. L., McGetchin,
T. R., and Walcott, R. I., eds., Dynamics of Plate Interiors: Washington, D.C., American
Geophysical Union, Geodynamic Series 1, p. 145-153.
GRIFFITHS, R. W, GURNIS, M., AND EITELBERG, G., 1989, Holographic measure-
ments of surface topography in laboratory models of mantle hotspots: Geophysical
Journal, v. 96, p. 477-495.
GURNIS, M., 1988, Large-scale mantle convection and the aggregation and dispersal of
supercontinents: Nature, v. 332, p. 695-699.
GURNIS, M., 1990a, Plate-mantle coupling and continental flooding: Geophysical Re-
search Letters, v. 17, p. 623-626.
GURNIS, M., I 990b, Bounds on global dynamic topography from Phanerozoic flooding
of continental platforms: Nature, v. 344, p. 754-756.
GURNIS, M., 1992a, Long-term controls on eustatic and epeirogenic motions by mantIe
convection: GSA Today, v. 2, p. 141-157 (not incl.).
GURNIS, M., 1992b, Rapid continental subsidence following the initiation and evolution
of subduction: Science, v. 255, p. 1556-1558.
HAGER, B. H., 1984, Subducted slabs and the geoid: constraints on mantIe rheology and
flow: Journal of Geophysical Research, v. 89, p. 6003-6015.
HAGER, B., CLAYTON, R. W, RICHARDS, M. A., COMER, R. P., AND DZIEWON-
SKI, A. M., 1985, Lower mantle heterogeneity, dynamic topography and the geoid:
Nature, v. 313, p. 541-545.
HALLAM, A, 1984, Pre-Quaternary sea-level changes: Annual Review of Earth and
Planetary Science, v. 12, p. 205-243.
242 T.1. Algeo, K.B. Seslavinsky
HAQ, B. U., HARDENBOL, J., AND VAIL, P. R., 1987, Chronology of fluctuating sea
levels since the Triassic: Science, v. 235, p. 1156-1167.
HARLAND, W. B., ARMSTRONG, R. L., COX, A. V., CRAIG, L. E., SMITH, A. G., AND
SMITH, D. G., 1990, A Geologic Time Scale 1989: Cambridge, Cambridge University
Press, 263 p.
HARRISON, C. G. A., 1988, Eustasy and epeirogeny of continents on time scales between
about 1 and 100 m.y.: Paleoceanography, v. 3, p. 671-684.
HARRISON, C. G. A., 1990, Long-term eustasy and epeirogeny in continents, in Revelle,
R. R., and others, Sea-Level Change: Washington, D.C., National Academy Press,
National Research Council, Studies in Geophysics, p. 141-158.
HARRISON, C. G. A., BRASS, G. w., SALTZMAN, E., SLOAN, J., II, SOUTHAM, J.,
AND WHITMAN, J. M., 1981, Sea level variations, global sedimentation rates and the
hypsographic curve: Earth and Planetary Science Letters, v. 54, p. 1-16.
HARRISON, C. G. A., MISKELL, K. J., BRASS, G. w., SALTZMAN, E. S., AND
SLOAN, J. L., II, 1983, Continental hypsography: Tectonics, v. 2, p. 357-377.
HARRISON, T. M., COPELAND, P., KIDD, W. S. E, AND YIN, A., 1992, Raising Tibet:
Science, v. 255, p. 1663-1670.
HECKEL, P. H., 1986, Sea-level curve for Pennsylvanian eustatic marine transgressive-
regressive depositional cycles along midcontinent outcrop belt, North America: Geol-
ogy, v. 14, p. 330-334.
HELLER, P. L., AND ANGEVINE, C. L., 1985, Sea-level cycles during the growth of
Atlantic-type oceans: Earth and Planetary Science Letters, v. 75, p. 417-426.
HELLER, P. L., WENTWORTH, C. M., AND POAG, C. w., 1982, Episodic post-rift
subsidence of the United States Atlantic continental margin: Geological Society of
America Bulletin, v. 93, p. 379-390.
HIRN, A., NERCESSIAN, A., SAPIN, M., JOBERT, G., XU, Z. X., GAO, E. Y., LU, D.
Y., AND TENG, J. w., 1984, Lhasa block and bordering sutures-a continuation of a
500-km Moho traverse through Tibet: Nature, v. 307, p. 25-27.
HURST, J. M., McKERROW, W. S., SOPER, N. 1., AND SURLYK, E, 1983, The rela-
tionship between Caledonian nappe tectonics and Silurian turbidite deposition in North
Greenland: Journal of the Geological Society of London, v. 140, p. 123-131.
HUTCHISON, C. S., 1989, The Palaeo-Tethyan realm and Indosinian orogenic system
of Southeast Asia, in Sengor, A.M.c., ed., Tectonic Evolution of the Tethyan Region:
Dordrecht, Kluwer, p. 585-643.
JELL, J. S., 1988, Lower and Middle Devonian of Queensland, Australia, in McMillan, N.
J., Embry, A. E, and Glass, D. J., eds., Devonian of the World, Vol. 1: Calgary, Canadian
Society of Petroleum Geologists, p. 755-772.
JOHNSON,1. G., KLAPPER, G., AND SANDBERG, C. A., 1985, Devonian eustatic
fluctuations in Euramerica: Geological Society of America Bulletin, v. 96, p. 567-587.
JOHNSON, M. E., BAARLI, B. G., NESTOR, H., RUBEL, M., AND WORSLEY, D.,
1991, Eustatic sea-level patterns from the Lower Silurian (Llandovery Series) of south-
ern Norway and Estonia: Geological Society of America Bulletin, v. 103, p. 315-335.
JORDAN, T. E., 1981, Thrust loads and foreland basin evolution, Cretaceous, western
United States: American Association of Petroleum Geologists Bulletin, v. 65, p. 2506-
2520.
KHAIN, V. E., AND SESLAVINSKY, K. B., 1991, Historical Geotectonics of the Paleozoic
(in Russian): Moscow, Nedra, 398 p.
KING, L. c., 1962, The Morphology of the Earth: Edinburgh, Oliver and Boyd, 699 p.
KLAPER, E. M., 1992, The Paleozoic tectonic evolution of the northern edge of North
America: A structural study of northern Ellesmere Island, Canadian Arctic Archipelago:
Tectonics, v. II, p. 854-870.
KOENEMANN, E H., 1993, Tectonics of the Scandian Orogeny and the Western Gneiss
Region in southern Norway: Geologische Rundschau, v. 82, p. 696-717.
KOMINZ, M. A., 1984, Oceanic ridge volumes and sea level change-an error analysis, in
Schlee, J., ed., Interregional unconformities and hydrocarbon accumulation: American
Association of Petroleum Geologists Memoir 36, p. 109-127.
KOMINZ, M. A., AND BOND, G. C., 1991, Unusually large subsidence and sea-level
events during middle Paleozoic time: new evidence supporting mantle convection mod-
els for supercontinent assembly: Geology, v. 19, p. 56-60.
KUO, B.-Y., 1993, Thermal anomalies beneath the Australian-Antarctic discordance: Earth
and Planetary Science Letters, v. 119, p. 349-364.
LAMBECK, K., 1983, The role of compressive forces in intracratonic basin formation and
mid-plate orogenies: Geophysical Research Letters, v. 10, p. 845-848.
LARSON, R. L., 1991, Latest pulse of the Earth: Evidence for a mid-Cretaceous super-
plume: Geology, v. 19, p. 547-550.
LEVY, M., AND CHRISTIE-BLICK, N., 1991, Tectonic subsidence of the early Paleo-
zoic passive continental margin in eastern California and southern Nevada: Geological
Society of America Bulletin, v. 103, p. 1590-1606.
LINDSAY, J. E, AND KORSCH, R. J., 1989, Interplay of tectonics and sea-level changes
in basin evolution: an example from the intracratonic Amadeus Basin, central Australia:
Basin Research, v. 2, p. 3-25.
McGETCHIN, T. R., BURKE, K. c., THOMPSON, G. A., AND YOUNG, R. A., 1980,
Mode and mechanisms of plateau uplifts, in BaIly, A. w., Bender, P. L., McGetchin, T.
R., and Walcott, R. I., cds., Dynamics of Plate Interiors: Washington, D.C., American
Geophysical Union, Geodynamic Series I, p. 99-110.
McKENZIE, D., 1984, A possible mechanism for epeirogenic uplift: Nature, v. 307, p.
616-618.
McKERROW, W. S., SCOTESE, C. R., AND BRASIER, M. D., 1992, Early Cambrian
continental reconstructions: Journal of the Geological Society of London, v. 149, p.
599-606.
MELOSH, H. J., AND EBEL, J., 1979, A simple model for thermal instability in the
asthenosphere: Geophysical Journal of the Royal Astronomical Society, v. 59, p. 419-
436.
METCALFE, I., 1991, Late Palaeozoic and Mesozoic palaeogeography of Southeast Asia:
Palaeogeography, Palaeoclimatology, PalaeoeCOlogy, v. 87, p. 211-221.
MITROVICA, J. X., BEAUMONT, c., AND JARVIS, G. T., 1989, Tilting of continental
interiors by the dynamical effects of subduction: Tectonics, v. 8, p. 1079-1094.
MOLNAR, P., 1989, The geological evolution of the Tibetan Plateau: American Scientist,
v. 77, p. 350-360.
MOORES, E. M., 1991, The Southwest U.S.-East Antarctic (SWEAT) connection: a hy-
p'0thesis: Geology, v. 19, p. 425-428.
MORNER, N.-A., 1976, Eustasy and geoid changes: Journal of Geology, v. 84, p. 123-151.
Nt J., AND BARAZANGI, M., 1983, High-frequency seismic wave propagation beneath
the Indian Shield, Himalayan arc, Tibetan Plateau, and surrounding regions: high up-
permost mantle velocities and efficient Sn propagation beneath Tibet: Geophysical J.
Royal Astronomical Society, v. 72, p. 665-689.
NIE, S.- Y., ROWLEY, D. B., AND ZIEGLER, A. M., 1990, Constraints on the locations of
Asian microcontinents in Palaeo-Tethys during the Late Palaeozoic, in McKerrow, W.
S., and Scotese, C. R., eds., Palaeozoic Palaeogeography and Biogeography: Geological
Society of London Memoir 12, p. 397-409.
OLD OW, J. S., BALLY, A. w., LALLEMANT, H. G. A., AND LEEMAN, W. P., 1989,
Phanerozoic evolution of the North American Cordillera; United States and Canada, in
244 T.!. Algeo, K.B. Seslavinsky
Bally, A. W., and Palmer, A. R., editors, The Geology of North America-An overview:
Boulder, Colorado, Geological Society of America, The Geology of North America,
vol. A, p. 139-232.
OPDYKE, N. D., SPANGLER, D. P., SMITH, D. L., JONES, D. S., AND LINDQUIST,
R. C., 1984, Origin of the epeirogenic uplift of Pliocene-Pleistocene beach ridges in
Florida and development of the Florida karst: Geology, v. 12, p. 226-228.
OS LEGER, D., AND READ, J. F., 1993, Comparative analysis of methods used to de-
fine eustatic variations in outcrop: Late Cambrian interbasinal sequence development:
American Journal of Science, v. 293, p. 157-216.
Paleogeographic Atlas of Australia, Vol. l-Cambrian, 1988: Canberra, Australian Gov-
ernment Publishing Service.
Paleogeographic Atlas of Australia, Vol. 2-0rdovician, 1992: Canberra, Australian Gov-
ernment Publishing Service.
PALMER, J., SEMPERE, 1. c., CHRISTIE, D., AND PHIPPS MORGAN, J., 1993,
Morphology and tectonics of the Australian-Antarctic Discordance between 123E and
128E: Marine Geophysical Research, v. 15, p. 21-152.
PEATE, D. w., HAWKESWORTH, C. J., MANTOVANI, M. S. M., AND SHUKOWSKY,
W., 1990, Mantle plumes and flood-basalt stratigraphy in the ParanO, South America:
Geology, v. 18, p. 1223-1226.
PELTIER, W. R., 1980, Models of glacial isostasy and relative sea level, in Bally, A. W.,
Bender, P. L., McGetchin, T. R., and Walcott, R. 1., eds., Dynamics of Plate Interiors:
Washington, D.C., American Geophysical Union, Geodynamic Series 1, p. 111-128.
PITMAN, W. c., III, 1978, Relationship between eustacy and stratigraphic sequences of
passive margins: Geological Society of America Bulletin, v. 89, p. 1389-1403.
POWELL, C.McA., 1986/87, Continental underplating model for the rise of the Tibetan
Plateau: Earth and Planetary Science Letters, v. 81, p. 79-94.
PYLE, D. G., CHRISTIE, D. M., AND MAHONEY, J. J., 1992, Resolving an isotopic
boundary within the Australian-Antarctic Discordance: Earth and Planetary Science
Letters, v. 112, p. 161-178.
QUINLAN, G. M., AND BEAUMONT, C., 1984, Appalachian thrusting, lithospheric
flexure, and the Paleozoic stratigraphy of the eastern interior of North America: Canadian
Journal of Earth Sciences, v. 21, p. 973-996.
RAMOS, V. A., JORDAN, T. E., ALLMENDINGER, R. w., MPODOZIS, C., KAY, S. M.,
CORTES, J. M., AND PALMA, M., 1986, Paleozoic terranes of the central Argentine-
Chilean Andes: Tectonics, v. 5, p. 855-880.
RIDING, R., 1984, Sea-level changes and the evolution of benthic marine calcareous algae
during the Palaeozoic: Journal of the Geological Society of London, v. 141, p. 547-553.
ROBARDET, M., PARIS, F., AND RACHEBOEUF, P. R., 1990, Palaeogeographic evo-
lution of southwestern Europe during Early Palaeozoic times, in McKerrow, W. S.,
and Scotese, C. R., cds., Palaeozoic Palaeogeography and Biogeography: Geological
Society of London Memoir 12, p. 411-419.
RONOV, A. B., 1980, The Earth's sedimentary shell (Quantitative patterns of its struc-
tures, compositions and evolution), in Yaroshevsky, ed., The Earth's Sedimentary Shell
(Quantitative patterns of its structures, compositions and evolution): Moscow, Nauka,
p. 1-80; also, American Geological Institute Reprint Series, v. 5, p. 1-73,1983.
RONOV, A. B., J994, Phanerozoic transgressions and regressions on the continents: a
quantitative approach based on areas flooded by the sea and areas of marine and conti-
nental deposition: American Journal of Science, v. 294, p. 777-801.
RONOV, A. B., KHAIN, V. E., BALUKHOVSKY, A. N., AND SESLAVINSKY, K. B.,
1980, Quantitative analysis of Phanerozoic sedimentation: Sedimentary Geology, v. 25,
p.311-325.
RONOV, A. B., KHAIN, V. E., AND SESLAVINSKY, K. B., 1984, Atlas of Lithologic-
Paleogeographic Maps of the World-Late Precambrian and Paleozoic of Continents (in
Russian): Leningrad, 70 p.
ROSENDAHL, B. R., 1987, Architecture of continental rifts with special reference to East
Africa: Annual Review of Earth and Planetary Sciences, v. 15, p. 445-503.
ROYDEN, L., SCLATER, J. G., AND VON HERZEN, R. P., 1980, Continental margin
subsidence and heat flow: important parameters in formation of petroleum hydrocarbons:
American Association of Petroleum Geologists Bulletin, v. 64, p. 173-187.
RUDDIMAN, W.E, PRELL, w.L., AND RAYMO, M.E., 1989, Late Cenozoic uplift
in southern Asia and the American West: rationale for general circulation modeling
experiments: Journal of Geophysical Research, v. 94, p. 18,379-18,391.
SAHAGIAN, D., 1987, Epeirogeny and eustatic sea level as inferred from Cretaceous
shoreline deposits: applications to the central and western United States: Journal of
Geophysical Research, v. 92, p. 4895-4904.
SAHAGIAN, D., 1988, Epeirogenic motions of Africa as inferred from Cretaceous shore-
line deposits: Tectonics, v. 7, p. 125-138.
SAHAGIAN, D., 1989, Epeirogeny of Europe and western Asia: Cretaceous Research, v.
10, p. 33-48.
SCHENK, P. E., 1991, Events and sea-level changes on Gondwana's margin: the Meguma
Zone (Cambrian to Devonian) of Nova Scotia, Canada: Geological Society of America
Bulletin, v. 103, p. 512-521.
SCOTESE, C. R., AND GOLONKA, J., 1993, PALEOMAP Paleogeographic Atlas: De-
partment of Geology, University of Texas, Arlington, 35 p.
SCOTESE, C. R., AND McKERROW, W. S., 1990, Revised World maps and introduc-
tion, in McKerrow, W. S., and Scotese, C. R., eds., Palaeozoic Palaeogeography and
Biogeography: Geological Society of London Memoir 12, p. 1-21.
SCOTESE, C. R., BAMBACH, R. K., BARTON, c., VAN DER VOO, R., AND ZIEGLER,
A. M., 1979, Paleozoic base maps: Journal of Geology, v. 87, p. 217-277.
SEMPERE, J.-C., PALMER, J., CHRISTIE, D. M., MORGAN, J. P., AND SHOR, A. N.,
1991, Australian-Antarctic discordance: Geology, v. 19, p. 429-432.
SENGOR, A. M. c., NATAL'IN, B. A., AND BURTMAN, V. S., 1993, Evolution of the
Altaid tectonic collage and Palaeozoic crustal growth in Eurasia: Nature, v. 364, p.
299-307.
SESLAVINSKY, K. B., 1987, Sedimentation and volcanism during the Caledonian stage
of Earth history (in Russian): Moscow, Nedra, 192 p.
SESLAVINSKY, K. B., 1991, Global transgressions and regressions of the Paleozoic (in
Russian): Izvestia of the U.S.S.R. Academy of Sciences, Geological Series, No. I, p.
71-79; English translation: International Geological Review, v. 2, p. 3-11.
SMITH, A. G., 1982, Late Cenozoic uplift of stable continents in a reference frame fixed
to South America: Nature, v. 296, p. 400-404.
STECKLER, M. S., AND WATTS, A. B., 1978, Subsidence of the Atlantic-type continental
margin off New York: Earth and Planetary Science Letters, v. 41, p. 1-13.
STERN, T. A., AND HOLT, W. E., 1994, Platform subsidence behind an active subduction
zone: Nature, v. 368, p. 233-236.
TORSVIK, T. H., AND TRENCH, A., 1991, The Ordovician history of the Iapetus ocean in
Britain: new palaeomagnetic constraints: Journal of the Geological Society of London,
v. 148, p. 423-425.
TORSVIK, T. H., RYAN, P. D., TRENCH, A., AND HARPER, D. A. T., 1991, Cambrian-
Ordovician paleogeography of Baltica: Geology, v. 19, p. 7-10.
TORSVIK, T. H., SMETHURST, M. A., VAN DER VOO, R., TRENCH, A., ABRA-
HAMSEN, N., AND HALVORSEN, E., 1992, Baltica: a synopsis of Vendian-Permian
palaeomagnetic data and their palaeotectonic implications: Earth-Science Reviews, v.

33, p. 133-152.
TRETIIN, H. P., 1989, The Arctic Islands, in Bally, A. W., and Palmer, A. R., eds., The
Geology of North America-An Overview: Boulder, Colorado, Geological Society of
America, The Geology of North America, v. A, p. 349-370.
TURCOTIE, D. L., AND BURKE, K., 1978, Global sea-level changes and the thermal
structure of the Earth: Earth and Planetary Science Letters, v. 41, p. 341-346.
VAIL, P. R., MITCHUM, R. M., JR., AND THOMPSON, S., III, 1977, Global cycles of
relative changes of sea level, in Payton, C. E., ed., Seismic Stratigraphy and Global
Changes of Sea Level: Tulsa, American Association of Petroleum Geologists Memoir
26, p. 83-97.
VEEVERS, J. J., cd., 1984, Phanerozoic Earth history of Australia: Oxford, Clarendon,
418 p.
WANG, H.-Z., comp., 1985, Alias of the Palaeogeography of China: Beijing, Cartographic
Publishing House, 143 maps + 28 p.
WESTAWAY, R., 1993, Forces associated with mantle plumes: Earth and Planetary Science
Letters, v. 119, p. 331-348.
WHITE, R. S., AND McKENZIE, D. P., 1989, Magmatism at rift zones: the generation of
volcanic continental margins and flood basalts: Journal of Geophysical Research, v. 94,
p.7685-7729.
WILKINSON, B. H., AND GIVEN, R. K., 1986, Secular variation in abiotic marine car-
bonates: Constraints on Phanerozoic atmospheric carbon dioxide contents and oceanic
Mg/Ca ratios: Journal of Geology, v. 94, p. 321-333.
WILKINSON, B. H., OWEN, R. M., AND CARROLL, A. R., 1985, Submarine hydro-
thermal weathering, global eustasy, and carbonate polymorphism in Phanerozoic marine
oolites: Journal of Sedimentary Petrology, v. 55, p. 171-183.
WISE, D. U., 1974, Continental margins, freeboard and the volumes of continents and
oceans through time, in Burk, C. A., and Drake, C. L., editors, The Geology of Conti-
nental Margins: New York, Springer, p. 45-58.
WORSLEY, T. R., NANCE, D., AND MOODY, J. B., 1984, Global tectonics and eustasy
for the past 2 billion years: Marine Geology, v. 58, p. 373-400.
WYATI, A. R., 1986, Post-Triassic continental hypsometry and sea level: Journal of the
Geological Society of London, v. 143, p. 907-910.
WYATI, A. R., 1987, Shallow water areas in space and time: Journal of the Geological
Society of London, v. 144, p. 115-120.
ZIEGLER, P. A., 1989, Evolution of Laurussia: Dordrecht, Kluwer, 102 p.
ZONENSHAIN, L. P., KUZMIN, M. I., AND KONONOV, M. V., 1987, Absolute recon-
structions of the state of continents in Paleozoic and Early Mesozoic: Geotectonics, v.
21, p. 199-212.
ZONENSHAIN, L. P., KUZMIN, M. I., AND NATAPOV, L. M. (Page, B. M., ed.), 1990,
Geology of the USSR: A Plate-Tectonic Synthesis, Geodynamics Series 21: Washington,
D.C., American Geophysical Union, 242 p.
CHAPTER 9
Depositional Sequences on Upper

Cambrian Carbonate Platforms:
Variable Sedimentologic Responses to
Allogenic Forcing
David A. OsJeger
ABSTRACT: Six third-order depositional sequences are documented for Late Cam-
brian time by interbasinal correlation of cyclic carbonates from tectonic settings in the
Appalachian and Cordilleran passive margins, the Texas cratonic embayment, and the
southern Oklahoma aulacogen. Paleobathymetric interpretation, integrated with graphic
correlation, is used to establish the relative synchroneity of Upper Cambrian depositional
sequences and is crosschecked with two quantitative techniques that provide an approxi-
mation of the accommodation history independent of fluctuations in carbonate sediment
production. The first technique, Fischer plots, graphically illustrates systematic changes
in the stacking patterns of meter-scale cycles that presumably reflect third-order changes
in accommodation potential. The second technique, subsidence analysis, determines the
accommodation remaining after the isostatic and thermo-tectonic components of total
decompacted subsidence have been removed. Integrating the three methods enhances
the interbasinal correlation of individual third-order depositional sequences and permits
the construction of a robust relative sea-level curve for the Upper Cambrian of North
America.
Comparison of the relative sea-level curve determined in this study with published
curves derived from different regions of North America suggests that all six sequences
have correlatives in other areas of the continent, supporting an allogenic control on se-
quence development. Detailed inspection of individual sequences from separate basins
illustrates the influence of intrinsic factors such as tectonic setting, platform morphol-
ogy, subsidence history, paleotopography, and prevailing oceanographic conditions on
the stratigraphic record. Even though each section is composed of different types of
meter-scale cycles and component lithofacies that reflect the environmental dynamics
of their depositional setting, similarities in the overall internal architecture of individual
sequences are clearly evident, and suggests that continent-wide depositional patterns
were controlled by a single allogenic mechanism, most likely eustasy.
247
B. U. Haq (ed.),
248 D.A. OsJeger
1. Introduction
Interbasinal correlations of stratigraphic successions have been increasingly re-

fined since the seminal efforts of Suess (1906), Schuchert (1916), and Grabau
(1936), primarily due to progressive improvement of the geologic time scale. De-
positional synchroneity at the second-order scale (10-100 m.y.) was recognized
by Sloss (1963) and Vail and others (1977), and subsequently, correlations at the
third-order scale (1-10 m.y.) have been proposed for several intervals throughout
the Phanerozoic (e.g., Hancock and Kaufmann, 1979; Lenz, 1982; Hallam, 1984;
Johnson and others, 1985; Haq and others, 1987; Scott and others, 1988; Ross and
Ross, 1988; Johnson and others, 1989). Many of these correlations are presented
in the form of comparative relative sea-level curves or paleobathymetric curves
that, arguably, establish the extent of depositional synchroneity between sequences
from separate basins. These interbasinal studies contribute toward resolving the
continuing controversy regarding the controlling processes behind sequence depo-
sition (tectonism, eustasy, environmental change), and correlated sequences may
ultimately prove to have chronostratigraphic significance. However, most of these
analyses do not illustrate sufficient lithologic detail from outcrop or core that
is necessary to directly compare sequence boundaries and systems tracts devel-
oped within separate basins. Moreover, most of these studies (with the notable
exception of the Exxon curves) make no attempt to quantitatively determine the
accommodation history that may have controlled sequence generation.
This paper presents the results of a detailed interbasinal analysis of Upper
Cambrian pericratonic cyclic carbonates from tectonic settings in the Appalachian
and Cordilleran passive margins, the Texas cratonic embayment, and the southern
Oklahoma aulacogen. Paleoenvironmental interpretations of lithofacies associa-
tions were determined for several complete sections of Upper Cambrian strata that
were measured and logged at the decimeter scale, accompanied by thin-section
analysis of >300 hand samples. The first objective of this paper is to document the
continent-wide correlations of six Upper Cambrian sequences that, within the lim-
its of the biostratigraphic resolution, appear to have been deposited synchronously
around the periphery of the continent. The second objective is to compare the
accommodation curve from this study, derived by combining the results of three
separate qualitative and quantitative methods, with other previously published
sea-level curves for Late Cambrian time. The third objective is to illustrate the
detailed lithologic composition of coeval sequence boundaries and systems tracts
that developed under unique environmental conditions in widely separated basins.
2. Tectonic and Stratigraphic Settings
During Late Cambrian time, the Earth was characterized by low-latitude conti-
nentality, greenhouse climates with associated high C02 concentrations (Berner,
1993), and globally high sea levels. Laurentia occupied an equatorial position dur-
ing Late Cambrian time, oriented with the Cordilleran passive margin extending
essentially east-west at about 10 to 15 N latitude and the Appalachian passive
margin stretching northwest-southeast at about 15 to 20 S latitude (Fig. 1, from
Upper Cambrian Depositional Sequences 249
0,
I
/-
I
Virginia- Pennsylvania
1000 km Tennessee Appalachians
Appalach ians
(after Witzke, 1990)
Fig. 1. Interpreted paleogeography for Late Cambrian Laurentia showing the location of field
localities investigated for this project. Continental margins and land areas (stippled) schematically
shown for times of maximum marine onlap. Dashed lines mark approximate 10 latitudes. Adapted
from Witzke (I 990).
Witzke, 1990). The Texas cratonic embayment and southern Oklahoma aulocogen
were located about 5 south of the equator (van der Voo, 1988). Warm, tropical
conditions permitted the accumulation of thick successions of shallow-marine car-
bonates and fine siliciclastics on broad passive margins that surrounded the North
American craton.
Both the Appalachian and Cordilleran passive margins originated in response
to breakup of a Late Proterozoic supercontinent (Dalziel, 1991) around 625 to
555 Ma (Stewart and Suczek, 1977; Bond and others, 1984; Levy and Christie-
Blick, 1991). The Appalachian passive margin amassed up to 1.6 km of Middle to
Upper Cambrian shallow-marine carbonates and intrashelf basin shale and siltstone
(Read, 1989). The Cordilleran passive margin accumulated approximately 2 km
of post-rift Middle to Upper Cambrian carbonates and fine siliciclastics (Poole
et a!., 1992). The Llano Uplift area of central Texas was part of the slowly-
subsiding craton and accumulated approximately 400 m of late Middle to Upper
Cambrian mixed carbonates and clastics. Sedimentation in the southern Oklahoma
aulacogen began during the Late Cambrian following a Late Proterozoic through
Middle Cambrian history of rifting and intrusive activity.
250 D.A. OsJeger
~o w TRILOBITE HOUSE LLANO WICHITA SW EASTERN

0:0 C) BIOMERE ZONE RANGE, UPLIFT, MTS. VIRGINIA PA & NJ
<C 0: ~
luw~O~~W~ ______ UTAH TEXAS OKLA.
-+~M~~g~Sg~OO~a~A~~~~--flr=~~-r-~~~-1
NE TENN.
t-
....
':'z
"'=>
::E
....
~
LAVA DAM
RED TOPS
::E SANSABA
....
SIGNAL
MOUNTAIN
r--------
~L3 rn
w z COPPER
Z ~ PTYCHASPID B.
ffi POINT
ROYER
RIDGE I
MRATOGIA
:r III PEAK
~ ~ ~ HELLNMARIA oJ FORTSIU
CONOCO-
a:: ~ z ~ CHEAGUE ALLENTOWN
m !i
TAENICEPHALUS
SNEAKOVER
MORGAN
CREEK
HONEY CREEK DOLOMITE
:E ~ E1.VW.. CORSET REAGAN r
~~
o
SPRING
PTERO- JOHNS
r-
DUNORBERGIA
~
w CEPHALIID
PREHOVSlA
WASH
z !E
~
OICANTHOPYGE
!I
:z: APHu.SPtS 0 CANDLAND
..J u f----+-----j LION MT.
~ CREPICEPHALUS ::E
ffi
a:
BIG HORSE .... CAP MT.
o >
MARJUMIID CEDAR'" t--'------j ~
t-B-a.-ASPIOfl--U.----i WEEKS FM. ii: HICS~~RY
Fig. 2. Biostratigraphic chart of Late Cambrian strata in this study. Note the three biomere boundaries
in the Late Cambrian. Adapted from several references listed in text.
2.1. TIME CONTROL AND ACCUMULATION RATES
Biostratigraphic control for Upper Cambrian strata consists of 10 major trilobite

zones, a few subzones, and conodont zonation in the later stages of the period
(Fig. 2). Relative age assignments were determined from published biostratigraphic
data (Palmer, 1954, 1965, 1971 a,b; Robison, 1964; Derby, 1965; Rassetti, 1965;
Longacre, 1970; Hintze, 1974; Hintze and Palmer, 1976; Barnes and Bell, 1977;
Stitt, 1971, 1977; Hintze, Miller and Taylor, 1980; Eby, 1981; Taylor and Miller,
1981; Stitt and others, 1982; Miller and others, 1982; Orndorff, 1988; Sundberg,
1990). Three biomere boundaries, cm-scale stratigraphic intervals showing abrupt
non-evolutionary changes in trilobite faunas, are unique to the Middle to Upper
Cambrian and enhance the relative time control. Biomere boundaries have been
recognized in several sections throughout North America and provide excellent
datums for chronostratigraphic correlation because they are generally interpreted
to manifest relatively isochronous extinction events (Palmer, 1965; 1984). The
recognition of biomere boundaries, as well as intervening trilobite biostratigraphy,
was the primary criterion in the selection of stratigraphic sections for this study.
The most recent estimates of Cambrian time have been determined by Bowring
and others (1993) from U-Pb zircon dating of Lower Cambrian rocks from north-
east Siberia. They have constrained the base of the Cambrian at ",544 Ma with
an estimated age for the beginning of Middle Cambrian time around 520 Ma. As-
suming a Cambro-Ordovician boundary date of 505 Ma (in accord with Palmer,
1983), and subequal durations for Middle and Late Cambrian time, these new dates
suggest time spans of7.5 m.y. each. Fig. 3 illustrates the burial histories of Middle
Cambrian to Middle Ordovician rocks for five of the studied localities, using the
Ee EO MO
540 Ma 520 500 480 460
~:::- _ _ _ _,Llano Uplift

(0.056)
E 2
~
L:
a
Q)
Cl 4
SWVA
Appalachians
6 (0.095)
Decompacted Subsidence House Range

(Late Cambrian rates in m/k.y.) Utah (0.211)
Fig. 3. Comparative decompacted subsidence curves for the Utah, Virginia, Tennessee, Texas,
and Oklahoma localities. Decompacted thicknesses were determined using software provided by M.
Kominz and G. Bond. Late Cambrian subsidence rates (in parentheses) were calculated from the
mid-point values between minimum and maximum decompaction estimates. The abrupt decrease
in apparent subsidence rates at the Cambro-Ordovician boundary is an artifact of the shortened
duration of Middle and Late Cambrian time proposed by Bowring and others (1993). The very rapid
subsidence rates of Late Cambrian ti me differ signi ficantly from those published in Osleger and Read
(1993) because of the new estimates for Cambrian time.
new age determinations from Bowring and others (1993). Delithified accumula-
tion rates for Late Cambrian strata range from 0.056 m/k.y. for the Texas cratonic
section to 0.211 m/k.y. for the House Range of Utah. These rates are considerably
more rapid than those calculated for most other passive margins (0.01 to 0.1 m/k.y.;
Wilson, 1975; Schlager, 1981), but are within the overall range of accumulation
rates for sections deposited over 106 to 107 year time spans (Sadler, 1981; Bosscher
and Schlager, 1993). These Late Cambrian rates appear suspiciously rapid com-
pared to Early Ordovician rates, likely due to the 32 m.y. uncertainty bounding
the 505 Ma date for the end of the Cambrian (Palmer, 1983). Thus, a conservative
50 percent margin of error is incorporated into all calculations involving total Late
Cambrian time.
3. Upper Cambrian Depositional Sequences
Depositional sequences are frequently interpreted within the context of a hierarchy

of spatial and temporal scales (e.g., Vail and others, 1977; Haq and others, 1987;
252 D.A. OsJeger
Goldhammer and others, 1990; Osleger and Read, 1991). The concept of "orders"
provides a convenient terminology to describe the hierarchical arrangement of
meter-scale cycles (usually spanning durations of 104 to 105 years; fifth-order)
within stacked sets of cycles (usuaIll105 to 106 years; fourth-order), which in
tum comprise sequences (usually 10 to 107 years; third-order). Sequences also
group into larger sequences (at the scale recognized by Sloss, 1963) that span
durations of 107 to 108 years (second-order). The boundaries between orders are
implicitly artificial and have no geological significance because cycles, cycle sets,
sequences and supersequences fall within a continuum of thicknesses and therefore
time spans. This is logical given the continuous, linear decrease in accumulation
rate with increasing time span (Sadler, 1981). As long as the inherent limitations
of the hierarchy of orders are understood, however, they provide a convenient
way of describing the various scales of depositional units recognized within many
stratigraphic successions.
Two second-order sequences have been recognized in Upper Cambrian strata
(Lochman-Balk, 1971; Palmer, 1971a,b), which are separated by a cratonwide un-
conformity at the Dresbachian-Franconian transition (the Sauk II-Sauk III bound-
ary of Palmer, 1981 b). Subaerial erosion or non-deposition associated with the
unconformity has removed the Dunderbergia through part of the Aphelaspis zones
from localities on the North American craton and on portions of the craton margin.
The boundary between the two long-term depositional sequences appears to be rel-
atively conformable on the more rapidly subsiding passive margins that surrounded
the Late Cambrian craton.
Several workers have recognized third-order depositional sequences in Upper
Cambrian strata throughout North America (Stitt, 1977; Aitken, 1978; Markello
and Read, 1982; Palmer, 1981a; King and Chafetz, 1983; Chow and James, 1987;
Bond and others, 1988, 1989; Read, 1989; Westrop, 1989; James and others, 1989;
Koerschner and Read, 1989; Kozar and others, 1990; Cooper and Edwards, 1991;
Osleger and Read, 1993; Cowan and James, 1993). Grand cycles, consisting of a
lower shaly half-cycle and an upper carbonate half-cycle spanning two or more
trilobite zones (Aitken, 1981), characterize Late Cambrian deposition in the south-
ern Canadian Rockies and have also been interpreted as depositional sequences.
Attempts at interbasinal correlation of Upper Cambrian grand cycles have relied
on the presumed synchroneity of grand-cycle boundaries as datums for correlation,
and Palmer (1981 a), Aitken (1981), and Bond and others (1989) were able to cor-
relate numerous third-order depositional sequences between grand cycles of the
southern Canadian Rockies and successions of Middle to Upper Cambrian strata
elsewhere on the continent. It should be made clear, however, that "grand cycles"
and "sequences" should not be unequivocally considered synonymous concepts;
grand cycles are lithostratigraphic units and would define depositional sequences
only if sequence boundaries coincide with the carbonate-to-shale transition at the
tops of grand cycles. Mount and others (1991) and Montanez and Osleger (1995)
have documented Lower and Middle Cambrian examples in which the sequence
boundary occurs in the middle of the shaly half-cycle, coincident with significant
falls in sea level and an influx of siliciclastics into the basin.
As opposed to many other outcrop-based sequence-stratigraphic studies that

focus on the areal distribution of sequences across a single platform, the interbasi-
nal nature of this project requires a dependence upon single outcrops that exhibit the
most complete exposure and biostratigraphic control. Thus a slight modification of
the standard definition of a depositional sequence is necessary. Rather than solely
being "a relatively conformable succession of genetically related strata bounded
by unconformities and their correlative conformities" (Mitchum, 1977), deposi-
tional sequences in this project are modified to include "interbasinally-correlative
upward-shallowing successions bounded by a relatively conformable transition
into upward-deepening successions" (cf. Goldhammer and others, 1993). The in-
ternal characteristics of correlative sequence boundaries and systems tracts will be
detailed in a later section. The following section briefly describes the field deter-
mination of six depositional sequences recognized in Upper Cambrian strata based
on vertical changes in interpreted paleoenvironments and paleobathymetry. This
fundamental method of stratigraphic analysis provides an overall approximation
of apparent changes in relative water-depths and long-term upward-deepening and
upward-shallowing trends for individual sections.
3.1. PALEOBATHYMETRIC TRENDS AND SEQUENCE CORRELATION
Paleobathymetric curves were constructed for Upper Cambrian stratigraphic sec-

tions in the House Range, the Virginia-Tennessee Appalachians, the Llano Uplift,
and the Wichita Mountains of Oklahoma (Fig. 4). Relative fluctuations in water
depth were determined from interpreted changes in depositional environments of
lithofacies based upon Holocene analogs, stratigraphic distance of lithofacies be-
low tidal-flat caps of peritidal cycles (Grotzinger, 1986; Koerschner and Read,
1989), and relative ordering of facies successions in upward-shallowing cycles.
Individual lithofacies and interpretations of estimated water depths are discussed
in detail in Osleger and Read (1991).
Paleobathymetric trends for individual sections should be interpreted with cau-
tion because they are reliable only for that particular locality and do not account
for variability across the basin of deposition. Several correlative sections from
widespread areas across the basin should ideally be used to construct a composite
paleobathymetric curve that can be used for interbasinal comparison. More impor-
tantly, paleobathymetric trends may be misleading indicators of accommodation
history because they include the effects of differential sediment production/supply
that are difficult to separate from the synchronously-occurring processes of sub-
sidence and eustasy (lervey, 1988; Osleger and Read, 1991; Bond and Kominz,
1991). Acknowledging these limitations, the strength of individual paleobathymet-
ric trends may be tested by evaluating their interbasinal recognition; if reasonable
correlations can be determined between basins, presumably the regional envi-
ronmental influence on sediment production/supply is diminished, supporting the
interpretation of paleobathymetric trends as recorders of accommodation history.
Six major third-order sequences were identified and correlated between the
individual sections using graphic correlation (Fig. 5; Shaw, 1964; Edwards, 1984).
Several smaller scale (10's of meters) upward-shallowing trends were recognized
254 D.A. OsJeger
HOUSE RAHGE, UTAH SW YlAGIMA HE n .... l.LANO lPlFT, TEXAS WICHIT II. MrS., OKU..
AHlALACHWtS
LC-6
-~ o-u_
TF ~ ::-:
LC3 IE:JF-.-
~ -
SS ~ n-..---
~ - .........
---
~,.. ,....
". 1::::1 - -
--
IE) " - , , . . - -
LC-l
...
Fig. 4. Plots of interpreted relative paleobathymetry of lithofacies for stratigraphic sections in the
House Range of Utah. the Virginia-Tennessee Appalachians. Threadgill Creek in the Llano uplift of
central Texas. and Kindblade Ranch. Wichita Mountains. Oklahoma. Lithofacies on the diagrams
are generalized; the shaded parts of the rectangles show which of the four lithofacies - tidal-flat
(TF). shallow subtidal (SS). deep subtidal (OS). and shaly deep subtidal (SH) - are dominant.
The section thicknesses have been normalized to the House Range section for ease of correlation;
notice how the vertical scale differs for each of the sections. The Cambrian-Ordovician boundary
is the upper datum. Biostratigraphy is noted to the right of each relative water depth curve. The
two long-term upward-shallowing cycles of the Late Cambrian are separated by the heavy dark
horizontal line crossing the middle of the diagram (wavy at unconformities). Other major episodes
of upward-shallowing are separated by thin subhorizontal tie lines. The Appalachian column is a
composite of three sections from the Conasauga intrashelf-basin and the peritidal platform. The key
to lithologies is applicable to all subsequent figures.
:Ii
.
:E
c
;:!

~
1
I_ REAGAN SANDSTONE -I~:; I- FORT SILL _~ROYER ---I-SIGNAL MOUNTAIN -
Fig. 5. A) Graphic correlation plot of the Utah (horizontal) and Texas (vertical) sections. Black dots
are the first occurrences of trilobite and conodont zones and the dots with the crosses represent biomere
boundaries; the line of correlation was determined by least-squares regression. The flat trend on the
line of correlation marks the major unconformity in the Texas section and its equivalent strata in the
upward-shallowing Utah section. Vertical and horizontal tie lines mark the tops of the six depositional
sequences discussed in the text. B) Graphic correlation plot for Franconian-Trempealeauan time for
the Oklahoma (horizontal) and Texas (vertical) sections. Tie lines mark the tops of sequences LC-5
and LC-6.
256 D.A. OsJeger
in individual sections but were found to have no apparent correlatives in other

localities; the origin of these local sequences will be discussed in a later section.
Depositional sequences were considered to be essentially coeval if their points
of intersection plotted reasonably close to the line of correlation. The six major
sequences were named on the basis of their relative time ranges according to the
trilobite biostratigraphy and are: late Cedaria (LC-1), early to mid-Crepicephalus
(LC-2), mid- to late Crepicephalus (LC-3), Aphelaspis to earliest Elvinia (LC-
4), Elvinia to early Saukia (LC-5), and early Saukia to the Cambro-Ordovician
boundary (LC-6). Based on the age model of Bowring and others (1993), the
sequences are estimated to range in duration from 0.8 to 3.0 m.y. (50%). The
internal attributes of each of the six major sequences can be found in Osleger
and Read (1993); a subsequent section of this paper will focus on the diverse
characteristics of sequence boundaries and systems tracts of sequence LC-5 that
spans Elvinia to early Saukia time.
4. Late Cambrian Accommodation History
Rather than relying solely on paleobathymetric analysis and the associated lim-
itations, two quantitative methods were used to determine the accommodation
history for Late Cambrian time that approach the problem from different scales.
Fischer plots use the stacking patterns of meter-scale cycles to identify long-term
changes in accommodation that may have generated the depositional sequences.
Subsidence analysis uses the opposite approach from Fischer plots: large-scale pat-
terns of total subsidence are incrementally reduced by progressively removing the
effects of isostatic subsidence, thermo-tectonic subsidence, and long-term eustatic
events to derive smaller scale (third-order) changes in accommodation.
4.1. STACKING PATTERNS OF METER-SCALE CYCLES
Upward-deepening or upward-shallowing trends within many Late Cambrian de-

positional sequences are characterized by systematically changing patterns of
meter-scale cycles (considered to be synonymous with parasequences). Thick
successions of hundreds of stacked peritidal cycles dominate the Appalachian
passive margin, whereas a variety of shallow to deep subtidal cycles (Osleger,
1991) characterize the Cordilleran passive margin near the House Range embay-
ment and the Conasauga intrashelf basin of the southern Appalachians. Cyclicity
is much more sporadic in the Texas cratonic and Oklahoma aulacogenic set-
tings. Meter-scale cycles (fifth-order scale) typically form systematic groupings
of upward-shallowing and upward-deepening successions that form fourth-order
cycle sets (0.1-1.0 my; tens of meters); third-order depositional sequences (1.0-
10.0 my; tens to hundreds of meters) are defined by an upward-deepening cycle
set followed by an upward-shallowing cycle set. Stacking patterns of meter-scale
cycles (variations in thickness and lithofacies composition) provide the intermedi-
ate connection between individual meter-scale cycles and larger scale depositional
sequences (Goldhammer and others, 1990; Osleger and Read, 1991; Montanez and
Osleger, 1993).
The relative thicknesses and compositions of stacked meter-scale cycles are

dependent on the amount of accommodation space generated by eustasy and sub-
sidence, modified by sediment accumulation and compaction. Stacking patterns
of meter-scale cycles can be graphically illustrated with Fischer plots, displays of
cumulative departure from mean cycle thickness versus cycle number (Fischer,
1964; Read and Goldhammer, 1988). The wave trains evident on the plots can
be correlated with plots from coeval stratigraphic intervals from other localities
and, if reasonable correlations can be recognized, matching patterns can be used
to interpret their common accommodation history. The conceptual basis for the
interpretation of Fischer plots is that stacks of cycles thicker than the mean plot as
positive slopes and are presumed to have formed under conditions of increasing
accommodation space provided by relative sea-level rise. Stacks of cycles thinner
than the mean plot as negative slopes and are interpreted to reflect reduced ac-
commodation space during relative sea-level fall. The integrity of interpretations
derived from Fischer plots has been exhaustively reviewed in Osleger and Read
(1993) and Sadler and others (1993).
Several Late Cambrian accommodation events can be recognized between
widely-separated localities on the basis of correlated Fischer plots (Figs. 6A
and 6B). The Fischer plots were constructed from meter-scale cycles recognized in
ten stratigraphic sections from the Appalachian and Cordilleran passive margins.
Relative time lines and correlation lines on Figs. 6A and 6B are based on avail-
able biostratigraphy, biomere boundaries, the presence of regional quartz sands,
and similarities between patterns of positive and negative slopes on the curves.
In general, several trends on the plots provide corroboration for interpretations of
falling and rising relative sea level determined from paleobathymetric analysis.
The plot for Dresbachian time (Fig. 6A) illustrates generally decreasing ac-
commodation toward the end of Cedaria time in both peritidal and subtidal cyclic
successions (LC-I). The subtidal cyclic sections of the Nolichucky Formation in
Virginia and Tennessee and the Big Horse Member in Utah show a well-defined in-
crease in accommodation at the beginning of Crepicephalus time followed by a rel-
ative decrease in accommodation toward the Crepicephalus-Aphelaspis biomere
boundary (LC-3). The long-term Crepicephalus cycle is not evident within the
peritidal Allentown and Elbrook localities, although covered intervals in the El-
brook may mask this event. Evidence for the paleobathymetrically-defined mid-
Crepicephalus accommodation event (LC-2) is equivocal but is hinted at by pulses
of increased accommodation (identified by a dashed vertical tie line) within the
overall Crepicephalus event in three of the five sections.
All five localities show the rapid increase in accommodation at the beginning
of Aphelaspis time that also was recognized on the paleobathymetric curves. Poor
exposures or a lack of cyclicity preclude making Fischer plots for Aphelaspis
through Dunderbergia time in four of the sections, but the Fischer plot for the
Thorn Hill section of Tennessee shows a complete cycle of accommodation. The
Aphelaspis-early Elvinia event (LC-4) culminates in thin, disconformity-capped,
restricted tidal-flat cycles of the basal Copper Ridge Formation that correspond to
the major unconformity between Dresbachian and Franconian strata recognized
cratonwide.
258 D.A. Osleger
BIG HORSE MBR, ifJi!1Mf~\h

ORR FM., HOUSE
RANGE, UTAH
Cedaria Crepicephalus Dunderbergia Elvinio
DRESBACHIAN RELATIVE TIME
8)
CONOCOCHEAGUE FM.
WYTHEVILLE, VA.
NONCYClIC DEEP RAMP

FACIES Of HEUNW-RIA MBA
Elvlnla Taenicephalus Saukia 505 Ma
FRANCONIAN - TREMPEALEAUAN RELATIVE TIME

Fig, 6. A) Correlation of Fischer plots for Dresbachian time. The horizontal axis is a relative time
scale with trilobite zonation, and the vertical axis is cumulative departure from mean cycle thickness.
The upper two Fischer plots (and the latest portion of the Tennessee Fischer plot) are from peritidal
cyclic sections, whereas the lower three Fischer plots are from shaIlow subtidal cyclic sections.
Cycles containing quartz sand laminae are denoted by black triangles. Covered intervals have been
left blank. Vertical dashed lines mark interpreted correlations of accommodation events. The vertical
wavy line indicates the position of the Dresbachian-Franconian unconformity as recognized from
graphic correlation with other sections. B) Correlated Fischer plots for Franconian - Trempealeauan
time. Note the position of Cambrian-Ordovician boundary and the appearance of quartz sandy cycles
(black) during the sea-level lowstand at the end of the Late Cambrian. Cycles from the Appalachian
sections are peritidal, whereas the Utah section is composed of shaIlow to deep subtidal cycles.
The correlated plots for Franconian-Trempealeauan time (Fig. 6B) illustrate

one long-term accommodation event in five sections that appears to correspond
to the second-order event recognized on the paleobathymetric curves (LC-5 and
LC-6). The expected increase in accommodation during Elvinia time is not well-
defined on any of the plots. Generally increasing accommodation is evident during
Taenicephalus time and is followed by a relative highstand during early Saukia
time. The depositional event recorded by paleobathymetric trends from earliest
Saukia time to the Cambrian-Ordovician boundary (LC-6) is not very well defined
on the correlated plots but may be masked by the overall highstand conditions
during early Saukia time. However, the abrupt rise occurring partly into Saukia
time (identified by a dashed vertical tie line between four of the plots) may manifest
renewed deepening and the onset of sequence LC-6.
4.2. SUBSIDENCE ANALYSIS
Subsidence analysis (Sleep, 1971; Steckler and Watts, 1978; van Hinte, 1978;
Bond and Kominz, 1984) is useful for determining the amount of accommodation
space remaining after the effects of tectonic subsidence and sediment loading are
incrementally removed from decompacted total subsidence. The remaining curve
(R2 curve of Bond and others, 1989) presumably reflects changes in accommoda-
tion generated solely by eustasy or local tectonism. A further reduction of the R2
curve removes the modulating effects of the longer term sea-level fluctuations and
produces a higher resolution record of third-order changes in accommodation (R3
curve). The method allows for a direct comparison of accommodation histories
between different geographic and tectonic settings. The subsidence analysis for
this Late Cambrian project follows techniques and parameters fully outlined in
Bond and Kominz (1984, 1991) and Bond and others (1988, 1989). A complete
discussion of the decompaction options, backstripping procedure, removal of the
thermo-tectonic component of subsidence, and strengths and limitations of the
technique can be found in Osleger and Read (1993).
Results of subsidence analysis techniques applied to Upper Cambrian rocks
from the House Range are shown on Fig. 7. The R3 curve is shaded to emphasize
third-order changes in accommodation and illustrates deviations from a best-fit
polynomial curve applied to the R2 curve above. R3 curves are very useful for
recognizing subtle variations in the relative rate of accommodation change. This is
critical for distinguishing third-order accommodation events, because even minor
changes in the rates of sea-level rise or fall can produce shifts in onlap or offtap
(Pitman, 1978; Christie-Blick, 1991). The R3 curve shows six major changes in
accommodation that correspond to the six depositional sequences recognized by
paleobathymetric analysis, a relationship that illustrates the lithologic basis for
both of the techniques. Comparison of accommodation changes as expressed on
R3 curves for each of the four major field localities show a reasonable degree of
correlation (Osleger and Read, 1993).
More instructively, R3 curves were compared with paleobathymetric curves
and Fischer plots derived from the same stratigraphic interval (Fig. 8). For the
House Range section, the various curves corroborate each other in general form
260 D.A. OsJeger
Late Cambrian ,
House Range,
Utah
LC -5
~ I
I
I...I------Orr Fm - - - - - ---I--Notch Peak Fm -I

Fig. 7. Late Cambrian R2 and R3 curves for the Upper Cambrian of the House Range of Utah. The
generalized decompacted stratigraphic column is oriented horizontally with Late Cambrian relative
time marked by trilobite zones. Magnitudes of the accommodation changes (vertical axis) are not
absolute due to the various assumptions implicit in the method (Bond and others, 1989). Solid
vertical lines connect equivalent inflection points on the R2 and R3 curves and separate third-order
depositional sequences. The R3 curves were derived by fitting a fifth-order polynomial to the R2
curve to maximize the number of deviations within the overall trend of the R2 curve.
and appear to define internal components of the depositional sequences. This should
not be a suprising result given that all three techniques are ultimately dependent
upon lithology and associated characteristics (e.g., biota, sedimentary structures,
degree of compaction, amount of internal cement) and are not strictly independent
approaches to determining a relative sea-level history. For the House Range section,
the interpreted deepest water lithofacies on the paleobathymetric curve (deep-ramp
argillaceous carbonates of the Big Horse Member and thick shales of the Candland
Shale) generally correspond to the rising portions of its Fischer plot and R3 curve.
The Fischer plot shows a net increase in accommodation created by the generally
thicker cycles that characterize the deeper water intervals whereas the R3 curve
shows increasing accommodation due to the lower densities of the argillaceous
carbonates and siliciclastic shales. In contrast, the shallowest water lithofacies on
the paleobathymetric curve for the House Range (shallow subtidal and peritidal
carbonates of the Big Horse and Johns Wash Members) correspond to falling
portions of the Fischer plot and R3 curve. The thinner cycles that comprise these
intervals generate a trend of decreasing accommodation on the Fischer plot, and
the dominantly marine-cemented, calcarenitic lithologies create a comparable trend
on the R3 curve. In sum, correlative trends expressed on paleobathymetric curves,
Fischer plots, and R3 curves supply strong corroborative evidence for determining
Late Cambrian accommodation history. When combined with graphic correlation,
House Range.
Utah Pal eobalhymetry Fischer Piol R3 curve
LC-4
..
.
E
F
,~
70
m
Cumu1al~~
from Mean Cycle ~
Fig. 8. Direct comparison of the three techniques used to derive accommodation history for the
Big Horse, Candland Shale, and Johns Wash Members of the Orr Formation, House Range, Utah.
Correlation lines connecting key horizons on the different plots are not perfectly horizontal because
the stratigraphic column and its paleobathymetric curve are plotted as thickness, whereas the other
two curves are plotted as time. The horizontal scale for the R3 curve has been expanded by a factor
of three to undo the backstripping procedure, so that direct comparison with the Fischer plot curve
can be made.
262 D.A. Osleger
Pericratonic USA Central and S. Canadian Tennessee

~ Tri1ob~e Northern
Southern
dCl '"co E'"
C)
Zone
Com~s~e Curve
this study) Ap~alachians
Rockies and Utah
(Bond and others, 1989)
Appalachians
(Kolar and others,
Appalachians
{James and others,
a: iii ,Q ( ead,1989) 1990) 1989)
0 CD
IliNJulquo/.
z
c
::> F
~ Sauki.
::I :l2 I
tw 0. I
I!: m
.s::: I
I
t"
!a,.tog/. I
I
I
z I
c
z !~
TNnlCflpNlu. I
I
oCt I
a:CD ~
Elvln/.
I
I
'iii I
:!: .s:::
0. Ound.m.rg;'
I
\
oCt
0 '"~ \
A_..
l'rehouaJ. \
0::'"
UJ Dlc.nthopyge \
I- p," \
\
oCt \
..oJ cz \
~ \
"c51 :l2
C,.plcepMIu. \
\
w
a:
0
'E::>
'C'
co
::< Cedari.
....Rise--Fall .....
Fig. 9. Comparative relative sea-level curves, scaled for time, for Upper Cambrian strata of
North America. The composite Late Cambrian relative sea-level curve of this study was derived
by qualitatively combining the paleobathymetric curves, Fischer plots, and R3 curves from the
Utah Cordillera, the Virginia-Tennessee Appalachians, the Texas craton, and the southern Oklahoma
aulacogen. The curves from other localities were generated from various combinations of regional
transgressive-regressive relations, paleobathymetry, Fischer plots, and R2 analysis. The magnitudes
of excursions on all of the curves are relative. Events have been scaled for relative Late Cambrian
time and are labeled according to the original authors (except for the letter designation on the curve
of Bond and others (1989) which was added for ease of discussion).
the three analytical methods collectively provide a useful set of crosschecking

techniques to recognize the degree of synchroneity of accommodation change
between basins.
4.3. COMPARISON OF LATE CAMBRIAN SEA-LEVEL CURVES
A composite relative sea-level curve for Late Cambrian time was constructed by
qualitatively combining the various curves derived from the three different methods
(Fig. 9). Only the six third-order depositional sequences that could be correlated
between all four localities are included on the composite curve. Magnitudes of
sea-level fluctuations and rates of sea-level rise and fall on the curve are strictly
relative. The second-order envelope (dashed line), upon which the third-order
signal is superimposed, approximates the form of the Sauk sequence (Sloss, 1963),
interpreted to be at its crest during the mid-Late Cambrian.
When compared with relative sea-level curves derived by other methods

from other localities, some distinct similarities and differences can be recognized
(Fig. 9). The four other Late Cambrian curves were rescaled from their original
source to roughly fit the biostratigraphic zonation so that temporal correlation of
events could be more accurately estimated. The curve of Read (1989) was derived
from chronostratigraphic correlation of transgressive-regressive relations in the
Appalachians from Pennsylvania to Tennessee, and Fischer plots of cyclic succes-
sions. The curve of Bond and others (1989) "is an approximation of the average
short-term eustatic curve derived by eye from the R2 curves", and was primarily
generated from sections in the southern Canadian Rockies and Utah. The curve of
Kozar and others (1990) was based upon sequence stratigraphic interpretation of
rocks in east Tennessee. Finally, the curve of James and others (1989) was derived
from detailed regional relationships in the northern Appalachians.
Other than biomere boundaries, Cambrian trilobite biostratigraphy is character-
ized by evolutionarily gradational faunal changes and a strong facies dependence
(A.R. Palmer, pers. comm., 1994). These attributes limit the temporal resolution
available for precise chronostratigraphic correlation of depositional sequences be-
tween basins and must be recognized in any study of this kind. Acknowledging
the constraints of the biostratigraphic resolution, some trends can be consistently
correlated between all of the curves whereas others appear to have limited recogni-
tion: 1) The lowstand at the end of the Cedaria biozone is shown by all five curves,
assuming that event A (Bond and others, 1989), event B-5 (Kozar and others,
1990), and the earlier subevent in 4 on the curve of James and others (1989) truly
do terminate near the end of Cedaria time. 2) The mid-Crepicephalus event recog-
nized in this study may be correlative with event B of Bond and others (1989), but
has not been identified by other workers. 3) The late Crepicephalus lowstand (LC-
3) appears to have been recognized by Read (1989) and Bond and others (1989)
but is not evident in the other two curves. 4) The major second-order lowstand at
the boundary between the Dresbachian stage and the Franconian-Trempea1eauan
stages is clearly distinguished on all five curves. 5) The Elvinia drowning is clear
in all curves with the early Saukia lowstand identified on four of the curves. An
additional event appears to have been recognized in east Tennessee, but the event
is not evident in the northern Appalachians. 6) All of the curves appear to illustrate
a gentle fall in relative sea-level toward the Cambro-Ordovician boundary. The
reasonable degree of correlation between events recognized in several regions of
North America suggests that the coeval sequences were generated by a common
mechanism, most likely eustasy. Epeirogenic upwarping and subsidence of the cra-
ton is, arguably, an alternative mechanism that may also explain the synchronous
deposition of Upper Cambrian sequences throughout North America, but modeling
of vertical motion driven by large-scale mantle convection (Gumis, 1990) shows
that epeirogenic movements occur at longer time scales (l 0 7_10 8 years) than sug-
gested by the shorter-term, third-order events recognized in this study. The only
true way to differentiate cratonal tectonism from eustasy as the primary controlling
mechanism, however, is to compare Upper Cambrian sections worldwide.
264 D.A. OsJeger
4.4. VARIABLE SEDIMENTARY RESPONSE TO ACCOMMODATION CHANGE
The recognition of specific depositional sequences in some basins but not in others
is a major problem in establishing an unequivocal controlling mechanism behind
sequence development (Hallam, 1992). The general imprecision of biostratigraphic
resolution, especially for the early Paleozoic, adds another element of uncertainty
to any interbasinal correlations (Miall, 1992). Moreover, shallow-marine deposits
such as those in this study are riddled with non-depositional and erosional gaps
of variable duration, a natural response to the unsteady process of sedimentation.
Accumulation rate decreases with increasing time span (Sadler, 1981) and the most
rapid decreases occur at time scales with a high proportion of hiatuses (Sadler,
1994). Cyclic successions within the 104 to 105 time band, such as those that
characterize the Upper Cambrian, tend to show rapidly decreasing accumulation
rates and thus, interpretively, a non-random spacing of hiatuses (Sadler, 1994).
These gaps in the stratigraphic record must affect the analytical results from Fischer
plots and the subsidence analysis to indeterminable degrees. Beyond the inherent
limitations of biostratigraphy and the stratigraphic record, though, other factors
may influence the presence or absence of third-order depositional sequences in any
individual basin, independent of eustatic fluctuations.
First, it has long been recognized that the initiation of onlap-offtap sequences
on a platform is dependent upon the rate of sea-level rise or fall relative to the sub-
sidence rate (Pitman, 1978). The sedimentologic response of individual platforms
to changes in the rate of eustatic rise or fall would differ based upon its rate of
subsidence. For example, the apparent lack of evidence for three of the six Upper
Cambrian depositional sequences in the northern Appalachians may be related to
its position on the inner flexural wedge (Cowan and James, 1993), and the atten-
dant low subsidence rates relative to the more rapidly subsiding passive margins
in the western Cordillera and the central and southern Appalachians. Some third-
order accommodation events may simply not have been recorded on the western
Newfoundland craton margin because the underlying, thick, stiff lithosphere was
not subsiding rapidly enough to feel the effects of some smaller scale on lap events.
A second factor that may influence the degree of synchroneity of coeval depo-
sitional sequences is platform morphology. Depositional slope and platform width
will affect the migration rates of facies belts (Pitman and Golovchenko, 1988).
Broad, flat, fully-aggraded platforms such as evident in the central and southern
Appalachians should exhibit rapid and widespread shifts in facies; steep, narrow
platforms such as flanked the southern Oklahoma aulacogen may exhibit a slower
sedimentologic response and have facies aligned in narrow belts. Thus, apparent
diachroneity in the timing of depositional sequences between basins may be the
result of different morphological profiles and associated sedimentologic responses
to the same eustatic change.
A third factor that may affect the presence, absence, or degree of diachroneity
of depositional sequences on any particular platform is the regional rate of sedi-
ment supply, production, or transport, perhaps related to the environmental regime
(Schlager, 1991, 1993). Environmental factors particular to individual platforms,
such as variations in sediment production and dispersal, wave and storm activity,
changing oceanic currents, oceanic productivity, and ambient ocean chemistry and
temperature, may conspire to influence, or even control, sequence deposition. Cli-

matic variations, perhaps related to orogenesis, may also affect deposition in one
basin that is not recorded in another basin. For example, Cowan and James (1993)
suggest that the supply of fine siliciclastics onto the Upper Cambrian platform of
the northern Appalachians occurred in response to climatic fluctuations, indepen-
dent of relative sea level. Environmental factors exert an important, if difficult
to recognize, influence on the deposition of sequences in any particular basin,
complicating attempts at interbasinal correlations.
All of these above factors suggest that even though depositional sequences
correlated between basins may be geneticaIly related, they are probably never
exactly synchronous. Limited biostratigraphic resolution, temporal gaps of vari-
able duration in the stratigraphic record, different tectonic subsidence rates and
platform morphologies, and variable sedimentary responses to regional environ-
mental change all may interact to affect the presence, absence, or synchroneity of
interbasinaIly-correlated sequences. Reynolds and others (1991) have suggested
that the timing of sequence boundaries may shift by up to one-quarter of the sea-
level cycle, depending on the magnitude of the accommodation signal. All of these
factors must be taken into consideration when attempting interbasinal correlations
to ultimately establish a controlling mechanism such as eustasy, tectonism, or
environmental change.
5. Sequence Stratigraphy
Compared to the traditional methods of defining sequences and their component

systems tracts by geometric relations of reflection contacts on seismic sections,
different approaches must be used to translate sequence stratigraphic concepts to
individual outcrops. In this section, the Elvinia to early Saukia sequence (LC-5)
will be used to illustrate the detailed lithologic composition of coeval sequence
boundaries and systems tracts that developed under unique environmental condi-
tions in widely separated basins (Fig. 10).
5.1. INTERBASINAL COMPARISON OF SEQUENCE COMPONENTS
The majority of sequence boundaries within Upper Cambrian strata of the four
outcrop localities are interpreted to be Type 2 boundaries (Van Wagoner and
others, 1987) because they lack evidence of extensive, long-term subaerial ero-
sion expressed along a single stratigraphic horizon, and they exhibit a relatively
conformable transition into overlying upward-deepening successions. The only
unequivocal Type 1 sequence boundary in Upper Cambrian rocks of the four locali-
ties is the widespread unconformity separating Dresbachian from Franconian strata
(Loch man-Balk, 1971; Palmer, 1981b). Most Upper Cambrian sequence bound-
aries are expressed as transitional zones of very thin cycles composed of restricted
lithofacies exhibiting evidence for prolonged exposure. These sequence boundary
zones (SBZs) are interpreted to have been generated during high-frequency rel-
ative sea-level fluctuations superimposed upon a longer term fall or stillstand in
relative sea-level (Montanez and Osleger, 1993).
IV
Orr and Notch Peak Fms. Interbaslnal Comparison of Upper 0\
House Range, Utah Cambrian Sequence LC-S 0\
Composite Section of the
Copper Ridge - Conococheague
Fms. from Thorn Hill. Tenn. and
Wilberns Fm. Klndblade Ranch
Wytheville. Virginia
Threadgill Creek, Wichita Mtns.
Texas Oklahoma
'.
'. '.
HST
.~ ,-
1~
.; . . .;. .:
~
c:
..... \,... ;.:.:.~:.:.;
l_ -
r~ -_ l
I - l
l --
I;'i@:m
~ .~ - .~
. , ..
,,- : ~
~
8 r
60m
1
~
~
Fig. 10. Interbasinal correlation of sequence LC-5 spanning Elvinia to early Saulda time as determined from graphic S?
correlation. The horirontaI datum is the biomere boundary separating the Pterocephaliid biomere from the overlying Ptychaspid
biomere (dashed where uncertain in the peritidal section from the Appalachians). The lower sequence boundary of LC-5 is
the Type 1 unconformity expressed craton wide and the upper sequence boundary rone is the shaded interval spanning several f
meters toward the top of the upward-shallowing succession. Field-measured thicknesses in meters.
Similar to the SBZs, it is difficult to unequivocally define the point of maximum

flooding at a single surface, especially in studies such as this one where the data
base consists of single sections where lateral expression of suspected critical
horizons has not been determined. The transition between transgressive systems
tracts (TST) and highstand systems tracts (HST) in cyclic successions such as these
Upper Cambrian examples is complicated by high-frequency sea-level events and
is considered to be a zone of maximum flooding (MFZ), similar conceptually to
the SBZs. MFZs in Upper Cambrian strata are characterized by vertical transitions
from thick cycles showing deepest water deposition to successively thinner cycles
with a concomitant increase in shallow-marine facies. This shift in stacking patterns
is interpreted to reflect the progressive decrease in accommodation space as the
rate of sea-level rise slows toward the eustatic peak.
Depositional sequence LC-5 illustrates unique characteristics depending upon
the tectonic and environmental settings of the individual localities (Fig. 10) . In the
House Range section, the sequence boundary zone at the top of LC-4 is marked by
fenestral mudstones of the Johns Wash Member of the Orr Formation, deposited
during a time of minimal accommodation across the entire continent (Lochman-
Balk, 1971; Palmer, 1981 b; Osleger and Read, 1993). The TST is represented by
thick shaly cycles of the Corset Spring Member; maximum flooding is interpreted
to occur below an interval of shale exhibiting evidence for deepest water condi-
tions. These shales grade upward into argillaceous carbonates of the Sneakover
Pass Member, the "subtidal blanket" of Brady and Rowell (1976) who recognized
its broad extent across the Utah-Nevada passive margin during middle-late Elvinia
time. Progressive shallowing within the thick HST is exhibited in the Hellnmaria
Member by a transition into cyclic alternations of biohermal thrombolites and
stromatolites and bedded, high-energy, peloidal grainstones. The upper sequence
boundary zone is marked by stromatolite bioherms, high-energy peloidal grain-
stones showing bidirectional cross-bedding, and fenestral dolomites capped by a
series of scalloped erosion surfaces. These microkarstic horizons provide evidence
for episodic subaerial exposure during times of minimum accommodation space
generated during a long-term third-order sea-level fall. Above the scalloped ero-
sion surfaces, a coarse oncolite lag deposit overlain by oolitic grainstones records
renewed flooding of the exposed platform and the initial development of sequence
LC-6.
In the peritidal Appalachian sections, the predominance of peritidal lithofacies
in the Copper Ridge-Conococheague Formations makes the qualitative recognition
of sequence components difficult, but systematic changes in stacking patterns of
peritidal cycles as illustrated on Fischer plots help to determine sequence boundary
zones and systems tracts. The basal sequence boundary separating LC-4 and LC-5
is expressed as several very thin cycles characterized by abundant quartz sand
and subtle erosional truncation of some cycle tops; unequivocal evidence for a
single major exposure surface is lacking. The TST is marked by the appearance of
thick cycles dominated by thrombolite bioherms and ooid grainstones. Maximum
flooding is difficult to determine in peritidal settings but is interpreted to occur in
an interval of the thickest cycles as expressed on the rising limbs of Fischer plots.
Upward shallowing in the HST is suggested by a change in cycle stacking pattern to
268 D.A. OsJeger
thin peritidal cycles characterized by extensive cryptalgallaminite caps and squat

thrombolite and stromatolite morphologies, indicative of reduced accommodation
space. The SBZ is interpreted to occur in a series of thin cycles that exhibit
evidence for prolonged subaerial exposure such as regolithic breccias, silicified
evaporite nodules, and low-relief tepees. Subsequent deepening into sequence
LC-6 is recorded by the abrupt appearance of thick, subtidal-dominated cycles
composed of ooid grainstones and digitate thrombolite bioherms.
In the Texas section, the predominance of shallow-subtidal and peritidal litho-
facies, complicated by the slow subsidence of its cratonal position, make transitions
between sequences and systems tracts tough to pinpoint. The basal unconformity
separating LC-4 and LC-5 is overlain by transgressive marine sandstones (Welge
Member of the Wilberns Formation) and open-marine skeletal packstones and
grainstones (Morgan Creek Member). Maximum deepening at the top of the TST
is recorded by a zone of cyclic fine siltstones and flat-pebble conglomerates of the
Point Peak Member, lithofacies indicative of deposition in the zone of storm-wave
reworking. Progressive shoaling in the HST is manifest by crossbedded skeletal
grainstones and thrombolite bioherms of the overlying San Saba Member. Maxi-
mum shallowing in early Saukia time is interpreted to occur within a 1 to 2 meter
unit of low-relief, elongate stromatolitic bioherms with intervening troughs filled
with coarse, tidal-deposited skeletal-pelletal grainstones. A return to open-marine,
shallow-subtidal conditions is marked by oolitic and skeletal grainstones with
abundant burrow mottling. LC-5 in the Texas section is punctuated by several
smaller-scale sequences (10-20 m thick; fourth-order scale) in the Point Peak
Member that consist of stacked thrombolitic bioherms above intervals of deeper
water laminated siltstones. These smaller-scale sequences cannot be correlated
with equivalent strata in the Cordilleran, Appalachian, or Oklahoma sections.
The southern Oklahoma aulacogen was first inundated in Elvinia time (Loch-
man-Balk, 1971; Stitt, 1971, 1977) as recorded by the deposition of the shallow
marine Reagan Sandstone over rhyolitic basement. In the Wichita Mountains
section, deepening in the TST is indicated by cyclic deposition of open-marine
skeletal limestones of the Honey Creek and lower Fort Sill Formations; maximum
flooding is interpreted to have occurred within deeper water shales of the lower
Fort Sill. The HST is interpreted as open-marine lithofacies of the upper Fort Sill
Formation that give way to highly restricted tidal-flat facies within the Royer and
lower Signal Mountain Formations (cryptalgallaminites and stacked thrombolite
bioherms). Coarsely crystalline dolomite within the Royer contains remnants of
fenestral mudstone and oncolitic packstone and probably represents a tongue of
peritidal facies that prograded out into the center of the aulacogen from the cratonic
margins (Stitt, 1971, 1977). Cryptalgal laminites in the lower Signal Mountain
Formation manifest the SBZ and minimum accommodation; the SBZ is overlain by
open-marine wackestone and packstone with stacked hardgrounds and glauconite,
recording renewed flooding at the base of LC-6.
The intention of this detailed inspection of sequence LC-5 is to show the distinct
sedimentologic responses of individual tectonic and environmental settings to a
single, third-order eustatic event. Even though each section is composed of different
types of meter-scale cycles and component lithofacies that reflect the environmental
dynamics of their depositional setting, the overall long-term trends within sequence
LC-5 are clearly evident. Factors intrinsic to a single depositional system such as
platform morphology, subsidence history, tectonic regime, paleotopography, and
prevailing oceanographic conditions all may contribute to a stratigraphic record
unique to that setting. Detailed interbasinal comparison of these variable Late
Cambrian sedimentary systems indicates, however, that they appear to have been
"synchronously" deposited in response to a single allogenic mechanism such as
eustasy.
Translating sequence stratigraphic principles originally defined for seismic-
scale siliciclastic systems to individual outcrops of cyclic carbonates requires a
de-emphasis of stratal geometries and an increased awareness of correlative vertical
changes in interpreted paleobathymetry and stacking patterns of component meter-
scale cycles. The major limitation of this method as applied to lower Paleozoic
successions is the quality and resolution of biostratigraphic control upon which
all interbasinal correlations are based. Chronostratigraphic control for this Upper
Cambrian study is significantly enhanced, however, by the occurrence of cm-scale
biomere boundaries which, arguably, provide unique time lines for continent-scale
correlations.
6. Summary and Conclusions
This study illustrates the utility of combining the results of three distinct tech-
niques to arrive at an accommodation history that can be tested as a global model
for Upper Cambrian deposition. I) Interbasinally-correlated trends of changing pa-
leobathymetry, integrated with a reasonably high degree of biostratigraphic control,
provide a good framework for establishing the relative synchroneity of depositional
sequences. 2) Systematic changes in the stacking patterns of meter-scale cycles,
as expressed on Fischer plots, furnish an independent test of the changing accom-
modation potential throughout the duration of a sequence. 3) Subsidence analysis
provides a third means of determining the remnant accommodation attributable
to eustasy and/or local tectonism after isostatic and thermo-tectonic components
of total subsidence have been removed. Each of these techniques has limitations
and perhaps should not be used individually to derive a eustatic sea-level history.
Rather they should be used in conjunction, if possible, as a set of crosscheck-
ing methods to document the degree of synchroneity of accommodation change
between basins.
The six third-order depositional sequences recognized in this study of outcrop
localities in the Utah Cordillera, the southern Appalachians, the Llano Uplift,
and southern Oklahoma show several similarities and a few differences when
compared with Late Cambrian relative sea-level curves of other researchers. The
reasonable degree of correlation between events recognized in several regions of
North America suggests that the coeval sequences were generated by a single
allogenic mechanism, most likely eustasy. If future correlations can be made
intercontinentally with a reasonable degree of synchroneity, the establishment of
a eustatic sea-level history will greatly increase our chronostratigraphic resolution
of Late Cambrian time beyond that afforded currently by biostratigraphy.
270 D.A. OsJeger
Acknowledgments
Several of the ideas generated in this study were the result of extensive interaction
with Fred Read, Isabel Montanez, and Maya Elrick. Expertise in the Cambrian and
guidance to field localities was provided by Jim Miller, Pete Palmer, Dick Faas,
and Lehi Hintze. Michelle Kominz and Gerard Bond graciously donated their
software for the subsidence analysis. Paul Enos, Pete Palmer, Gerard Bond, and
John Rodgers contributed important editorial commentary to an earlier associated
manuscript. Tom Algeo and Bilal Haq considerably improved the final version
through their careful editing.
References
AITKEN, J. D., 1978, Revised models for depositional grand cycles, Cambrian of the
southern Rocky Mountains, Canada: Bulletin Canadian Petroleum Geology, v. 26, p.
515-542.
AITKEN, J. D., 1981, Generalizations about grand cycles, in Taylor, M. E., editor, Short
papers for the Second International Symposium on the Cambrian System: United States
Geological Survey Open-File Report 81-743, p. 8-14.
BERNER, R. A., 1993, Paleozoic atmospheric C02: Importance of solar radiation and
plant evolution: Science, v. 261, p. 68-70.
BOND, G. c., CHRISTIE-BLICK, NICHOLAS, KOMINZ, M. A. and DEVLIN, W. J.,
1984, An early Cambrian rift to post-rift transition in the Cordillera of western North
America: Nature, v. 316, p. 742-745.
BOND, G. C. and KOMINZ, M. A., 1984, Construction of tectonic subsidence curves
for the early Paleozoic miogeocline, southern Canadian Rocky Mountains: Implications
for subsidence mechanisms, age of breakup and crustal thinning: Geological Society of
America Bulletin, v. 95, p.155-173.
BOND, G. C. and KOMINZ, M. A., 1991, Some comments on the problem of using
vertical facies changes to infer accommodation and eustatic sea-level histories with
examples from Utah and the southern Canadian Rockies, in Franseen, E. K., Watney,
W. L., Kendall, C. G. St. c., and Ross, W., editors, Sedimentary modelling: Computer
simulations and methods for improved parameter definition: Kansas Geological Survey
Bulletin 233, p. 273-292.
BOND, G. c., KOMINZ, M. A. and GROTZINGER, J. P., 1988, Cambro-Ordovician
eustasy: evidence from geophysical modelling of subsidence in Cordilleran and Ap-
palachian passive margins, in Paola, c., and Kleinsephn, K., editors, New Perspectives
in Basin Analysis: New York, Springer-Verlag, p. 129-161.
BOND, G. c., KOMINZ, M. A., GROTZINGER, J. P. and STECKLER, M. S., 1989, Role
of thermal subsidence, flexure and eustasy in the evolution of Early Paleozoic passive
margin carbonate platforms, in Crevel\o, P., Wilson, J. L., Sarg, J. E, and Read, J. E,
editors, Controls on Carbonate Platform and Basin Development: Society Economic
Paleontologists Mineralogists Special Publication 44, p. 39-62.
BOSSCHER, H. and SCHLAGER, W, 1993, Accumulation rates of carbonate platforms:
Journal of Geology, v. 101, p. 345-355.
BOWRING, S. A., GROTZINGER, J. P., ISACHSEN, C. E., KNOLL, A. H.,
PELECHATY, S. M. and KOLOSOV, P., 1993, Calibrating rates of Early Cambrian
evolution: Science, v. 216, p. 1293-1298.
BRADY, M. J. and ROWELL, A. J., 1976, Upper Cambrian subtidal blanket carbonate of
the Cordilleran miogeocline, eastern Great Basin: Brigham Young University Geological
Studies, v. 23, p. 153-163.
CHOW, N. and JAMES, N. P., 1987, Cambrian Grand Cycles: a northern Appalachian
perspective: Geological Society of America Bulletin, v. 98, p. 418-429.
CHRISTIE-BLICK, NICHOLAS, 1991, Onlap, offiap, and the origin of unconformity-
bounded depositional sequences: Marine Geology, v. 97, p. 35-56.
COOPER, J. D. and EDWARDS, J. c., 199 I, Cambro-Ordovician craton-margin carbonate
section, southern Great Basin: A sequence-stratigraphic perspective, in Cooper, J. D.,
and Stevens, C. H., editors, Paleozoic Paleogeography of the Western United States-II:
Pacific Section Society Economic Paleontologists Mineralogists, v. 67, p. 237-252.
COWAN, C. A. and JAMES, N. P., 1993, The interactions of sea-level change, terrigenous-
sediment influx, and carbonate productivity as controls on Upper Cambrian Grand
Cycles of western Newfoundland, Canada: Geological Society of America Bulletin, v.
105, p. 1576-1590.
DALZIEL, I.W.D., 1991, Pacific margins of Laurentia and East Antarctica-Australia as
a conjugate rift pair: Evidence and implications for an Eocambrian supercontinent:
Geology, v. 19, p. 598-601.
DERBY, J. R., ms, 1965, Paleontology and stratigraphy of the Nolichucky Formation in
southwest Virginia and northeast Tennessee: Ph.D. dissertation, Virginia Polytechnic
Institute and State University, Blacksburg, Virginia, 468 p.
EBY, R. G., ms, 1981, Early Late Cambrian trilobite faunas of the Big Horse Limestone
and correlative units in central Utah and Nevada: Ph.D. dissertation, State University of
New York, Stony Brook, 613 p.
EDWARDS, L. E., 1984, Insights of why graphic correlation (Shaw's method) works:
Journal of Geology, v. 92, p. 583-597.
FISCHER, A. G., 1964, The Lofer cyclothems of the Alpine Triassic, in Merriam, D. F.,
editor, Symposium of cyclic sedimentation: Kansas Geological Survey Bulletin 169, p.
107-150.
GOLDHAMMER, R. K., DUNN, P. A. and HARDIE, L. A., 1990, Depositional cycles,
composite sea-level changes, cycle stacking patterns, and the hierarchy of stratigraphic
forcing: Examples from Alpine Triassic platform carbonates: Geological Society Amer-
ica Bulletin, v. 102, p. 535-562
GOLDHAMMER, R. K., LEHMANN, P. J. and DUNN, P. A., 1993, The origin of high-
frequency platform carbonate cycles and third-order sequences (Lower Ordovician El
Paso Gp, west Texas): constraints from outcrop data and stratigraphic modeling: Journal
of Sedimentary Petrology, v. 63, p. 318-359.
GRABAU, A.W., 1936, Oscillation or pulsation?: Report of the 16th International Geo-
logical Congress, v. I, p. 539-552.
GROTZINGER, J. P., 1986, Cyclicity and paleoenvironmental dynamics, Rocknest plat-
form, northwest Canada: Geological Society of America Bulletin, v. 97, p. 1208-1231.
GURNIS, M., 1990, Bounds on global dynamic topography from Phanerozoic flooding of
continental platforms: Nature, v. 344, p. 754-756.
HALLAM, A., 1984, Pre-Quaternary sea-level changes: Annual Review of Earth and
Planetary Sciences, v. 12, p. 205-243.
HALLAM, A., 1992, Phanerozoic Sea-Level Changes: Columbia University Press, New
York, 266 p.
HANCOCK, J. M. and KAUFMANN, E. G., 1979, The great transgressions of the Late
Cretaceous: Journal of the Geological Society of London, v. 136, p. 175-186.
HAQ, B. U., HARDENBOL, JAN and VAIL, P. R, 1987, Chronology of fluctuating
sea-levels since the Triassic: Science, v. 235, p. 1156-1167.
272 D.A. Osleger
HINTZE, L. F., MILLER, J. F. and TAYLOR, M. E., 1980, Upper Cambrian-Lower

Ordovician Notch Peak Formation in western Utah: United States Geological Survey
Open-File Report 80-776, p. 67.
HINTZE, L. F. and PALMER, A. R., 1976, Upper Cambrian Orr Formation: its subdivisions
and correlatives in western Utah: United States Geological Survey Bulletin 1405-G, p.
1-25.
JAMES, N. P., STEVENS, R. K., BARNES, C. R. and KNIGHT, IAN, 1989, Evolu-
tion of a Lower Paleozoic continental-margin carbonate platform, northern Canadian
Appalachians, in Crevello, Paul, Wilson, J. L., Sarg, J. F., and Read, J. F., editors, Con-
trols on Carbonate Platform and Basin Development: Society Economic Paleontologists
Mineralogists Special Publication. 44, p. 123-146.
JERVEY, M. T., 1988, Quantitative geological modeling of siliciclastic rock sequences
and their seismic expression: in Wilgus, C. K., Posamentier, H. W., Ross, C. A., and
Kendall, C. G. St. C., editors, Sea-Level Changes: An Integrated Approach: Society of
Economic Paleontologists and Mineralogists Special Publication 42, p. 47-69.
JOHNSON, J. G., KLAPPER, G. and SANDBERG, C. A., 1985, Devonian eustatic fluc-
tuations in Euramerica: Geological Society of America Bulletin, v. 96, p. 567-587.
JOHNSON, M. E., RONG JIA-YU and FOX, W. T., 1989, Comparison of Late Ordovician
epicontinental seas and their relative bathymetry in North America and China: Palaios,
v. 4, p. 43-50.
KING, D. T. and CHAFETZ, H. S., 1983, Tidal-flat to shallow-shelf deposits in the Cap
Mountain Limestone Member of the Riley Formation, Upper Cambrian of central Texas:
Journal of Sedimentary Petrology, v. 53, p. 261-273.
KOERSCHNER W. F. and READ, J. F., 1989, Field and modelling studies of Cambrian
carbonate cycles, Virginia Appalachians: Journal of Sedimentary Petrology, v. 59, p.
654-687.
KOZAR, M. G., WEBER, L. J. and WALKER, K. R., 1990, Field and modelling studies of
Cambrian carbonate cycles, Virginia Appalachians - Discussion: Journal of Sedimentary
Petrology, v. 60, p. 790-794.
LENZ, A. c., 1982, Ordovician to Devonian sea-level changes in western and northern
Canada: Canadian Journal of Earth Science, v.19, p. 1919-1932.
LEVY, M. and CHRISTIE-BLICK, NICHOLAS, 1991, Tectonic subsidence of the early
Paleozoic passive continental margin in eastern California and southern Nevada: Geo-
logical Society of America Bulletin, v. 103, p. 1590-1606.
LOCHMAN-BALK, CHRISTINA, 1971, Cambrian of the craton: in Holland, E.R. editor,
Cambrian of the New World, London, Wiley-Interscience, p. 79-167.
LONGACRE, S. A., 1970, Trilobites of the Upper Cambrian Ptychaspid Biomere, Wilberns
Formation, central Texas: Paleontological Society Memoir 4, 70 p.
MARKELLO, J. R. and READ, J. E, 1982, Upper Cambrian intrashelfbasin, Nolichucky
Formation, southwest Virginia Appalachians: American Association of Petroleum Ge-
ologists Bulletin, v. 66, p. 860-878.
MIALL, A. D., 1992, Exxon global cycle chart: An event for every occasion?: Geology, v.
20, p. 787-790.
MILLER, J. F., 1984, Cambrian and earliest Ordovician conodont evolution, biofacies, and
provincialism, in Clark, D. L., editor, Conodont Biofacies and Provincialism: Geological
Society of America Special Paper 196, p. 43-68.
MILLER, J. E, 1987, Upper Cambrian and basal Ordovician conodont faunas of the
southwest flank of the Llano uplift, Texas, in Grayson, R. C., editor, Early and Late
Paleozoic conodont faunas of the Llano uplift region, central Texas - Biostratigraphy,
systemic boundary relationships, and stratigraphic importance: Geological Society of
America, Guidebook for Field Trip I, 1987 Meeting South-Central Section, p. 1-22.
MILLER, 1. F., TAYLOR, M. E., STITT, 1. H., ETHINGTON, R. L., HINTZE, L. F. and
TAYLOR,1. F., 1982, Potential Cambrian-Ordovician stratotype sections in the western
United States, in Bassett, M.G., and Dean, W.T., editors, The Cambrian-Ordovician
boundary: Sections, fossil distributions, and correlations: National Museum of Wales
Geological Series no. 3, p. 155-180.
MITCHUM, R. M., 1977, Glossary of terms used in seismic stratigraphy, in Payton, C.
E., editor, Seismic stratigraphy - applications to hydrocarbon exploration: American
Association of Petroleum Geologists Memoir 26, p. 205-212.
MONTANEZ, I. P. and OSLEGER, D. A., 1993, Parasequence stacking patterns. third-
order accommodation events and sequence stratigraphy of Middle to Late Cambrian
platform carbonates, Bonanza King Formation, southern Great Basin. in Loucks. R. B.,
and Sarg. J. F.. editors, Carbonate sequence stratigraphy - Recent developments and
applications: American Association of Petroleum Geologists Memoir 57. p. 305-326.
MONTANEZ, I. P. and OSLEGER, D. A., 1995, Contrasting sequence boundary zones
developed within cyclic carbonates of the Bonanza King Formation, Middle to Late
Cambrian, southern Great Basin: in Witzke, B., Ludvigson, G., and Day. T., eds., Pale-
ozoic Sequence Stratigraphy: North American Perspectives, GSA Special Publication,
(in press).
MOUNT, J. F., HUNT, D. L., GREENE, L. R. and DIENGER, J., 1991. Depositional
systems, biostratigraphy and sequence stratigraphy of Lower Cambrian grand cycles,
southwestern Great Basin: in Cooper, J.D., and Stevens, C.H., eds., Paleozoic Paleo-
geography of the Western United States II: Pacific Section SEPM, v. 67, p. 209-229.
ORNDORFF, R. c., 1988, Latest Cambrian and Earliest Ordovician conodonts from the
Conococheague and Stonehenge Limestones of northwestern Virginia: United States
Geological Survey Bulletin 1837, A 1-18.
OSLEGER, D. A., 1991, Subtidal carbonate cycles: Implications for allocyclic versus
autocyclic controls: Geology, v. 19, p. 917-920.
OSLEGER, D. A. and READ, 1. F., 1991, Relation of eustasy to stacking patterns of meter-
scale carbonate cycles, Late Cambrian, U.S.A.: Journal of Sedimentary Petrology, v. 61,
no.7. p. 1225-1252.
OSLEGER, D. A. and READ, 1. F., 1993, Comparative analysis of methods used to
define eustatic variations in outcrop: Late Cambrian interbasinal sequence development:
American Journal of Science, v. 293, p. 157-216.
PALMER, A. R., 1954, The faunas of the Riley Formation in central Texas: Journal of
Paleontology, v. 28, p. 709-786.
PALMER, A. R., 1965, Trilobites of the Late Cambrian Pterocephaliid biomere in the
Great Basin, United States: United States Geological Survey Professional Paper 493,
105 p.
PALMER, A. R., 1971 a, The Cambrian of the Great Basin and adjacent areas, western
United States, in Holland, E. R.. editor, Cambrian of the New World: London, Wiley-
Interscience, p. 1-79.
PALMER. A. R., 1971 b, The Cambrian of the Appalachian and eastern New England
regions, eastern United States, in Holland, E.R . editor. Cambrian of the New World:
London, Wiley-Interscience, p. 289-332.
PALMER. A. R., 1981 a, On the correlatibility of Grand Cycle tops, in Taylor, M. E . editor,
Short Papers for the Second International Symposium on the Cambrian System: United
States Geological Survey Open File Report 81-743, p. 156-157.
PALMER, A. R., 1981 b, Subdivision of the Sauk sequence. in Taylor, M. E., editor, Short
Papers for the Second International Symposium on the Cambrian System: United States
Geological Survey Open File Report 81-743, p. 160-163.
274 D.A. OsJeger
PALMER, A. R., 1983, The Decade of North American Geology 1983 geologic time scale:
Geology, v. II, p. 503-504.
PALMER, A. R., 1984, The biomere problem: evolution of an idea: Journal of Paleontology,
v. 58, p. 599-611.
PITMAN, W. C. III, 1978, Relationship between eustasy and stratigraphic sequences of
passive margins: Geological Society of America Bulletin, v. 89, p. 1389-1403.
PITMAN, W. C. and GOLOVCHENKO, X., 1988, Sea-level changes and their effect on
the stratigraphy of Atlantic-type margins, in Sheridan, R. E., and Grow, J. A., editors,
The Atlantic continental margin: U.S., The Geology of North America Volume 1-2,
Geological Society of America, p. 429-436.
POOLE, E G., STEWART, J. H., PALMER, A. R., SANDBERG, C. A., MADRID, R.
J., ROSS, R. J., HINTZE, L. E, MILLER, M. M. and WRUCKE, C. T., 1992, Latest
Precambrian to latest Devonian time; Development of a continental margin: in Burchfiel,
B.C., Lipman, P.w., and Zoback, M.L., eds., The Cordilleran Orogen: Coterminous U.S.:
Boulder, Colorado, Geological Society of America, The Geology of North America, v.
G-3.
RASSETTI, FRANCO, 1965, Upper Cambrian trilobite faunas of northeastern Tennessee:
Smithsonian Miscellaneous Collections, v. 148, 140 p.
READ, J. E, 1989, Controls on evolution of Cambrian-Ordovician passive margin, U.S.
Appalachians, in Crevello, Paul, Wilson, J. L., Sarg, 1. E, and Read, J. E, editors, Controls
on Carbonate Platform and Basin Development: Society Economic Paleontologists
Mineralogists Special Publication. 44, p. 147-166.
READ, J. E and GOLDHAMMER, R. K., 1988, Use of Fischer plots to define 3rd order
sea-level curves in peritidal cyclic carbonates, Early Ordovician, Apalachians: Geology,
v. 6, p. 895-899.
READ, J. E, OSLEGER, D. A. and ELRICK, M. A., 1991, Two-dimensional modeling
of carbonate ramp sequences and component cycles: in Franseen, E. K., Watney, W.
L., Kendall, C. G. St. C., and Ross, w., editors, Sedimentary modeling: Computer
simulations and methods for improved parameter definition: Kansas Geological Survey
Bulletin 233, p. 473-488.
REYNOLDS, D. J., STECKLER, M. S. and COAKLEY, B. J., 1991, The role of the sedi-
ment load in sequence stratigraphy: The influence of flexural isostasy and compaction:
Journal of Geophysical Research, v. 96, no. B4, p. 6931-6949.
ROBISON, R. A., 1964, Upper Middle Cambrian stratigraphy of western Utah: Geological
ROSS, C. A. and ROSS, J. R. P., 1988, Late Paleozoic transgressive-regressive deposition,
in Wilgus, C., Posamentier, H., Ross, C., and Kendall, C. G. St. C., editors, Sea-Level
Changes: An Integrated Approach: Society Economic Paleontologists Mineralogists
Special Publication 42, p. 227-247
SADLER, P. M., 1981, Sediment accumulation rates and the completeness of stratigraphic
sections: Journal of Geology, v. 89, p. 569-584.
SADLER, P. M., 1994, The expected duration of upward-shallowing peritidal carbonate
cycles and their terminal hiatuses: Geological Society of America Bulletin, v. 106, p.
791-802.
SADLER, P. M., OSLEGER, D. A. and MONTANEZ, 1. P., 1993, On the labelling, length,
and objective basis of Fischer plots: Journal of Sedimentary Petrology, v. 63, p. 360-368.
SCHLAGER, w., 1981, The paradox of drowned reefs and carbonate platforms: Geological
SCHLAGER, W., 1991, Depositional bias and environmental change - important factors
in sequence stratigraphy: Sedimentary Geology, v. 70, p. 109-130.
SCHLAGER, W., 1993, Accommodation and supply - a dual control on stratigraphic

sequences: Sedimentary Geology, v. 86, p. 111-136.
SCHUCHERT, c., 1916, Correlation and chronology in geology on the basis of paleo-
geography: Geological Society of America Bulletin, v. 27, p. 491-514.
SCOTT, R. w., FROST, S. H. and SHAFFER, B. L., 1988, Early Cretaceous sea-level
curves, Gulf Coast and southeastern Arabia: in Wilgus, C. K., Posamentier, H. W.,
Ross, C. A., and Kendall, C. G. S1. c., editors, Sea-Level Changes: An Integrated
Approach: Society of Economic Paleontologists and Mineralogists Special Publication
42, p. 275-284.
SHAW, A. B., 1964, Time in Stratigraphy, New York, McGraw-Hili, 365 p.
SLOSS, L. L., 1963, Sequences in the cratonic interior of North America: Geological
STECKLER, M. S. and WATTS, A. B., 1978, Subsidence of the Atlantic-type continental
margin off New York: Earth and Planetary Science Letters, v. 41, p. 1-13.
STEWART, J. H. and SUCZEK, C. A., 1977, Cambrian and latest Precambrian paleo-
geography and tectonics in the western United States: in Stewart, 1. H., Stevens, C. H.,
and Fritsche, A. E., editors, Paleogeography of western United States: Society of Eco-
nomic Paleontologists and Mineralogists, Pacific Section, Pacific Coast Paleogeography
Symposium, 1st, p. 1-18.
STITT, J. H., 1971, Late Cambrian and earliest Ordovician trilobites, Timbered Hills and
Lower Arbuckle Groups, western Arbuckle Mountains, Murray County, Oklahoma:
Oklahoma Geological Survey Bulletin 110, 83 p.
STITT, J. H., 1977, Late Cambrian and earliest Ordovician trilobites, Wichita Mountains
area, Oklahoma: Oklahoma Geological Survey Bulletin 124,79 p.
STITT, J. H., BARNES, V. E., MILLER, J. E and TAYLOR, 1. E, 1982, Cambrian and
lowest Ordovician stratigraphy and biostratigraphy of southern Oklahoma and central
Texas, in Taylor, M.E., and Palmer, A.R., editors, Guidebook for Field Trip 3, 2nd
International Symposium on the Cambrian System, p. 1-56.
SUESS, E., 1906, The Face of the Earth, v. 2, Oxford, Clarendon Press.
SUNDBERG, EA., ms, 1990, Morphological Diversification of the Ptychopariid Trilobites
in the Marjumiid Biomere (Middle to Upper Cambrian): Ph.D. dissertation, Virginia
Polytechnic Institute, Blacksburg, Virginia, 425 p.
TAYLOR, M. E. and MILLER, J. E, 1981, Upper Cambrian and lower Ordovician stratig-
raphy and biostratigraphy, southern House Range, Utah, in Taylor, M. E., and Palmer,
A. R., editors, Cambrian stratigraphy and paleontology of the Great Basin and vicinity,
western United States: Guidebook for Field Trip 1, 2nd International Symposium on the
Cambrian System, p. 73-77.
VAIL, P. R., MITCHUM, R. M. and THOMPSON, s.m, 1977, Seismic stratigraphy and
global changes of sea-level, Part 4; Global cycles of relative changes of sea-level, in
Payton, C. E., editor, Seismic stratigraphy - applications to hydrocarbon exploration:
American Association Petroleum Geologists Memoir 26, p. 83-97.
VAN DER VOO, R., 1988, Paleozoic paleogeography of North America, Gondwana, and
intervening displaced terranes: Comparisons of paleomagnetism with paleoclimatology
and biogeographical patterns: Geological Society of America Bulletin, v. 100, p. 311-
324.
VAN HINTE, J. E., 1978, Geohistory analysis - application of micropaleontology in
exploration geology: American Association of Petroleum Geologists Bulletin, v. 62,
p.201-222.
VAN WAGONER, J. C., MITCHUM, R. M., POSAMENTIER, H. W. and VAIL, P.
R., 1987, The key definitions of sequence stratigraphy, in Bally, A. W., editor, Atlas
276 D.A. OsJeger
of Seismic Stratigraphy, v. 1: American Association Petroleum Geologists Studies in

Geology 27, p. 11-14.
WESTROP, S. R., 1989, Facies anatomy of an Upper Cambrian grand cycle: Bison Creek
and Mistaya formations, southern Alberta: Canadian Journal of Earth Science, v. 26, p.
2292-2304.
WILSON, J. L., 1975, Carbonate Facies in Geologic History: New York, Springer-Verlag,
471 p.
WITZKE, B., 1990, Palaeoclimatic constraints for Palaeozoic palaeolatitudes of Laurentia
and Euramerica: in McKerrow, W.S., and Scotese, c.R., eds., Palaeozoic Palaeogeog-
raphy and Biogeography, Geological Society Memoir No. l2, p. 57-73.
CHAPTER 10
Sequence-Stratigraphy of
Early-Middle Cambrian Grand Cycles
in the Carrara Formation, Southwest
Basin and Range, California and
Nevada
Roy D. Adams
ABSTRACT: The Early to Middle Cambrian Carrara Formation includes two partial
third-order sequences and three complete third-order sequences. The first partial se-
quence, composed of the basal two Carrara Formation members, is completed by the
addition of the underlying Early Cambrian Zabriskie Quartzite. The middle six Carrara
Formation members compose three complete sequences. The second partial sequence
incorporates the uppermost Carrara Formation member and is completed by the addition
of the basal part of the overlying Middle to Late Cambrian Bonanza King Formation.
These third-order sequences consist of a basal siliciclastic portion and an upper carbonate
portion, corresponding to Grand Cycles. Minimal evidence of erosion or exposure along
Carrara Formation sequence boundaries indicates that they are Type 2 unconformities.
Carrara Formation Grand Cycles allow sequence-stratigraphic modeling of a depositional
system with two sediment sources that respond differently to changes in accommodation.
Nearshore marine and non-marine, fine-grained siliciclastics prograded basinward from
the craton and were deposited in shelf-margin systems tracts or by-passed the shelf if
little to no accommodation space was available on the shelf. In either case, subsequent
siliciclastic deposition was retrogradational forming transgressive systems tracts. Silici-
clastics were distributed widely in shelf-margin and early transgressive systems tracts,
but were not able to maintain wide distribution during later transgressive and highstand
systems tracts. Subtidal and peritidal carbonates prograded craton ward from an offshore
carbonate bank and were deposited in transgressive and highstand systems tracts. The
carbonate bank experienced decreased sediment production rates during accumulation of
the shelf-margin systems tract and ever-increasing rates during accumulation of the trans-
gressive and highstand systems tracts. This interpretation of shelf-margin systems tract
siliciclastics and highstand systems tract carbonates differs from some interpretations of
Grand Cycles in the southern Canadian Rockies.
277
B. U. Haq (ed.),
278 R.D. Adams
Shelf-margin systems tracts in most (common) Grand Cycles are either very thin and
difficult to distinguish from transgressive systems tracts or are missing entirely. In
anomalous Grand Cycles, the shelf-margin systems tract is well developed with intertidal
to supratidal lithofacies. Both the common and the anomalous types of Grand Cycles are
present in the Carrara Formation.
1. Introduction
Sequence stratigraphic models, developed for depositional systems with a single

sediment source, track the interplay between sediment supply, tectonics, and eu-
stasy (Jervey, 1988; Posamentier and others, 1988; Posamentier and Vail, 1988;
Sarg, 1988; Van Wagoner and others, 1988, 1990). In contrast, Grand Cycles, large-
scale depositional couplets of siliciclastics overlain by carbonates (Aitken, 1966),
are produced by complex, large-scale depositional systems with two different sed-
iment sources. Grand Cycles are composed of two half-cycles, one dominantly
of carbonate rocks overlying another dominantly of siliciclastic rocks, developed
on a scale of tens to hundreds of meters in thickness (Aitken, 1966, 1978, 1981).
Therefore, a sequence stratigraphic model of Grand Cycle deposition must ac-
count for the influence of these dual sediment sources and their different responses
to changes in tectonics and eustasy. In the analysis of depositional systems, the
influences of tectonics and eustasy are commonly combined and are termed "ac-
commodation", i.e. the amount of vertical space available below sea level in which
sediments may accumulate(Jervey, 1988; Posamentier and others, 1988; Posamen-
tier and Vail, 1988). Lateral and vertical arrangement of facies and cycles reflects
the interplay through time between rate of change in accommodation space and
varied rates of sediment supply.
This paper provides a detailed description of lateral and vertical distribu-
tion of lithofacies groups within Carrara Formation Grand Cycles and presents
a sequence-stratigraphic interpretation in terms of the interplay between rate of
change in accommodation space and varied rates of sediment supply from two
sediment sources. The model relates accommodation changes and half-cycles in
similar fashion to older pre-sequence-stratigraphy models (Aitken, 1966, 1978;
Halley, 1974; Palmer and Halley, 1979; Mount and Rowland, 1981). It differs
from these older models and from most previous attempts (Grotzinger, 1986a,b;
Chow and James, 1987; Bond and others, 1988, 1989; Aitken, 1989; Demicco
and Spencer, 1990; Bond and Kominz, 1991; Demicco and others, 1991; Mount
and others, 1991; Cowen and James, 1993) to reconcile Grand Cycles to sequence
stratigraphy in several aspects: 1) a Grand Cycle is a single sequence and the
boundary between two successive Grand Cycles is a sequence boundary; 2) the
initial spread of siliciclastics is related to the decrease in accommodation during
and immediately after formation of the sequence boundary; 3) in a common Grand
Cycle the siliciclastic half-cycle is mostly in the transgressive systems tract (TST
) and the shelf-margin systems tract (SMST ) is too thin to be differentiated; 4)
in anomalous Grand Cycles a thick SMST is recognizable by the abundance of
supratidal and nonmarine facies in the lower part of the siliciclastic half-cycle
Early-Middle Cambrian Grand Cycles in California and Nevada 279
and the TST is in the upper part of the siliciclastic half-cycle; 5) the carbonate
half-cycle is predominantly in the highstand systems tract (HST ).
2. Previous Work and Regional Geology
The Carrara Formation, of Early to Middle Cambrian age (Barnes and Palmer,
1961; Barnes and others, 1962; Palmer and Campbell, 1976; Palmer and Halley,
1979), crops out in the Basin and Range province in southeastern California and
southern Nevada, within the area shown in Fig. 1. A 300-500 m thick succession
of mixed nearshore marine carbonate and siliciclastic shelf sediments composes
the Carrara Formation, whereas to the northwest and north correlative formations
are predominantly carbonates (Stewart, 1970; Albers and Stewart, 1972; Palmer
and Halley, 1979; Palmer and Nelson, 1981; Palmer and Rowland, 1989; Signor
and Mount, 1989). To the east and south correlative formations are either of
mixed composition or are predominantly siliciclastic (Longwell and others, 1965;
Stewart, 1970; Tschanz and Pampeyan, 1970; Palmer and Halley, 1979; Palmer
and Rowland, 1989). Correlations are summarized in Fig. 2.
Cornwall and Kleinhampl (1961) named the formation and designated the
type section on Bare Mountain near the ghost town of Carrara, Nevada. The
general lithology, age, extent of the formation, and regional correlations are de-
scribed by previous workers (Barnes and others, 1962; Stewart, 1965; Stewart and
Barnes 1966; Barnes and Christiansen, 1967; Stewart, 1970; Palmer and Halley,
1979). Trilobite biostratigraphy was used to refine regional correlations (Barnes
and Palmer, 1961; Palmer and Halley, 1979). Important studies of Carrara For-
mation sedimentology, stratigraphy, and biostratigraphy have been conducted by
Bates (1965), Halley (1974), and Palmer and Halley (1979). The Carrara Formation
was described by Palmer and Halley (1979) as a collection of Grand Cycles anal-
ogous to Grand Cycles of the southern Canadian Rockies (Aitken, 1966, 1978).
Members defined by Halley (1974) and Palmer and Halley (1979), are used in
this paper and are illustrated in the generalized Carrara Formation stratigraphic
sections shown in Fig. 2.
Siliciclastics in Carrara Formation Grand Cycles were derived from the craton
to the east and south of the outcrop area (Stewart, 1970; Stewart and Suczek,
1977; Stewart, 1991). A major recurring carbonate bank formed an in situ source
of carbonate sediments and was located along the edge of the continental margin
several tens to a few hundred kilometers from the mainland. The bank is represented
by part of the Carrara Formation (Palmer and Halley, 1979; Adams, 1993) and
by age-equivalent carbonate strata to the northwest (Stewart and Suczek, 1977;
Palmer and Halley, 1979). Between bank and mainland was a lagoon that received
sediments from both sources and also produced carbonate sediments in situ. Craton-
sourced siliciclastics tended to accumulate in and prograde across the lagoon to the
carbonate bank. Periodically the bank was inundated by siliciclastics, terminating
carbonate deposition and resulting in member-scale interbedding of siliciclastics
and carbonates that led Palmer and Halley (1979) to apply Aitken's (1966) Grand
Cycle model to the Carrara Formation.
280 R.D. Adams
r "-, tl T
,
, .3 E eGR
" ,,18
LeC""',,," ill NEVADA
118" "1.&.1
',' + +
1170 116" I 1150
f-
() '" fl
37",
LIncoln County
~ .~C 8,- - ""Cj;,kCounty -
"'I;. '" e
1- " SoH DR
0
:"1
"-
CALIFORNIA
eEC",,- If, LV

i
N,
<:::
10
~
'
EM -",MH '5
N~e ",- I
of
FM
LASi+ e
VEGAS
I
+ -< e MR+,-, 36",
CS, N "
_ I~y~'t 2~ _e_ _ "-.
San Bernardino County
____ I WP_
",,-
o km 80-L
J
A. MEASURED SECTION LOCATIONS
NV

IfIl oGR
I UT
1---
NQIfS: 1. Base map locadona from the
palinspasUc r... oraUon 01 Levy and ChrlsUe-Blick
(198D). St8to boundaries Ire not restored.
DR
2. Palmer and Hanoy (1979) locollon. 'PH' and o
'RS(Sr Ire Dqulvalanl to thl. paper"s locetlona _ _- LV_ _,FM.I
'MH' ond 'MR' ro.pecU ly,
3. Tho Chappo SprIng. (CS) secUon.localed 3 km
north of ,h. southern R.otlng Spring Ronge (SR)
.Hellon & shown on Figures 5 Bnd 6, I. not placed
on the map. due to ecale limitations. CA
t
B. CROSS SECTION LOCATIONS N
Fig. 1. A) Location map showing unrestored positions of stratigraphic columns measured in

this study (solid circles) and four stratigraphic columns measured by Halley (1974) and Palmer
and Halley (1979) (open squares), B) Location map based on the palinspastic map of Levy and
Christie-Blick (1989) showing restored positions of stratigraphic columns and cross sections. The
depositional-dip cross sections extend west to east, from the Last Chance Range (downdip) to
the Frenchman Mountains (updip). The depositional-strike cross sections extend north to south,
from Echo Canyon (proximal to the carbonate bank) to Winters Pass (distal to carbonate bank).
Abbreviations: BR = Belted Range; CS = Chappo Springs; DR Desert Range; EC Echo Canyon; = =
EM = Eagle Mountain; FM = Frenchman Mountain; GR = Groom Range; LC = Last Chance Range;
LV = Las Vegas Range; MH = Montgomery Hills; MR = Middle Resting Springs Range; N Nopah =
Range; NR =Northern Resting Springs Range; SH =Striped Hills; SR = Southern Resting Springs
Range; TC = Titus and Titanothere Canyons; WP = Winters Pass.
A. GENERALIZED STRATIGRAPHIC COLUMN AND GRAND CYCLES.
WH~~;O~YO SOUTHWESTERN NEVADA. DEATH VALLEY REGION ~~N~~N

Upi'. Tamarack Fm Nopah Formallon
5
amb Fm
'1
Mow
Emigrant 8andtd Mountain Membw Jangle
Bonanza King FormatIOn- ~-- - -- -- - - -- -- - ---
Limestone
FOl'TNllion e Llmlone
Mdl. Papoou U.k. M.".~r .!! Member
Camb
U
Mono"
Fm.
I
.
.Ii
E
U Pahrump Hilla
Limestone .!! Sna .. Mambtlr
.g'"
"0
:2 z
tl
:Ii
AedP...
..
a: liMestone
~ """bo<
Wood Canyon a:
Formation e
a: P').,.mldSMIe
It
Camplto a: M_
Formation e
<.>
Deep Spring Gold Ace
Fonnatlon e lhn..toM
~ Me..... 2
..... Stirling Quartzite
E
13
11
Member
Comb Wyman
Formation Mombo<
(ModlUtd from Palmer and Hillley, 1979; Palmer and Nelson, 1981) ~~.M:=
zabriskie Quartzite
B. REGIONAL CORRELATIONS AND TRILOBITE BIOZONES.

Trilobite Southem Nevada Marble Inyo Mountains, Esmeralda Highland Grand Canyon.
11 Mountains,
'C Biozones and Southeastem California County. Nevada Range, Nevada Arizona
aI California f.~I~~~~~~ (to northwest) (10 northwest) (to northeast) (to east)
Glosso- Desert Range Chisholm
Limestone
I pleura Member Shale
Jangle
Limestone Lyndon
Member Limestone
e Albertell
~E
.
Emigrant
Pahrump Hills
..
Monola Formation Bright Angel
Shale Member Cadiz
u Formation (part) Shale
z Formation
~~e
:Ii e
::E Red Pass
a: Limeslone
0 Member
Plaglura "-
Pollella ~
e Pyramid Shale
a:
a: Member Pioche Shale
e -7-
u f--7-
..
e GoldAco
Llmeslone
Chambless Mule Spring
..
'C Limestone Mule Spring
Sl Bonn/. Member Limestone
E OIenelius Limestone Tapeats
0
~
Ei!':...~~'. Sandstone
hinble LImes"iii
~ Member Latham Shale
-' Saline Valley Saline Valley
Eagle Mountain
Shale Member fu~~~~~~~ fu~~~:1:~
(Modified from Palmer and Halley, 1979)
Fig. 2. Precambrian through Cambrian stratigraphy of the Death Valley region is presented with
generalized regional correlations. (A) Top: A generalized uppermost Precambrian through Cambrian
stratigraphic column in the Death VaHey region showing correlative formations to northwest and the
east (from Palmer and Nelson, 1981). The White-lnyo region lies northwest of the limit of the Carrara
Formation, and Frenchman Mountain lies at the eastern limit. At Frenchman Mountain, some workers
utilize Grand Canyon stratigraphic terminology as shown, however Death Valley terminology is used
in this study. The five Grand Cycles and their constituent members and formations are shown on the
right. (8) Bottom: Regional stratigraphic column with trilobite biozones and members of the Carrara
Formation showing correlative formations to north, south, east, and west (from Palmer and Halley,
1979).
282 R.D. Adams
Siliciclastic sediments, derived from the craton, range from shales to pebbly
sandstones in the Carrara Formation and the underlying Zabriskie Quartzite (Prave,
1992). Siliciclastics generally become progressively finer to the west and north and,
in the Carrara Formation, sandstone is subordinate to shale and siliceous mudstone.
Most siliciclastics were deposited in shallow, nearshore environments, probably
under conditions of low day-to-day wave energy, though some were deposited in a
nonmarine setting, either in a supratidal environment or on a coastal plain. Marine
carbonate sediments were mostly produced in relatively shallow subtidal depths,
above storm wave base. Intertidal carbonates are subordinate in volume to subtidal
lithofacies.
The Sauk transgression, a major eustatic rise, occurred from the Late Precam-
brian through the Cambrian and into the Ordovician (Sloss, 1963), and is reflected
in long-term changes within the five Grand Cycles of the Carrara Formation.
There is a general and progressive decrease upwards through the five successive
Grand Cycles in the percentage of coarse-grained siliciclastics, the proportion of
nonmarine siliciclastics, and the maximum distance of siliciclastic progradation.
In addition, throughout the study area there is an increase in the thickness and
craton ward extent of each successive carbonate half-cycle.
In Carrara Formation Grand Cycles, the effects of tectonics and eustatic fluctu-
ations cannot be isolated and tested independently. Carrara Formation deposition
took place on a mature passive margin (Stewart, 1970; Stewart and Suczek, 1977;
Stewart, 1991). Geophysical modeling indicated that subsidence followed an ex-
ponential decay path which is characteristic of thermal subsidence rather than
episodic subsidence (Bond and Kominz, 1984; Bond and others, 1988, 1989; Levy
and Christie-Blick, 1991). Both thermal subsidence and the Sauk transgression
occur over longer time periods than required to cause Grand Cycle deposition,
therefore a shorter duration tectonic and/or eustatic mechanism is needed. Pre-
vious sedimentologic studies have suggested the main mechanism to determine
position and geometry of Grand Cycles was shorter-duration eustatic fluctuation
(Mount and Rowland, 1981; Grotzinger, 1986a, 1986b; Chow and James, 1987;
Aitken, 1989; Mount and others, 1991). However, geophysical modeling by Bond
and others (1988, 1989) suggested that tectonic mechanisms could not be ruled out
even though third-order eustatic fluctuations might be the most plausible mecha-
nism.
Sequence-stratigraphic terminology conforms to usage of Mitchum and others
(1977) and Van Wagoner and others (1987, 1988, 1990). "Parasequence" is equiv-
alent to a "shallowing-upwards cycle" in many carbonate platforms, e.g., James
(1984), as well as the "Punctuated Aggradational Cycle" (PAC) of Goodwin and
Anderson (1985). In this paper "parasequence" is synonymous with "meter-scale
shallowing-upwards cycle". Parasequence terminology is preferred because the
accompanying concepts of sequence stratigraphy provide a larger framework in
which to place observations. The term "Grand Cycle" is used in the sense of Aitken
(1966, 1978) for couplets tens to hundreds of meters in thickness composed of
dominantly siliciclastic sediments overlain by dominantly carbonate sediments. A
sharp contact occurs at the base of the shale half-cycle and a gradational contact
at the base of a carbonate half-cycle. In the Carrara Formation a Grand Cycle

corresponds to a sequence.
3. Lithofacies and Meter-Scale Cycles/Parasequences
Subtidal to intertidal siliciclastic and carbonate lithofacies as well as some suprati-

dal siliciclastic lithofacies occur in the Carrara Formation. Carrara Formation
lithofacies are summarized in Table 1. Comprehensive descriptions are reported in
Adams (1993).
"Meter-scale" shoaling cycles vary in thickness from less than 1 m to about
20 m. It is possible that thicker cycles are actually composites of several smaller,
but indistinguishable, cycles. Twenty-three generalized shoaling cycles were rec-
ognized in the stratigraphic sections. These generalized cycles emphasize the wide
variation of vertical succession of facies and lithologies present in the Carrara For-
mation and are reported in detail in Adams (1993). Boundaries between subtidal,
intertidal, and supratidal facies are normally gradational within a cycle. Bound-
aries between cycles are sharp, record a small-scale relative deepening, and are
parasequence boundaries.
Recognition of meter-scale shallowing-upward cycles containing both car-
bonate and siliciclastic sediments demonstrates coeval deposition of these two
sediments. This is suppolted by regional correlations (Figs. 3 through 6) which
document a widespread lagoonal zone of mixed lithofacies between carbonate bank
and mainland. Mount (1984) listed four possible mixing processes on carbonate-
bank-rimmed continental margins with siliciclastics shed from the craton. Three
of the four processes affected Carrara Formation deposition. Emplacement of thin
layers of ooids onto siliciclastic intertidal fiats is interpreted as evidence for "punc-
tuated mixing" driven by storm events. "Facies mixing", driven by diffusion of
sediments into adjacent lithofacies belts, was postulated by Mount (1984) to be
present across relatively narrow zones. In the Carrara Formation, given the width
of the mixed-lithofacies intervals between clean carbonate intervals and clean sili-
ciclastic intervals (Figs. 3 through 6), this mixing zone was relatively broad, on
the order of 100-200 km down depositional dip. "In-situ mixing", caused by storm
winnowing of fine-grained siliciclastics leaving a coarse lag of death assemblages
of calcareous organisms, was commonplace and is recorded by numerous fossil-
iferous packstone/grainstone event beds in thick intervals of siliceous shale and
mudstone. In some Carrara Formation parasequences, mixing can also be shown to
have taken place at the scale of bedsets, beds, and even laminae. Most commonly
intimate admixtures are found between quartz sandstone/siltstone, and carbonate
arenite/siltite. However, other carbonate lithofacies (e.g., ooid grainstone) also
are associated with siliceous beds. The Red Pass Member oolites (Fig. 4) are
commonly interbedded with shale to sandstone.
Groups of meter-scale cycles may form larger overall shoaling cycles wherein
each successive cycle tends to shoal to shallower depths or to have a greater pro-
portion of shallower facies than does the preceding cycle. The Jangle Limestone
Member (Fig. 5) is made up of two of these larger shoaling parasequence sets;
meter-scale cycles lower in the parasequence set tend to have more subtidal facies,
TABLEl IV
00
Lithofacies in the Canara formation ~
LEGEND:
Grain types: Sedimentary structures: Sedimentary structures:
o ooids H homogenous beds (no structures) -....,."r scoured surfaces
o peloids --., current ripples "VVVV" hardgrounds
It) undifferentiated ooids & peloids ~ climbing current ripples -U- load structures
111 oncoids ,,-- flaser ripples N contorted beds
@J general fossils ~ wave ripples 'UI burrows
01 trilobites ~W climbing wave ripples 'UI~ burrowed-to-bioturbated
ti. hyolithoids
~, combined flow ripples ~ bioturbated
:lv'lithoclasts
.7Z planar cross beds ~ soil-churned
= vertical trend ~ trough cross beds 4 S mudcracks
changing In A:? hummocky cross beds / \ tepees
f upward direction == horizontal planar laminae <lI:> fenestrae
~ wavy parallel laminae % microfauRing
~ wavy nonparallel laminae pyrite
Carbonate sedimentary structures and textures: ~W wrinkly laminae
= A planar microbial laminae + gutter casts
~A wavy microbial laminae
00 R well-layered ribbon rock texture Abbreviations:
00 NR nodular ribbon rock tecture bndst =boundstone rndst =mudstone wkst =wackestone pkst = packstone grst = grainstone
00 BR bioturbated ribbon rock texture wI =with % =per cent sh =shale sts siltile or siltstone
=
IA stromatolitic boundstone ss =sandstone aren =arenite Is =limestone dol = dolostone or dol- or dolo- or dolomitic
thrombolitic boundstone
ITt VF =very fine F =fine C =coarse VC = very coarse M = medium
f[)I digitate miocrobial boundstone m =meter cm =centimeter mm =millimeter qtz = quartz crsn = coarsen
incr =increase decr =decrease =
ps =parasequence talc altered sediments
ul, u2, u3 etc. = informal units in shallowing-upwards cycles. skol Skolithos burrows
=
CDAS = calcarenite, calcisiltite, dolarenite, and dolosiltite
:::0
~
:>
fr
s
fIl
TABLE 1
Lithofacies in the Carrara fonnation
~
:::L
'<:
VERTICAL TRENDS (In an
UTHOFACIES UTHOLOGIES & GRAIN TYPE SEDIMENTARY STRUCTURES BED THICKNESS (Range only upward direction) ASSOCIATED UTHOFACIES k
SUBTIDAL CARBONATE FACIES ~
Carbonate mud. Bulbous heads, squat Individual heads & columns: heights 5-
120 cm & widths 5 - 200 cm. Planoconvex Calcitic mdst, ribbon rock, Q
t"V BNDST columns, planoconvex lens wi
lens: ~75 cm high & ~3 m wide. Tabular dolsts wkst, VF arenite,@/0
convex side down, tabular
pklgrst, 0 pklgrst, dolsts and
(Uncommon.) units: 25 -190 em high & > severall0's of
s0-
units, irregular mounds and a
m wide. Irregular mounds: ,,280 cm high & dolSts~ may be found :::!.
reefal complex. Il:l
> 3 m widths. Bioherm complex: ~ 6.3 m between heads.
high & ~ 100 -150 m wide.
a=
Includes arenitlc calcitic wkst V' Doists-filled. 'U'-~ Bed Thickness: $ 1 - 10 cm. None. @ pkst, dolsts. P:l
& silty mdsl =
MDIWKST H 0..
@fragments. Q --
(Less common.) Q
0$ 1/2 mm. 0 $ 1/8 mm. t")
"""" $lxl em. ~

en
S 1/2 - 4 cm, generally .........--, fi)) Bed Thickness: 1- 40 cm. s
WKlPKl spheroidal, but also
V' - No consistent changes. Dolsts, mdst, & tRt & rn
heads.
GRST cylindrical, ellipsoidal, & Q
Some siliceous mdsl ....
discoidal. @fragments.
(Uncommon.) 0$1/2-1 mm. "'$lX5cm. ~
VC & smaller qtz sand grains :3t;j.
Muddy & sandy matrix, dolsts Bed Thickness: $ 1- 20 cm. None. All 3 varieties of ribbon rock, Il:l
"""" PKlGRST matrix, mdst matrix. -- V' ->.
Is & dol stslarenite, @;pklgrsl 0..
=
"""" II & dol sillite/arenite, $ Generally beds are either
(Uncommon.) 112xl/2 mm - 8xl0 cm. composed of small clasts ($
0$1/2mm. $3cm. 112 cm) in lenticular beds $10
~
~$O-1/2mm. cm thick, or of larger clasts in ~
tabular beds 10 - 20 cm thick. fr
@WKlPKI @fragments of trilobites, V' = H $35 cm.
I...CJ Bed Thickness: 1 - 35 cm. Generally fine & thin, wi an Dolsts wlin interlaminae,
echinoids, hyolithoids, increase in % dolsts. May interbeds, & burrow-fills.
GRST ~~.........--, Some siliceous mdsl Often
(Common. brachiopods. extend across several
0$ 1/4 - 2 mm. $ 1/2 - 3 cm. ""VVV" at base & internal to occurs as $1-10 cm thick,
Also may unit adjacent bed sets. lenticular interbeds in
extend into very """" $1 112 x 20 cm. "VVVV rare, at top of bedsets. siliceous mdst & sh, mostly wi
lowermost VF - C qtz sand grains Q fragments. Interbedded WI
intertidal.) Unidentified coated grains. ~ CII:> in one bedse!. ss, sts, 0 pklgrs!.
- - - - - - - - _.. _ - - - - - ~-
tv
00
VI
TABLE 1 Continued tv
00
Lithofacies in the Carrara fonnation 0'1

LITHOFACIES ITHOLOGIES & GRAIN TYPES SEDIMENTARY STRUCTURES BED THICKNESS (Range only upward direction) ASSOCIATED LITHOFACIES
SUBTIDAL CARBONATE FACIES
o GRST o,,1/8-2mm, wi one \..(; 5 - 50 cm. 7Z' H Bed Thickness: ,,1 - 50 em.
= Increase or decrease in % dol- Dolsts @)Icoated grain grst,
occu renee of 8 mm. st~ain size, @) fragments, & Is arenite.
(Common. @) fragments, some coated. ~, ~ .--...>. 'Ul N
Also may & .
"'VV'" both at base &
extend into very Ay ~ lxl mm -" 3xl0 cm. internally. en::> very rare.
lowermost ,,2 cm, rare.
intertidaL) O/@sl/8mm.
Liths: mdst, or 0 grst, wkst, 'Ul spar-filled & dolsts-filled. Bed Thickness: Ls s 1 -10 cm Decrease % dolsts, rarely Dolsts in all beds wlin
COBR rare mdlWkst, @)pklgrst, Dol" 1 - 6 cm. increase % dolsts. intenaminae, burrow-fills, &
o grst, calci-sts, calcarenite. 'Ul-~ ~ H irregular, discontinuous
(Less common.) OSl/8mm. .--...>. & == uncommon. patches. Some wkstto grst of
s 1/8 - 3/4 mm. ,,2 cm.
o o @)Ay , and some rn .
@)oOIfragments. ~6. rare.
Liths: pkst, grst, aren/sts Bed Thickness: Ls" 1 - 30 cm Decrease % dolsts in ~ 1/2 of Dolsts wlin interlaminae,
CONR (sometimes mixed wi qtz ss & 'Ul 'Ul-~ ~ H #? bedsets, sometimes increase burrow-fillS, & irregular,
Dol S 1 - 6 em.
(Common. sts), wi wkst, mdst (orOgrst?). == ~, Internal"VVVV? % dolsts, rarely increase & discontinuous patches.
Also may @)fragments. 0 s 1/8 - 1 mm . .--...>. ~ r - ~ then decrease % or vice verss. Uncommon occurrences of
extend into very ~: 1/8 mm. O/@ S 1/4 mm. Some beds thin & fine., some
beds coarsen & thicken, @) or Jlf wklpkst, 0 or ~ ,
lowermost S 1.5xl.5 -lx6 cm. 001 \..(; S 30 cm. ~ (j)\ rare.
others increase or decrease pklgrst, VF-Fqtz ss, carbonate
intertidaL) S 4 cm. 6. rare. I\.rvery rare. very rare.
err::> amount of burrowing. arenite, & siliceous mdsl. rn
Liths: grst & arenlsts, rare Bed Thickness: Ls S 1/2 -15
COR 'Ul = ~''--'''>' H Sometimes decrease % dolsts. Dolsts in all beds wlin
wkst, qtz ss & s cm. Dol S 0.5 - 8 cm.
(Common. 0118 -112 mm. fragments. \..(;S20cm. iii Rarely fines up. intenaminse, burrow-fillS, &
-u- Rare channels S 20 cm high X
Also may ~S 2 mm - 2x15 em. irregular, discontinuous
2-t0-20 m wide.
extend into Some unidentified coated "'VV'" 'Ul-~ uncommon. patches. Some arenite, @)
lowermost grains. en::>? & 4' both uncommon. wkst, ~@)0 pklgrst, 0
intertidaL) grsl.
S2cm. O? 001 ~rare. Graded beds.
CDAS Liths: Generally sillite is = ~, .--...>. '--"'>"'?PBed Thickness: S 1 - 45 cm. Decrease amount of Calc-arenite & calcists &
LITHOFACIES dolomitic & arenite is calcitic. , - - flaser, wavy & lenticular. siliciclastics andlor dolsts. dolsts often together in one
(Common. Unidentified carbonate grains. \..(;Sc~ ~ ~~-u- Thickens and coarsens. bedsel. Often interbedded wi
Also may Qtz silt & VF sand. Overall change from'--"'>' to ~ ss &0 grsl. Some@)& ::0
0s 1/8 -1/4 mm. O? S 1/4 mm ~ 5-45cm. #?rare.N
extend into ~ in some beds. ..., pk/grst, calcitic mdst, ~
~ at base of bedsets.
lowermost ~Il cm.@) 001 fragments. silty, siliceous mdst & sandy
'Ul-~ en::> rare & at top of ~
intertidal) rare, S 1/2xl0 cm. bedsets. mdst, & sts.
~
fI.l
TABLE 1 Continued
Lithofacies in the Carrara fonnation ~
f
LITHOFACIES ITHOLOGIES & GRAIN TYPES SEDIMENTARY STRUCTURES BED THICKNESS (Ranga only) ASSOCIATED LITHOFACIES
upward direction) ~
INTERTIDAL CARBONATE FACIES
e::
CD
Orange dolsts, dol-mdsL Gray =A~A Bulbous (1x1 -10x10 cm.) to From broad up to bulbous. g
-- /\1 very broad (HxW .1x5 -10x55 = A & ~Adolsts.
~BNDST Is arenite, sts, mdst. From~upto = A . From S
(Uncommon.
CD::> S 1x1 mm. N cm) heads. Intermediate Is up to dol. Decrease up in ~ occur twice with fT\.
F qtz sand grains. cr
Lowest ""VV"" at base of bedsets. shapes (HxW) 4x10 - 25x40 size, e.g. 10x10 cm to 2x2 cm. :=!.
0? ~ often occur between Il>
intertidal.) cm. Synoptic relief generally
bedsets of siliceous shale to i:l
less than half the height.
sh to sis. C)
01/S - 114 mm. None. Occasional plelgrst, dolsts. r:!
CD::> -- VI H Bed Thickness: s 1 em. i:l
CALCAREOUS @.I fragments. 0. Cl...
MDIWKST
(Uncommon. " ' s 1112x2 cm. .Q
Lowest JZ1 S 1 cm.
intertidal.) ~
C"I>
Generally orange, but may be ->. -'"', = ~W Bed Thickness: mm-Iaminated Calcitic mdst, calcarenite, S
DOLSTS, red, tan, cream. brown, green. Increase in % = in some calcists, & some
~ "VVVV"? "vvV" VI t020cm.
CALCARENITE Unidentified carbonate grains. bedsets, rarely decrease % of
""\.r loaded into UnderlYin~ @)l0/0 wlelpkst,0/coated
g.....
& Unidentified coated grains. CD::>. grain pk/grst, and occasional
CALCISILTITE Qtz silt & VF sand.
lithologies. N H VI-
~-like microbial laminae, stromatolites. ~
(Common. Low O? IlJ s 1 em. @.I fragments. :3
intertidal.) & incipient ~. = A
o it,of dolsls, s 112x10 cm. ~.
# CD::> ~ A rare.
Sl)
. . , of dolsts & = A mdst, CD::> S2mm. Bed Thickness: 2 - 215 cm. Increase in % dolsts or in % Orange dolsts, & rarely i:l
CALCAREOUS S5X5 mm. dolarenite or JZ1 wk/pkst. Cl...
= A discontinuous & = A in some beds.
CD::> MOST
@.I fragments rare. indistinct.
(Less common. ~ ===s 1-5cm.
~
Low to mid
JZ1 rare & s 1em. ~
intertidal.) O? ->.? VI? # rare.
g.
Unidentified carbonate grains. =A~A= #/\ Bed Thickness: mm-Iaminated Increase in % calcarenite, sts, Calcitic mdst, calcarenite,
ORANGE calcists, & some
Qtz silt & VF sand. CD::> ""VV"" ttl VI -;t:. to 15cm. ormdst.
DOLSTS @.I/0/0 wkJpkst,0/coated
~ -like microbial laminae, & grain pkJgrst, and occasional
(Less common. ..,of dolsts, s 112x10 cm.
Mid to high incipient tftt, S 2x2 cm. stromatolites.
o O? @.I fragmenls. ".. .... -'"', uncommon.
intertidal.)
->. uncommon.
N
00
-...l
TABLE 1 Continued N
00
Lithofacies in the Carrara fonnation 00

LITHOFACIES LITHOLOGIES & GRAIN TYPE SEDIMENTARY STRUCTURES BED THICKNESS (Range only upward direction) ASSOCIATED LITHOFACIES
SUBTIDAL SILICICLASTIC FACIES

Fissile sh, silty sh, silty mdst, Bed Thickness: mm~aminated Coarsen & thicken up, often
often only horizontal.
7.P, @Jwk/pkigrst, wklpkst, M-
SHlMDST all often chloritic (green, to 15 cm. changing to sts &lor ss. VF ss, calc- & dolarenite,
greenish-tan, yellowish-green, 7.P-~ ~ N Sometimes increase the
(Most common khaki-green, tan, yellowish-tan dolsts.
amount of bioturbation
siliciclastic gray, light gray, purple, purple == ,...>. ~ upward, or fine up from mdst
lithofacies.) brown). Finer lithologies often + filled wi VF qtz ss; ,; to sh.
cleaved. 2x2 em - 10x30 cm.
Muddy SIS, muddy ss, M-VF ss ---- ~'~5-25cm. Bed Thickness: ,; 1/2 - 40 em. Coarsen & thicken up most
STs/SS @J pkigrst, calcarenite,
(green, khaki-green, tan, tan- common, but sometimes fine dolsts, silly siliceous mdst.
,...>., ,...>. ~W == & thin up. Bioturbation may
orange, brown, light gray,
(Common.) white, pink). Muddy ~5-20em. decrease or increase up.
lithologies often chloritic. -vvv..r at base of bedsets.
@) Q O? .., ,;5x10em. 7.P 7.P-~ ~
INTERTIDAL SILICICLASTIC FACIES
Fissile sh, mdst, silty mdst ~, ,...>. # = Bed Thickness: ,; 1/2 - 2 cm. None. F-VF ss & calcarenite
SHIMDST (red, pink, orange, tan, brown, interbeds.
mostly horizontal & ,; 1
7.P,
greenish-tan, green). Often
mm diameter.
(Less common. cleaved.
7.P-~
OIz sts & VF - F ss (pink, = ~, ,...>., ,...>. Bed Thickness: ,; 1/2 - 25 cm. None. Silly siliceous mdst, calc- &
STSISS orange, tan, brown, khaki- dolarenite/sillite,0 grst in
green, gray,). ~W H 7.P 7.P-~ interlaminae & interbeds.
(Common.)
\.T CID / \ #
SUPRATIDAL SILICICLASTIC FACIES

Blocky, non-fissile sh, slightly Bed Thickness: ,; 1/2 cm. May alternate soil-churned Doists, ss interbeds.
# 1\ H ~
SHlMDST silty sh, and not very silty ~S intervals wi
=='lJI?#I\'lJI-l"
mudstone (red, red-purple, == & crude, coarse ==. intervals vertically, on a l-to- ~
(Uncommon.) purple, gray-purple). few meter scale. Often a ~S__ Gradational contact.
7.P? 7.P-~ uniform soil-churned texture >
throughout a bedset. ~S Sharp contact.
I ~
til
whereas meter-scale cycles in the upper part tend to have more intertidal facies
(Figs. 3 through 6). No deepening-upward meter-scale cycles were found in the
Carrara Formation. Where deepening of the section is recorded, it is due to ret-
rogradational stacking of shoaling cycles. Successive cycles do not shoal to as
shallow a depth as did the preceding cycle.
4. Correlation Strategy
Correlations used in this study, and shown on Figs. 3 through 6, are not lithostrati-
graphic correlations and do not follow formation and member boundaries exactly.
Rather, correlations are guided by regionally extensive surfaces (sequence bound-
aries) that define regionally extensive depositional sequences. Within Grand Cy-
cles, lateral changes in lithofacies reflect regional lateral changes in depositional
environments. Although only sequence boundaries and gross lithofacies group
boundaries are shown on cross sections (Figs. 3 through 6), smaller-scale parase-
quence boundaries were used to guide placement of many lithofacies boundaries.
On Fig. 5, two parasequence set boundaries are marked that subdivide lithofa-
cies groups and offset lithofacies boundaries. Parasequence correlations have not
been shown because: 1) they are too numerous and close together, and at this
scale obscure the overall internal stratigraphy of Grand Cycles; and 2) laterally
discontinuous, fifth-order cycles (parasequences) reduce the confidence level of
parasequence boundary correlations given the spacing of stratigraphic sections
(See Adams [1993] and Adams and Grotzinger [submitted] for detailed parase-
quence correlations in part of the Carrara Formation).
Biostratigraphic (trilobite) correlations (Palmer and Halley, 1979) are not vi-
olated by the Grand Cycle/sequence-level stratigraphic correlations of this paper.
However, most of the correlations within Grand Cycles are below resolution of
trilobite biostratigraphy. Chronostratigraphic aspects of correlations within Grand
Cycles emphasize coeval centers of carbonate and of siliciclastic deposition with
mixed lagoonal sedimentation between.
s. Carrara Lithostratigraphy
Member Descriptions
Members of the Carrara Formation were defined as lithostratigraphic units (Halley,
1974; Palmer and Halley, 1979), with each member dominantly carbonates or sili-
ciclastics. These member designations are followed in this paper. Within a Grand
Cycle, there is normally a transitional interval of mixed carbonate-siliciclastic
lithofacies between basal clean siliciclastic lithofacies and upper clean carbonate
lithofacies (Figs. 3 through 6). This vertical and lateral transitional interval blurs
the contact between members, causing some member designations to be some-
what arbitrary as noted in following discussions. However, between members of
successive Grand Cycles there is less difficulty defining vertical contacts because
of sharp juxtaposition of clean siliciclastic lithofacies overlying clean carbonate
lithofacies.
290 R.D. Adams
Figs. 3 through 6. Stratigraphic cross sections.
DEPOSITIONAL DIP SECTIONS

3A = First Grand Cycle/sequence of the Zabriskie Quartzite-Eagle Mountain Shale-Thimble Limestone and
second Grand Cycle/sequence of the Echo Shale-Gold Ace Limestone
4A =Third Grand Cycle/sequence of the Pyramid Shale-Red Pass Limestone
5A =Fourth Grand Cycle/sequence of the Pahrump Hills Shale-Jangle Limestone
6A =Fifth Grand Cycle/sequence of the Desert Range-Papoose Lake Member
DEPOSITIONAL STRIKE SECTIONS

3B = First Grand Cycle/sequence of the Zabliskie Quartzite-Eagle Mountain Shale-Thimble Limestone and
second Grand Cycle/sequence of the Echo Shale-Gold Ace Limestone
48 =Third Grand Cycle/sequence of the Pyramid Shale-Red Pass Limestone
58 =Fourth Grand Cycle/sequence of the Pahrump Hills Shale-Jangle Limestone
6B =Fifth Grand Cycle/sequence of the Desert Range-Papoose Lake Member
Distances between sections, as shown on Fig. 18, were estimated from the palinspastic base map of Levy and
Chlistie-Blick (1989). "CARB" is for clean carbonate lithofacies intervals. "MXY' stands for mixed lithofacies
intervals. "SIL" is for the clean siliciclastic intervals, with "SAND" indicating that the interval is dominantly sand-
stone and siltstone. "NON-MAR" indicates the nonmaline intervals of supratidal or coastal plain shale/mudstone
red beds of probable paleosol origin. Other abbreviations are listed below. Death Valley stratigraphic terminology
is used in the Frenchman Mountain (FM) section for clarity of correlations rather than Grand Canyon terminology
(See also Fig. 2A).
Silicicla~tic lithologies, shown to the left of the center line, have been grouped into two categories: "shale,
mudstone" is close to the center line and "siltstone plus sandstone" is farther away. Carbonate lithologies, shown
to the light of the center line, are grouped into "mudstone, wackestone" close to the center line, and "pack.stone,
grainstone and/or siltite, arenite" fruther away. Inteltidal and supratidal lithofacies are filled in with black. Single
lines represent interbedded thin beds, commonly of coarser-grained lithologies, in finer-grained lithologies. In
some instances, the interbeds are carbonates in siliciclastics or vice versa.
Most of the lithofacies boundru'ies shown are time transgressive. If parasequence boundalies were shown, then
lithofacies lines would truncate against parasequence boundalies and be offset laterally. Para~equence boundalies
were omitted for clarity in showing gross lithologic changes and how those relate to Grand Cycles/sequences
and the valious systems tracts. Sequence boundalies are numbered from I to 5 and are abbreviated "SB#". The
sequence boundary at the base of each sequence is a~sociated with that sequence. The sequence boundary at the
top of each sequence is a~sociated with the subsequent sequence.
The datum for each of Figs. 3 to 5 is the upper Grand Cycle/sequence boundary, chosen as the best available
approximation of a time-line. Fig. 6 uses the lower sequence boundary for a datum because the upper sequence
boundary wa~ not mea~ured.
Abbreviations: Stratigraphic Units: DR =Desert Range Limestone; E =Echo Shale; EM =Eagle Mountain Shale;
GA =Gold Ace Limestone; J =Jangle Limestone; P =Pyramid Shale; PH =Pahrump Hills Shale; RP =Red Pass
Limestone; T = Thimble Limestone; Z =Zabriskie Qurulzite. Locations of stratigraphic sections: BR = Belted
Range; CS =Chappo Springs; DR =Desert Range; EC =Echo Canyon; EM =Eagle Mountain; FM =Frenchman
Mountain; GR = Groom Range; LC = La~t Chance Range; LV = La~ Vega~ Range; MH = Montgomery Hills;
MR =Middle Resting Springs Range; N = Nopah Range; NR = Northern Resting Springs Range; SH =Striped
Hills; SR =Southern Resting Springs Range; TC =Titus and Titanothere Canyons; WP =Winters Pass. Sequence
Stratigraphic Units: SMST =Shelf-Margin Systems Tract; TST =Transgressive Systems Tract; HST =Highstand
Systems Tract.
~
"<
WI 62km 17km 56km

....~
I I I ~;
LC TC ""'" EC MH SB3 L~ e;
(b
CARB II < 11= I I ...". ~I ':> II

g
S
0-
GA
---- :::!.
:n ""::3
~ Cl
w CARB iii
?> ::3
SB~ _1 a..
'"Tl
0 Q
""'::!l MXT
-- (")
"-
(JQ (b
c: en
(0
""' ....::3
CARB
'"'"
'E. g
O "-
....
::>
en C3'
(0
(0 MXT :3....
'0 A. Depositional Dip Section of the First Grand
(JQ
'" Cycle/Sequence (Zabriskie Quartzite-Eagle
(0
Mountain Shale-Thimble Limestone) and
N
""::3""
\0 Second Grand Cycle/Sequence (Echo Shale-
o V.E. = 600:1
a..
0 Gold Ace Limestone).
SIL(SAND) km
-- 24
~
Z
Z
r- ~
-
t
S:~~
tv
\0
-
IV
NI--- 17 km ~ 24km I 15 km --t- 7 km I 15 km ~ B km I S8 26 km -------1 \0
EC MH EM NR MR SA N 3 IV
HST
1MXT E
-r
TST
FAULT
."
riQ.
VJ
c:I
."
..,o
:::l"> MXT EM
~
@
<">
-g
'" SIL
o
=:I
en
g TST
't:I
(lQ
("0
'"
~ SIL(SAND)
o
B. Depositional Strike Section of the First Grand

Cycle/Sequence (Zabriskie Quartzite-Eagle
Mountain Shale-Thimble Limestone) and
Second Grand Cycle/Sequence (Echo Shale-
Jr o V.E. = 300:1
Gold Ace Limestone). SMST o - km 12 ~

r- ~
~
1, ~ fr
s
fIl
E
~~ :
ui
>'
o
.-c:JI-IC:J-ICl-.
c.
""
...
CD
C/)
III
a:
<I:
U
l: :-; ; ;I=:l: :;:~;:~ : p-\o.~:;:=t- - ->\=l

Fig. 4A. For figure caption see page 290
N
10
~
- \ - - - 15 km - - - ] - 8 km 26 km -------1
N1----. 17 km rvlH 24 km
EC
!1 CARB
~
~
~
5..,
::n
()Q
c::
a
n
'"
'S
o
g
"0
~ ~ 40
()Q
'" m
~ ~ *Y B. Depositional Strike Section of the Third Grand

~.E.':';OO::'" Cycle/Sequence (Pyramid Shale-Red Pass
o km 12 Limestone).
::t1
~
>
~
CIl
~
"<
~
WLrIC- - - - - -------T+b---- 39km J_ 11k.n\ .J.. 89km 56km------:J'
~
g
E3
0-
CARB ::!.
Il:l
'Tl :::J
~.
(systems Tract . __
Boun~~CARB 7
Cl
VI .---V !h
-->---=-~=t-::~~r=::::::~~:---~
;> P.i
:::J
MXT 0..
5..., ~
::tl
~ ~
en
@ _(systems Tract Boundary)
__7
CARB 5
n
o
'"
"9. g
o
::>
r ....
'"~ CARB ;
[ V.E.= 600:1
"0
9t;;.
(JQ
'" km 24
(I> Il:l
IV :::J
Depositional Dip Section of the Fourth Grand 0..
8 PH Cycle/Sequence (Pahrump Hills Shale-Jangle
Limestone). ~
~
fr
N
10
Ut
tv
\0
0'1
26 km -------1
24 km I
EM
CARB 1
--"'1, ~
::!1 _~~:!~~!'.!!!~.l~~::...~~.~~?,:r.y.L............
q<> CARB
VI
CO
MXT--'~
"I1
o..,
::l>
OQ
c: MXT
~ .~.,[~y.:~:.ms Yr.
(') ......!'!.~~Und.ry)
'"
"'g ~NON-MAlf
o
:0
~ MXT
'0
o;s'"
N
~ m
B. Depositional Strike Section of the Fou;th Grand t jj JL o V.E. = 300:1

Cycle/Sequence of the Pahrump Hills Shale- -- km 12
Jangle Limestone).
~
~
~
~
til
~
f
~
0-
S
(1)
WI U!un 39 !un -+----------- ..... I ..... - - - - - - 1 E
LC TC EC MH LV FM
Q
S
0-
CARB CARB ::!.
~
::J
::n
q<l
ro a
P' ~
::J
0-
61
... Z '" $ \ffiLII Q
::n (")
~
IJ_ M: D.~ffiO""DlJtlo"~IftJ"d$ J j
- -
km
- 24 Cycle/Sequence (Desert Range-Papoose Lake
Member)
~ .....
(1)
CIl
~
n
~
....::J
~
g. N!--- 17km I Ullin 115m ----f- 7km -t-- 121l1n ---+31un+- 8km I 26km IS Q
EC MH EM NR MR CS SA N WP .........
~ b'
"0 :3....
~
OQ If ~
('1> ~
tv ::J
\0 0-
o CARB ro ~
~
go
MXT SB5 MXT [ V.E. = 300:1
B. Depositional Strike Section of the Fifth Grand
-- km
. 12
Cycle/Sequence of the Desert Range-Papoose
Lake Member).
N
IQ
--.l
298 R.D. Adams
In general, within a Grand Cycle, member boundaries do not coincide with time
lines but are time-transgressi ve (Palmer and Halley, 1979; Aitken, 1981; Adams,
1993). Carbonate members generally are cleaner and contain less mixed lithofacies
to the north and west, whereas siliciclastic members are cleaner and contain less
mixed lithofacies to the south and east. These lateral changes may make it more
difficult to pick out sequence/Grand Cycle boundaries in the central part of the
lagoonal area where mixed lithofacies are present on either side of a sequence
boundary. This is most noticeable between the Montgomery Hills (MH) and the
Nopah Range (N) of Fig. 3B where mixed lithofacies are present on either side of
the sequence boundary between the first and second Grand Cycles, resulting in a
somewhat arbitrary placement of the boundary.
On the cross sections (Figs. 3 through 6), an attempt has been made to as-
sign lithostratigraphic member designations, but these assignments are somewhat
arbitrary. A strict adherence to member designations tends to obscure sedimento-
logic and chronostratigraphic relationships within Grand Cycles. Because of this,
the member picks on the cross sections are de-emphasized and should be used
primarily in comparing this study to previous studies on the Carrara Formation.
At TituslTitanothere Canyon (TC), from the Eagle Mountain Shale through
the Gold Ace Limestone, the section is only 90 m thick (Adams, 1993). At the
adjacent Last Chance Range (LC) and Echo Canyon (EC) locations, this interval is
approximately 150 m thick (Fig. 3A). Carbonate intervals assignable to the Gold
Ace Limestone and the Thimble Limestone based on trilobite occurrences (Palmer
and Halley, 1979) are both anomalously thin, as is the siliciclastic interval below
the "Thimble Limestone" (Adams, 1993). One fault was recognized below the
Gold Ace Limestone and the quality of outcrop precludes determination of exactly
where, and how many, other faults may slice the section (Adams, 1993). Because
of these problems with the data, the lower part of the TituslTitanothere Canyon
(TC) section has been left off of Fig. 3A.
5.1. FIRST SEQUENCE/GRAND CYCLE: ZABRISKIE QUARTZITE-EAGLE MOUNTAIN

SHALE-THIMBLE LIMESTONE
5.1.1. Zabriskie Quartzite

Prave (1992) di vided the Zabriskie Quartzite into the lower Resting Springs Mem-
ber and the upper Emigrant Pass Member. The base of the Resting Springs Member
lies atop a thick, intensely Skolithos-bioturbated sandstone and is composed of a
coarsening-up series, from marine shale to marine sandstone to braid-plain sand-
stone. A transgressive lag separates this lower member from the Emigrant Pass
Member, composed of interbedded subtidal sandstone, siltstone, mudstone, and
shale (Halley, 1974; Palmer and Halley, 1979; Prave, 1992). Parts of the member
may be intertidal (Prave, 1992). Beneath clean Zabriskie Quartzite sandstone and
above the last carbonates in the Wood Canyon Formation, are a few tens of meters
of marine shale, siltstone, and sandstone (Fig. 3B: base of the Nopah Range [N])
lithostratigraphically assigned to the Wood Canyon (Stewart, 1970; Prave, 1992).
These siliciclastics are part of the basal Grand Cycle, and here are informally
grouped with the Zabriskie Quartzite.
5.1.2. Eagle Mountain Shale

Lithology. The contact of the Eagle Mountain Shale Member of the Carrara Forma-
tion with underlying Zabriskie Quartzite lies at the top of the last clean quartzite
greater than one meter thick. This definition aIIows recognition and differen-
tiation of the upper Zabriskie Quartzite member where it crops out. However,
the lithostratigraphic boundary between the Zabriskie Quartzite and Carrara For-
mation is strongly time-transgressive (Fig. 3), becoming younger to south and
east. Shale and siliceous mudstone dominate the Eagle Mountain Shale with less
common thin sandstone and siltstone interbeds in its lower part. Overlying these
clean siliciclastics are shale and mudstone with thin interbeds offossiliferous pack-
stone/grainstone. The top of the member is within the mixed carbonate-siliciclastic
lithofacies interval across most of the study area (Fig. 3). On the cross sections,
most of the mixed interval is lithostratigraphicaIIy designated as Eagle Mountain
Shale.
Lateral Change. In the most western and northern location, the Last Chance
Range (LC), the member is marked by mixed carbonate and siliciclastic lithofacies.
At the most southern location, Winters Pass (WP), and at the two most eastern
locations, the Las Vegas Range (LV) and Frenchman Mountain, the member has
changed lateraIIy to sandstone and is no longer distinguishable from Zabriskie
QuartzitelTapeats Sandstone (Fig. 3).
5.1.3. Thimble Limestone

Lithology. This is the smaIIest and least lateraIIy extensive carbonate member of the
Carrara Formation (Fig. 3). Clean carbonate lithofacies, mostly subtidal nodular
ribbon rock, are present only in the uppermost part and only at the northwestern part
of the study area; Last Chance Range (LC) and Echo Canyon (EC). Lower in the
member, mixed carbonate and siliciclastic lithofacies are indistinguishable from
upper Eagle Mountain Shale Member. Shale and siliceous mudstone are the back-
ground sediments of the mixed lithofacies interval of the combined Eagle Mountain
Shale-Thimble Limestone with mostly fossiliferous and/or lithoclast packstone in-
terbeds. Only at the Last Chance Range (LC) and Echo Canyon (EC) are siliciclastic
and carbonate lithofacies approximately equal. In the upper part of the member, ad-
ditionallithofacies include the mixed calcarenite-calcisiltite-dolarenite-dolosiltite
(CDAS) lithofacies, oolitic grainstone, weII-layered and nodular ribbon rock, and
oncolitic packstone/grainstone (Table 1). An upper contact for the Thimble Lime-
stone and for the top of the Grand Cycle in the mixed lithofacies is somewhat
arbitrarily picked at an oncolitic and fossiliferous packstone 1 to 2 m thick, or
at the top of one of the thicker carbonate units where the packstone is missing
(Fig. 3).
Lateral Change. To the south and east clean carbonates change laterally into
mixed lithofacies and then to Zabriskie QuartzitelTapeats Sandstone (Fig. 3). Hal-
ley (1974) and Palmer and HaIIey (1979) also reported these lateral and vertical
variations in the Thimble Limestone. Further, they assigned the mixed lithofacies
to the upper Eagle Mountain Shale and stated that where the Thimble Lime-
stone is missing, it is difficult to differentiate between Eagle Mountain Shale and
300 R.D. Adams
Echo Shale. This paper concurs for lithostratigraphic designations, but recognizes
that the upper Eagle Mountain Shale, where not overlain by clean carbonates, is
isochronous with the Thimble Limestone.
5.2. SECOND SEQUENCE/GRAND CYCLE: ECHO SHALE-GOLD ACE LIMESTONE
5.2.1. Echo Shale

Lithology. This member's clean siliciclastic portion is restricted to the most south-
ern and eastern parts of the study area: Winters Pass (WP) and the Las Vegas
Range (LV) (Fig. 3). It is very similar to the basal part of the Eagle Mountain Shale
with dominantly shale and siliceous mudstone and less sandstone and siltstone
interbeds. Over the rest of its extent the member, composed of mixed carbonate
and siliciclastic lithofacies, is very similar to the mixed lithofacies of the Eagle
Mountain Shale-Thimble Limestone. If there are not overlying clean carbonate
lithofacies of the Gold Ace Limestone, no attempt is made to pick a top contact
to this member due to similarities in the mixed lithofacies of the Echo Shale and
Gold Ace Limestone Members. On the cross sections, most of the mixed interval
is lithostratigraphically designated as Echo Shale.
In the middle of the mixed interval at Eagle Mountain (EM), approximately
20 m of subtidal, muddy siltstone mixed with carbonates are present and can be
tentatively correlated to approximately 3 m of siltstone at the northern Resting
Spring Range (NR). These siltstones are the only coarse siliciclastics within the
member, aside from interbedded thin sandstone and siltstone event beds.
Lateral Change. At least part of the Echo Shale Member is laterally equivalent
to the Zabriskie Quartziterrapeats Sandstone at the Las Vegas Range (LV) and
Frenchman Mountain (FM). At Winters Pass (WP) (Fig. 3B) a basal tongue of
shale and siliceous mudstone lies above the Zabriskie/Carrara boundary. That
tongue is assigned to the Echo Shale, although part of the underlying Zabriskie
Quartzite sandstone may be chronostratigraphically equivalent to the Echo Shale.
The lower part of the mixed lithofacies above the tongue is also likely to be part of
the Echo Shale. On the strike section, Fig. 3B, the area between the Montgomery
Hills (MH) and Nopah Range (N) is entirely of mixed lithofacies in both Echo
Shale and the Gold Ace Limestone. In both cases, no attempt is made to precisely
place the boundary between members in a mixed lithofacies interval. To the west
and north there is an overall increase in the abundance of carbonate lithofacies in
the mixed interval.
5.2.2. Gold Ace Limestone

Lithology. Clean subtidal carbonates are present in the northern and western parts of
the study area (Fig. 3). The mixed lithofacies interval is beneath clean carbonates,
and at Last Chance Range (LC) and Echo Canyon (EC) the contact between them
may be considered as the contact between the Gold Ace Limestone and underlying
Echo Shale. Across the rest of the study area, clean carbonates are replaced by the
upper part of the mixed interval. One additional carbonate lithofacies present in
the Gold Ace is fossiliferous mudstone interbedded with siliceous mudstone.
Lateral Change. The lateral extent of clean carbonate lithofacies in the Gold
Ace Limestone is approximately equal to that in the Thimble Limestone, however,
Gold Ace clean carbonates are up to 3.5 times thicker. In addition the laterally
equivalent part of the mixed interval extends farther east and south, to the Las
Vegas Range (LV) and to Winters Pass (WP), where the top 7 to 8 m of clean
carbonates is lithostratigraphically designated Gold Ace Limestone. Coeval shale
and siliceous mudstone at Frenchman Mountain (FM) pass laterally down-dip
into mixed lithofacies and clean siliciclastics at the Las Vegas Range (LV). At the
Montgomery Hills (MH), equi valent strata consist only of mixed lithofacies, which
are predominantly siliciclastic (Fig. 3A). Seventeen km west at Echo Canyon (EC)
the section is almost entirely clean carbonates.
5.3. THIRD SEQUENCE/GRAND CYCLE: PYRAMID SHALE-RED PASS LIMESTONE
5.3.1. Pyramid Shale

Lithology. In contrast to the two lower siliciclastic members, the Pyramid Shale
contains a moderate percentage of subtidal, and some intertidal, siltstone and sand-
stone. However, the lower, clean siliciclastic interval extending across the entire
dip section contains only subtidal facies (Fig. 4A). The mixed lithofacies interval is
richer in carbonates than in previous members, and is grouped lithostratigraphically
with the overlying Red Pass Limestone Member. At the Las Vegas Range (LV),
intertidal sandstone is more common than subtidal ooid grainstone, fossiliferous
packstone/grainstone, and lithoclast packstone/grainstone. The mixed interval is
assigned to the Pyramid Shale Member.
Lateral Change. Three of the thickest sandstone and siltstone intervals in
the Pyramid Shale are located in a down-dip position at the Last Chance Range
(LC) and TituslTitanothere Canyons (TC) (Fig. 4A). This somewhat unexpected
positioning may be due to lateral movement of fluvial-deltaic systems to pro-
duce a non-uniform distribution of sandstone and siltstone. This interpretation is
supported by sandstone/siltstone concentrations along strike in the Resting Spring
Range (NR, MR, and SR) and Nopah Range (N) (Fig. 4B). At the top of the Pyramid
Shale, over 20 m of clean siliciclastics at Eagle Mountain (EM) change laterally to
mixed lithofacies at the Montgomery Hills (MH) (Fig. 4B). To the south at Win-
ters Pass (WP), 25 m of mixed lithofacies, dominantly subtidal shale/mudstone
with interbeds of fossiliferous packstone/grainstone, are present at the base of the
member.
5.3.2. Red Pass Limestone

Lithology. The Red Pass Limestone is divided into a lower mixed lithofacies in-
terval and an upper clean carbonate interval, with an additional mixed interval
at the top of the section in the Las Vegas Range (LV). The faulted-out inter-
val at the Montgomery Hills (MH) is interpreted as a clean carbonate interval,
based on lateral correlations to Echo Canyon (EC) to the west and the Las Vegas
Range (LV) to the east. Carbonate lithofacies common to both mixed lithofacies
intervals and to clean carbonate intervals are: ooid grainstone, nodular ribbon
302 R.D. Adams
rock, fossiliferous packstone/grainstone, subtidal CDAS lithofacies, and low in-

tertidal dolosiltite-calcarenite-calcisiltite lithofacies (Table 1). In addition, subtidal
mudstone/wackestone, thrombolitic boundstone, and stromatolitic boundstone are
found in both intervals but are rare. The most common carbonate lithofacies in both
intervals are nodular ribbon rock and ooid grainstone. Nodular ribbon rock is more
common in the clean intervals than is ooid grainstone, but the situation is reversed
in the mixed interval. In the mixed interval additional carbonate lithofacies include
lithoclast packstone/grainstone, and subtidal-to-Iowest-intertidal well-layered and
nodular ribbon rock, as well as subtidal siliciclastic sandstone, siltstone, and
shale/mudstone. Additional lithofacies found only in the clean carbonate interval
include bioturbated ribbon rock, calcic fenestral mudstone, rare oncolitic pack-
stone, and rare mid-to-high-intertidal, microbially laminated, orange dolosiltite.
Lateral Change. To the south and east, the base of the mixed interval is
generally higher in the stratigraphic section, particularly between the Montgomery
Hills (MH) and Eagle Mountain (EM), and between the Nopah Range (N) and
Winters Pass (WP) (Fig. 4B). Clean siliciclastics replace both mixed lithofacies
and clean carbonates between the Las Vegas Range (LV) and Frenchman Mountain
(FM) (Fig. 4A). However, an 18 m interval of mixed lithofacies at the top of the
section at the Las Vegas Range (LV) displaces clean carbonates westward. Within
the mixed interval, siliciclastics decrease in abundance from east to west, forming
a majority of the interval at the Las Vegas Range (LV) (and designated as Pyramid
Shale), but making up a minority at all other locations. The clean interval extends
updip to somewhere between the Las Vegas Range (LV) and Frenchman Mountain
(FM) (Fig. 4A), but ends between the Montgomery Hills (MH) and Eagle Mountain
(EM) along strike (Fig. 4B).
5.4. FOURTH SEQUENCE/GRAND CYCLE: PAHRUMP HILLS SHALE-JANGLE LIME-

STONE
5.4.1. Pahrump Hills Shale

Lithology. Lateral facies changes between mUltiple tongues of siliciclastic and
mixed lithofacies intervals add complexity to the Pahrump Hills Shale Member.
The lithostratigraphic boundary with the overlying Jangle Limestone Member is
at the top of the mixed interval above the upper nonmarine siliciclastic tongues
(Fig. 5). However, a mixed interval in the vicinity of the Montgomery Hills (MH),
lithostratigraphically designated as the Pahrump Hills Shale Member (Fig. 5A),
changes laterally to the lower half of the Jangle Limestone between Echo Canyon
(EC) and the Las Vegas Range (LV).
Siliciclastic tongues are composed of subtidal and minor intertidal shale, silt-
stone, and sandstone, as well as supratidal shale/mudstone. This is the only sili-
ciclastic member in which a demonstrable supratidal lithofacies is found. Blocky
weathering, slickensides, intense disruption of sedimentary structures, mudcracks,
and small tepee structures indicate development of a supratidal paleosol. Intertidal
lithofacies with few mudcracks are associated with supratidal shale/mudstone. The
basal part of many parasequences in the supratidal lithofacies includes thin (~1 0-
15 cm), intertidal dolosiltite beds. Most siliciclastic supratidal lithofacies intervals
(labeled "NON-MAR", i.e. nonmarine) delineated on Fig. 5 commonly encompass

two or more parasequences with lithofacies boundaries parallel to parasequence
boundaries.
Mixed lithofacies intervals incorporate a variety of siliciclastic and carbon-
ate lithofacies, including subtidal and intertidal siliciclastic shale to sandstone,
and all subtidal and intertidal carbonate lithofacies, except bioturbated ribbon
rock and calcareous fenestral mudstone. Low-intertidal stromatolitic boundstone
is more common than elsewhere in the Carrara Formation, and crops out in distinc-
tive orange-to-gray ledges that can be correlated for tens of kilometers. Intertidal
carbonate lithofacies are moderately common throughout tongues of mixed litho-
facies, and are more common than in previous members. At Titusrritanothere
Canyons (TC) they make up most of the outcrop.
Lateral Change. The overall pattern of lateral lithofacies change shows sili-
ciclastics built out from east and south, and carbonates from west and north. In
the basal part of the dip section (Fig. 5A), two siliciclastic tongues, separated by
a mixed tongue, onlap the basal sequence boundary to the east. A higher silici-
clastic interval is mainly restricted to the Frenchman Mountain (PM) area, with
only the upper part extending west to at least the Las Vegas range (LV). The lower
mixed interval intertongues and grades eastward into the lower siliciclastic tongues
(Fig. 5A). It is, in part, overlain by the middle tongue of supratidal siliciclastics that
continues down-dip past the Last Chance Range (LC) as subtidal sandstone. The
upper mixed interval crops out at the Las Vegas Range (LV), grading eastward into
clean siliciclastics and westward into clean carbonates at the Montgomery Hills
(MH) and Echo Canyon (EC) areas. Intertonguing relationships are more complex
in the strike section (Fig. 5B) with the lowest two marine siliciclastic tongues ex-
tending across the entire section and five nonmarine siliciclastic tongues defining
a progradational-to-retrogradational stacking couplet. Mixed lithofacies intervals
grade laterally with all of the siliciclastic tongues.
5.4.2. Jangle Limestone

Lithology. Intervals of carbonate lithofacies and mixed lithofacies intertongue in
the Jangle Limestone Member. In addition, at Winters Pass (WP) the northern
terminus of a siliciclastic tongue crops out, underlain by a carbonate interval and
overlain by a mixed interval. The Jangle contains the greatest variety of car-
bonate lithofacies, with all carbonate lithofacies present in carbonate intervals
and only bioturbated ribbon rock, stromatolitic boundstone, and intertidal mud-
stone/wackestone missing in mixed intervals. Mixed intervals also contain subtidal,
with subordinate intertidal, shale/mudstone and siltstone/sandstone facies. Inter-
tidal carbonate lithofacies are common in all mixed intervals and in the three upper
carbonate tongues, but are lacking in much of the basal carbonate tongue. The two
middle carbonate tongues grade laterally into a mixed lithofacies interval and this,
combined with the lack of intertidal lithofacies in the lowest carbonate tongue, al-
lows subdivision of the Jangle Limestone into two fourth-order shallowing-upward
cycles composed of fifth-order shallowing-upward cycles. This subdivision can be
recognized at all locations except Frenchman Mountain (PM) where siliciclastics
304 R.D.Adams
are contemporaneous with the lower fourth-order cycle (Fig. SA). The siliciclastic
interval at Winters Pass (WP) is composed of subtidal shale to sandstone.
Lateral Change. The pattern of siliciclastics derived from the east and south
and carbonates derived from the west and north repeats in the Jangle Limestone,
but is somewhat complicated by the presence of the two fourth-order shallowing-
upwards cycles. Three carbonate lithofacies tongues in the lower fourth-order cycle
contrast with a single tongue in the upper (Fig. S). The basal carbonate tongue is the
major tongue in the lower cycle and correlates across the strike section (Fig. SB).
In contrast, the next two tongues grade laterally into a mixed interval between
Eagle Mountain (EM) and the northern Resting Spring Range (NR). These two
thin carbonate tongues are separated by a thin 2.S m) mixed-interval tongue.
In tum, the main mixed interval, overlying the basal carbonate tongue, grades
laterally into the siliciclastic interval at Winters Pass (WP).
In the dip section (Fig. SA), the lower carbonate tongues pass into the mixed
interval east of the Montgomery Hills (MH), whereas the thin upper (third) car-
bonate tongue extends farther east through the Las Vegas Range (LV). Because
members are defined as lithostratigraphic units, the extension of the mixed interval
in the Las Vegas Range (LV) is assigned to the Pahrump Hills Member (cf. Fig. S)
due to dominant siliciclastic lithofacies and despite apparent isochroneity with the
lower fourth-order cycle of the Jangle Limestone.
The upper fourth-order cycle is less complicated, consisting of a thick, clean
carbonate interval extending updip (Fig. SA), but merging southward along strike
into a mixed interval between the middle part of the Resting Spring Range (MR) and
Chappo Spring (CS) (Fig. 5B). At the Las Vegas Range (LV) and at Frenchman
Mountain (FM), a minor tongue of siliciclastics displaces the lower part of the
carbonate interval westward. In the strike section (Fig. 5B), the basal part of
the carbonate interval extends south of the Nopah Range (N), whereas the upper
part grades laterally into mixed lithofacies. A thin tongue of mixed lithofacies
also extends into the carbonate interval between Eagle Mountain (EM) and the
northern Resting Spring Range (NR). Lastly, a 15-m thick interval of subtidal
mixed lithofacies at the Montgomery Hills (MH) is without laterally equivalent
mixed intervals in neighboring ranges (Fig. 5).
Intertidal carbonate lithofacies tend to be concentrated at the northern and
western parts of the sections and in the upper half of each of the fourth-order cycles.
However, lateral continuity of intertidal lithofacies varies considerably, indicating
that large areas of the carbonate bank did not simultaneously shoal to the intertidal
zone (Adams and Grotzinger submitted). When lithofacies correlations are further
constrained by fifth-order parasequence boundaries, the lateral discontinuity of
intertidal facies becomes more pronounced (Adams and Grotzinger, submitted).
5.5. FIFfH SEQUENCE/GRAND CYCLE: DESERT RANGE LIMESTONE-PAPOOSE

LAKE MEMBER OF THE BONANZA KING FORMATION
5.5.1. Desert Range Limestone

Lithology. In the Desert Range Member a basal siliciclastic-rich mixed interval, S.S
to 30 m thick, is overlain by a carbonate interval. The lithostratigraphic boundary
with the overlying Papoose Lake Member of the Bonanza King Formation is picked
at a lithofacies change from nodular ribbon rock to oncolitic, bioturbated ribbon
rock (Halley, 1974; Palmer and Halley, 1979; Adams, 1993). As a result, thickness
of the Desert Range Member varies considerately (Fig. 6).
In the mixed interval subtidal sandstone and siltstone are dominant lithofa-
cies, with lesser proportions of subtidal shale/mudstone and intertidal shale to
sandstone. Subtidal carbonate facies are mudstone/wackestone, lithoclast pack-
stone/grainstone, fossiliferous packstone/grainstone, oolitic grainstone, well-lay-
ered and nodular ribbon rock, and CDAS lithofacies. Intertidal carbonate facies
are dolosiltite-calcarenite-calcisiltite lithofacies, and calcic fenestral mudstone.
The siliciclastic interval at Frenchman Mountain (FM) is composed of subtidal
shale/mudstone and the carbonate interval most commonly contains subtidal nodu-
lar ribbon rock and subtidal-to-Iowest-intertidal well-layered ribbon rock. Other
subtidal carbonate lithofacies include oncolitic wackestone-to-grainstone, litho-
clast packstone/grainstone, fossiliferous packstone/grainstone, oolitic grainstone,
and CDAS lithofacies. Other intertidal carbonate lithofacies include dolosiltite-
calcarenite-calcisiltite lithofacies and microbially laminated dolosiltite. Intertidal
facies are sparse and laterally discontinuous in both siliciclastic and carbonate
intervals.
Lateral Change. Both mixed and carbonate intervals essentially extend across
the Carrara Formation (Fig. 6). The mixed interval grades laterally into the clean
siliciclastic interval at Frenchman Mountain (FM) and there is a moderate amount
of lateral intergradation along the contact between mixed and carbonate intervals,
as shown on both sections (Fig. 6). A second, small tongue of mixed lithofacies
on the southern end of the strike section (Fig. 6B) thins northward in the Nopah
Range (N). It is more carbonate-rich than the lower mixed interval, but is neither as
thick nor as extensive. Lateral variation in thickness of the Desert Range Member
indicates lateral intergradation with basal carbonates of the overlying Bonanza
King Formation.
5.5.2. Papoose Lake Member of the Bonanza King Limestone

Although widely recognized (Barnes and Palmer, 1961; Barnes and Christiansen,
1967; Gans, 1974; Kepper, 1981), no detailed sedimentologic descriptions of the
Papoose Lake Member have been published. Burchfiel and others (1982) stated
that member thickness varies from approximately 33050 m to 460 m at the
Montgomery Hills, Nopah, and Resting Spring Ranges. They described the member
as gray and dark gray, mottled limestone and dolostone, with two thickly bedded
intervals of white dolostone, one about 130 m above the base and the other about
130 m below the top. The middle part of the member also has numerous thin beds
of white dolostone, generally laminated. Kepper (1981) interpreted the member
as a shelf lithofacies, with alternation of bioturbated and wavy-bedded or current-
rippled beds. He also stated that the base of the overlying Banded Mountain
Member is an isochronous, rusty-colored, dolomitic siltstone/sandstone capping
the overall shallowing-upward trend of the Papoose Lake Member.
306 R.D. Adams
Oncolitic, bioturbated ribbon rock has been identified in this paper in the basal
Papoose Lake Member and is interpreted as a quieter-water lithofacies that may
represent deeper subtidal deposition than the underlying Desert Range Member.
6. Carrara Sequence Stratigraphy
6.1. FIRST SEQUENCElGRAND CYCLE: ZABRISKIE QUARTZITE-EAGLE MOUNTAIN

SHALE-THIMBLE LIMESTONE
6.1.1. Type 2 Sequence Boundary

At the base of this Grand Cycle, a Type 2 sequence boundary is recognized at
the top of the last subtidal fossiliferous packstone/grainstone in the upper Wood
Canyon. Below the contact are mixed, subtidal carbonate and siliciclastic sediments
and above are fine-grained subtidal siliciclastic sediments that coarsen up into
clean Zabriskie Quartzite sandstones. In the Nopah Range, the contact is eroded,
has a few centimeters of relief, and carbonate grains are truncated at the contact
implying lithification prior to erosion (Adams, 1993). Mixed carbonate/siliciclastic
subtidal sediments below the sequence boundary are displaced by clean, subtidal
siliciclastic sedimentation above the sequence boundary due to a strong basinward
shift in deposition. This sequence boundary is similar to the sequence boundary
between the Echo Shale-Gold Ace Grand Cycle and the Pyramid Shale-Red Pass
Limestone Grand Cycle (Figs. 3 and 4) discussed below.
6.1.2. Shelf-Margin Systems Tract

Progradational stacking of lithofacies and parasequences in the Resting Springs
Member (Prave, 1992) reflects progradational stacking patterns common in SMSTs
(Posamentier and Vail, 1988). Prave's (1992) description of the Zabriskie Quartzite
facies succession, Resting Springs Member (shoreface overlain by braid-plain)
separated by a flooding surface from overlying Emigrant Pass Member (nearshore
deposits), does not show evidence of the basin ward shift in facies needed to define
a Type 2 sequence boundary. Zabriskie Quartzite outcrops generally contain more
than one significant bed of Skolithos-bioturbated sandstone ("piperock"). This
indicates that caution should be used in defining a maximum flooding surface based
on this lithofacies. However, Prave (1992, p. 512) maintains that a distinctly thicker
unit " ... occurs only at that particular stratigraphic position (although thinner
piperock beds are present in the lowermost Zabriskie Quartzite and uppermost
Wood Canyon Formation) ... ". The transgressive surface with its lag deposits
separates the SMST from the overlying TST.
The documented widespread nature of Resting Springs Member siliciclastics
indicates the carbonate bank was not established within the study area during
deposition of the SMST. Further, the lack of carbonate interbeds in the member
indicates that if the bank was active outside the study area, it was sufficiently
distant that carbonate sediments were not transported into the area.

The Emigrant Pass Member exhibits retrogradational stacking of parasequences
(Prave, 1992) as does the Eagle Mountain Shale Member (Fig. 3). Retrogradational
stacking in the Emigrant Pass Member is not illustrated in Fig. 3 because of a lack
of measured stratigraphic sections through the member. However, retrogradation
in the Eagle Mountain Shale is apparent in the dip cross section (Fig. 3A). In the
lowermost part of the cross section, between the Last Chance Range (LC) and Echo
Canyon (EC), mixed lithofacies of the Eagle Mountain Shale interfinger laterally
with both Zabriskie Quartzite sandstone and with shale-rich siliciclastics assigned
to the Eagle Mountain Shale. Farther east, shale-rich siliciclastics grade laterally
into Zabriskie Quartziteffapeats Sandstone near the Las Vegas Range (LV) and
possibly as far east as Frenchman Mountain (FM). This series of lateral changes, as
recorded in the succession of lithofacies groups, defines retrogradational stacking
of lithofacies and parasequences. Retrogradational stacking does not show very
well on the strike cross section, Fig. 3B, except in the lateral transition from the
shale-rich siliciclastic interval of the Eagle Mountain Shale into Zabriskie Quartzite
sandstone and shale between the Nopah Range (N) and Winters Pass (WP).
An overall dominance of carbonate over siliciclastic deposition in mixed litho-
facies at the Last Chance Range (LC) indicates that carbonate bank deposition
was active outside the study area during the early TST. The active bank was near
enough to provide carbonate sediments to the western end of the study area.

The Thimble Limestone Member and the upper part of of the Eagle Mountain
Member make up the HST (Fig. 3). Clean carbonate lithofacies of the Thimble
Limestone are exposed over a relatively small part of the study area due to location
of most carbonate bank deposition farther north and west. A clean carbonate
lithofacies interval is not present east and south of Echo Canyon (EC) and only at
the Montgomery Hills (MH) is the laterally equivalent mixed interval sufficiently
rich in carbonates to be considered part of the Thimble Limestone. Farther north
and west, at the Last Chance Range (LC), the mixed interval beneath the carbonate
lithofacies interval is predominantly carbonate. Clean carbonate lithofacies of the
bank existed northward and westward of this location. Late in the HST , final
progradation of the bank away from its center and towards the craton was not able
to extend past the Montgomery Hills area.
6.1.5. An Alternate Interpretation of the Zabriskie Quartzite

Prave (1992) developed a sequence stratigraphic model for the Zabriskie Quartzite
in which the lower Resting Springs Member and the upper Emigrant Pass Member
belong to a different sequences. He interpreted the lower Resting Springs Member,
composed of braid-plain siliciclastics overlying marine siliciclastics, as a HST
separated from its underlying TST , composed of uppermost Wood Canyon For-
mation deposits, by an intensely bioturbated sandstone thought to be a condensed
interval. A Type 2 sequence boundary between the two members coincides with a
308 R.D. Adams
transgressive surface characterized by a transgressive lag recording flooding of the

braid plain. Nearshore marine siliciclastics of the Emigrant Pass Member form a
retrogradational stack of thin parasequences; hence, the member is interpreted as a
TST. The Eagle Mountain Shale Member of the Carrara Formation is also viewed
as part of this TST.
A less complex interpretation of the Resting Springs Member is preferred.
The surface between the Zabriskie Quartzite members interpreted by Prave (1992)
as a combination sequence boundary/transgressive surface, is reinterpreted as a
transgressive surface only and not as a sequence boundary. The Resting Springs
Member is reinterpreted as a SMST and the Zabriskie Quartzite is part of only
one Grand Cycle/sequence in the Montgomery HillslResting Spring Range/Nopah
Range area. Either interpretation of the lower Zabriskie Quartzite is compatible
with the sequence stratigraphic interpretation of the Carrara Formation proper.
Only in the larger viewpoints of regional, multi-formational stratigraphy and of
sequence stratigraphic models of Grand Cycles does the lower Zabriskie Quartzite
interpretation become important.
6.2. SECOND SEQUENCE/GRAND CYCLE: ECHO SHALE-GOLD ACE LIMESTONE

The sequence boundary separating the Thimble Limestone from the overlying Echo
Shale marks the beginning of the second Grand Cycle/sequence. In the lagoonal
area, from the Montgomery Hills (MH) to the Nopah Range (N) (Fig. 3B), it is
difficult to pick the sequence boundary due to a lack of a clean carbonate interval
and a lack of a definitive bounding surface. The lack of a definite bounding surface
indicates that sedimentation in the lagoon was not interrupted by the basin ward
shift in deposition for any appreciable interval of time. This is consistent with
a Type 2 boundary and also may reflect continued distal bank deposition. The
decrease in accommodation that marks the initiation of this Grand Cycle was a
stall in the long-term increase of accommodation. The basinward shift in deposition
marking this sequence boundary is not as pronounced as for the previous sequence
boundary. On Fig. 3A, westward displacement of lithofacies groups denotes this
shift: between the Montgomery Hills (MH) and the Las Vegas Range (LV) clean
siliciclastic lithofacies displace mixed lithofacies; between the Last Chance Range
(LC) and Echo Canyon (EC) mixed lithofacies displace clean carbonate lithofacies.
On Fig. 3B the shift is denoted by northward displacement: between the Nopah
Range (N) and Winters Pass (WP) clean siliciclastic lithofacies displace mixed
lithofacies; between Echo Canyon (EC) and the Montgomery Hills (MH) mixed
lithofacies displace clean carbonate lithofacies.

Within the study area, there is little evidence for or against a SMST in this Grand
Cycle. Clean siliciclastic lithofacies are only found in the easternmost locations,
the Las Vegas Range (LV) and Frenchman Mountain (PM), and the southernmost
location, Winters Pass (WP) (Fig. 3). This limited extent does not allow definitive
recognition of either a progradational or retrogradational stacking pattern of litho-

facies and parasequences to differentiate between a SMST and a TST. If a SMST
is present, it is likely restricted to the basal 10 to 20 m of the Grand Cycle, is
composed of only marine lithofacies, and is lithostratigraphicaIIy part of the Echo
Shale. As such, it would be very difficult to distinguish from the TST. Because
of the widespread mixed lithofacies interval at the base of the Grand Cycle, it
is inferred that the carbonate bank remained active outside the study area during
SMST and/or TST deposition.

A vertical change from clean siliciclastics to mixed lithofacies at Winters Pass
(WP) and the Las Vegas Range (LV) (Fig. 3) is interpreted as a retrogradational
stacking of lithofacies and parasequences. This interpretation avoids the question
of whether the clean siliciclastic interval represents a progradational (SMST ) or
retrogradational stacking (TST ). If a SMST is not present in this Grand Cycle,
then the TST will incorporate the clean siliciclastic interval at the southern and
eastern parts of the study area, as well as the lower portion of the mixed lithofacies
interval. In the study area, the clean carbonate interval is relatively restricted areally
(Fig. 3). Its thickness, approximately 70 m from the Last Chance Range (LC) to
Echo Canyon (EC), indicates the carbonate bank was relatively stable during much
of the Grand Cycle. It is possible that part of the clean carbonate interval belongs
in the TST. The relative stability of the bank over most of its duration precludes
definitive differentiation of TST and HST.

The clean carbonate interval comprises the Gold Ace Limestone and much or all
of the HST in the northern and western part of the study area. The part of the
mixed interval assigned to the Echo Shale that is lateraIIy equivalent to the Gold
Ace Limestone forms the HST to the east and south. In the Las Vegas Range (LV),
only the upper seven meters of section are mixed lithofacies and represents the
greatest cratonward progradation of mixed deposition in the lagoon. It is interpreted
as deposited during the latest HST , when the rate of accommodation decreased
(Fig. 7). At least part of the shale-rich siliciclastic interval at Frenchman Mountain
(FM) is assigned to the HST.
6.3. THIRD SEQUENCPlGRAND CYCLE: PYRAMID SHALE-RED PASS LIMESTONE

At the base of the Pyramid Shale, in locations lacking the Gold Ace Limestone, the
sequence/Grand Cycle boundary is placed at the contact between mixed lithofacies
and overlying clean siliciclastic lithofacies (Figs. 3 and 4). No erosion or evidence
for subaerial exposure was recognized along the contact, and although there is
a relatively abrupt change from mixed to clean siliciclastic deposition across
the contact, the depositional environment remains marine. However, a basinward
310 R.D. Adams
-ACCOMMODATION CURVE-"
1
1
1
1 1 ACCOM~~ATION
MINiMUM - - - - 1- - - - - -I - - - - - I .
ACCOM"10DATION MAXIMJM RATE 1
-1- - - - - -, - - - - - r - - - - - - - - -I - - - - - -r .
1 1 1
POSITIVE
1 RATESOF
<>",~ 1 CHANG E
il.'- ___
,;:.'"
o - - _INL~~?!QN~O!!!T~ _ _ 1 1_ _ _ _ ~ 1_ .!.NF.bEgltW J. '0
1 ~ 1 \l}1< POINT 1
u~'f' 1 ~~ 1 NEGATIVE
~ 1 1 {.. 1 RATESOF
CHANGE
1 1 1
-1- - - - - - - - - +- - - - - - 1- - - - - -I - - - - -
MINIMUM
RATE
-RATE OF CHANGE CURV~
Fig. 7. The upper curve represents an idealized, symmetrical cycle of accommodation fluctuation.
The horizontal axis represents ti me and the cycle progresses from left to right. Starting at the far
left, the high point of the curve represents maximum accommodation. Accommodation decreases
until the low point of the curve where minimum accommodation is reached. Accommodation then
begins to increase until the high point of the curve is again reached. The positions on the cycle
where the various systems tracts are deposited are designated on the curve, as is the position where
the sequence/Grand Cycle boundary forms. The derivative of the upper curve is drawn as the lower
curve, and this shows the rate at which accommodation changes. See text for further discussion.
shift in lithofacies is pronounced, with westward and northward displacement of

clean carbonate lithofacies and mixed lithofacies by clean siliciclastic lithofacies
(Fig. 4). These features indicate a Type 2 boundary separates the two Grand Cycles.
At Eagle Mountain (EM) and the Montgomery Hills (MH), the boundary has been
drawn 30 and 12.5 m above the contact between mixed and clean siliciclastic
deposits. At Eagle Mountain, there is a final 1O-cm thick trilobite packstone at the
boundary, and the underlying shale contains numerous trilobite fragments, though
without carbonate interbeds. The overlying shale is devoid of both trilobites and
carbonate interbeds. At the Montgomery Hills (MH), the boundary pick is based
on maintaining thickness for the two Grand Cycles as there are no trilobites in
the upper 12.5 m of the preceding Grand Cycle and there is no final carbonate
interbed at the boundary. The contact could be moved down by the 12.5 m with
little affect on the overall stratigraphic interpretation. At Winters Pass (WP), the
sequence boundary pick is also somewhat arbitrary and is chosen for maintenance
of relative thickness along strike cross sections. It could easily be moved from its
present position at the top of the 8-m thick carbonate bed set to the top of the last
carbonate interbed 25 m above.

As with the previous Grand Cycle, there is not definitive evidence for or against
the presence of a SMST at the base of this Grand Cycle. However, in this cycle the
clean siliciclastic interval, the Pyramid Shale Member, does extend all across the
study area and tends to be greater than 40 m thick. Retrogradational stacking in
the siliciclastic interval has been interpreted as present between the Montgomery
Hills (MH) and Eagle Mountain (EM) (Fig. 4B). If correct, this indicates that at
least the upper part of the siliciclastic interval is part of the TST. The decrease
in accommodation associated with initiation of this sequence was of sufficient
amplitude and/or duration to allow distribution of siliciclastic sediments across a
wide area. Given this and the siliciclastic interval thickness, it is likely that part
of the interval was deposited during the SMST. However, without recognition of
either a transgressive surface within the interval to mark the change from SMST to
TST , or additional stratigraphic sections down depositional dip to demonstrate a
change from a progradational to retrogradationallithofacies stacking, no definitive
assignment can be made.
As in the first Grand Cycle, the lack of carbonate interbeds indicates that if
the carbonate bank remained active in the SMST , then it was distal enough to not
distribute carbonate sediments into the study area.

As discussed above, at least the upper part of the Pyramid Shale was deposited dur-
ing the TST. Laterally equivalent parts of the mixed lithofacies interval, especially
at Echo Canyon (EC) and the Montgomery Hills (MH) (Fig. 4B), and the lower
parts of the mixed interval in other locations (Fig. 4) were also deposited during
the TST. Because much of the mixed interval is predominantly carbonate, it is
lithostratigraphically assigned to the Red Pass Limestone Member. The presence
of carbonate interbeds is indicati ve of re-establishment of the carbonate bank early
in the TST , within distance to contribute carbonate sediments.

In the study area the clean carbonate interval, assigned to the Red Pass Limestone,
was deposited entirely during the HST. Along strike this interval grades laterally
into mixed lithofacies between the Montgomery Hills (MH) and Eagle Mountain
(EM) (Fig. 4B). However, the clean carbonate interval extends updip east of the
Las Vegas Range (LV) (Fig. 4A) and represents a substantially greater eastward ex-
pansion, without increased southward expansion, of the carbonate bank than in the
previous two Grand Cycles. At the Las Vegas Range (LV), in the uppermost HST ,
a mixed lithofacies interval overlies the carbonates. Rapid progradation commonly
coincides with the late HST because the rate of accommodation becomes increas-
ingly negative (Fig. 7). If the rate of basin ward siliciclastic progradation is greater
than the rate of cratonward progradation of the carbonate bank, then a geometry
such as this would develop.
312 R.D. Adams
6.4. FOURTH SEQUENCE/GRAND CYCLE: PAHRUMP HILLS SHALE-JANGLE LIME-

STONE

No evidence of exposure or erosion has been found along this boundary and
whereas the initial sediments of the Grand Cycle are principally clean siliciclas-
tics, the depositional environment remains peritidal/marine across the boundary.
Because of this, the sequence boundary is interpreted as a Type 2, demarcated
by a change from carbonate or mixed lithofacies into clean siliciclastic lithofa-
cies caused by a strong basinward shift in siliciclastic deposition. At Frenchman
Mountain (PM) positioning of the boundary is based upon red coloration of over-
lying sandstone and by trilobite occurrences and zonations reported by Palmer and
Rowland (1989). The boundary at the Las Vegas Range (LV) is somewhat more
arbitrary and could be moved up or down a few meters with little influence on
correlations and interpretations. The choice at Titusrritanothere Canyons (TC) is
based on first influx of siliciclastics, even though the change in lithofacies across
the boundary is minor.
No attempt has been made to pick the boundary at the Last Chance Range
(LC), where the section is unusually thin. No faulting was recognized for over 50 m
above or below the division between third and fourth Grand Cycles. Unpublished
stratigraphic sections of the Carrara Formation above the Red Pass Limestone,
measured by Halley (1974) and by I. Montanez and D. Osleger (personal com-
munication, 1993) are in close agreement with this study regarding Grand Cycle
thickness and position of the contact between the Carrara Formation and the Bo-
nanza King Formation. Neither section contained any faulting, erosion, or major
exposure/dissolution surfaces through the upper Carrara Formation. If the corre-
lation of sandstone between the Last Chance Range (LC) and Titusrritanothere
Canyons (TC) is correct, then missing section would be placed below the sandstone
(Fig. 5A). The anomalously thin section at the Last Chance Range (LC) for this
Grand Cycle currently is not explained.

The SMST is within the Pahrump Hills Shale, but does not involve the entire
member (Fig. 5). Relative positions of nonmarine red beds of shale/mudstone
define the progradational SMST , the transgressive surface (a parasequence set
boundary), and the retrogradational TST. The overall progradation of the lower
three nonmarine, siliciclastic tongues (Fig. 5B) characterizes the SMST. The first
two nonmarine tongues are separated by a thin incursion of mixed marine lithofa-
cies and the third nonmarine tongue is the most widespread, placing the farthest
basin ward position of siliciclastics at the top of the SMST (Fig. 5A). The marine
extension of the third tongue is delineated on Fig. 5 as "marine siliciclastics"
(designated as "SIL" on Fig. 5), despite associated carbonates, to distinguish it
from adjacent mixed lithofacies intervals. The dip section (Fig. 5A) shows overall
progradational stacking of the two basal marine siliciclastic tongues, whereas the
strike cross section (Fig. SB) emphasizes overall progradational stacking of the
three lower nonmarine tongues.
The SMST at Titusrritanothere Canyons (TC) is predominantly mixed litho-
facies, except for IS m of carbonates underlying the final siliciclastic tongue.
Carbonates in the SMST imply the carbonate bank was never completely overrun
by siliciclastics and if carbonate production was shut off, it was only briefly. Ex-
tension of the uppermost siliciclastic tongue to the Last Chance Range (LC), the
thin 2 m) nature of the siliciclastic interval, and the thick carbonate intervals
(>SO m) on either side of the siliciclastics support this. Between the Montgomery
hills (MH) and the Las Vegas Range (LV), the entire SMST appears to onlap the
underlying sequence boundary (Fig. SA), as predicted for a SMST (Van Wagoner
and others, 1990), but lack of accessible outcrop precludes verification of this
relationship.

The lower parasequence set in the Jangle Limestone and part of the Pahrump Hills
Shale makes up a TST that is well defined by a transgressive surface at the top
of the third nonmarine tongue and by retrogradational stacking of the fourth and
fifth nonmarine tongues. The TST top is less clearly defined, but is placed within
the Jangle Limestone at the parasequence set boundary between two fourth-order,
shallowing-upward cycles. Each of these fourth-order cycles, which can be rec-
ognized over almost all of the Carrara Formation, are composed of meter-scale,
fifth-order shallowing-upward cycles. A relatively pronounced deepening across
this boundary, in many cases going from intertidal facies to deeper subtidal fa-
cies (Adams and Grotzinger, submitted) represents the maximum flooding surface
of the sequence/Grand Cycle. This interpretation represents a significant depar-
ture in facies relationships from the standard siliciclastic sequence-stratigraphic
model. In the standard model, a position many tens of kilometers offshore from the
cratonal shoreline will record subtidal sediments on either side of the maximum
flooding surface (Jervey, 1988; Posamentier and others, 1988; Posamentier and
Vail, 1988; Van Wagoner and others, 1990). In the Jangle Limestone below the
maximum flooding surface, carbonate bank sediments and adjacent parts of the
lagoon shoaled into the intertidal zone (Fig. SB). In a distal position cratonwards
from the bank, e.g., the Las Vegas Range (LV) and Frenchman Mountain (FM),
lagoonal mixed sediments and siliciclastic sediments remained subtidal below the
maximum flooding surface (Fig. SA). This results in intertidal sediments underly-
ing the maximum flooding surface across the bank and into the adjacent lagoon,
but with relatively deeper subtidal sediments overlying it. Farther cratonward, the
maximum flooding surface has subtidal sediments on either side as in the standard
model.
Within the TST , in addition to the two nonmarine tongues at the base, there are
three carbonate tongues expanding south and east from the bank, displacing and
overriding mixed lithofacies tongues. In turn, mixed-lithofacies tongues coalesce
and laterally grade south and east into marine siliciclastics between the Las Vegas
Range (LV) and Frenchman Mountain (FM). Cratonward expansion of a carbonate
314 R.D. Adams
tongue and underlying mixed lithofacies may be interpreted as continued retrogra-

dational stacking of lithofacies and parasequences (Fig. 5A). This retrogradational
stacking pattern is not apparent in the strike section (Fig. 5B) because the low-
est carbonate tongue extends farther south than subsequent carbonate tongues.
A higher rate of siliciclastic influx from the south than from the east during the
middle and late portions of the fourth-order cycle could account for this difference
of apparent stacking patterns between the two sections.
The nonmarine tongues and the lower mixed-interval tongue are lithostrati-
graphically Pahrump Hills Shale. In the Las Vegas Range (LV) the coalesced
mixed interval and the upper carbonate tongue are placed entirely within the
Pahrump Hills Shale, as is the siliciclastic interval at Frenchman Mountain (FM)
(Halley, 1974). At Winters Pass (WP) the siliciclastic tongue at the top of the TST
is assigned to the Jangle Limestone due to the underlying 14 m of clean carbonates.

The HST is composed of the upper parasequence set and, in the easternmost part of
the study area, a part of the Pahrump Hills Shale. One carbonate interval makes up
the HST in the dip section, except in the Las Vegas Range (LV) and at Frenchman
Mountain (FM) where less than 10m of marine siliciclastics are at the base.
Twenty-six meters of mixed lithofacies are present only at the Montgomery Hills
(MH), apparently the result of a local influx of siliciclastics. A mixed lithofacies
interval is present south of the middle Resting Spring Range (MR) and a 6-m thick
mixed lithofacies tongue continues northward into the northern Resting Spring
Range (NR) before terminating in the plane of the cross section. This tongue may
connect with the isolated mixed interval at the Montgomery Hills (MH) outside
the plane of the cross section. Displacement of the carbonate interval reflects
an increased rate of siliciclastic sediment influx from the south in the middle to
late HST. All of the HST is assigned to the Jangle Limestone, except the basal
siliciclastics at the Las Vegas Range (LV) and at Frenchman Mountain (FM) which
are lithostratigraphically Pahrump Hills Shale.
6.5. FIFTH SEQUENCE/GRAND CYCLE: DESERT RANGE LIMESTONE-PAPOOSE

LAKE MEMBER OF THE BONANZA KING FORMATION
6.5.1. Type 1 or Type 2 Sequence Boundary?

The boundary between the fourth and fifth Grand Cycles, with similar periti-
dal/marine lithofacies on either side, shows no evidence for exposure. However,
at Eagle Mountain, there is equivocal evidence for 5 to 9.5 m of local relief over a
lateral distance of less than 200 m (Adams, 1993; Adams and Grotzinger, submit-
ted). Locally, the sequence boundary is marked by subtidal siliciclastic lithofacies
overlying subtidal carbonate lithofacies. No rip-up clasts of underlying subtidal
lithofacies are incorporated into overlying subtidal lithofacies and there are no
obvious scoured surfaces. However, when contacts are walked out and correlated
using the lower parasequence set boundary as a datum, the section is found to be
shorter in the middle of the outcrop. Nodular ribbon rock adjacent to a large throm-
bolitic complex is missing, but no obvious channels or scoured geometries can be
recognized. Some apparent relief also may relate to subtle, small-scale faulting
lower in the section that causes loss of a few meters of stratigraphic thickness
and/or may be due to lateral discontinuity of nodular ribbon rock prior to boundary
formation. Such lateral discontinuity of subtidal lithofacies has been demonstrated
in other intervals at Eagle Mountain (see Adams, 1993; Adams and Grotzinger
submitted). No evidence has been found to indicate that erosion occurred along
the boundary at other locations. Because of these reasons, the sequence boundary
is considered to be a Type 2 boundary. Erosion along sequence boundaries has
been reported from Grand Cycles at other areas (Aitken, 1978; Mount and others,
1991) and a Type 1 boundary could develop if a decrease in accommodation was
of sufficient amplitude and/or rapidity.
6.5.2. Systems Tracts

Stacking patterns are not apparent due to a lack of lateral variation in the Desert
Range Limestone, the absence of measured sections in the Papoose Lake Mem-
ber, and a paucity of published information regarding Papoose Lake lithofacies. A
lower mixed lithofacies interval is present along the sequence boundary, and grades
laterally into the siliciclastic interval exposed at Frenchman Mountain (PM). These
intervals are overlain by a clean carbonate interval of subtidal and intertidal litho-
facies that crops out in all sections. All of these intervals, assigned to the Desert
Range Limestone, may be part or all of the SMST and/or the TST.
If the Papoose Lake represents deeper subtidal deposition than the Desert
Range, the two members may record overall development of an TST. This does
not preclude a thin SMST at the base of the Jangle. The thick bedded white
dolomite located 130 m above the basal formation contact (Burchfiel and others,
1982) may represent the uppermost beds of a sequence. In combination with the
Desert Range they would make up a Grand Cycle about 160 to 200 m thick. If
these white dolomite beds are intertidal (published descriptions are too general to
allow determination), then they may be analogous to intertidal beds found in the
upper part of the Pyramid ShalelRed Pass Limestone Grand Cycle (Echo Canyon
[EC], and TituslTitanothere Canyons [TC]; Fig. 4), and the upper beds of the
Pahrump Hills Shale/Jangle Limestone Grand Cycle (Eagle Mountain [EM], the
southern Resting Spring Range [SR] and TituslTitanothere Canyons [TC]; Fig. 5).
The combined thickness of the Papoose Lake Member (~330 m) and the Desert
Range Limestone (30 to 70 m) makes it likely that the two members encompass
parts of at least two Grand Cycles.
6.6. OVERVIEW
The three complete and two partial Grand Cycles of the Carrara Formation record
interactions of sediments deri ved from two very different sources; a cratonal source
of siliciclastic sediment transported into the study area, and an offshore, in situ
source of carbonate sediment. Sediment from both sources was deposited on a
rapidly subsiding passive margin during long-term eustatic sea level rise (Sloss,
316 R.D. Adams
1963, 1988; Stewart, 1970, 1991; Bond and others, 1988; Levy and Christie-Blick,
1991). In addition, shorter-term changes in accommodation, possibly eustatic in
origin, also affected sedimentation (Bond and others, 1988). Sediments from each
source responded somewhat differently to this setting as discussed below.
In the basal Grand Cycle (Wood Canyon-Zabriskie Quartzite-Eagle Moun-
tain Shale-Thimble Limestone) (Fig. 3), siliciclastics apparently were introduced
equally from east and south; carbonates were relatively thin and somewhat distal
to the study area. In the second Grand Cycle (Echo-Gold Ace) (Fig. 3), the less
abundant siliciclastics were derived predominantly from the east. The third Grand
Cycle (Pyramid-Red Pass) (Fig. 4) was relatively balanced in proportions of car-
bonates to siliciclastics; siliciclastics were again apparently equally derived from
south and east. However, the carbonate bank was better developed in the north-
ern part of the area with little change in its western position. The fourth Grand
Cycle (Pahrump Hills-Jangle) (Fig. 5) also was balanced in proportion of carbon-
ates to siliciclastics, and the main source area for siliciclastics was to the south.
The carbonate bank remained well established in the northern position, though it
narrowed near the end of the Grand Cycle due to siliciclastics derived from the
southern source area. The fifth and last Grand Cycle (Desert Range-Papoose Lake)
(Fig. 6) is almost entirely carbonate sediments, except for the basal 5 to 30 m which
are mostly mixed lithofacies. Siliciclastic sediments again were derived from the
east and south, with perhaps a greater easterly contribution as shown by 30 m
of clean siliciclastics at Frenchman Mountain (PM) versus the second tongue of
mixed lithofacies present in the upper part of the Jangle Limestone at Winters Pass
(WP). These were the last siliciclastic sediments deposited in the Death Valley
region for the next several hundred meters of section (Stewart, 1970). Overall, the
five Grand Cycles record an increasing contribution of carbonate sediments at the
expense of siliciclastic sediments as a result of long-term flooding of the craton.
The changing importance of a southern versus an eastern siliciclastic source may
reflect changing cratonal drainage patterns.
Another general trend encompassed by the first four Grand Cycles involves a
progressively greater percentage of intertidal lithofacies in each successive Grand
Cycle. The first two Grand Cycles have almost no intertidal lithofacies, whereas the
third contains intertidal lithofacies in both mixed and clean carbonate intervals. The
fourth Grand Cycle contains intertidal facies in all three lithofacies groups, and
is especially noteworthy for its nonmarine shale/mudstone red beds interpreted
as paleosols. The fifth Grand Cycle, Desert Range-Papoose Lake, has not been
examined above the basal few meters of the Papoose Lake Member of the Bonanza
King Formation. In the Desert Range Limestone, intertidal lithofacies are found
in both mixed and carbonate intervals. The basal part of the Papoose Lake is
subtidal, but published literature on the member does not differentiate subtidal from
intertidal(Barnes and Palmer, 1961; Barnes and Christiansen, 1967; Gans, 1974;
Kepper, 1981; B urchfiel and others, 1982). Lithology aside, intertidal lithofacies
are present lower in this Grand Cycle than in the bottom three, and at comparable
levels to the fourth. Hence, the trend of increasing proportion of intertidal deposits
may continue into the fifth Grand Cycle.
Two of the Grand Cycles, the Zabriskie Quartzite-Eagle Mountain Shale-

Thimble Limestone and the Pahrump Hills Shale-Jangle Limestone, are interpreted
to have siliciclastic SMST deposits at their bases. The presence of the SMST places
these two cycles into an uncommon group of Grand Cycles that includes the Gog
Sandstone and the Arctomys-Waterfowl Grand Cycle of Aitken (1978). The other
three Grand Cycles, the Echo Shale-Gold Ace Limestone, the Pyramid Shale-
Red Pass Limestone, and the Desert Range Limestone-Papoose Lake Member of
the Bonanza King Formation, either have no SMST deposits or very thin SMST
deposits that cannot be distinguished from the thicker siliciclastic TST deposits.
This results in placement of the transgressive surface on or very close to the
sequence boundary at the base of the Grand Cycle. These three Grand Cycles
resemble most other Grand Cycles described in the literature (see Table 2).
7. Sequence Stratigraphic Model of Grand Cycles
7.1. INTRODUCTION
Aitken's (1966, 1978) model of Grand Cycles was originally defined for Cambrian
rocks in the southern Canadian Rocky Mountains and most other reported Grand
Cycles were deposited along North American passive margins during the long-term
eustatic rise associated with the Sauk transgression (Sloss, 1963). Grand Cycles are
depositional couplets composed of two half-cycles, one dominantly of carbonate
rocks overlying another dominantly of siliciclastic rocks, developed on a scale of
tens to hundreds of meters in thickness (Aitken, 1966, 1978, 1981).
There are no implicit or explicit reasons to restrict Grand Cycle deposition to
the Cambrian and several of the papers listed in Table 2 discuss Ordovician Grand
Cycles. Grotzinger (1986a,b) applied the concept of Grand Cycles to Paleopro-
terozoic rocks of the Rocknest Formation. Grand Cycles in the Carrara Formation
are approximately 106 years duration or third-order cycles (Adams, 1993). The
model that follows is developed for Cambrian-style, third-order Grand Cycles.
Early papers on Grand Cycles (Aitken, 1966, 1978; Palmer and Halley, 1979)
favored a mechanism of variable subsidence rates to explain the two different litho-
logic half-cycles. Subsequent papers (Aitken, 1981; Mount and Rowland, 1981;
Palmer, 1981; Palmer and Nelson, 1981; Bond and Kominz, 1984; Grotzinger,
1986a,b; Bond and others, 1988; Demicco and Spencer, 1990; Demicco and oth-
ers, 1991) have recognized that variable rates of eustatic sea level change could
also be invoked as a mechanism, with most favoring a eustatic origin for Grand Cy-
cles (Grotzinger, 1986a,b; Chow and James, 1987; Bond and others, 1988, 1989;
Aitken, 1989; Bond and Kominz, 1991; Mount and others, 1991). Cowan and
James (1993) presented a model that relates Grand Cycle deposition to changes
in accommodation and climatic changes. Changes in accommodation affect the
robustness of carbonate production: the carbonate factory is robust during the LST
and is less so during the HST (Cowan and James, 1993). Influx of siliciclastics
is related to climatic change, not to accommodation (Cowan and James, 1993).
Many papers (see Table 2) recognized that the siliciclastic half-cycle was deposited
318 R.D. Adams
under conditions of increased rates of accommodation (relative sea level rise due
to subsidence and/or eustasy) compared to the carbonate half-cycle.
Most previous workers referenced above and in Table 2 agreed on facies inter-
pretations, facies relationships, recognition of meter-scale shoaling cycles within a
larger Grand Cycle, and the occurrence of greater rates of accommodation (relative
sea level rise) during deposition of the siliciclastic half-cycle relative to the carbon-
ate half-cycle, and this study also agrees with these observations. Some anomalous
Grand Cycles are recognized with supratidal facies in the siliciclastic half-cycle
and have generated disagreement over their sea-level history; e.g., Demicco and
Spencer (1990) and Demicco and others (1991) versus Bond and Kominz (1991).
Disagreement between previous workers has also arisen over the interpretation of
the eustatic history recorded in Grand Cycles and assignment of sequence strati-
graphic terminology to Grand Cycles (cf. Mount and Rowland, 1981; Chow and
James, 1987; Demicco and Spencer, 1990; Bond and Kominz, 1991; Demicco
and others, 1991; Cowan and James, 1993). A new sequence stratigraphic model
of Grand Cycle development encompasses both common and anomalous Grand
Cycles and is summarized here. However, most studies, including an earlier study
on the Carrara Formation (Halley, 1974; Palmer and Halley, 1979), have ascribed
deposition of the siliciclastics to high stands of sea level and deposition of part
or all of the carbonates to low stands of sea level (Aitken, 1966, 1989; Chow and
James, 1987; Bond and others, 1988, 1989; Bond and Kominz, 1991; Cowan and
James, 1993).
7.2. COMPREHENSIVE MODEL
The proposed comprehensive model (Figs. 7 and 8) differs from most previous
attempts to reconcile Grand Cycles and sequence stratigraphy in that it equates
high rates of accommodation during siliciclastic deposition with the TST , and
the subsequent decrease in rate of accommodation with the HST. In a common
Grand Cycle the SMST is either very thin and not differentiatable from the lower
TST or is missing altogether (Fig. 8), e.g. the Echo Shale-Gold Ace Limestone
Grand Cycle (Fig. 3) or the Pyramid Shale-Red Pass Limestone Grand Cycle
(Fig. 4). Anomalous Grand Cycles are interpreted to have unusually thick SMST s
which incorporate intertidal and supratidal facies in most or all of the siliciclastic
half-cycle (Fig. 8), e.g., the Zabriskie Quartzite-Eagle Mountain Shale-Thimble
Limestone Grand Cycle (Fig. 3), the Pahrump Hills Shale-Jangle Limestone Grand
Cycle (Fig. 5), or the Arctomys-Waterfowl Grand Cycle (Aitken, 1981). Lastly,
boundaries of Grand Cycles are considered to be equivalent to sequence boundaries
(Table 2).
The comprehensive sequence stratigraphic model for Grand Cycle deposition
equates a Grand Cycle with a sequence (commonly a third-order cycle with a dura-
tion of approximately 106 years). Grand Cycle boundaries are sequence boundaries,
normally Type 2, but Type 1 boundaries are possible and may be the cause of mi-
nor erosion noted on several Grand Cycle boundaries (Aitken, 1978; Mount and
others, 1991; this paper). Siliciclastic sediments were initially spread across the
lagoon and possibly over the carbonate bank by the decrease in accommodation
TABLE 2
A summary of Grand Cycle models and interpretations arranged chronologically. The first column is of an
ideal Grand Cycle, based on the literature. In the last column is the model presented in this paper. Note the
~
large amount of overlap for each systems tract reflecting the variability of Grand Cycles. Also note that a f
Type I sequence boundary may occur in some Grand Cycles, but that none were identified in the Carrara ~
Formation Grand Cycles.
IDEAUZED GRAND CYCLE I AITKEN 1978 PALMER & MOUNT & GROTZINGER CHOW & BOND.
~
,~"'~ j HALLEY 1979 ROWLAND 1986a, b JAMES 1987 KOMINZ,& Q
PULSATORY l1l11NO 0 PULSES OF 1981
CRATON SUBSIDENCE
GROTZINGER
(EUSTASY AS AN VARY RATE OF IIASIN AATEOF EUSrATlCSEA EUSTASY
1988
ALTERNATE SUBSIDENCE OR OF LEVEL CHANGE
sCT
MECHANISM) SEA LEVEL RISE CHANGES IN RATE
OF EUSTATIC RISE & Of EUSTAS"f' ~"
BASIN suaaDEMCE t:>
::J
C)
SHARP
t3
::J
CONTACT Sllliinn"Po"...blY - - CWbOnated."',.,..;;; e;;;..P;;(poHibl." - - - - - -
1=1 'I . . ~,o ... by ,. . . 10" __ on ............ -.-... ' .... '01' .',,"'on, Q..
.po"'on 01 eyel. In .... of 'onorwte
_poeilion t.ctory'
r1~1
of pwilldlll
oerborwtee
P.rlUdel
.rbonet.
Q
v~~:-r!!-:r.!'r...':'. Slow... me of ...
".'av. rI.. levelrI ... Inllll1l" Jcnon... IhoIII exp.,ch ~
.... on&l.. .... - en
V.,.IhtIII_to ~.r... of ~:::::j..:':. O'o...... nt.ot ... lw. .xpo....
In_'.....y ..... "on
-I . . . . . ""'.,o.o'.". ...... _"'... I 'n ... " ''''''....... 01
S"
0"-"* MI'.cyole upper . . ._m p.-t 01
c.rb0fWle memben * Q
rapid bulnw.d apreect
Progr.cs. SUbtld.. of .."c,cl ....c. D_ rll. of reid'll.
--.- j --. --- .ublllcfenc.w .....I.tlc
-aa;
Q SHI...,elri.. ..,
~ -= - I ~ ,------ t.;"
- - - - - - - - - - - - rnor.r.pdlyI 4 - - -
:ia:
CJ
Deer Nd tlNflgenoue
d"-='":"w:rrL~
-:'=1 depoeItion kHp.
~
Q.
due to prevlou.
LLI d'po.ltton
Hlgl,..n.rg ri In Wlde.preect ..lIclcl....c .:::r.C::~nL.t1 (

~
MeI..
':nt.. ....~!':.... dltpo.. tlon ....... deI y ~
~ ~=1.~~::.1. Incr fIhI 01 Inere... r of ... I...,eI Shllflow eublldel I .... eItv~~1.:!.... 01 ~
or 0'''' rl .. ord.c...... r... of tlRtflltdlpoaition 2
, ..."v.Y~~&mu=
... lloor wi ncr..
tllf1'lgtnoutl dlmlnt
r-
lnor rate of
rlN (...... deepentl
IV"".,ct.nc. Incr r. of r.....v.
1..., .. 'III
--
A more 9IlIclcl...lc depolillon Acc...,etton In r 01
gr.dUll oontInUl. rt. . .hul1l off c_bonsl.
Sudden .......g,... Pfoductlon wi
...... P
CONTACT
--""- 1-
Aood''''' .w"', duo 10 .'Mg,'"
down....:!JOPP..!!. ...!.......!ion
.'on. ,",, ,
2e!d ___
---
.!!!P'd~l~ pI~onn
nO",_ not oubm".....
_
0 ..... Ono"
_ 1
.'n __
_I "ploo.mon' ohoI.
.e...byO!!....
LID!!.. RII.o' ... ..t!!..gh IJ.)

.....t.;~
10
~ 1. Aitken 1981: Attribules mechanism 10 euslatic variation in rale andlor direction of sea level fluctuation. 2. Palmer 1981: Euslatic for some. bUI nol all
based on correlatibilily.
-
TABLE 2 Continued VJ
Comparison of Grand Cycle Models (right halO ~
BOND, AITKEN 1989 IDEMICCO & I BOND & DEMICCO, MOUNT, HUNT, GREEN, & COWAN & ADAMS 1993 &
KOMINZ, SPENCER 1990 KOMINZ 1991 SPENCER, DIENGER 1991 JAMES 1993 THIS PAPER
STECKLER, & EUSTATIC PULSES (Abstract)
OF SEA LEVEL RISE
WATERS,& EUSTASY
GROTZINGER EUSTASY INflUENCE OF INFLUENCE OF
NON-EUSTATIC SEA
CLOYD 1991 flUCTUATIONS IN
I ACCOMMODATION
1989 LEVEL FLUCTUATION MODEL 1 MOOEL2 ACCOMMODATION & FLUCTUATION ON
RELATIVE SeA LEVEL
FLUCTUA'TION CliMATE ON SEDlMENTA'TlON RATES
EUSTASY ED MENTATION RATES
~2 qu.nc. Type 2 ~nc. .!)'pI_ 2 qu.nc.

-. T";;;;-::7rt':~-;;' 1--- -1- boundlrY&'& trlfl'- boundary & A. tr...- boundery
g"".' .Trt.ol gr .:Trtec.
Tr.....gr. . . . SYI""'I
.... l1 l1
~.~.nc.
[ tc.ep-up
Accomrnoddon retto
boundar,
Slower.ubslct.no, HOT
d.c,.. but
Not dlscud ..dilm.ntdon,"'.c L T
ICcommodlltion ...... H ........ ("r. . ., H_
T,enr.!t~;:~ItHn.
I
CaIc"-
Incr.... cwbonatto ....
~tl~;:~'::!~Oc~::. ''''U'~f
~o ..... rtmNglMto 1
up.-.d Into I~n Tr_gr.,.' IY""n,
Ir.c:t(mwln'tt.polll")
F.Ung w.tetlc "rent Sloweu.t.tlcrf..

~c~ ... ~.e.
.u....,'~I " ...". I IType 2 "qu"'CI Slrne. bolXldwy St~ 1--1 _
boundlltY
,,' - ' r-
, ', - 1
A
TSr
SllIciojNtic mud IWItpt

lOre lagoon TST
Accomrnoddon r~.
TIfT .... ,.. pI.y.-
Incr but mlrglnll m.1ne tid" HST
Low.t.nd Hdlmenl.tlon rltl I>
eccommodltlon rM .
.... I
Hlgh.1Ind or low.tlWld TST
(P285)
Subtldlll IIIllolel . .tlcl
HSf
S"'' "r S ST
(Often thin
Du:~:::~~gC::P.~ 01
aublld.nu" rllflld &
.out. 0" cllf'bon.t, I Abrupt ""Inl rl.. &
rJT
M
-.:!.po~n _ ~".tr~ ~ Typ.12 qu.nc ~12 qt.nC.

I- - - - boundery& boundwy& boundery ~
tr.-.g........urf.c. tr_g,....Iv.lUrlec.
1:~. ~
>
fr
s
(/l
A. COMPLETE (ANOMALOUS) GRAND CYCLE
BASINWARD CRATONWARD
B. INCOMPLETE (COMMON) GRAND CYCLE
BASINWARD CRATONWARD
Intertidal
Carbonat as
Subtidal
Carbonates ~- -- MIxed
_: _: Carbonates &
r. . .':) Subtidal
.\,::\,;:. Siliciclastics
Nonmarine
SIIlclclaslics
- - Siliciclastics --'.'.'.'.'
Fig. 8. Diagrams showing the two varieties of Grand Cycles. A) A complete Grand Cycle with a
well developed shelf-margin systems tract (SMST ) with nonmarine and marine lithofacies, as well
as a transgressive systems tract (TST ) and a highstand systems tract (HST ). The presence of the
SMST is uncommon and complete Grand Cycles have been called anomalous Grand Cycles. B) An
incomplete Grand Cycle with the SMST thin to nonexistent. This is the common expression of a
Grand Cycle. See text for elaboration and Table 2 for a summary of published models of Grand
Cycles.
322 R.D. Adams
(fall in relative sea level) associated with development of the sequence boundary
(Fig. 7). Carbonate production was either shut off or severely diminished by the
decrease in accommodation and/or the spread of siliciclastics. The SMST may be
present, though in most cases if present it is apparently quite thin and easily over-
looked (Fig. 7). However, many of the anomalous siliciclastic half-cycles in Grand
Cycles with an abundance of intertidal to nonmarine facies may be interpreted as
thick SMST deposits: e.g., the Zabriskie Quartzite-Eagle Mountain Shale-Thimble
Limestone Grand Cycle and the Pahrump Hills Shale-Jangle Limestone Grand Cy-
cle, or the Arctomys-Waterfowl Grand Cycle (Aitken, 1966, 1981; Demicco and
Spencer, 1990; Bond and Kominz, 1991; Demicco and others, 1991) and the Gog
Group of the Canadian Rockies (Aitken, 1968,1978). Commonly, the transgressive
surface at the base of the TST either overlies the sequence boundary or is in close
proximity to the boundary. In the common case, much or all of the siliciclastic
half-cycle was deposited during the time of maximum rates of accommodation,
the TST , and recorded retrogradational stacking of meter-scale shoaling cycles in
the siliciclastic half-cycle (Figs. 7 and 8). It is hypothesized that coincident with
the cratonward movement of the siliciclastic sediments the carbonate bank was
reestablished in its most basin ward locations as the water column cleared. During
the time of maximum rates of accommodation, the carbonate sediments were in a
catch-up mode (Kendall and Schlager, 1981; Schlager, 1981). As the siliciclastics
retreated cratonward, the carbonate bank slowly expanded cratonward with its rate
of progradation increasing as the rate of accommodation slowed, i.e. as the TST
passed into the HST. The carbonate bank achieved its greatest expansion during
the HST, and may have prograded basinward as well as cratonward. In the late
HST , siliciclastics may have prograded towards the bank sufficiently to begin
to encroach upon the area of carbonate deposition. The initiation of the decrease
in accommodation (fall in relative sea level) also may have reduced the area of
the subtidal carbonate factory, slowing the rate of carbonate sediment production
and progradation. Because siliciclastic sediment production would not be slowed
by decreasing accommodation (falling relative sea level), the siliciclastics could
have prograded faster and this double response to decreasing accommodation may
explain the encroachment of siliciclastics over carbonates. Encroachment would
have further decreased the area of the carbonate factory.
8. Summary
The Lower to Middle Cambrian mixed carbonate-siliciclastic Carrara Formation,

deposited on the western passive margin of North America, records complex
interplay between long-term eustatic sea level rise, long-term passive margin sub-
sidence, shorter-term (third-order?) fluctuation in accommodation, and varying
sediment production and delivery rates of both a cratonic source of siliciclastics
and an offshore source of carbonates. Prior lithostratigraphic analysis of the Carrara
Formation identified nine members of the formation and those member designa-
tions have been used in this study. Sequence stratigraphic principles have been
followed in describing and correlating stratigraphic sections in eastern California
and southern Nevada, and five Grand Cycles have been recognized.
Grand Cycles are couplets, ranging in thickness from less than 100 m to over
200 m, of underlying siliciclastic sediments overlain by carbonate sediments. Top
and bottom boundaries of Grand Cycles are sharp and relatively isochronous within
resolution of trilobite biostratigraphy, whereas the internal boundary between the
lower siliciclastic half-cycle and the upper carbonate half-cycle is very gradational
and diachronous. Three complete Grand Cycles are totally contained within the
Carrara Formation and the basal and uppermost members of the formation form
parts of two additional Grand Cycles. The basal Grand Cycle is completed by
addition of the uppermost fine-grained siliciclastics above the last carbonates of
the Upper Member of the Wood Canyon Formation and the Zabriskie Quartzite.
The uppermost Grand Cycle is completed by the addition of part or all of the
Papoose Lake Member of the overlying Bonanza King Formation.
A new sequence-stratigraphic model of Grand Cycle deposition places the se-
quence boundary at the Grand Cycle boundary which is characterized by the sharp
contact between the upper carbonate half-cycle of the preceding Grand Cycle and
the lower siliciclastic half-cycle of the succeeding Grand Cycle. The sequence
boundary is a Type 2 sequence boundary in the Carrara Formation Grand Cycles,
but there is no restriction against development of a Type 1 boundary in other Grand
Cycles. The SMST is dominantly siliciclastic, and may be well developed or thin
to nonexistent. The basinward influx of siliciclastics associated with the SMST
may have overrun the offshore carbonate bank or may have displaced it farther
basinward. There is little or no evidence of erosion, diagenesis, or karst devel-
opment to indicate that the bank was subaerially exposed for any period of time.
Some Grand Cycles, described as anomalous due to the presence of intertidal to
nonmarine sediments with progradational stacking of parasequences and facies in
the lower half-cycle, have generated much discussion and disagreement over the
years. Those anomalous basal intervals are interpreted to be thick SMST s based on
the presence of two such units in Carrara Formation Grand Cycles. Thin or miss-
ing SMST s are the normal condition for Grand Cycles and three of the Carrara
Formation Grand Cycles are of this type. Thin SMST s would be hard to recog-
nize and differentiate from similar lithofacies of the overlying TST. The TST was
deposited during the period of maximum rate of accommodation associated with
increasing accommodation. In many sections measured through Grand Cycles, the
TST is dominantly siliciclastic with a retrogradational stacking of parasequences
and facies. During the retrogradation of siliciclastics, the carbonate bank tended to
become re-established and/or prograde cratonward. This coexistence of carbonate
and siliciclastic sources results in a transitional internal boundary between lower
and upper half-cycles. The complementary reciprocating movement of carbonates
and siliciclastics during the TST produced the marked diachroneity of the internal
boundary. As accommodation reached its maximum, the rate of accommodation
development decreased facilitating progradation of both siliciclastics and carbon-
ates during the HST. The predominance of carbonates in the upper half-cycle of
Grand Cycles indicates that proximity to the more basinal parts of the lagoon and
higher rates of carbonate progradation tended to allow for deposition of a greater
proportion of carbonate sediments. More cratonal positions of the lagoon recorded
324 R.D.Adams
increased concentration of siliciclastics in the HST. In this model, there is only

one sequence per Grand Cycle.
Acknowledgments
This paper is part of a Ph.D. thesis done under John P. Grotzinger at the Mas-
sachusetts Institute of Technology, with funding from National Science Founda-
tion Grant EAR-8916870 to John Grotzinger and John Southard, National Science
Foundation Grant EAR-9058 I 999 to John Grotzinger, and grants from Exxon
Production Research Company and Union Pacific Resources Company to John
Grotzinger. Reviews by Dave Osleger, Pete Palmer, staff at the Utah Geological
Survey, and Chris Kendall improved and clarified the manuscript. Any errors and
omissions remain the responsibility of the author.
References
ADAMS, R. D., 1993, Sequence-stratigraphic analysis of mixed carbonate-siliciclastic
Cambrian sediments, Carrara Formation, southwest Basin and Range, California and
Nevada: Unpublished Ph. D. thesis, Massachusetts Institute of Technology, Boston, MA,
750p.
ADAMS, R. D. and GROTZINGER, J. P., submitted, Lateral continuity of fifth-order
and fourth-order cycles in the middle Cambrian Jangle limestone member, Carrara
formation, death valley region, California: Journal of Sedimentary Research.
AITKEN, J. D., 1966, Middle Cambrian to Middle Ordovician cyclic sedimentation, south-
ern Rocky Mountains of Alberta: Bulletin of Canadian Petroleum Geology, v. 14, p.
405-441.
AITKEN, J. D., 1968, Cambrian sections in the easternmost southern Rocky Mountains
and the adjacent subsurface, Alberta: Geological Survey of Canada, Paper 66-23, 96 p.
AITKEN, J. D., 1978, Revised models for depositional Grand Cycles, Cambrian of the
southern Rocky Mountains, Canada: Bulletin of Canadian Petroleum Geology, v. 26, p.
515-542.
AITKEN, J. D., 1981, Generalizations about Grand Cycles: in Taylor, M. E., ed., Short
Papers for the Second International Symposium on the Cambrian System, Golden, CO,
United States Geological Survey, Open-File Report 81-743, p. 8-14.
AITKEN, J. D., 1989, Birth, growth and death of the Middle Cambrian Cathedral carbonate
lithosome, Southern Rocky Mountains: Bulletin of Canadian Petroleum Geology, v. 37,
p.316-333.
ALBERS, J. P. and STEWART, J. H., 1972, Geology and mineral deposits of Esmeralda
County, Nevada: Nevada Bureau of Mines and Geology Bulletin 78, 80 p.
BARNES, H. and CHRISTIANSEN, R. L., 1967, Cambrian and Precambrian rocks of
the Groom district, Lincoln County, Nevada: United States Geological Survey, Bulletin
1244-G, 34 p.
BARNES, H., CHRISTIANSEN, R. L. and F. M. BYERS, J., 1962, Cambrian Carrara
Formation, Bonanza King Formation, and Dunderberg Shale east of Yucca Flat, Nye
County, Nevada: United States Geological Survey, Professional Paper, v. 450-D, p.
D27-D31.
BARNES, H. and PALMER, A. R., 1961, Revision of stratigraphic nomenclature of Cam-
brian rocks, Nevada Test Site and vicinity, Nevada: United States Geological Survey,
Professional Paper, v. 424-C, p. CIOO-CI03.
BATES, E. E., 1965, Stratigraphic analysis of the Cambrian Carrara Formation, Death
Valley region, California-Nevada: Unpublished Master's thesis, University of California
at Los Angeles, Los Angeles, CA, 182 p.
BOND, G. C. and KOMINZ, M. A., 1984, Construction of tectonic subsidence curves for
the early Paleozoic miogeocline, southern Canadian Rocky Mountains: Implications for
subsidence mechanisms, age of breakup, and crustal thinning: Geological Society of
BOND, G. C. and KOMINZ, M. A., 1991, Some comments on the problem of using vertical
facies changes to infer accommodation and eustatic sea-level histories with examples
from Utah and the southern Canadian Rockies, in Franseen, E. K., Watney, W. L.,
Kendall, C. G. S. C. and Ross, W., ed., Sedimentary Modeling: Computer simulations
and methods for improved parameter definition: Lawrence, KS, Kansas Geological
Survey, Bulletin 233, p. 273-291.
BOND, G. c., KOMINZ, M. A. and GROTZINGER, J. P., 1988, Cambro-Ordovician
eustasy: Evidence from geophysical modelling of subsidence in Cordilleran and Ap-
palachian passive margins, in Kleinspehn, K. L. and Paola, C., ed., New Perspectives in
Basin Analysis: New York, Springer-Verlag, p. 129-160.
BOND, G. c., KOMINZ, M. A., STECKLER, M. S. and GROTZINGER, J. P., 1989, Role
of thermal subsidence, flexure, and eustasy in the evolution of early Paleozoic passive-
margin carbonate platforms, in Crevello, P. D., Wilson, J. L., Sarg, J. F. and Read, J. F.,
ed., Controls on carbonate platform and basin development: Tulsa, OK, SEPM Special
Publication 44, p. 39-61.
BURCHFIEL, B. C., HAMILL, G. S., IV and WILHELMS, D. E., 1982, Stratigraphy
of the Montgomery Mountains and the northern half of the Nopah and Resting Spring
Ranges, Nevada and California: The Geological Society of America, Map and Chart
Series MC-44, 2 sheets, 9 p.
CHOW, N. and JAMES, N. P., 1987, Cambrian Grand Cycles: A northern Appalachian
perspective: Geological Society of America Bulletin, v. 98, p. 418-429.
CORNWALL, H. R. and KLEINHAMPL, F. J., 1961, Geology of the Bare Mountain
Quadrangle, Nevada: United States Geological Survey, Quadrangle Map GQ-157.
COWAN, C. A. and JAMES, N. P., 1993, The interactions of sea-level change, terrigenous-
sediment influx, and carbonate productivity as controls on Upper Cambrian Grand
Cycles of western Newfoundland, Canada: Geological Society of America Bulletin, v.
105, p. 1576-1590.
DEMICCO, R. V. and SPENCER, R. J., 1990, Depositional environments of the Cambrian
Arctomys Formation, Alberta Rockies: Implications for third-order sea level history:
Geological Society Abstracts with Programs 1990 Annual Meeting, v. 22, p. A283.
DEMICCO, R. V., SPENCER, R. J., WATERS, B. B. and CLOYD, K. C., 1991, Two-
dimensional computer models of a Cambrian carbonate shelf, in Franseen, E. K., Watney,
W. L., Kendall, C. G. S. C. and Ross, w., ed., Sedimentary Modeling: Computer
simulations and methods for improved parameter definition: Lawrence KS, Kansas
Geological Survey, Bulletin 233, p. 463-472.
GANS, W. T., 1974, Correlation and redefinition of the Goodsprings dolomite, southern
Nevada and eastern California: Geological Society of America Bulletin, v. 85, p. 189-
200.
GOODWIN, P. W. and and ERSON, E. J., 1985, Punctuated aggradational cycles: A general
hypothesis of episodic accumulation: The Journal of Geology, v. 93, p. 515-533.
GROTZINGER, J. P., 1986a, Cyclicity and paleoenvironmental dynamics, Rocknest plat-
form, northwest Canada: Geological Society of America Bulletin, v. 97, p. 1208-1231.
326 R.D. Adams
GROTZINGER, J. P., 1986b, Evolution of Early Proterozoic passive-margin carbonate

platform, Rocknest Formation, Wopmay Orogen, Northwest Territories, Canada: Journal
of Sedimentary Petrology, v. 56, p. 831-847.
HALLEY, R. B., 1974, Repetitive carbonate bank development and subsequent terrigenous
inundation; Cambrian Carrara Formation, southern Great Basin: Unpublished Ph.D.
thesis, State University of New York at Stoney Brook, 377 p.
JAMES, N. P., 1984, Shallowing-upward sequences in carbonates, in Walker, R. G., ed.,
Facies models (Second Edition): Geoscience Canada, Reprint Series 1, Toronto, Ontario,
Geological Association of Canada, p. 213-228.
JERVEY, M. T., 1988, Quantitative geological modeling of siliciclastic rock sequences
and their seismic expression, in Wilgus, C. K., Hastings, B. C., Posamentier, H. P.,
Van Wagoner, J. c., Ross, C. A. and Kendall, C. G. S. c., ed., Sea-level changes: an
integrated approach: Tulsa, Oklahoma, SEPM Special Publication 42, p. 47-69.
KENDALL, C. G. S. C. and SCHLAGER, w., 1981, Carbonates and relative changes in
sea level: Marine Geology, v. 44, p. 181-212.
KEPPER, J. C., 1981, Sedimentology of a Middle Cambrian outer shelf margin with
evidence for syndepositional faulting, eastern California and western Nevada: Journal
of Sedimentary Petrology, v. 51, p. 807 -82l.
LEVY, M. and CHRISTIE-BLICK, N., 1989, Pre-Mesozoic palinspastic reconstruction of
the eastern Great Basin (western United States): Science, v. 245., p. 1454-1462.
LEVY, M. and CHRISTIE-BLICK, N., 1991, Tectonic subsidence of the early Paleo-
zoic passive continental margin in eastern California and southern Nevada: Geological
LONGWELL, C. R., PAMPEYAN, E. H., BOWYER, B. and ROBERTS, R. J., 1965,
Geology and mineral deposits of Clark County, Nevada: Nevada Bureau of Mines and
Geology Bulletin 62, 218 p.
MITCHUM, R. M." JR., VAIL, P. R. and THOMPSON, S., III, 1977, Part Two: The
depositional sequence as a basic unit for stratigraphic analysis, in Payton, C. E., ed.,
Seismic stratigraphy - applications to hydrocarbon exploration: AAPG Memoir 26,
Tulsa Oklahoma, American Association of Petroleum Geologists, p. 53-62.
MOUNT, J. E, 1984, Mixing of siliciclastic and carbonate sediments in shallow shelf
environments: Geology, v. 12, p. 432-435.
MOUNT, J. E, HUNT, D. L., GREENE, L. R. and DIENGER, J., 1991, Depositional
systems, biostratigraphy and sequence stratigraphy of Lower Cambrian Grand Cycles,
southwestern Great Basin, in Coopcr, J. D. and Stevens, C. H., ed., Paleozoic Paleogeog-
raphy of the Western United States - II: Los Angeles CA, The Pacific Section SEPM, p.
209-226.
MOUNT, J. E and ROWLAND, S. M., 1981, Grand Cycle A (Lower Cambrian) of the
southern Great Basin: A product of differential rates of relative sea-level rise, in Taylor,
M. E., ed., Short Papers for the Second International Symposium on the Cambrian
System, Golden, CO, United States Geological Survey, Open-File Report 81-743, p.
143-146.
PALMER, A. R., 1981, On the correlatability of Grand cycle tops, in Taylor, M. E.,
ed., Short Papers for the Second International Symposium on the Cambrian System:
Open-File Report 81-743, Golden, CO, United States Geological Survey, p. 156-159.
PALMER, A. R. and CAMPBELL, D. P., 1976, Biostratigraphic implications of trilobite
biofacies: Albertella Zone, Middle Cambrian, western United States, in Robison, R.
A. and Rowell, A. J., ed., Paleontology and depositional environments: Cambrian of
western North America: Provo, UT, Brigham Young University, Geology Studies 23:
Part 2,p. 39-50.
PALMER, A. R. and HALLEY, R. B., 1979, Physical stratigraphy and trilobite biostratig-
raphy of the Carrara Formation (Lower and Middle Cambrian) in the southern Great
Basin: United States Geological Survey, Professional Paper 1047, 131 p.
PALMER, A. R. and NELSON, C. A., 1981, Lower and Middle Cambrian stratigraphy of
the southern Great Basin, California and Nevada: in Taylor, M. E. and Palmer, A. R., eds.,
Guidebook for Field Trip I: Cambrian stratigraphy and paleontology of the Great Basin
and vicinity, western United States, Second International Symposium on the Cambrian
System, Golden, Colorado, United States Geological Survey and the Subcommission on
Cambrian Stratigraphy, Commission on Stratigraphy, International Union of Geological
Sciences, p. 1-49.
PALMER, A. R. and ROWLAND, S. M., 1989, Early Cambrian stratigraphy and pale-
ontology, southern Great Basin, California-Nevada: in Taylor, M. E., ed., Field Trip
Guidebook Tl2S: Cambrian and Early Ordovician stratigraphy and paleontology of
the Basin and Range Province, western United States, 28th International Geological
Congress, Washington, D. c., American Geophysical Union, p. 17-27.
POSAMENTIER, H. w., JERVEY, M. T. and VAIL, P. R., 1988, Eustatic controls on clastic
deposition I-conceptual framework, in Wilgus, C. K., Hastings, B. C., Posamentier, H.
P., Van Wagoner, J. C., Ross, C. A. and KendaIl, C. G. S. C., ed., Sea-level changes: an
integrated approach: Tulsa, Oklahoma, SEPM, Special Publication 42, p. 109-124.
POSAMENTIER, H. W. and VAIL, P. R., 1988, Eustatic controls on clastic deposition II -
sequence and systems tract models, in Wilgus, C. K., Hastings, B. C., Posamentier, H.
P., Van Wagoner, J. c., Ross, C. A. and Kendall, C. G. S. c., ed., Sea-level changes: an
integrated approach: Tulsa, Oklahoma, SEPM, Special Publication 42, p. 12S-lS4.
PRAVE, A. R., 1992, Depositional and sequence stratigraphic framework of the Lower
Cambrian Zabriskie Quartzite: Implications for regional correlations and the Early Cam-
brian paleogeography of the Death VaIley region of California and Nevada: Geological
Society of America Bulletin, v. 104, p. SOS-SIS.
SARG, J. F., 1988, Carbonate sequence stratigraphy, in Wilgus, C. K., Hastings, B. C.,
Posamentier, H. P., Van Wagoner, J. C., Ross, C. A. and KendaIl, C. G. S. C., ed., Sea-
level changes: an integrated approach: Tulsa, Oklahoma, SEPM, Special Publication 42,
p. ISS-181.
SCHLAGER, W., 1981, The paradox of drowned reefs and carbonate platforms: Geological
Society of America BuIletin, v. 92, p. 197-211.
SIGNOR, P. W. and MOUNT, J. F., 1989, Paleontology of the Lower Cambrian Waucoban
Series in eastern California and western Nevada: in Christie-Blick, N. and Levy, M.,
eds., Field Trip Guidebook T331: Late Proterozoic and Cambrian tectonics, sedimenta-
tion, and record of metazoan radiation in the western United States: 28th International
Geological Congress, Washington, D. c., American Geophysical Union, p. 47-S3.
SLOSS, L. L., 1963, Sequences in the Cratonic Interior of North America: Geological
Society of America BuIletin, v. 74, p. 93-113.
SLOSS, L. L., 1988, Tectonic evolution of the craton in Phanerozoic time, in Sloss, L.
L., ed., Sedimentary Cover-North American Craton: The Geology of North America,
Boulder, CO, The Geological Society of America, v. D-2, p. 2S-S1.
STEWART, J. H., 1965, Precambrian and Lower Cambrian formations in the Last Chance
Range area, Inyo County, California: United States Geological Survey, Bulletin, v.
1224-A, p. 60-70.
STEWART, J. H., 1970, Upper Precambrian and Lower Cambrian strata in the southern
Great Basin, California and Nevada: United States Geological Survey, Professional
Paper 620, 206 p.
328 R.D.Adams
STEWART, J. H., 1991, Latest Proterozoic and Cambrian rocks of the western United
States-an overview, in Coopcr, J. D. and Stevens, C. H., ed., Paleozoic paleogeography
of the western United States-II: 1, Pacific Section of the SEPM, p. 13-37.
STEWART, J. H. and BARNES, H., 1966, Precambrian and Lower Cambrian formations
in the Desert Range, Clark County, Nevada: United States Geological Survey, Bulletin,
v. 1244-A, p. A35-A42.
STEWART, J. H. and SUCZEK, C. A., 1977, Cambrian and latest Precambrian paleo-
geography and tectonics in the western United States, in Stewart, J. H., Stevens, C. H.
and Fritsche, A. E., ed., Paleozoic paleogeography of the western United States: Pacific
Coast Paleogeography Symposium 1: Pacific Section of SEPM, p. 1-17.
TSCHANZ, C. M. and PAMPEYAN, E. H., 1970, Geology and mineral deposits of Lincoln
County, Nevada: Nevada Bureau of Mines and Geology, Bulletin 73, 188 p.
VAN WAGONER, J. c., MITCHUM, R. M., JR., CAMPION, K. M. and RAHMANIAN, V.
D., 1990, Siliciclastic sequence stratigraphy in well logs, cores, and outcrops: concepts
for high-resolution correlation of time and facies: American Association of Petroleum
Geologists, Methods in Exploration Series 7, 55 p.
VAN WAGONER, J. C., MITCHUM, R. M., JR., POSAMENTIER, H. W. and VAIL, P. R.,
1987, Key definitions of sequence stratigraphy, in Bally, A. W., ed., Atlas of sequence
stratigraphy: AAPG Studies in Geology #27, American Association of Petroleum Ge-
ologists, p. 11-14.
VAN WAGONER, J. c., POSAMENTIER, H. w., MITCHUM, R. M., JR., VAIL, P.
R., SARG, J. E, LOUTIT, T. S. and HARDENBOL, J., 1988, An overview of the
fundamentals of sequence stratigraphy and key definitions, in Wilgus, C. K., Hastings,
B. C., Posamentier, H. P., Van Wagoner, J. C., Ross, C. A. and Kendall, C. G. S. C.,
ed., Sea-level changes: an integrated approach: SEPM, Special Publication 42, Tulsa,
Oklahoma, p. 39-45.
CHAPTER 11
Milankovitch Fluctuations in Sea Level

and Recent Trends in Sea-Level
Change: Ice may not always be the
answer
David K. Jacobs and Dork L. Sahagian
ABSTRACT: Studies of short term sea-level change have emphasized the effects of
climate on the volume of water tied up in continental ice. Here we discuss two different
time scales of non-ice-related storage of water on the continents and their implication
for sea-level change. Human activities generate a flux of water from continental reser-
voirs, such as aquifers and wetlands, to the sea. Our calculations suggest that this flux is
currently in excess of 113 of the sea-level rise rate inferred from tide gauge records. This
observation has implications for interpretation of 20th century sea-level rise. Secondly,
on time scales of orbital variations, climatically driven changes in non-ice-related conti-
nental water storage can produce geologically significant cyclic change in sea-level. This
mechanism of sea-level change may have been the dominant source of Milankovitch fre-
quency eustatic fluctuation during periods of Earth history that lacked continental scale
ice sheets. In a final comment we consider the impact of fluctuation in lake area on
climate models, and on the abundance of modern lake related fauna.
1. Introduction
In the absence of the dramatic changes in ice volume that characterize the Quater-
nary record, changes in intensity of monsoonally driven precipitation provides the
strongest natural driving mechanism for changing water storage on the continents.
Lake basin filling chronologies and global climate models indicate that precession
of the equinoxes had a controlling influence on the waxing and waning of mon-
soonal intensity and areal distribution in the Quaternary. We used this Quaternary
record of changing monsoonal influence as a proxy to evaluate the possible mag-
nitude of sea-level change caused by fluctuations in monsoonal intensity. In this
initial work (Jacobs and Sahagian, 1993) our intent was not to reconstruct events
in the Quaternary, but to assess the limits of continental water storage pertaining to
329
B. U. Haq (ed.).
Sequence Stratigraphy and Depositional Response to Eustatic. Tectonic and Climatic Forcing. 329-366.
1995 Kluwer Academic Publishers. Printed in the Netherlands.
330 D.K. Jacobs, D.L. Sahagian
periods of time when the Earth lacked continental ice. Our calculations suggest that
2 to 8 meters of sea-level change could be produced periodically via Milankovitch
driven changes in water storage in the absence of changing ice volume. The mag-
nitude of this change appears to be sufficient to explain the meter scale eustatic
cycles observed in near-shore carbonate sediments during periods of Earth history
lacking continental ice sheets. For example, Triassic lacustrine deposits document
monsoonal fluctuation of water storage. These fluctuations in continental water
storage correlate with carbonate packages generated by eustatic fluctuation. Spec-
tral analyses indicate a precession dominated Milankovitch signature from lake
deposits as well as in littoral carbonates. Thus the Late Triassic provides strong
support for this ice-free mechanism of Milankovitch sea-level change.
In this work we revisit the Quaternary proxy and the Late Triassic example,
and explore some of the variables critical to this mechanism of documenting
Milankovitch frequency sea-level fluctuation, including: the absence of ice, a
near shore carbonate record of sea-level fluctuation, tectonic generation of large
internally drained basins, and the continental configurations necessary to enhance
monsoonal flow. We then suggest that more work can be done demonstrating
similar phenomena in other geologic periods. Much of the Mesozoic, especially
the Early Cretaceous, as well as the Late Permian and Devonian look promising
in terms of evidence for cyclic sedimentation potentially driven by the monsoonal
eustatic mechanism. We then discuss how human activities may change the amount
of water stored on the continents and thereby influence sea-level. We also briefly
discuss the implications of Quaternary fluctuation in lake area for climate models,
and for conservation biology. The large scale of change in lake environments may
have strongly influenced the evolution and distribution of the modern fauna.
2. Ice
In the Quaternary and late Tertiary changes in ice volume generated sea-level
change at Milankovitch frequencies, and during this period ice sheets reaching
the sea generated ice-rafted debris resulting in diamictite deposits over very broad
areas. Continental glacial deposits or ice-rafted debris provide evidence for conti-
nental ice sheets in the Late Ordovician-Early Silurian as well as the Carboniferous.
However, no strong evidence of this type is available for the Cambro-Ordovician,
much of the Devonian, or the Late Permian through the Early Eocene (Hambrey
and Harland, 1981). In addition, uniformitarian arguments based on the tempera-
ture requirements of modern taxa provide positive evidence for equable climates
at the poles. For example, crocodile fossils are known from Ellesmere island in the
Eocene. Such evidence strongly suggests the absence of glacial-scale ice for the
entire Mesozoic and early Tertiary. However, during the Cambro-Ordovician, De-
vonian, Late Triassic and Early Cretaceous, small scale Milankovitch frequency
sea-level changes are well documented by nearshore carbonate deposits which
show repeated cycles of flooding and subaerial exposure (Fig. 1). These cyclic
changes in sea-level have often led researchers to infer fluctuations in ice volume
during these time periods. However, if Milankovitch-driven fluctuations in the
Sea-Level Change - No Ice? 331
Schematic of Peritidal Carbonate Cycle

Indicating Subaerial Exposure
T
Subtidal-Oolites, Peloides, Thrombolites
Flooding surface-Rip up clasts, Lags

Subaerial exposure-Desication Cracks, Quartz sands, Regoliths
Tidal Flat-Algal Laminates, Stromatolites
Fig. 1. Sedimentary cycles documenting subaerial exposure and subtidal flooding provide the best
documentation of sea-level fluctuation. Such cycles are often well preserved on carbonate platforms.
In such cycles flooding of the platform is often documented by rip up clasts or other lag deposits,
followed by any of a number of subtidal carbonate facies, such as oolites, peloides, ribbon rock or
thrombolitic algal heads. These are usually followed by intertidal stromatolitic or algal laminated
deposits, and then ideally, by features documenting subaerial exposure, such as mud cracks, regolith,
or a quartz sand sheet. Obviously, sea-level also strongly effects sedimentary processes in shallow
marine environments that do not experience exposure, and deep water facies may also be effected
by sea-level change through changing sediment bypass. However, flooding and subaerial exposure
cycles provide the strongest evidence for sea-level fluctuations prior to the Quaternary (after Osleger
and Read, 1991: Goldhammer and others, 1990).
monsoon can also produce sea-level changes, the glacial explanation for rapid
sea-level fluctuation may not always be appropriate.
3. Quaternary Proxy
Monsoonal l circulation results from summertime insolation over large subtropical

and mid-latitude continental areas. During northern hemisphere summer the con-
tinental configuration of Eurasia and Africa generates the largest monsoonal effect
on Earth today, resulting in the deflection of wind patterns and dramatic sum-
mertime precipitation referred to as the Asian monsoon. Solar heating of elevated
surfaces, such as Tibet, accentuate the effect, leading to lower pressure and greater
convection than would occur otherwise (Kutzbach and others, 1993). The orbit of
the Earth is elliptical. When proximity of the Earth and Sun, perihelion, occurs
during northern hemisphere summer the strength of the Asian monsoon dramat-
ically increases. Perihelion last coincided with the northern hemisphere summer
in the early Holocene, roughly 9,000 years ago. Lake basin filling chronologies
and global climate models indicate an early Holocene expansion of monsoonal
precipitation over large areas of North Africa and South Asia (Fig. 2). Much
greater precipitation occurred in these desert regions at that time than occurs to-
day. Perihelion coincides with the northern hemisphere summer solstice at roughly
21,000 year intervals as a consequence of the precession cycle of the Earth's orbit.
Consequently, fluctuations in monsoonal intensity are expected to recur with an
approximately 21,000 year repeat time. Thus, evidence from both modeling and
the early Holocene geologic record document a mechanism by which periodic
expansion of the monsoon could result in fluctuating continental water storage.
However, the relevance of this mechanism depends on the magnitude of sea-level
change it could generate. In particular, were sea-level changes generated by the
mechanism large enough to produce the meter scale eustatic carbonate cycles ob-
served during periods of earth history lacking evidence for continental ice sheets?
To explore the potential size of the effect, we assessed the water storage potential
associated with the region of increased monsoonal precipitation produced in the
early Holocene (Jacobs and Sahagian, 1993).
3.1. EVIDENCE FOR INCREASED EARLY HOLOCENE MONSOONAL PRECIPITATION
The 9,000 year ago COHMAP (1988) model indicates increased monsoonal pre-
cipitation over North Africa and South Asia, roughly 1/4 of the Earth's land surface
area (Fig. 2). Note that much of this area is internally drained (Fig. 3). Increased
precipitation in these regions tends not to flow to the sea, but forms lakes and
increases water storage in aquifers. Several kinds of geologic evidence support
these model results. Lake level chronologies in closed basins provide the strongest
evidence. In a research program given early impetus by the work of Butzer and oth-
ers (1972) strand line deposits in African lakes have been dated using Carbon 14.
IThe term monsoon comes from an arabic root and was originally associated with seasonal wind
shifts on the Indian Ocean. Some authors use the term just to refer to seasonal shifts in wind or
only to the seasonal climate of South Asia. We use the "monsoon" in a more general sense refer to
convection and precipitation associated with continental heating in the summer. We do not restrict
our usage as to time or place, but are most interested in large effects that involve displacment of the
Intertropical Convergence Zone (ITCZ) from the equator
Sea-Level Ch ange-No~ceo? 333
.-
-----_. -------------c-~'.
()
p-Ig. 2. Area of greater p ..

Th'
Th~t ba,I~< 'h~g COH~A~O"""""
IS pattern of gre t reClpnation in th
numbers indicate tah er precipitation is suep early Holocene period f
precipitatio~ an~
".'" ' too I".
pond to the ,"tn,,;" po,"d by,mod' of 0th'"pa"d d
, I (,,.,,wlthl"
'",ub"th'nd;'" of ,"""1993).
ahagian,
I"ftu,,,,,,
..od monsoonal(1988) gmup 10'
60
(l.
~
()~ .
o
.. ;-~
~
Fig 3 A 0
. . reas of'
~frica are curren~nte~nal drainage on the e th
e~~
mfluence in the mternally drained. T::
thot ocou"", ot tlmt;'Ime
HoI.",,",
m lakes,"d
andwooid .~
aq UI ers
tod~y.Note that lar
" regmn< f,11
necessarily havewlthl"
(after Street
~~ .
e,nd
st ored much '"'"
area
and Grove, 1979).
of Eo",I, , nd North
ofthexpanded
of monsoonal
e extra precipitation
Using these and other dating techniques lake level chronologies for a large number
of lake basins have been developed. Virtually all North and Central African Lakes
tabulated by Street and Grove (1979) and Street-Perrott and others (1990) were at
highstand about 9,000 years ago. Before 10,000 years ago and since 5,000 years
ago fewer than 20% of these lakes were at highstand. Much ancillary evidence
supports the increased precipitation and water storage evident from higher lake
levels in the early Holocene. Charophyte algae, now only found in Holarctic lakes,
document the presence of cool oligotrophic lakes in locations which are now some
of the hottest and driest areas on Earth (Kropelein and Soulh~-Marsche, 1991).
Buried channels evident from Landsat images demonstrate runoff in currently arid
regions of the Sahara (McCauley and others, 1986; Breed and others, 1979). An-
cient human activity documents wetter conditions in North Africa and South Asia.
Petroglyphs of giraffes and other organisms typical of savannas are found in the
center of the Sahara and Stone Age artifacts abound in areas where today even
camel caravan cannot travel due to the absence of water (Petit-Maire and others,
1990). Early agrarian civilization flourished in regions of the Indus vaHey that are
desert today.
Evidence from the Atlantic Ocean, as well as from the Arabian, Red and
Mediterranean Seas, documents increased precipitation over an area extending
from North Africa through South Asia. Marine cores containing freshwater algae
and forest pollen document wetter conditions in the Niger drainage (Hooghiemstra,
1989). Cores from the eastern margin of the Arabian Sea contain early Holocene
spikes of mangrove suggesting more freshwater influx in the coastal regions (Van
Campo, 1986). Marine cores from the north-eastern Arabian Sea document greater
upwelling in the early Holocene as well as in earlier Milankovitch generated periods
of increased northern hemisphere summertime insolation. Such upwelling results
from increased strength of monsoonal circulation (Prell and Van Campo, 1986).
In the Red Sea corals from Ethiopia and Aqaba show fluorescent banding dating
to the early Holocene (Klein and others, 1990). Humic acids derived from the
water produce this fluorescence suggesting increased terrestrial runoff. Increased
influx from the Nile generated sapropels in the Mediterranean. These organic
rich sedimentary layers correlate not only with the early Holocene expansion of
the monsoon, but also with model predictions of earlier increases in monsoonal
intensity (Rossignol-Strick, 1983). Thus, many terrestrial and marine observation
confirm the interpretation of expanded monsoonal precipitation inferred from lake
highstand data.
If one restricts the analysis to lake basin chronologies, data from Asia are less
uniformly distributed, and less well documented than those in Africa. However,
several lakes in the region that have been studied had higher water levels in the
early Holocene. Of special interest is Qinghai Lake (Lister and others, 1991)
where lake levels rose rapidly from 10,000 years ago and subsequently declined
after 6,000 years ago. Qinghai Lake's drainage area at the northeast extremity of
the Tibetan Plateau is intimately associated with rivers that flow into Quidam,
Tarim and several large Tibetan basins. Consequently, substantial flow to these
other basins must also have occurred in the early Holocene. In addition, cores from
lake basin sediments in westernmost Tibet also document pluvial conditions in the
TABLE 1
Volumes of some of the larger empty closed basins in the area of expanded monsoonal influence
in early Holocene (after Jacobs and Sahagian, 1993). Volumes were calculated using topographic
data from the Defense Mapping Agency (ETOP05U) and represent the currently unfilled
volume of the basin up to the outflow elevation. East Iran, Afghan-Zabol and one of the basins
in Baluchistan proved to be contiguous and are treated together in the analysis.
Basin Area Volume Sea Level Outflow elevation

(I 010 m2 ) (1013 m3 ) (cm) (m)
1. Tarim 87 39 108 1500 (NE rim)

2. Caspian 232 I3 36 100 (Don River)
3. East Iran 57* 7* 19" 900 (NW)*
4. Afghan-Zabol
5. Baluchistan
6. Chad 67 5.9 16400 (Benue River)
7. Urs Nor 22 5.2 14 1600 (Samaltagatay)Quidam
8. Quidam 16 3.3 9 3500 (Lapiquan)
9. Balkash-Zungar 30 1.9 5.3 400 (Caspian)
10. Esfahan 5.1 5.8 1.6 1800 (NE to E. Iran)
early Holocene (Gasse and others, 1991; Van Campo and Gasse, 1993). Thus, the
available data are consistent with the widespread early Holocene pluvial conditions
in South Asia suggested by the COHMAP (1988) climate models (Fig. 2).
3.2. QUANTIFICATION OF LARGE BASIN VOLUMES
To determine how much water could be stored in the region of expanded monsoonal
influence 9,000 years ago, we first calculated the volume often of the largest basins
in the region. For this analysis we contoured digital topographic data (Table 1), and
calculated the volume of each basin from the bottom of the basin, or current lake
level if a lake was present, to the overflow elevation. The ten basins considered
could hold a volume of water equivalent to 2 meters of sea-level. The Tarim basin
alone, if filled, would lower sea-level I meter.
In addition to these large basins, many other basins of moderate size occur
throughout the region of the expanded early Holocene monsoon. Graben structures
form many large internally drained basins in Tibet. The orogenic belt extending
from Anatolia south through the Zagros and Makran into Baluchistan contains
many internally drained basins of substantial size, such as Lake Tuz and Lake Van in
Turkey and large dry basins further south. Many internally drained tectonic features
are associated with the East African Rift, including basins in Kenya, such as lake
Natron and Lake Turkana, as well as basins extending north through Ethiopia,
ultimately including related structures such as the Dead Sea. Even Lake Tanganyika
was as much as 350 m below highstand during late Pleistocene precipitation minima
(Gasse and others, 1989). Many broad shallow basinal features extend across North
Africa. These include the Quattara depression, Chott Melhrir in Tunisia, and the
Tanzrouft, Taodenni and Quarane regions of Algeria, Mali and Mauritania. These
basins approach in size the smaller basins we examined. Thus many of them may
accommodate volumes equivalent to a few centimeters of sea-level. The majority
of these basins were clearly wetter in the early Holocene when numerous smaller
North African lakes were at highstand (for example,Gasse and others, 1987; Street-
Perrott and others, 1990).
3.3. IMPORTANCE OF GROUNDWATER
A number of field studies link highstands in medium and small lakes with large
aquifer systems. In Lake Natron in the rift valley of Kenya highstand elevation,
documented by dating of stromatolites, appears to be controlled by interaction with
the groundwater table in the local volcanic terrain (Hillaire-Marcel and Casanova,
1987). At Wadi Howar (Kropelein and others, 1991), Selima oasis (Haynes and
others, 1989) and Oyo (Ritchie and others, 1985) in western Sudan, lakes formed
in depressions in porous Mesozoic sedimentary rocks during the early Holocene
pluvial period suggesting the possibility of large groundwater responses associated
with changes in lake level. Presumably this is also true of other basinal oasis
systems such as El Kharg, the Fayum and Dakleh of Egypt. Perhaps most intriguing
are inter-dune lakes in large dune systems. Clear documentation is available that
bodies of water were associated with the Great Western Erg of Algeria (Fontes and
others, 1985; Gasse and others, 1987) and the Grand Erg of Bilma (Grunnert and
others, 1991) in the early Holocene. Landsat images document lacustrine deposits
associated with dune fields throughout North Africa and the Empty Quarter of
Arabia (for example, Breed and others, 1979). This evidence is consistent with
the interpretation that all these dune fields had higher water tables in the early
Holocene. Today, the grass-covered Sand Hills of Nebraska harbor numerous
interdunallakes providing an example of a dune field in a hydrated condition.
Lakes are essentially a surface expression of the ground water table, and many
of the lakes in the region of increased early Holocene monsoonal precipitation
interact with large oasis or dune field aquifer systems. We modeled this complex
system of lake basins and groundwater as a continuous layer of uniform porosity
across the region of increased monsoonal precipitation in the early Holocene
(Jacobs and Sahagian, 1993). The actual water storage reservoir is not uniform
because of variation in porosity and thickness of the hydrated layer. Within the
area of increased precipitation in the early Holocene there are empty basins which
have a porosity of unity; large areas of eolian deposits, such as loess and sand,
with porosities in excess of 0.4; Tertiary fluvial and lacustrine deposits, as well as
Mesozoic sediments with porosities in excess of 0.3; and small areas of basement
rock with porosities of about 0.1. Assuming a conservative average porosity of
0.3 over the area of monsoonal variations, hydration of a layer 65 and 200 meters
thick would lower sea-level 2 and 6 meters, respectively. In combination with the
result from the calculation of volumes of the ten large basins, these calculations
suggest a potential sea-level change of 4 to 8 meters.
3.4. CONTEXT OF THE VOLUME CALCULATIONS
Sea-level reduction approaching the calculated amount may well have occurred
in the early Holocene. However, the large sea-level movements associated with
changes in Quaternary ice volume overwhelm such small changes. In addition,
the presence of continental ice in the Quaternary makes it difficult to isolate the
effects of precession-related monsoonal effects from the climatic consequences
imposed by the ice sheets. For example, although the internal basins of the Aral
and Caspian may have been influenced by the increased monsoonal activity in
the early Holocene through the Amu Darya and other drainages, these basins may
also have received glacial meltwater through the Volga or Turgai during the late
Pleistocene and earliest Holocene. This sort of complexity further increases the
difficulty of assessing the influence of Quaternary pluvials on sea-level.
As we have indicated, this exercise is not intended to provide an explanation
for actual sea-level change in the Quaternary, rather it is intended to illustrate the
potential sea-level change that could be induced by this mechanism during periods
of Earth history when continental ice was not a factor. During many periods of
Earth history, the potential for periodic water storage via this monsoonal mech-
anism was likely to have been much greater than it is today. Larger continental
areas in the subtropics were typical of the late Paleozoic and Mesozoic. These could
have generated more extensive monsoonal circulation, and more broadly ranging
fluctuations of monsoonal influence than occurred in the Quaternary. Sedimentary
cover with high porosities, such as eolian deposits, as well as regions of basinal
topography were also more extensive in the past than they are today. These differ-
ences in configuration of continents and water storage may have resulted in greater
potential for fluctuating water storage than the Quaternary evidence supports.
4. Orbital Variation, Continental Configuration and the Monsoons: Theo-

retical Considerations
Monsoons result from solar heating of continental areas. The intensity of the
monsoon depends on continental size and arrangement. Here we explore some
of the possible interactions between orbital variation and continental configura-
tion that could conceivably generate periodic fluctuations in monsoonal intensity.
Berger (1978) extrapolated from recent Solar System dynamics to produce the
Milankovitch spectra pertinent to the Quaternary. More recently Berger and oth-
ers (1989) have examined the question of how these cycles varyed through the
Peanerozoic. As one might expect, given the slowing of Earth's rotation doc-
umented by increasing day length, these astronomical cycles have also slowed
over geologic time. For example, the 41 ka obliquity cycle had approximately
32.9 ka period at around 300 Ma in the Pennsylvanian and the 23 ka precession
cycle component had an approximately 20.5 ka periodicity at that time (Berger
and others, 1989). These calculations suggest that, with only minor adjustments,
Milankovitch periods can applied in a uniformitarian manner to the interpretation
of cyclic sedimentary deposits throughout the Phanerozoic.
CONTINENTAL CONFIGURATIONS AND PERIODICITY OF MONSOONAL WATER STORAGE
Continental Area and Water Storage Continental Area and Water Storage
Subtropical One Hemisphere Temperate One Hemisphere
Precession (21,OOOy) Dominant Precession (21,OOOy) and
Eccentricity (100,OOOy) Influential Obliquity (41,OOOy) Influential
e WATER
STORAGE
D LAND
~~~
JJJ
SEA
Continental Area and Water Storage Continental Area and Water Storage
Tropical Subtropical Both Hemispheres
Double Precession Cycles (@10,OOOy) Double Precession Cycles (@10,OOOy)
Forced by Perigee at Equinoxes Forced by Perigee at Solstices
Eccentricity (100,OOOy) Influential Eccentricity (100,OOOy) Influential
Fig. 4. End member continental configurations expected to respond to particular Milankovitch

parameters. Precession driven eustatic cycles are well documented in the Quaternary and Triassic
and may have occurred in other times. Monsoonal changes driven by double beats, c, may also have
occurred in the Cretaceous. See text for more details.
4.1. PRECESSION, ECCENTRICITY, AND OBLIQUITY AND IDEALIZED CONTINENTS
Prior to examining the Phanerozoic record for sea-level fluctuations generated

by the monsoon, we consider Milankovitch-monsoonal responses for several ide-
alized continental configurations (Fig. 4A-4D). The three Milankovitch orbital
variables of precession, eccentricity, and obliquity interact with one another. As
discussed previously the precession cycle has an impact on the monsoon by mod-
ulating the intensity of the seasonal cycle. Given the modem configuaration of
the continents, when perihelion coincides with the northern hemisphere summer
solstice, greater summertime heating generates stronger monsoonal flow. This pre-
cessional response should generally pertain when continents are aggregated in the
low and middle latitude of a single hemisphere (Fig. 4A). Solar system dynamics
generate a complex set of precessional harmonics. Spectral peaks occur at 23.7,
22.4 and 18.9 ka (Berger, 1978). As a practical matter, these multiple peaks are
conceived of, and tend to resolve in the record as, a general 21 ka periodicity.
Although multiple precessional harmonics have been recognized in isotopic data

from Pleistocene age marine cores (Hays and others, 1976).
Major periods of the eccentricity determined from celestial mechanics occur
at 412.8 ka, 94.9 ka, 128.2 ka and 99.5 ka, as well as additional smaller peaks
(Berger, 1978). For practical purposes, peaks in eccentricity are said to occur at
about 100 ka and 400 ka. Although eccentricity has only a small impact on total
insolation, the 100 ka eccentricity cycle has a dominant effect on changes in ice
volume in the Pleistocene (Hays and others, 1976). Eccentricity modulates the
intensity of the precession signal. The more oblate the Earth's orbit, the larger the
precessional effect. Given the importance of precession to the monsoon, eccen-
tricity would be expected to influence the intensity of the monsoonal fluctuations.
As a consequence of this interaction between precession and eccentricity, a pre-
cession driven climatic signature, such as the storing of water on large subtropical
continents, should include frequency components of eccentricity as well (Fig. 4A).
The generalized relationship sought in the rock record is the repeated grouping
of @21 ka precession cycles into bundles of 4 or 5. Spectral analyses showing
eccentricity and precessional peaks combined with sedimentary units showing
consistent bundling are often viewed as confirmatory of a Milankovitch signature
(for example, Schwarzacher, 1993), and have been observed in association with
monsoonal activity in the Late Triassic (Olsen, 1986).
The angle of Earth 's rotational axis relative to the ecliptic, obliquity, is currently
at 23 0 and varies from 22.10 to 24.4 o There are mUltiple spectral peaks for obliquity
variation periods of 41.0 to 39.7 to 53.6 ka (Berger, 1978). The largest peak at
41 ka is the feature of geologic interest. Large angles of obliquity result in greater
insolation at high latitudes in the summer and less in the winter. High angles
of obliquity, in association with precessional cycles, could lead to more extreme
poleward excursions of the monsoon. If large continents and water storage potential
were located in the temperate region, an expanded monsoon, and consequent sea-
level change, might be limited to times with high angles of obliquity (Fig. 4B).
The 21 ka precession cycle and 41 ka obliquity cycle come in and out of phase
with a periodicity of about 800,000 years so any pattern generated could be long
term and complex. Sea-level fluctuations that depended on the obliquity cycle
would require large water storage capacity at the poleward edge of monsoonal
influence. An anomalous highstand is evident in Qinghai Lake well above the last
precessional highstand (Lister and others, 1991). This suggests that more intense or
more extensive poleward migration of monsoonal precipitation may have occurred
earlier in the Quaternary.
4.2. NULL MONSOONS AND DOUBLE PRECESSION BEATS
It is somewhat counterintuitive, but equatorial continents could also generate peri-

odic monsoon related signatures. These effects would be due to accentuation of the
Hadley cell or intertropical convergence (lTCZ) circulation at or near the equator.
Normally there is some monsoon related departure of the ITCZ from the equator,
however, such departure would be minimized, and equatorial precipitation accen-
tuated, when perihelion coincides with either the spring or fall equinox. At these
points in time insolation in the tropics should be greatest, focusing convection

and precipitation. The two equinoxes in the annual cycle should produce a double
precessional beat. Rainfall, and potentially water storage, would occur roughly
every 10,000 years (Fig. 4C). This concentration of rainfall in the tropics could
then be referred to as an "anti" or "null monsoon".
Climate models based on modern geography show double beat precession
peaks in temperature over equatorial continental areas (Short and others, 1991).
In the Cretaceous double beat precessional cycles have also been documented
in marine cores (Park and others, 1993). These results suggest that double beats
may be a frequent response of large continental areas in the tropics. Double beats
could also be generated if large continental areas were located in the subtropics
both north and south of the equator (Fig. 4D). In this case precessional forcing
would be associated with the coincidence of perihelion with each solstice. Pangean
continental configuration of the late Paleozoic and early Mesozoic times had
considerable continentality both north and south of the equator. Consequently,
double precessional beats could have occurred at these times. However, to be
relevant to our discussion of sea-level fluctuation, sufficient continental water
storage capacity would have to be located both north and south of the equator in
the region of monsoonal influence. We will return to the various frequencies at
which monsoons may have stored water in the past when we consider the evidence
for such behavior from the pre-Quaternary geologic record.
4.3. OROGRAPHIC EFFECTS
In addition to continental configuration, orographic effects can greatly influence

the monsoon. The Tibetan Plateau provides a high elevation surface which when
heated by the sun generates anomolously high temperatures for that altitude. Insta-
bility and convection result from this high elevation heating. This intensifies the
Asian summer-time low pressure systems strengthening the monsoon (for exam-
ple, Kutzbach and others, 1993). The high topography of South Asia also serves
to separate temperate and tropical air masses preventing larger scale mixing. This
maintains temperature gradients localizing and focusing convective processes.
Monsoonal flow over other continents is accentuated to varying degrees by topog-
raphy (see Meehl, 1992).
5. Evolution of Internal Drainage
Large areas of internal drainage are necessary if large amounts of water are to
be stored on the continents. The maintenance of internally drained basins can be
viewed as a competition between processes that disrupt drainage systems, and those
processes, such as stream capture, that increase the efficiency of drainage networks.
Many modern high latitude lakes result from the relatively recent disturbance of
drainages caused by glacial advance and retreat. In a world without continental
ice sheets, tectonics should be the primary generator of internally drained basins.
Eolian processes may provide a secondary mechanism of drainage disruption.
Drainage will organize faster near the sea or other relevant base level, and when
frequent precipitation generates substantial and relatively frequent runoff. Atmo-

spheric transport of marine moisture is limited by continental size and mountain
barriers. In addition, in the arid interiors of large continents eolian processes are
more active. Intermittently active dune fields disrupt drainage in the Sahara today.
Due to these factors, evolution of effective drainage systems should take longer
on larger continents. Strength of monsoonal circulation also relates to continental
area. Collisions generate large continents and orogenic thickening of continental
crust. These factors increase the volume of ocean basins, lowering sea-level and
exposing continental shelves, effectively increasing continental area and elevation.
As pointed out by Hay and Leslie (1990) the margins of continents exposed at low-
stands contain a disproportionately large amount of porous sedimentary rocks ideal
for aquifer water storage. The relative elevation of the continents can accentuate
the aridity of continental interiors and may accentuate monsoonal activity (Meehl,
1992). Thus, a number of factors effecting the ability to store water on continents
via fluctuations in monsoonal intensity closely correlate with continental size.
Many of the tectonic processes likely to produce closed internal basins are
active during the periods of continental collision that produce large continents, or
during the initial break up of such continents. Tectonic processes associated with
the accretion of continental blocks into larger continents generate large closed
basins in a variety of ways. We tabulated modern and ancient regions of closed
or nearly closed contours on the basis of the tectonic processes that produce them
(Table 2). Plate convergence can generate a number of basin types. For example, as
continents approach each other collision is often initiated at separated points along
the margins. Areas of oceanic crust are trapped between these loci of collision.
This mechanism formed the Mediterranean, Black and Caspian Seas (Table 2). Of
these, only the Caspian basin is partially empty today. However, in the Messinian
(6 Ma) the Mediterranean was completely isolated from the other oceans of the
world. Its volume equivalent of 15 meters of sea water evaporated (Hsu and
others, 1973). Lower sea levels during the Pleistocene similarly isolated the Black
Sea. Tethyan closure may have generated closed basins in the late Mesozoic and
Cenozoic as portions of the Asiatic land mass accreted. The colliding terranes
include the continental blocks that make up China, Afghanistan, Iran, and India. In
the Paleozoic trapping of ocean crust may have isolated basins in the early stages of
the Acadian and Alleghanian orogeny. Island arc formation isolates oceanic basins
today. Examples include the Aleutian Basin, Sea of Okhotsk, and the Caribbean
(Table 2).
Continental collision, and other convergent tectonic processes, produce re-
gions of basinal storage in two additional ways. Thrust loading produces basins
as material from one continental block overrides another, and crustal thickening
results in plastic deformation and extensional tectonics in the thickened region.
Thrust loading due to the collision of India and Asia, by Tibet from the south and
by the Tien Sien from the north, has depressed Tarim producing the largest empty
basin on Earth today. Thrust loading by foreland fold and thrust belts has generated
extremely large basins, such as the Western Interior Seaway of Cretaceous age, and
the Late Paleozoic Appalachian Basin of the Eastern US. Comparable basins are
few on Earth today, although the Gangetic Plain, Tigris Valley-Persian Gulf, and
TABLE 2
Basins and Water Storage Potential. Modern and ancient basins and other regions of water storage
potential. Most basins are, or were closed at some point in the past, other modern basins that are
nearly closed are included for comparison. Basins are grouped by tectonic type, and area, as well as
volume in meters of sea-level change, are reported in the following columns. In the last column the
time of basin closure or desiccation is reported, and if the basin is partially empty today an estimate
of the sea-level equivalent is reported in parentheses. Calculations are rough and are based on a
number of almanacs and atlases.
Basin Type Area Volume DesiccationlIsolation

Basin (106 km2 ) ~ Sea-level
(m)
Convergent
Collision Trapped Ocean crust
Mediterranean 2.51 10.4 Messinian desiccation
Black Sea 0.51 1.68 Pleistocene lowstand
Caspian-Aral Basin 2.32 1.42 Currently (0.36)
Arc Trapped
Caribbean 2.52 17.95
Bering Sea 2.26 9.34
Sea of Okhotsk 1.39 3.75
East China Sea 0.66 0.34
Thrust loaded basins
Tarim 0.87 1.08 Jurassic, Currently (1.08)
Balkash-Zungar 0.30 0.053 Currently (0.53)
Zungar 0.30+ 0.42 Permian, Trias, Maim, Dogger
Persian Gulf 0.23 0.06 Pleistocene lowstand?
Karoo 1.9? 2.1O? Beaufort-Permian Lakes
Green River 0.36 0.30 Eocene-Precession cycles
Chinle (Owl Rock) 1.0+ 0.55+ Norian Lake Deposits
Transpression
Central Valley 0.10 0.14
Quidam 0.16 0.09 Eocene-Recent (0.09)
Crust dilation
Tibet 1.22 1.66? Currently (1.5?)
Andes 0.12 0.17? Currently (0.114)
Basin and Range 0.86 1.2.? Miocene-Quaternary, Bonneville
Falkland-Aghulas 2.5 3.46? Mid Jurassic-Late Jurassic
E. North Am. 7.0? 5.8+ Late Triassic-Early Jurassic
Keuper 6.0? 5.0+ Late Triassic
E. Siberian Lowlands 4.3 3.6 Triassic
TABLE 2
Continued
Basin Type Area Volume DesiccationlIsolation

Basin (106 km 2 ) ~ Sea-level
(m)
Divergent
Rifts
Red Sea 0.45 0.67 Miocene-Pliocene Desiccation
Tanganyika 0.32 0.13 Late Pleistocene (0.31)
Parana Benue 2.0 3.46 Aptian-Albian
Zechstein and rifts 4.1 3.4 Late Permian
Rift Shoulder bounded Basins
Victoria 0.07 0.015
Congo 1.44 1.20 Early Cretaceous,
K wango Lake beds
Pull Apart
Gulf of Mexico 1.51 6.74 Dogger, Luanne SaIt
Shanxi 0.24 0.33 Triassic-Tertiary
Orcadian Basin 0.80 0.88 Devonian, Old Red
Back Arc Basins
Andaman 0.57 1.75
Sea of Japan 1.01 4.68
Transform Traps
Walvis Ridge 1.25 \0
Hotspot Trap
Arctic Ocean 13.23 38.05 Isolated in Oligocene
Chad 0.67 0.16 Currently (0.16)
Non Tectonic
Reefs
Permian Basin
Ergs
N. Africa-Arabia 5.4 0.75? Currently (0.75?)
Navaho-Kayenta 1.0+ 0.20+ Norian-Lias
Botucatu-Sambaiba 3.8+ 0.53+ Aptian-Albian
Coconino, Lyons. 3.0+ 0.60+ Permian
Rotliegendes 3.5+ 0.72+ Permian
Barun Guyot Cretaceous
the Adriatic Sea provide some examples. Both the Adriatic and the Persian Gulf
have submerged closed contours forming discrete closed basins (Table 2). Loading
of the South African craton by the Cape Fold Belt in the Pennian produced very
large lake basins (Yemane, 1993; Yemane and Kelts, 1990). Thrust loading associ-
ated with the inversion of Uintan trough during Laramide compression produced
the Green River and other Eocene lake basins. Interestingly Green River deposits
have long been thought to contain Milankovitch period fluctuations in sedimenta-
tion (Bradley, 1929; Fischer and Roberts, 1991). Similar thrust loading can occur
along transform margins. Transpression produces subsidence and closed contours
in the southern portion of the Central Valley of California (Buena Vista Lake).
Continental collision and subduction related compressional tectonics can re-
sult in large regions of thickened continental crust. As such crust reaches thennal
equilibrium, it undergoes ductile deformation, spreading out and producing many
large graben structures in the thinned brittle layer of overriding crust. The Tibetan
plateau contains numerous large extensional structures of this type with substan-
tial water storage potential (Table 2). Similarly, the axial graben of the Andes
contains Lake Titicaca and large areas of dry lake beds. With the dissipation of
compressional stress in the latest stages of orogeny, extension may accelerate
producing even larger regions of dilated crust with complex basinal topography.
Post-Laramide extension of the western US produced the basin and range topog-
raphy of the Great Basin and adjacent areas. This region contained two very large
lakes, Lahontan and Bonneville, as well as many smaller ones, as recently as
early Holocene times. Similar regions of extension followed Paleozoic Pangean
orogenies. In Gondwana active extension produced basin and range topography
in thinned crust that now constitutes parts of Mozambique, coastal South Africa,
the Aghulas Bank and Falkland Plateau. A similar extensional regime followed
the late Paleozoic Appalachian orogeny; many basins formed in the early Meso-
zoic along the Eastern North American coastal plane, shelf, and rise. A roughly
contemporaneous extensional regime formed in West Siberia adjacent to the Urals.
During continental break up, graben formation, rift shoulders, back arc basins,
and pull-apart structures associated with transforms can all form isolated basins.
The East African rift provides a modern analog for continental break up. Lake
levels in all the large lakes, including Malawi (Scholz and Finney, 1994), Tan-
ganyika (Gasse and others, 1989), Natron (Hillaire-Marcel and Casanova, 1987)
and Turkana (Johnson and others, 1991) have fluctuated in response to climate
changes associated with the precession cycle. The Red Sea provides an example of
an even larger rift basin of closed contours (Table 2). Similar basins were present in
the early stages of most continental rifting events. Elevated shoulders of adjacent
rifts can also isolate basins; Lake Victoria occupies such a basin today and the
Congo Basin was similarly isolated by rift shoulders in the Cretaceous.
In later stages of rifting in vol ving oceanic crust, spreading ridges, hotspots and
"leaky transforms" interact with other features to produce closed contours. The
Arctic Ocean is perhaps the largest region of well defined closed contours today.
It was isolated from the Atlantic by hotspot volcanism in the Iceland region in the
Oligocene (Table 2). Other basalt edifices that have played a role in isolating basins
include the Walvis Ridge in the South Atlantic, and Florida which is founded on
transform associated basalts. Thus, a variety of distinct tectonic processes have

played a role in isolating basins. We discuss specific examples in further detail
when we examine the potential for basinal response to monsoons in the various
periods of the Phanerozoic.
6. Carbonate Models and Sea Level Change
Nearshore carbonate deposits document periodic subaerial exposure and provide

the most compelling evidence for periodic small scale sea-level change in the Mi-
lankovitch frequency spectrum (Goldhammer and others, 1990; Osleger and Read,
1991). Following the work of Fischer a number of workers generated synthetic se-
quences of nearshore carbonates to compare with those they find in the field (Fig. 1)
(for example, Read and others, 1986; Read and Goldhammer, 1988). These mod-
els require subsidence rates, carbonate production rates, and sea-level curves as
input parameters. This approach strongly supports a causal linkage between the
nearshore carbonate packages observed and orbital variation driven fluctuations
in sea-level. As such, they tend to confirm power spectra analyses that document
orbital forcing periods for many carbonate sedimentary units. However, ten me-
ters of relative sea-level rise is often invoked in models that produce meter-scale
carbonate packages. These amplitudes result from the interplay between carbonate
production terms and the sinusoidal shapes of the sea-level curves employed in
such modeling. A number of measurements of modern reefs suggest that carbonate
production can be very rapid (for example, Schlager, 1981). With large carbonate
production terms in the models it becomes difficult to flood a surface because car-
bonate production keeps pace with the sea-level rise. If this were indeed the case
only tidal flat facies would be preserved. The models could not produce the subtidal
facies evident at the bases of meter scale beds (Fig. 1). We argue that carbonate
production need not have been uniformly high, and that climatic responses and
consequent sea-level change may have been relatively abrupt. If either were the
case, the rock record could have been produced with smaller amplitude sea-level
rises than previously recognized.
6.1. CARBONATE PRODUCTION RATE
Carbonate production rates in sedimentary models derive from modern reefs (for
example, Schlager, 1981). However, carbonate production rates may not pertain
uniformly to times in the past because modem reef building taxa that produce
carbonate at high rate were not present during the time period in question, for
example scleractinian corals were not present in Paleozoic reefs. Alternatively,
changes in ocean chemistry may have altered calcification rates (for example,
Holland, 1984). Temperate climate, clastic influx, salinity change, and organic
input are all known to limit carbonate production today. All these factors changed
in association with the early Holocene monsoon. Thus, carbonate production need
not be constant. It is likely to change periodically in response to the same climatic
factors governing sea-level. In addition, hurricanes transport sequestered carbonate
material to deeper water in many reef systems today. The assumption of a high
carbonate production and retention on an ancient reef or platform indicates that the
researcher can recognize in the record, and discount, those variables that are known
to limit carbonate production at present. If carbonate production and retention on
the platform was in fact lower than values assumed in the models, then amplitudes
of sea-level change need not be as high to produce the eustatically-generated
meter-scale carbonate beds observed in the rock record.
6.2. RATE OF PLUVIALLy-INDUCED SEA LEVEL CHANGES
Even if carbonate production values were high, sea-level change may not conform
to the sinusoidal curve employed in many carbonate models. Ice core evidence
suggests that climate changes of considerable magnitude can occur relatively
abruptly, with dramatic changes in hemispheric temperatures occurring in tens of
years (for example, Dansgaard and others, 1993; Anklin and others, 1993). Tropical
precipitation decreased abruptly in the early Holocene, and in the last century, in
response to perturbation of the Atlantic thermohaline circulation (Street-Perrott
and Perrott, 1990). Thus, relatively rapid fluctuations in monsoonal precipitation
is to be expected. If climate change is abrupt the rise and fall of sea-level as a
consequence of sequestration of water by the monsoons could also be relatively
rapid. Isotopic data from some North African lakes are taken to indicate extremely
rapid filling, a long high stand period, followed by very rapid emptying, resulting
from dramatic changes in monsoonal rainfall (Mckenzie, 1993). Very large flux
rates of water are also suggested by early Holocene flood stages of the Nile.
Thus, water storage and small-scale sea-level change associated with changes in
monsoonal conditions could have been quite rapid (see Gasse and Vancampo,
1994).
Figures quoted by Hay and Leslie (1990) indicate that annual precipitation
on the continental areas of Earth today exceeds 105 km 3 per year. This volume
equates to roughly a 30 cm change in sea-level. With dramatic changes in climate
it may be possible to store a large fraction of this amount, lowering sea-level by
several centimeters a year. Evaporation rates from continents are roughly 2/3rd of
the precipitation rate. Annual fluxes of water into and out of aquifer storage are
currently about l/Sth the continental precipitation rate and a little more than 1I4th
the evaporation rate (Hay and Leslie, 1990). Hsu and others (1973) calculated that
modern evaporation rates would desiccate the Mediterranean in about a thousand
years, leading to 1 cm sea-level rise per year. The area of increased precipitation in
the early Holocene was 4 xl 0 7 km 2 , roughly 16 times the area of the Mediterranean.
If decline in rainfall in this area was sudden, the resulting sea-level rise rate could
have been substantially larger than that calculated by Hsu. After a dramatic climate
change initially high sea-level rise rates should decline as the wetted surface area
where evaporative loss could occur decreases and deeper subsurface aquifer layers
drain with falling water tables. Water storage would similarly be initially rapid
and then slow as more and more basins came to equilibrium with evaporation or
overflowed to the sea. The processes leading to water storage on and water loss
from the continents are not identical. Consequently, some asymmetry between rise
and fall might be expected. However, a rapid 100 y) change in sea-level of one to
a few meters following shifts in monsoonal climate appears consistent with current
flux rates and our understanding of the rapid onset of climate change in the past.
Carbonate modelers (for example, Hinnov and Goldhammer, 1991) have assumed
maximum sea-level change rates on the order of 1 to 2 meters per thousand years.
It may be that pluvially-induced sea level changes of an order of magnitude higher
rate occurred regularly during some period of the Phanerozoic.
If carbonate productions were lower than those used by modelers, or if sea-
level rise rates were higher, much smaller changes in sea-level, on the order of
three meters, could be sufficient to account for meter scale carbonate beds with
algal laminated or subaerially-exposed tops, rather than the ten meters of relative
sea-level change that has often been invoked.
7. The Late Triassic
Several lines of evidence support a Milankovitch driven monsoonal water storage

mechanism for Late Triassic sea-level fluctuations. During the Triassic the land
masses of the Earth were still associated into one large continent, Pangea; a large
percentage of the land mass was located in subtropical to temperate latitudes;
sea-level was low resulting in a large continental area; and there was a large
tropical seaway - the Tethys - adjacent to Pangea. All these factors are thought
to contribute to monsoonal circulation. In addition, widespread sedimentologic
evidence indicates a strongly seasonal climate (Dubiel and others, 1991). Computer
models of global climate based on Triassic continental configurations also indicate
very strong monsoonal flow (Kutzbach and Gallimore, 1989) leading to use of the
term "Megamonsoon".
During the Triassic, post orogenic (Alleghanian-Variscan) rifting and incipient
continental breakup produced vast regions of graben and pull-apart basins. These
basins extend from what is now the Gulf of Mexico along the east coast of North
America to northern Europe and the Arctic (Holzer and others, 1988; Ziegler,
1982, 1988; Uchapi, 1989). Triassic evaporites are more extensive than those of
any other period (Robinson, 1973) testifying to the frequent desiccation of these
basins.
That filling and emptying of basins was subject to both seasonal and 104 -year
fluctuations attributable to Milankovitch forcing is more directly evident from an-
nual varves in lacustrine basins in the Newark Supergroup (Van Houten, 1964).
Spectral analyses of these sedimentary cycles were performed on Carnian age
rocks of the Locatong Formation and Norian age Rocks of the overlying Pas-
saic Formation. These analyses suggest a dominant cyclicity of approximately
20,000 years, as would be expected if precession of the equinoxes (Olsen, 1986)
was the primary driving Milankovitch parameter. In addition, lower frequencies,
attributable to eccentricity, are also observed. Olsen (1986) attributes these spec-
tral peaks to Milankovitch controlled variation in the monsoon and consequent
variation in fluvial processes and lake level.
The Dachstein carbonates of the Northern Alps record small scale changes in
eustasy termed Lofer cycles. These deposits were first interpreted as cyclic in the
1930's; subsequently, Fischer did his classic work on cyclicity and eustasy on these
sections (Sander, 1936; Fischer, 1964). In the Dolomite mountains of northern

Italy meter-scale "Latemar cycles" also record sea-level changes (Goldhammer
and others, 1987, 1990; Hinnov and Goldhammer, 1991). The meter scale units
in both regions consist of progradational packages with subarially exposed caps
thought to be indicative of repeated eustatic flooding in conjunction with relatively
continuous subsidence. Spectral analysis (Goldhammer and others, 1987) indicate
a dominant 20,000-year periodicity and bundling of these cycles units of 4 or 5, a
result similar to the spectral signature in the Newark supergroup (Olsen, 1986).
The Lofer cycles roughly correlate with the cyclic Norian age Passaic For-
mation of the Newark Supergroup. More Van Houten cycles are reported from
the preceding late Carnian age of the Locatong Formation (Olsen, 1986). Latemar
eustatic cycles occur in the Ladinian stage that precedes the Carnian (Hinnov
and Goldhammer, 1991). Thus the observations on eustatic Lofer and Latemar
cycles bracket and overlap observations of the lacustrine Milankovitch cycles in
the Newark Super Group. The potential for water storage in modem basins in the
region of fluctuating monsoonal influence is sufficient to produce four to eight
meter changes in sea-level, Table 1, Fig. 2 (Jacobs and Sahagian, 1993).
Extraordinarily large areas of internal drainage, evaporite deposition and eolian
deposits are evident in reconstructions of Late Triassic paleogeography. Graben
structures containing lake and fluviatile deposits extend from the Gulf of Mexico
to Europe with associated basins in Morocco (Brown, 1980). These basins may
have covered 13 millions square km (Uchapi, 1989; Ziegler, 1982). Additional
basinal settings of Late Triassic age occur on the extended crust of East Siberia
and the Donetz Basin (Ziegler, 1988) as well as large areas of lake (Owl Creek
Member) and eolian deposits in the Chinle Formation of the southwestern U.S.
(Dubiel and others, 1991). Thus rift graben, lakes, and evaporite deposits had a
far greater areal extent in the Late Triassic than they do today (Table 2). Given
the larger areas involved, storage potential may have been considerably in excess
of the 4 to 8 meter potential calculated for the Quaternary (Jacobs and Sahagian,
1993). In addition, storage potential was concentrated in the Northern Hemisphere,
particularly in the subtropics and would have been effected by the megamonsoonal
circulation predicted for this time period (Kutzbach and Gallimore, 1990; Dubiel
and others, 1991). Given precessional forcing of the monsoon, fluctuation of sea-
level sufficient to accommodate meter scale eustatic packages would appear to
be relatively easy to generate. Standard carbonate models (e.g. Goldhammer and
others, 1990) suggest 10 meters of relative sea-level change, 7 meters with isostatic
adjustment. Eustatic fluctuations of this magnitude may have been produced in the
Late Triassic. However, as we argued earlier, smaller sea-level changes may have
been sufficient to produce these carbonate packages.
To claim that ice did not play a role in Late Triassic sea-level change ne-
cessitates a negative argument, always a dangerous task in geology.2 However,
2Just ask Darwin: in his negative or exclusionary argument, "Observations on the Parallel Roads
of Glen Roy, and other parts of Lochaber in Scotland, with an attempt to prove that they are
of marine origin", Darwin overlooked the possible effects of glacial ice in this geomorphologic
question. Developments in glaciology convinced the scientific establishment of the glacial origin of
these features. This led Darwin to later disparage exclusionary argument (Hull, 1973).
the Greenland Ice Cap contains only enough water to raise sea-level 5 meters
(Rowley and Markwick, 1992). Currently, ice sheets that reach the sea distribute
dropstones over wide high latitude regions. No such characteristic deposits occur
in the Late Triassic (Hambrey and Harland, 1981). Thus there is no evidence for a
continental ice sheet at all, not even one a fraction of the size of Greenland. Today,
mountain glaciers contain only the equivalent of a 70 cm rise in sea-level, and
previous expansions of mountain glaciers in the Quaternary were closely tied to
cold conditions at the poles (for example, Dawson, 1992). In the Late Triassic, cy-
cads grew at high paleolatitudes demonstrating equable climates at the poles (Ash
and Bassinger, 1991). Thus there is a large body of evidence suggesting cyclic
lacustrine water storage in the Late Triassic, and no evidence indicating glacia-
tion. Clearly, glaciation is the dominant influence on sea-level in the Quaternary;
however, fluctuations in lake level and groundwater storage appears to provide an
alternative mechanism for small scale sea-level fluctuations in periods of Earth
history which are thought to lack continental glaciers such as the Late Triassic.
8. Other Time Periods
The Late Triassic meets all the expectations of a time period when Milankovitch
driven monsoonal water storage generated high frequency sea-level change. How-
ever, in several other periods evidence suggests Milankovitch-monsoonal sea-level
fluctuations.
8.1. CAMBRIAN-ORDOVICIAN
From the Late Cambrian to Middle Ordovician times there is little evidence of
ice (Hambrey and Harland, 1981), and a well documented record of Milankovitch
driven sea-level cyclicity recorded in carbonates on the massive platforms of
eastern North America (for example, Goldhammer and others, 1993; Osleger and
Read, 1991). During this period a precession driven monsoonal cycle could have
developed in response to the Gondwanan continental landmass in the Southern
Hemisphere. Evaporite deposits are reasonably abundant, especially in Siberia
(for example, Frakes and others, 1992). However, we are not aware of any cyclic
fluviatile or lacustrine sediments that document periodic pluvial conditions or
water storage on the continents in this time period.
8.2. DEVONIAN
In the Early and Middle Devonian there is no evidence for glaciation and cli-
mates were warm and equable (Hambrey and Harland, 1981; Frakes and others,
1992). Early Devonian carbonates of the Helderberg Group contain meterscale
thrombolitic, stromotoporoid rich beds toped by flat laminated or stromatolite
caps. These deposits have been interpreted as products of a precession dominated
Milankovitch frequency oscillation in sea-level (Goodwin and Anderson, 1985;
Anderson and others, 1984, Goodwin and others, 1986; Brett, 1986). Upper Fras-
nian carbonates from western Canada also contain meter scale shallowing upward
cycles, some of a subtidal origin and others with biolaminated tops that exhibit
subaerial exposure (Fejer and Narbonne, 1992).
Middle Devonian lake deposits from Scotland contain repeated 8 meter thick
transitions from lacustrine to fluviatile deposits inferred to be precession dominated
on the basis of spectral analyses (Kelly, 1992; Rogers and Astin, 1991). Henrik
Olsen (1990) argued that Late Devonian sediments of the Kap Grath Group in
eastern Greenland document cyclic fluctuations in stream hydraulics resulting
from precession-mediated fluctuation of monsoonal precipitation. Late Devonian
fluvial deposits from southwest Ireland record periods thought to equate with the
eccentricity cycle (Sadler and Kelly, 1993). These deposits suggest periodic storage
of water in the Orcady Basin. Ziegler's (1982) European reconstructions indicate
areas approaching 1 million square kilometers of fluviatile and eolian deposits
with some evaporites. Given the distribution of continents in the sUbtropics and
the uplift of the Acadian Orogeny a strong monsoon seems likely. Old Red facies
and evaporite deposits suggest aridity on a continental scale (Frakes and others,
1992). Continental interactions between Gondwana and Laurasia (Scotese, 1987)
may have trapped ocean crust or generated basins through crustal dilation and pull-
apart structures. Little evidence of these basins would have survived subsequent
orogenic activity. Additional basins could have formed via rifting along the South
Tethyan margin during the Devonian (Scotese, 1987).
In summary, there is evidence for Milankovitch frequency fluctuations, in
sea-level in the Early and Late Devonian, and evidence for similar frequency
fluctuations in fluvial and lacustrine sediments in the Middle and Late Devonian
from the Orcady Basin and Greenland. Precessional signals of both a pluvial and
eustatic nature occur in the Late Devonian. However, there is also limited evidence
for glacial ice in the Famminian (Frakes and others, 1992). Thus, the case for
monsoonally generated sea-level fluctuations in the Devonian is provocative, but
not comprehensive.
8.3. LATE PERMIAN-EARLY TRIASSIC
Gondwanan ice sheets presumably dominated any Milankovitch driven sea-level

fluctuations in the Early Permian. By the Late Permian abundant plant remains are
evident from the same regions covered by glacial ice sheets in the lower Permian
(Crowell, 1978). Thus, high frequency sea-level fluctuations late in the Permian
could be due to fluctuation in water storage rather than ice sheet advance and
retreat. As discussed previously, there is no evidence for large scale glacial activity
in the Triassic. The Late Permian Zechstein salts and associated deposits suggests
the intermittent filling of basins in a large region. Including associated graben
structures containing terrigenous deposits, these basins total over 4 million square
kilometers in Europe alone (Ziegler, 1982). Anderson (1982) documented preces-
sion driven cycles of evaporite deposition in the Castile Formation of the Permian
Basin of West Texas. Lake basins of Permian age occur in the Zungar region of
North China (Binjie and others, 1980) and very extensive lake basins are evident
in the Karoo Deposits of Southern Africa (Yemane and Keltz, 1990). First order
sea-level was low in the Late Permian, and Pangean continental configurations
appear conducive to Monsoonal activity. Despite these supporting factors, there

are few reports in the literature of high frequency sea-level fluctuations. Perhaps
given low overall sea-level and high productivity few carbonate platfonns had
developed in which to record eustatic fluctuations.
As was discussed previously, the Late Triassic has considerable potential for
monsoonally driven water storage and very strong evidence indicating that such a
phenomenon actually occurred in the Carnian and Norian. Earlier in the Triassic
continental water storage potential also seems large. However, it is not until the
Ladinian that Tethyan carbonate platforms develop and provide a clear record of
sea-level fluctuation.
8.4. JURASSIC
The Early Jurassic transgression flooded much of the European region conducive
to water storage. However, basins in the extensional terrain along the eastern mar-
gin of North America contain Hettangian and Sinemurian fluviatile and lacustrine
deposits, as well as substantial evaporites. The Middle Jurassic Louann salt docu-
ments that the Gulf of Mexico as an evaporite basin. By Middle and Late Jurassic
times the incipient break up of Gondwana generated large areas of extensional
crust in what has become coastal Mozambique, the Aghulas Bank and Falkland
Plateau (Dingle and others, 1983). Rifting also took place along the Tethyan margin
of Gondwana. Eolian deposits of the western U.S. extend in to the Early Jurassic,
and lake deposits of Jurassic age in China also suggest additional water storage
potential (Table 2). Although not as large as in the Late Triassic, Early Jurassic
storage capacity appears to be substantial. However, this capacity may not have
been ideally located to respond to the monsoon, some being in the Northern Hemi-
sphere and some in the Southern. In addition, the opening of the North Atlantic in
the Middle Jurassic, broke up the continentality that drove Late Triassic northern
hemisphere monsoonal circulation. Milankovitch cycles have been reported from
the Liassic (Weedon, 1985) and spectral analyses of limestone marl couplets from
the Kimmeridgian of southern Iberia (Ol6riz and others, 1991, 1992) suggest pre-
cessional and eccentricity forcing. Bosellini and Hardie (1985) indicate that cyclic
carbonate sedimentation extends from the Triassic to the Eocene in some of the
carbonate platforms of Northern Italy. However, spectral analyses, or other work
detailing the eustatic-Milankovitch nature of these cycles, have not been reported.
Thus, monsoon ally driven eustatic cycles may have occurred in the Jurassic, but
have not been well documented.
8.5. EARLY CRETACEOUS
Next to the Late Triassic, the Early Cretaceous exhibits the best support for eustatic
cycles generated by a monsoonal mechanism. Basinal water storage is substantial
and the marine record contains many precession related cyclic deposits and most
authors agree that there is no evidence of continental glaciation from the Late
Permian to the Early Tertiary. However, some poorly dated drops tones in Northern
Australian deposits suggest seasonal climates at some point in the Jurassic or Early
Cretaceous (Frakes and Krassay, 1992).
A large amount of basinal storage potential was generated by rifting activity
associated with the opening up of the South Atlantic. In the early stages of opening
the a basaltic edifice known as the Walvis Ridge isolated the northern portion of
the South Atlantic, generating a basin containing evaporites of Aptian age (for
example, Uchapi, 1989). Burke and Sengor (1988) calculate the volume of this
basin, and argue that catastrophic marine flooding of the basin lowered sea-level
by 10 meters in the middle Aptian. Rifts contemporaneous with, and presumably
genetically-related to South Atlantic opening are widespread in South America and
Africa. These include the Parana and Maranhao Basins which contain fossiliferous
lake deposits of the Aptian and Albian Santana Formation (Maisey, 1991). These
lake sediments are locally interbed with eolian deposits that may have covered
as much as 3.8 million square miles in South America (McKee, 1979). In Africa,
the Benue Trough of Nigeria contains Aptian lake deposits. The Benue is in tum
connected to large structures of similar age in Chad and Niger (Petters, 1991) as
well as several en echelon graben structures in the Sudan that contain lake deposits
of Early Cretaceous age that exhibit cyclicity (Wycisk and others, 1990). Activity
in these rift zones, opening of the South Atlantic, and incipient uplift along the
western branch of the East African Rift isolated a large lake basin, the remains of
which form the Congo drainage today. Tectonically this basin was similar to Lake
Victoria in that it formed between uplifted rift shoulders rather than in a rift graben.
The Congo Basin is 20 times the areal extent of Lake Victoria and overflowed to
the Benue graben, possibly at relatively high elevation, in the Cretaceous. The
extent of the Kwango beds, lake deposits of Early Cretaceous age (Cahen, 1983),
suggests that a large volume of water was stored in the Congo Basin.
Gilbert (1895) may have been the first to argue for orbital control of sedi-
mentation in the Cretaceous. Examination of marine cores has led other workers
to infer Milankovitch forcing of sedimentary processes in the Cretaceous (see
Schwarzacher, 1993 for review). Many of these deposits are suggestive of mon-
soonally generated changes in productivity that occur at precessional intervals in
the Quaternary of the Arabian Sea (Prell and Van Campo, 1986), or the Quaternary
age precession driven sapropel deposits of the Mediterranean (Rossignol-Strick,
1983). The Scisti a Fucoidi, Aptian-Albian marine sediments ofltaly, record pre-
cession frequency modulated delivery of eolian material, an observation consis-
tent with the alternating wet and dry conditions expected from precession driven
changes in the monsoon (Pratt and King, 1986). Direct evidence for Milankovitch
frequency sea-level fluctuation is available from the Berriasian of southern Iberia
where meter scale carbonate cycles with submarine bases are capped by supratidal
deposits. Spectral analyses of these sections (De Cisneros and Vera, 1993) suggest
precession cycle forcing.
Tropical continents could produce an increase in rainfall twice with each
precession cycle as each equinox coincides with perihelion resulting in maximal
tropical heating. In the Early Cretaceous such a double beat precession cycle might
be expected. The proximity of North Africa and Northern South America in the
Early Cretaceous provides a large equatorial continent region that could have re-
sponded to perihelion at the equinoxes. In addition, a large portion of Gondwana

extends into southern sub-tropical and temperate latitudes. So one might expect a
tropically-generated equinox-related double beat, or subtropically-generated sin-
gle beat, precession response. In fact, there is evidence in the rock record for
both. As previously discussed, there are number of precession driven deep marine
records from Cretaceous deposits. In one study of Lower Cretaceous rhythmic
black shales and limestones of the Maiolica formation of central Italy, precession
cycle forcing is associated with the limestone-shale couplets, but the limestones
show an additional shale parting suggestive of an additional half precessional 10 ka
signature (Herbert, 1992). More concrete evidence for double periodicity is avail-
able from the Campanian of the Central Atlantic where cores show a clear double
precession 12 and 10 ka peaks (Park and others, 1993). At this point the Atlantic
is still narrow in the tropics and this behavior may reflect the large amount of
continent near the equator. As discussed previously, such equatorial continentality
may generate double beats (for example, Short and others, 1991; Fig. 4C).
The evidence from the Early Cretaceous strongly suggests that the precession
driven monsoonal water storage mechanism may be in operation. Studies of cyclic-
ity in lake deposits of Cretaceous age from South America and Africa could greatly
strengthen the argument. The Early Cretaceous, especially up through the Aptian,
seems promising in terms of water storage (Table 2). In the Upper Cretaceous
there is evidence of Milankovitch forcing of deep marine deposits and there are
large eolian deposits with associated lakes in Mongolia (McKee, 1979). However,
with higher first order sea-level, and Gondwanan break up, large continents are
no longer available to generate monsoons and store water. So the Late Cretaceous
may not be as promising for monsoonally-generated water storage.
9. Anthropogenic Sea Level Change
Human impact on climate has the potential to induce sea-level change. However,
human activities can more directly effect sea level without global climate me-
diation. Sahagian and others (1994a) performed some preliminary calculations
documenting activities that ultimately result in transfer of water to the sea includ-
ing: the pumping of non-recharging aquifers, draining of wetlands, deforestation,
desertification, and diversion of surface waters. Unquantified activities that could
also lead to sea-level rise include: construction of poulders, removal of fluid in
oil fields, and burning of fossil fuels. On the other hand, dams retain water that
would otherwise flow to the sea. Dam construction has a net negative effect on
sea-level (Chao, 1988, 1991; Newman and Fairbridge, 1986). Clearly it is impor-
tant to quantify the human generated fluxes of water between the continents and
the oceans. Such an exercise is necessary before the true effects of climate change
on sea-level can be assessed.
Sahagian and others (1994a) emphasized the current rates or fluxes of water
between land and sea. Other authors have commented on the flux rates calculated,
considered other hydrologic sources and sinks, or have focused on total sea-level
change rather than current rate of sea-level rise (Chao, 1994; Rodenburg, 1994;
GruelI, 1994; Gornitz and others, 1994) leading to replies from Sahagian and others
(1994b,c). We will discuss the fluxes to the sea, and the controversy surrounding
dam construction and the need for further assessment.
9.1. FLUXES OF WATER TO THE SEA
Pumping of non-recharging aquifers constitutes mining of water. Some aquifers

filled during earlier periods of wetter climate. In other cases, the pumping of
water simply exceeds natural recharge rates even though some precipitation and
runoff occurs in the region. This is the case in the High Plains Aquifer where
200,000 pumps currently operate (Weeks and Gutentag, 1988). Depression of the
groundwater table, beginning in the 1930s, clearly indicates that aquifer pumping
rates have exceeded recharge rates. Mining of three U.S. aquifers has already
produced a 3.2 mm rise in sea-level: the High Plains regional aquifer, inclusive
of the Ogallah Formation; the Southwest U.S. aquifer, encompassing the alluvial
basins of southern Arizona and California and northern Sonora (Anderson and
others, 1988); and the Central Valley of California (Fig. 5; Table 3). Aquifers in
Arabia are being exploited at an increasing rate. The Arabian aquifers (Balek,
1977; AI-Ibraham, 1991) have recoverable reserves sufficient to change sea-level
over 1 meter each, and Africa in general has recoverable reserves in non-recharging
aquifers equivalent to over 4 meters of sea-level change (Castaney, 1991; Table 3).
Exploitation of the Arabian aquifer is currently rising, and may have implications
for sea-level change in the next century.
In the Aral Basin, the Amu Darya and Syr Darya rivers have been diverted to
the Karakum Desert for the cultivation of rice and cotton (Korlyakov, 1991). The
greater evaporation and transpiration resulting from these diversions has effectively
removed large volumes of water to the sea. The Aral Sea now occupies half
the surface area it did in 1960 (Micklin, 1992). This desication has resulted in
ecological disaster and economic dislocation. Groundwater levels in the region are
tied to declining lake levels. Continuing rapid water removal from the Aral Basin
generates a flux to the sea equivalent to a sea-level rise rate of 0.18 mm of sea
water per year (Table 3). This is the largest single flux from the continents to the
oceans; however, this contribution to sea-level change will diminish as the Aral
approaches complete desiccation early in the next century.
The Caspian has also suffered water loss from diversion of the Volga leading
to lowering of the level of the Caspian Sea. Water sufficient to raise sea-level
4 mm was diverted from the Caspian Basin during the middle part of this century.
However, presumably as a consequence of climatic changes, this trend has reversed
and Caspian lake and groundwater level are currently rising. Other than these large
basins, the largest sources of water extraction from continents involves the burning
of tropical forests. This activity is currently raising sea-level at 0.14 mm per year.
Draining of wetlands also makes a contribution to current sea-level rise (Sahagian
and others, 1994a). The rate of sea-level rise currently generated by these activities
combined is approximately 0.54 mm per year. This is roughly 113 of the sea-level
rise rate inferred from the 20th century global tide gauge records (Fig. 5).
TABLE 3
~
Q)
Anthropogenic contributions to sea-level rise. Total removable volume is the amount of water that can potentially be withdrawn from each aquifer with current technology
given reasonable economic assumptions. Methods of volume estimation and economic a~sumptions vary. so these figures should be considered rough approximations. Sea level ~
equivalent is the amount of eustatic sea-level rise predicted if the volume in question were added to the oceans (area of the oceans taken to be 3.6 x 10 14 m2 ). Present net
extraction rate is the rate of water removal after accounting for recharge rates of 6.75 x 109 m 3/yr. 1.05 x 109 m 3/yr and 7.8 x 108 m 3/yr for the High Plains. SW U.S . and
~
California aquifers. respectively. Saharan and Arabian aquifers are not recharging significantly. "Projected sea-level change" is the volume of water that would be withdrawn Q
Q)
in 50 years if the extraction rate remained the same as it is at present. "Sea level change to date" is the contribution of each aquifer to the 20th century sea-level rise. For most
cases. this was calculated by a~suming a linear increase in extraction rate from 0 to the present rate. The increa~e is a~sumed to have begun in 1930 for American aquifers. 1950 ~
for Saharan and Arabian aquifers. 1960 for Sahel desertification. and 1940 for deforestation. Deforestation figures include only losses of tropical forests. Wetland reduction I
includes the total global wetland area in total removable volume. but the rates and projections include only the loss of wetlands in the U.S. If the rate of reduction in the rest
of the world is equal to the U.S. rate. these figures should be doubled. The Aral and Caspian histories are well recorded by their fluctuating levels. We assume a specific yield ~
of 0.2 in the sands in the surrounding desert. over an area of about 5 times the area of the lake itself for the AraI. and 3 times the area for the Caspian. At the present rate of ~
(1)
reduction. the Aral will be gone in about 20 years. The negative values for damed reservoirs reflect anthrpogenic water storage on land. and assumes that all reservoirs are .~
constantly filled to capacity. However see discussion on dams in the text.

Water Total volume Sea level Present net Sea level Projected Estimated
Reservoir removable equivalent extraction rate rise rate Sealevel change Sealevel change
next 50 yrs to date
(xlO J2 m 3 ) (cm) (x 10 10 m 3/yr) (mmlyr) (mm) (mm)
High Plains 4.0 1.1 1.2 0.03 1.6 1.1
SWU.S. 3.0 0.83 1.0 0.03 1.5 0.92
California 10.0 2.7 1.3 0.04 1.9 1.2
Sahara 600 167.0 1.0 0.03 1.4 0.56
Arabia 500 140.0 1.6 0.04 2.2 0.89
Aral (lake)
1960 1.1 0.3 2.7 0.08 3.0 2.2
1990 0.3 0.08
AraI (ground water) 2.2 0.6 3.7 0.1 5.1 3.1
Caspian (lake) 56.0 15.4 0.77 0.02 1.1 1.3
Caspian (grdwtr.) 220.0 61.2 0.47 0.01 0.65 0.78
Sahel (soil water) 0.1 0.03 0.34 0.01 0.5 0.28
Deforestation 3.3 0.9 4.9 0.14 6.8 3.4
Wetland reduction 8.6 2.4 0.2 0.006 0.3 1.3
Darns -1.9 -0.52 -5.2
TOTAL 1406.7 392.1 19.2 0.54 26.1 11.8

w
VI
VI
12.---------------------~----------------~
-
9
x Sahara
E
E
-6
Arabia
Q)
... Sahel
C)
c:: + Deforestation
cu 0 Wetlands
J:
(,) 3
t:. Dams
Q)
> ... total
~
cu
Q)
0 US Aquifers
c Aral
(J)
Caspian
-3
-6+---~--~----~--~--~--~----~--~---r
1900 1920 1940 1960 1980
Fig. 5. Anthropogenic contributions to sea-level change in the 20th century after Sahagian and
others, 1994. The largest contributions to anthropogenic sea-level rise include: diversion of water
from the Aral and Caspian basins, tropical deforestation, and pumping of Aquifers in the U.S. Note
that the Caspian is now refilling perhaps due to climatic change. Dam construction generates an
opposite effect. lowering sea level. Total sea-level rise attributable to anthropogenic sources has
been about 12 mm, although other authors disagree as to the total volume associated with dams.
Assuming that dam construction is now virtually nil, the sea-level rise rate due to these sources is
now 0.55 mm/year. A figure roughly one third of the sea-level rise rate inferred from tide gauge data.
9.2. DAMS
Impounding water behind dams reduces sea-level. Sahagian and others (l994a,c)
argued that current dam building activities are much diminished from activities at
mid century. On this basis they assume that dams currently are having little effect
on sea-level rise rate. Other authors disagree (Chao, 1994; Rodenburg, 1994) and
feel there has been greater storage of water behind dams than recognized, and
that there may be continued increase in storage capacity. This may be the case,
however, there is a wide range in storage capacities cited by the authors and
the actual timing of reservoir filling is difficult to assess. We agree with Chao
(1994) that the global water storage in impoundments needs to be reexamined.
However, the relationship of manmade impoundments to sea-level is likely to be
more complex. Sedimentation reduces water storage in reservoirs. On the other
hand, if the impoundment had not been constructed, that same sediment might
have found its way to the sea and displaced sea water raising sea-level. On yet
other time scales the absence of those sediments in the nearshore environment may
reduce coastal subsidence, possibly reducing water storage in the sea. These factors
in combination with infiltration and evaporation suggest that we need more than
just a better understanding of impoundment volumes. These additional theoretical
considerations must be addressed before the influence of impoundments on sea-
level is completely understood.
9.3. TOWARD MORE COMPREHENSIVE ANALYSES
Sahagian and others's (1994a) analysis was not intended to be all inclusive. Only
the best documented instances in each category were examined, for example:
only the Sahel was considered in desertification; only the U. S. was considered
in wetland reduction; and only tropical, not temperate, forests were considered
in calculating the effects of forest destruction on sea-level. Gornitz and others
(1994) point out additional fluxes of water not considered by Sahagian (1994a).
In particular, they argue that a 10% fraction of evapotranspiration from irrigated
plants and evaporation from reservoirs is deposited permanently in the atmosphere
on an annual basis. In other words they suggest an atmospheric sink receives the
equivalent of more than 0.6 mm of sea-level per year. This number is 158% of
our figure (Table 3) for annual mining of all aquifers. In fact, it seems exceedingly
unlikely that a flux of water of this magnitude to an atmospheric sink can be sus-
tained. In addition, the loss of atmospheric moisture from desertification, burning
of tropical forests and loss of surface water, in the Aral Basin for example, were not
considered and may be of opposite sign to the feature they chose to examine. This
analysis points out the need for more comprehensive and inclusive analyses than
those performed to date. Many variables could be analyzed further, for example:
the pattern of forest destruction and regrowth in temperate regions is not simple,
and the amount of water retained may depend on details of precipitation, soil type
and plant community; erosional downcutting effects water tables removing surface
water to the sea; and complex interactions ensue when sediment delivery to the sea
changes with changing land use and flood control practices. Thus, more analyses
are necessary before direct anthropogenic effects on sea-level are understood.
10. Lakes, Climate Models and Environmental Concerns
10.1. LAKES AND CLIMATE MODELS
Expansion and contraction of lakes provides robust documentation of climate

change. Lakes influence climate on local and regional scales. Lakes may also ef-
fect climate on a global scale. Late Permian lakes are thought to have ameliorated
the effects of the Gondwanan landmass (Yemane, 1993). Climate models that do
not include these lacustrine features generate much colder winters and tend to
produce glaciers (Kutzbach and Ziegler, 1993). Thus, incorporation of large lakes
appears essential to adequate modeling of paleoclimate. Street-Perrot and others
(1990) incorporated albedo of the paleo-lake and vegetation distribution in a recon-
struction of the early Holocene climate of North Africa. Actual lake distribution
in Asia is less well known (Dawson, 1992), but may be more critical for Holocene
climate reconstruction. The chronology of ice-dammed lakes in large regions such

as the West Siberian lowlands as well as a good chronology for filling of the
Caspian-Aral system could have critical implications for climate reconstruction.
The late onset of the Asian Monsoon relative to modeling expectations has been
attributed to late melting of glacial ice in Tibet (Socci, 1991; Lautenschlager and
Santer, 1991). However paleo-lakes have not been incorporated into these climate
models. Reconstruction of the history oflakes in Asia might permit a more detailed
and realistic modeling of Quaternary climate
10.2. LAKES, FAUNAS AND ENVIRONMENTAL POLICY
The effects of ice sheet advance and retreat is thought to have strongly influ-
enced the biogeography and community structure of modern floras leading to the
communities of plants and animals we know today (for example, Webb, 1992).
Waxing and waning of lake area in response to the precession cycle may have had
a comparable effect on the population size and distribution of modern organisms
(Jacobs and Hertel, 1994). Large lakes at temperate and subtropical latitudes were
pervasive throughout much of the Quaternary. Today such lakes are at a minimum
(for example, Street and Grove, 1979). Many organisms presumably evolved in the
context of this larger average lake area in the Quaternary. White pelicans inhabiting
Pyramid Lake in the Great Basin provide an example. Pyramid lake is a remnant of
much larger lakes that occupied the region during much of the Pliestocene. Lakes
were as much as three orders of magnitude more extensive in the Pleistocene
of the Great Basin than they are today. Presumably the white pelican population
became established in the basin when lakes were more extensive, and the pelican
population was much larger at that time as well.
Biologists often presume that large aquatic birds, such as the Whooping Crane,
were never numerous based on historic records of there abundance and distribution
(for example, Matheissen, 1967). These historic observations form the basis for
conservation objectives. However, on average during the Quaternary, the habitat
available to these birds was much larger than it has been in historic times. Thus
the population reductions from the long term average has been much more severe
than generally recognized. Presumably populations of pelicans, flamingos, storks
and cranes, groups which include some of the most endangered species of birds on
Earth today, waxed and waned with the lake environments. When human activity
destroys these environments as is currently the case in the Aral Sea, Mono Lake,
and Pyramid Lake, and the Tigris delta marshes, it accentuates processes already at
natural minimum on time scales of 104 to 106 Most species of crane are endangered
by precisely this combination of natural loss of habitat and accompanied by human
habitat destruction and depredation. Recognition that certain taxa were already in
bottlenecks prior to the effects of human activities, and that they may suffer
continued pressure from natural dynamics, suggests that much larger areas of
wetlands and other comparable habitats must be preserved if viable populations of
taxa such as Storks, Cranes, and Flamingoes are to be preserved.
11. Conclusions
1) Fluctuations of the monsoon driven by the precession cycle can store water
sufficient to influence sea-level during geologic periods that lacked continental
ice. This mechanism of sea-level change is very likely to have produced sea-
level fluctuations documented in Late Triassic carbonate packages of the Alpine
region. Similar sea-level changes produced at other time periods when evidence
for continental ice is limited may also have been generated in this way. Such times
include the Early Cretaceous and Devonian, and possibly the Cambro-Ordovician,
Late Permian-Triassic, and Jurassic.
2) Human activity can directly effect sea-level as a result of dam building,
aquifer pumping, diverting surface water for irrigation and destruction of forests,
among others. These activities change the relative balance and flux of water be-
tween the land and sea. Preliminary estimates of sea-level change resulting directly
from human activities suggest a current flux equivalent to at least 0.54 mm/year of
sea-level rise per year. More research on this topic is needed if the climate driven
component of sea-level change is to be isolated climatic.
3) Policy decisions are often based on the premise of a stable, unchanging
system. Low and mid latitude lake environments were widespread during the
Quaternary. They are now rare as a consequence of climate change driven by
orbital fluctuations. Recognition of this may lead to a better understanding of the
critical nature of the remaining habitats.
Acknowledgments
We greatly appreciate the effort and insight of the reviewers Bill Hay and Chris
Kendall, as well as comments by F. Gasse.
References
AL !BRAHAM, A. A., 1991, Excessive use of groundwater resources in Saudi Arabia:

impacts and policy options: Ambio, v. 20, p. 34-37.
ANDERSON, E. J., GOODWIN, P. W. AND SOBIESKI, T. H., 1984, Episodic accumula-
tion and the origin of formation boundaries in the Helderberg Group of New York State:
Geology, v. 12, p. 120-123.
ANDERSON, R. Y., 1982 A long geoclimatic record from the Permian, Journal of Geo-
physical Research v. 87, P. 7285-7294.
ANDERSON, T. G., WELDER, G. LESSER, G., AND TRUJILLO, A., 1988, Region 7:
Central Alluvial Basins in Back, w., Rosheshein, 1., and Soaber, P., Hydrogeology, The
Geology of North America vol. 0-2, Geological Society of America, Boulder Colorado,
524 pp.
ANKLIN, M., BARNOLA, J. M., BEER, J., AND BLUNIER, T., 1993, Climate instability
during the last interglacial period recorded in the grip ice core: Nature, v. 364, p. 203-207.
ASH, S. R., AND BASINGER, J. E, 1991, A high latitude Upper Triassic flora from the
Heiberg Formation, Sverdrup Basin, Arctic Archipelago. in Contributions to Canadian
Paleontology, Geological Survey of Canada, Bulletin 412, p. 101-131.
BALEK, J., Hydrogeology and water resources of tropical Africa: Developments in Water
Science, v. 8, Elsevier, 208 p.
BERGER, A., 1978, Long term variation of daily insolation and Quaternary climatic
changes: Journal of Atmospheric Science, v. 35, p. 2362-2367.
BERGER, A, LOUTRE, M. E AND DEHANT, V., 1989, Pre-Quaternary Milankovitch
frequencies: Nature, v. 342, p. 133.
BERGER, A. L., 1992, Astronomical theory of paleoclimates and the last glacial-
interglacial cycle: Quaternary Science Review, v. II, p. 571-581.
BINJIE, L., RUCHANG, R., PING, S., RONGYAD, H., HONG SHUN, S., YOUXIAO,
w., AND GUANGHUA, R., 1980, Formation and migration of continental oil and gas
in China: Scientia Sinica, v. 23, p. 1286-1295 and 1417-1427.
BOSELLINI, A AND HARDIE, L. A, 1991, Facies e cicli della Dolomia principalle delle
Alpi Venete: Mem. Soc. Geol. It. , v. 30, p. 245-266.
BRADLEY, W. H., 1929, The varves and climate ofthe green river epoch: U. S. Geological
Survey Professional Paper, v. 158-E, p. 87-110.
BRETT, C. E., 1986, Symmetrical and upward shallowing cycles in the Middle Devonian of
New York State and their implications for the punctuated aggradational cycle hypothesis:
Paleoceanography, v. 1, p. 431-445.
BREED, C. S., FRYBERGER, S. C., ANDREWS, S., MCCAULEY, c., LENNARTZ, E,
GEBEL, D., AND HORSTMAN, K., 1979, Regional studies of sand seas using Landsat
(ERTS) imagery, U. S. Geological Survey Prof. Pap. v. 1052, p. 305-398.
BROWN, R. H., 1980, Triassic rocks of the Argana Valley, Southern Morocco, and their
regional structural implications: The American Association of Petroleum Geologists, v.
64, p. 988-1003.
BURKE, K., AND SENGOR, A. M. c., 1988, Ten meter global sea-level change associated
with South Atlantic Aptian salt deposition, Marine Geology, v. 83, p. 309-312.
BUTZER, K. w., ISAAC, G.L., RICHARDSON, 1. L. AND WASHBOURN-KAMAU,
C., 1972, Radiocarbon dating of East African lake levels, Science, v. 175, p.1069-1076.
CAHEN, L., 1983, Breves precisions sur rage des groupes creataciques post-Wealdien
(Loia, Bokungu, Kwango) du Bassin interieur du Congo (Republique du Zaire): Mus.
R. Afr. Centro Tervuren (Belg.), Dept. Geol. Min.
CASTENY, G. , 1981, Hydrogeology of dcep aquifers: The hydrological basin as the basis
of groundwater management: Episodes, V. 15, p. 18-22.
CHAO, B. E, 1994, Man-made lakes and sea-level rise: Nature, V. 370, p.258.
CHAO, B., 1990, Man water and global sea-level: Eos, V. 72, p. 492.
CHAO, B., 1988, Excitation of the Earth's polar motion due to mass variation in major
hydrological reservoirs: Journal of Geophysical Research, V. 93, p. 13811-13819.
CLEMMENSEN, L. B., 1987, Complex star dunes and associated aeolian bedforms, Hope-
man Sandstone (Permo-Triassic), Moray Firth Basin, Scotland. in L. Frostick and I. Reid,
ed., Desert Sediments: Ancient and Modern: Geological Society ... , p. 213-231.
COHMAP, 1988, Climatic changes of the last 18,000 years: Observations and model
simulations, Science V. 241, p. 1043-1051.
CROWELL, J. C., 1978, Gondwanan glaciation, cyclothems, continental positioning, and
climate change: American Journal of Science, V. 278, p. 1345-1372.
CROWLEY, T. 1., BAUM, S. K. AND HYDE, W. T., 1992, Milankovitch fluctuations on
supercontinents: Geophysical Research Letters, v. 19, p. 793-796.
DANSGAARD, w., JOHNSEN, S. J., CLAUSEN, H. B., AND DAHLJENSEN, D., 1993,
Evidence for general instability of past climate from a 250-kyr ice-core record: Nature,
v. 364, p. 218-220.
DARWIN, C., 1839, Observations on the Parallel Roads of Glen Roy, and other parts
of Lochaber in Scotland, with an attempt to prove that they are of marine origin:
Philisophical Transactions of the Royal Society of London v. 129, p. 39-81.
DAWSON, A. G., 1992, Ice Age earth: late Quaternary geology and climate: Routledge,
London; New York, 293 p.
DE CISNEROS, C. J. AND VERA, J. A., 1993, Milankovitch cyclicity in Purbeck peritidal
limestones of the Prebetic (Berriasian, southern Spain): Sedimentology, v. 40, p. 513-
537.
DINGLE, R. V., SIESSER, W. G., AND NEWTON, A. R., 1983, Mesozoic and Tertiary
Geology of Southern Africa: Balkema, Rotterdam, 375p.
DRUMMOND, C. N. AND WILKINSON, B. H., 1993, Carbonate cycle stacking patterns
and hierarchies of orbitally forced eustatic sea-level change: Journal of Sedimentary
Petrology, v. 63, p. 369-377.
DUBIEL, R. F., PARRISH, 1. T., PARRISH, J. M. AND GOOD, S. C., 1991, The Pangean
megamonsoon - evidence from the Upper Triassic Chinle Formation, Colorado Plateau:
Palaios, v. 6, p. 347-370.
FAIRHEAD, F. D., 1988, Late Mesozoic rifting. in W. Manspeizer, ed., Triassic-Jurassic
Rifting: Continental Breakup and the Origin of the Atlantic Ocean and Passive Margins,
Part B: Elsevier, Amsterdam, p. 821-831.
FEJER, P. E. AND NARBONNE, G. M., 1992, Controls on Upper Devonian, metre-
scale carbonate cyclicity, Icefall Brook, southeast British Columbia, Canada: Bulletin
of Canadian Petroleum Geology, v. 40, p. 363-380.
FISCHER, A. G., 1964, Lofer cyclothems of the of the alpine Trias: Kansas geological
Survey Bulletin, v. 169, p. 107-148.
FISCHER, A. G. AND ROBERTS, L. T., 1991, Cyclicity in the Green River Formation
(lacustrine Eocene) of Wyoming: Journal of Sedimentary Petrology, v. 61, p. 1146-1154.
FONTES, J. c., GASSE, F., CALLOT, Y., PLAZIAT, J.-C., CARBONEL, P., DUPEU-
BLE, P. AND KACZMARSKA, I., 1985, Freshwater to marine-like environments from
Holocene lakes in northern Sahara: Nature, v. 317, p. 608-610.
FRAKES, L. A., FRANCIS, J. E., AND SYKTUS, J. I., 1992, Climate Modes of the
Phanerozoic: Cambridge University Press, Cambridge, 274 p.
FRAKES, L. A., AND KRASSAY, A. A., 1992, discovery of probable ice-rafting in the
Late Mesozoic of the Northern Territory and Queensland: Australian Jornal of Earth
Sciences, v. 39, p. 115-119.
GASSE, F., FONTES, J. c., PLAZIAT, J. C., CARBONEL, P., KACZMARSKA, I.,
DE DECKKER, P., SOULIE-MARSCHE, I., CALLOT, Y. AND DUPEUBLE, P. A.,
1987, Biological remains, geochemistry and stable isotopes for the reconstruction of
environmental and hydrological changes in the Holocene lakes from North Sahara:
Palaeogeography, Palaeoclimatology, Palaeoecology, v. 60, p. 1-46.
GASSE, F., ARNOLD, M., FONTES, J. C., FORT, M., AND OTHERS, 1991, A 13,000-
year climate record from western Tibet: Nature, v. 353, p. 742-745.
GASSE, F., LEDEE, V., MASSAULT, M. AND FONTES, J.-C., 1989, Water-level fluctu-
ations of Lake Tanganyika in phase with oceanic changes during the last glaciation and
deglaciation: Nature, v. 342, p. 57-59.
GASSE, F. AND VANCAMPO, E., 1994, Abrupt post glacial climate events in West Asia
and North Africa monsoon domains: Earth and Planetary Science Letters, v. 26. p.
435-456.
GILBERT, G. K., 1895, The sedimentary measurement of Cretaceous time: Journal of
Geology, v. 3, p.121-127.
GOLDHAMMER, R. K., DUNN, P. A. AND HARDIE, L. A., 1987, High frequency
glacio-eustatic sealevel oscilations with Milankovitch characteristics recorded in Middle
Triassic platform carbonates in Northern Italy: American Journal of Science, v. 287, p.

853-892.
GOLDHAMMER, R. K., DUNN, P. A. AND HARDIE, L. A., 1990, Depositional cycles,
composite sea-level changes, cycle stacking patterns, and the hierarchy of stratigraphic
forcing: Examples from Alpine Triassic platform carbonates: Geological Society of
GOLDHAMMER, R. K., LEHMANN, P. J. AND DUNN, P. A., 1993, The origin of high-
frequency platform carbonate cycles and third-order sequences (Lower Ordovician El
Paso Group, West Texas): Constraints from outcrop data and stratigraphic modeling:
Journal of Sedimentary Petrology, v. 63, p. 318-359.
GOODWIN, P. AND ANDERSON, E. J., 1985, Punctuated aggradational cycles: a general
hypothesis of episodic stratigraphic accumulation: Journal of Geology, v. 93, p. 515-527.
GOODWIN, P. w., ANDERSON, E. 1., GOODMAN, W. M. AND SARAKA, L. J., 1986,
Punctuated aggradational cycles: implications for stratigraphic analysis: Paleoceanog-
raphy, v. 1, p. 417-429.
GORNITZ, V., ROSENZWEIG, C. AND HILLEL, D., 1994, Is sea-level rising or falling:
Nature, v. 371, P.481.
GREUELL, W., 1994, Sea-level rise: Nature, v. 369, p. 615-616.
GRUNERT, J., BAUMHAUER, R. AND VOLKEL, J., 1991, Lacustrine sediments and
Holocene climates in the southern Sahara: The example of paleolakes in the Grand Erg
of Bilma (Zoo Baba and Dibella, eastern Niger): Journal of African Earth Sciences, v.
12, p. 133-146.
HAMBREY, M. J. AND HARLAND, W. B., 1981,. Earth's pre-Pleistocene glacial record:
Cambridge university Press, 10004 p.
HARDIE, L. A., BOSELLINI, A. AND GOLDHAMMER, R. K., 1986, Repeated subaerial
exposure of subtidal carbonate platforms, Triassic, northern Italy: Evidence for high-
frequency sea-level oscillations on a 104 year scale: Paleoceanography, v. 1, p. 447-457.
HAY, W. W. AND LESLIE, M. A., 1990 Could possible changes in global groundwater
reservoir cause eustatic sea-level fluctuations? p. 161-171 in Revelle, R. R. ed., Sea-
Level Change, National Academy Press, Washington, D.C.
HAYNES, C. V., EYLES, C. H., PAVLISH, L. A., RITCHIE, J. C. AND RYBAK, M.,
1989, Holocene paleoecology of the Eastern Sahara; Selima Oasis: Quaternary Science
Reviews, v. 8 pp. 109-136.
HAYS, J. D., IMBRIE, J., AND SHACKLETON, N. J. 1976, Variations in the Earth's
orbit: pacemaker of the Ice Ages: Science, v. 194, p. 1121-1132.
HERBERT, T. D., 1992, Paleomagnetic calibration of Milankovitch cyclicity in Lower
Cretaceous sediments: Earth and Planetary Science Letters, v. 112, p. 15-28.
HILLAIRE-MARCEL, C. AND CASANOVA, 1., 1987, Isotopic hydrology and paleo-
hydrology of the Magadi (Kenya)-Natron (Tanznia) basin duting the late Quaternary:
Palaeogeography, Palaeoecology, Palaeoecology, v. 58, p. 155-181
HILLAIRE-MARCEL, C., CARRO, O. AND CASANOVA, 1.,1986, 14C and Th/U dating
of Pleistocene and Holocene Stromatolites from east African Paleolakes: 71, v. ??, p.
312-329.
HINNOV, L. A. AND GOLDHAMMER, R. K., 1991, Spectral analysis of the Middle
Triassic Latemar Limestone: Journal of Sedimentary Petrology, v. 61, p. 1173-1193.
HOLZER, W. T., CLEMENT, G. P., JANSA, L. F. AND WADE, J. A., 1988, Evaporite
deposits of the North Atlantic rift. in W. Manspeizer, ed., Triassic-Jurassic Rifting:
Continental Breakup and the Origin of the Atlantic Ocean and Passive Margins, Part B:
Elsevier, Amsterdam, p. 711-755.
HOOGHIEMSTRA, H., 1989, Variations of the Northwest African trade wind regime dur-
ing the last 140,000 years: Changes in pollen flux evidenced by marine sediment records.
in M. Leinen and M. Sarthein, ed., Paleoclimatology and Paleometeorology: Modern

and Past Patterns of Global Atmospheric Transport: Kluwer Academic Publishers, p.
733-770.
HOLLAND, H., 1984, The Chenical Evolution of the Atmosphere and Oceans, Princeton
University Press, 582 p.
HSO, K. J., RYAN, W. B. E AND CITA, M. B., 1973, Late Miocene desiccation of the
Mediterranean: Nature, v. 242, p. 240-244.
HULL, D. L., 1973, Darwin and his critics; the reception of Darwin's theory of evolution
by the scientific community: Harvard University Press, Cambridge, 473 p.
JACOBS, D. K., AND HERTEL, E, 1984, Milankovitch forced fluctuations in biome size,
implications for evolution, coservation biology and interpretation of the fossil record:
EOS Supplement 1994, p. 79.
JACOBS, D. K. AND SAHAGIAN, D. L., 1993, Climate-induced fluctuations in sea-level
during non-glacial times: Nature, v. 361, p. 710-712.
JADOUL, E, BERRA, E AND FRISIA, S., 1992, Stratigraphic and paleogeographic
evolution of a carbonate platform in an extensional tectonic regime: The example of the
Dolomia Principale in Lombardy (Italy): Riv. It. Paleont. Strat., v. 98, p. 29-44.
JOHNSON, T. C., HALFMAN, J. D. AND SHOWERS, W. J., 1991, Paleoclimate of the
past 4000 years at Lake Turkana, Kenya, based on the isotopic composition of authigenic
calcite: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 85, p. 189-198.
KELLY, S. B., 1992, Milankovitch cyclicity recorded from Devonian non-marine sedi-
ments: Terra Nova, v. 4, p. 578-584.
KLEIN, R., LOYA, Y., GVIRTZMAN, G., ISDALE, P. J. AND SUSIC, M., 1990, Seasonal
rainfall in the Sinai Desert during the late Quaternary inferred from fluorescent bands
in fossil corals: Nature, v. 345, p. 145-147.
KORLYAKOV, V. M., 1991, The Aral Sea basin: a critical environmental zone: Environ-
ment, v. 33, p. 4-9.
KROPELEIN, S. AND SOULIE-MARSCHE, I., 1991, Charophyte remains from Wadi
Howar as evidence for deep mid-Holocene freshwater lakes in the eastern Sahara of
Northwest Sudan: Quaternary Research, v. 36, p. 210-223.
KUTZBACH, J. E. AND STREET-PERROTT, EA., 1985, Milankovitch forcing of fluc-
tuations in the level of tropical lakes from 18 to 0 kyr B.P.: Nature, v. 317, p. 130-134.
KUTZBACH, J. E. AND GALLIMORE, R. G., 1989, Pangaean climates: Megamonsoons
of the megacontinent: Journal of Geophysical Research, v. 94, p. 3341-3357.
KUTZBACH, J. E., PRELL, W. AND RUDDIMAN, W. E, 1993, Sensitivity of Eurasian
climate to surface uplift of the Tibetan plateau: Journal of Geology, v. 101, p. 177-190.
KUTZBACH, J. E. AND ZIEGLER, A. M.,1993, Simulation of Late Permian climate and
biomes with an atmosphere ocean model- comparisons with observations: Philosophical
Transactions of the Royal Society of London Series B-biological sciences, v. 341, p.
327-340.
LAUTENSCHLAGER, M., AND SANTER, B. D., 1991, Atmospheric response to a
hypothetical Tibetan ice sheet: Journal of climate, v. 4 , p. 386-394.
LAUTENSCHLAGER, M., AND SANTER, B., D., 1991, Atmospheric response to a
hypothetical Tibetan ice sheet: Journal of Climate, v. 4, p. 386-394.
LISTER, G. S., KELTS, K., ZAO, C. K., YU, J.-Q. AND NIESSEN, E, 1991, Lake
Qinghai, China: Closed-basin lake levels and the oxygen isotope record for ostracoda
since the latest Pleistocene: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 84,
p. 141-162.
MAISEY, J. G., 1991, Cretaceous sedimentary basins in Brazil. in J. G. Maisey, ed., Santana
Fossils: An Illustrated Atlas: T. E H. Publications, Inc., Neptune City, N.J., p. 31-33.
MATHIES SEN, P., 1967, Wildlife in America: Viking, New york, 332p.
MCCAULEY, C., BREED, C. S., SCHABER, G. G., MCHUGH, W. P., ISSAWI, B.,
HAYNES, C. V., GROLIER, M. J., AND KILANI, A. E., 1986, Paleodrainages of
the eastern Sahara-The radar rivers revisited (SIR-alb implications for a Mid-Tertiary
trans-African drainage system, EEE Transactions on Geoscience and Remote Sensing,
GE-24,624-647
MCKEE, E. D., 1979, Ancient sandstones considered to be eolian: U. S. Geological Survey
Prof. Pap. v. 1052, p. 189-238.
MCKENZIE, J. A., 1993, Pluvial conditions in the eastern Sahara following the penultimate
deglaciation: implications for changes in atmospheric circulation paterns with global
warming: Palaeogeography, Palaeoecology, Palaeoecology, v. 103, p. 95-105.
MEEHL, G. A., 1992, Effect of tropical topography on global climate: Annual Review of
Earth and Planetary Sciences, v. 20, p. 85-112.
MICKLIN, P., The Aral Crisis: Introduction to the special issue: Post-Soviet Geography,
v. 33, p. 269-282.
NEWMAN, W. AND FAIRBRIDGE, R., 1986, The management of sea-level rise: Nature,
v. 320, p. 319-321.
OLORIZ, F., MARQUES, B. AND RODRIGUEA-TOVAR, F. J., 1991, Eustatism and
faunal associations. Examples from the South Iberian Margin during the Late Jurassic
(Oxfordian-Kimmeridgian): Eclogae Geologicae Helvetiae, v. 84, p. 83-106.
OLORIZ, F., RODRIGUEA-TOVAR, F. J., CHICA-OLMO, M. AND PARDO, E., 1992,
The marl-limestone rhythmites from the Lower Kimmeridgian (Platynota Zone) of the
central Prebetic and their relationship with variations in orbital parameters: Earth and
Planetary Science Letters, v. 111, p. 407-424.
OLSEN, P. E., 1986, A 40-million-year lake record of early Mesozoic orbital climatic
forcing: Science, v. 234, p. 842-848.
OLSEN, H., 1990, Astronomical forcing of meandering river behaviour: Milankovitch
cycles in Devonian of East Greenland: Palaeogeography, Palaeoclimatology, Palaeoe-
cology,v.79,p.99-115.
OSLEGER, D. AND READ, J. F., 1991, Relation of eustasy to stacking patterns of meter-
scale carbonate cycles, late Cambrian, U.S.A.: Journal of Sedimentary Petrology, v. 61,
p.1225-1252.
PARK, J., D'HONDT, S. L., KING, J. W. AND GIBSON, C., 1993, Late Cretaceous
precessional cycles in double time: A warm-earth Milankovitch response: Science, v.
261, p. 1431-1434.
PATZKOWSKY, M. E., SMITH, L. H., MARKWICK, P. J., ENGBERTS, C. J. AND
GYLLENHAAL, E. D., 1991, Application of the Fujita-Ziegler paleoclimate model:
Early Permian and Late Cretaceous examples: Palaeogeography, Palaeoclimatology,
Palaeoecology, v. 86, p. 67-85.
PETIT-MAIRE, N. COMMELIN, D., FABRE, J., AND FONTUGNE, M., 1990, First
evidence for Holocene rainfall in the Tanzerouft hyperdesert and its margins: Palaeo-
geography, Palaeoecology, Palaeoecology, v 79, p. 333-338
PETIERS, S. w., 1991, Regional Geology of Africa: Springer-Verlag, New York, 722 p.
PRATT, L. M. AND KING, 1. D., 1986, Variable marine productivity and high eolian input
recorded by rhythmic black shales in mid-Cretaceous pelagic deposits from central Italy:
Paleoceanography, v. 1, p. 507-522.
PRELL, W. L. AND VAN CAMPO, E., 1986, Coherent response of Arabian Sea upwelling
and pollen transport to late Quaternary monsoonal winds: Nature, v. 323, p. 526-528.
RITCHIE, J. C., EYLES, C. H. AND HAYNES, C. v., 1985, Sediment and pollen evidence
for an early to mid-Holocene humid period in the eastern Sahara: Nature, v. 314, p. 352-
355.
READ, J.E, AND GOLDHAMMER, R. K., 1988, Use of Fischer plots to define 3rd
order sea level curves in peritidal cyclic carbonates, Early Ordovician, Appalachians:
Geology, v. 16, p. 895-899.
READ, J. F, GROTZINGER, 1. P., BOVA J. A. AND KROERSCHNER, W. E, 1986,
Models for generation of carbonate cycles; Geology, v. 14, p. 107-110.
ROBINSON, P. L.,1973, Paeleoclimatology and continental drift: in Tarling, D. H. and
Runcorn, S. K., eds. Implications of Continental Drift to the Earth Sciences, I, Academic
Press, London, p. 449-476.
RODENBURG, E. 1994, Man-made lakes and sea-level rise: Nature, v. 370, p.258.
ROGERS, D. A., AND ASTIN, T. R. 1991, Ephemeral lakes, mud pellet dunes and
windblown sand and silt: Reinterpretation of Devonian lacustrin cycles in N. Scotland:
Special Publication International Association of Sedimentology v. 13, p. 199-221.
ROSSIGNOL-STRICK, M., 1983, African monsoons, an immediate climate response to
orbital insolation: Nature, v. 304, p. 46-48.
ROWLEY, D. B. AND MARKWICK, P. J., 1992, Haq and others eustatic sea-level curve:
Implications for sequestering water volumes: Journal of Geology, v. 100, p. 703-715.
SADLER, S. P. AND KELLY, S. B., 1993, Fluvial processes and cyclicity in terminal
fan deposits: An example from the Late Devonian of southwest Ireland: Sedimentary
Geology, v. 85, p. 375-386.
SAHAGIAN, D. L., SCHWARTZ, E W AND JACOBS, D. K., 1994a, Direct anthro-
pogenic contributions to sea-level rise in the twentieth century: Nature, v. 367, p. 54-57.
SAHAGIAN, D. L., JACOBS, D. K., AND SCHWARTZ, E W, 1994b, Sea-level rise-
reply. Nature, v. 369, p. 616.
SAHAGIAN, D. L., JACOBS, D. K., AND SCHWARTZ E W., 1994c, Man-made lakes
and sea-level rise - reply: Nature, v. 370, p. 258.
SANDERS, B., 1936, Beitrage zur Nenntnnis der Anlagerungsgefuge: Minerol. Pet-
rogr.MitL, v. 48, p. 27-139.
SCHOLZ, C. A. AND FINNEY, B. P., 1994, Late Quaternary sequence stratigraphy of
Lake Malawi (Nyasa), Africa: Sedimentology, v. 41, p. 163-179.
SCHWARZACHER, W, 1993, Cyclostratigraphy and the Milankovitch Theory: Develop-
ments in Sedimentology: Elsevier, Amsterdam, p.225.
SCHLAGER, W, 1981, The paradox of drowned reefs and carbonate platforms: Geological
Society of America Bulletin, v. 92, p.197-21 1.
SCOTESE, C. R., 1987, Phanerozoic plate tectonic reconstructions; Paleoceanographic
Mapping Project, Technical Report, no. 90, 56p.
SHORT, D. A., MENGEL, J. G., CROWLEY, T. J., HYDE, W T. AND NORTH, G. R.,
1991, Filtering of Milankovitch cycles by earth's geography: Quaternary Research, v.
35, p. 157-173.
SOCCI, A. D., 1992, Climate, glaciation and deglaciation, controls, pathways, feedbacks,
rates and frequencies: Modern Geology, v. 16, p. 279-316.
STREET, E A., AND GROVE, A. T., 1979, Global maps of lake level fluctuations since
30,000 bp: Quaternary Research, v. 12, p. 83-118.
STREET-PERROTT, EA., MITCHELL, J. E B., MARCHAND, D. S. AND BRUNNER,
1. S., 1990, Milankovitch and albedo forcing of the tropical monsoons: a comparison of
geological evidence and numerical simulations for 9,OOOyBP: Transactions of the Royal
Society of Edinburgh: Earth Sciences, v. 81, p. 407-427.
TEN KATE, W G. H. Z. AND SPRENGER, A., 1993, Orbital cyclicities above and
below the Cretaceous/Paleogene boundary at Zumaya (North Spain), Agost and Relleu
(Southeast Spain): Sedimentary Geology, v. 87, p. 69-101.
UCHAPI, E., 1989, The tectonic style of the Atlantic Mesozoic rift system: Journal of
African Earth Science, v. 8, p. 143-164.
VAN CAMPO, E., 1986, Monsoon fluctuations in two 20,000 yr B.P. oxygen-isotope/pollen
records off Southwest India: Quaternary Research, v. 26, p. 376-388.
VANCAMPO, E. AND GASSE, F., 1993, Pollen-inferred and diatom-inferred climatic and
hydrological changes in Sumxi co basin (western Tibet) since 13,000 yr bp: Quaternary
Research, v. 39, p., 300-313.
VAN HOUTEN, F., 1964, Cyclic lacustrine sedimentation, upper Triassic Lackatong for-
mation, central New Jersey and adjacent Pennsylvania: Kansas Geeological Survey
Bulletin, v. 169, 497 -532.
WEBB, T., 1992, Past changes of vegetaion and climate: lessons for the future: in Peters, R.
1., and Lovejoy, T. E., eds., Global Warming and Biological Diversity, Yale University,
p.59-72.
WEEDON, G. P., 1985, Hemipelagic shelf sedimentation and climatic cycles: the basal
Jurassic (Blue Lias) of South Britain: Earth and Planetary Science Lettesa, v. 76, p.
321-335.
WEEKS, J., AND GUTENTAG, E., 1988, Region 17: High Plains, in Back, w., Rosheshein,
J., and Soaber, P., Hydrogeology, The Geology of North America, Geological Society
of America, Boulder Colorado, 524 pp.
WYCISK, K., KLITZSCH, E., JAS, c., AND REYNOLDS, 0., 1990, Intracratonic se-
quence development and structural control of Phanerozoic strata in Sudan: Berliner
geowiss. Abh. A. 120, p. 45-86.
YEMANE, K., 1993, Contribution of Late Permian palaeogeography in maintaining a
temperate climate in Gondwana: Nature, v. 361, p. 51-54.
YEMANE, K., AND KELTS, K., 1990, A short review of palaeoenvironments for Lower
Beaufort (Upper Permian) Karoo sequences from southern to central Africa: A major
Gondwana lacustrine episode: Journal of Afriacan Earth Sciences, v. 10, p. 169-185.
ZIEGLER, P. A., 1982, Geological atlas of Western and Central Europe: 130 p. Shell
International Petroleum.
ZIEGLER, P. A., 1988, Post-Hercynian plate reorganization in the Tethys and Arctic-North
Atlantic domains. in W. Manspeizer, ed., Triassic-Jurassic Rifting: Continental Breakup
and the Origin of the Atlantic Ocean and Passive Margins, Part B: Elsevier, Amsterdam,
p.711-755.
CHAPTER 12
The Enigma of Third-Order Sea Level

Cycles: A Cosmic Connection?
Christopher G.St.c. Kendall, Phillip A. Levine and Robert Ehrlich
ABSTRACT: We propose the hypothesis that hypervelocity asteroids and/or comets that
collide with the Earth may be responsible for most third-order global sequences. Colli-
sions with asteroids and/or comets greater than 10 km in diameter can produce 1032 to
1033 ergs of energy which is capable of inducing global release of stress at plate bound-
ary faults. In response. continental margins adjust to near isostatic equilibrium. inducing
global marine transgressions. The frequency of terrestrial impact craters coincident with
eustatic sea level events lends support to this hypothesis.
Previously. third order (1-3 million years) depositional sequences which characterize
most continental margins had been attributed to either glacial eustasy. regional release of
stress at continental plate boundaries. tectonically driven variations in sediment supply,
or geoidal distortion. However. these mechanisms do not adequately explain global
third-order cycles. as much of the stratigraphic record cannot be equated with periods of
glaciation and the latter three effects are local.
1. Introduction
The sedimentary record is globally divided into unconformity bounded sequences

that progressively onlap and offlap the continental margins in a quasi-regular man-
ner (Vail and others, 1977; Vail and Todd, 1981; Haq and others, 1987; Revelle,
1990). These global onlapping sequences are subdivided on the basis of order of
duration into four types of sequences (Vail and others, 1977; Worsley and others,
1984; Revelle, 1990; Plint and others, 1992). First-order sequences, which last
in excess of 50+ million years, are related to continental breakup; second-order
sequences, which range in duration from 3-50 million years, are related to ridge
spreading and crustal cooling (Vail and others, 1991); third-order stratigraphic se-
quences, which last 1-3 million years, have been attributed to glacial eustasy (Haq
and others, 1987; Vail and others, 1977) for specific geologic time periods, stress
release at plate margins (Cloetingh, 1992), tectonism with subsequent sediment
erosion (Galloway, 1989a,b), and geoidal eustasy (Marner, 1981, 1987; Sabadini
367
B. U. Haq (ed.),
1995 Kluwer Academic Publishers. Printed in the Netherlands.
368 C.O.St.c. Kendall, P.A. Levine, R. Ehrlich
and others, 1990); and fourth-order sequences, which range in duration from 0.5-1
million years, seem to be related to tectonics and variation in climate (Vail and
others, 1991). This high frequency climatic variation is ascribed to perturbations in
the Earth's orbit and the inclination of its axis of rotation (Plint and others, 1992).
With the exception of third-order sequences, the mechanisms previously dis-
cussed appear to adequately account for global sedimentary cycles identified in
the geological record. However, excluding glaciation, the mechanisms invoked to
explain third-order sedimentary sequences do not affect the global sea level sig-
nal and only produce local or regional effects. Therefore, they do not adequately
explain global third-order cycles.
2. Third-Order Sequences Driven by Glaciation
Glacial eustatic signals are restricted to the Pleistocene, Late Tertiary (Fig. 1),
Pennsylvanian, Permian, Ordovician and Precambrian (Revelle, 1990; Plint and
others, 1992). Third-order stratigraphic sequences can be linked to these glacial
events (Miller and others, 1990), however, the numerous Mesozoic and early
Tertiary sequences (Haq and others, 1987) do not appear to be the product of
glaciation, since the climate during these times was warmer and there is no evidence
of global glaciation (Revelle, 1990).
3. Third-Order Sequences Driven by Stress Release
An examination of the Haq and others (1987) eustatic sea level curves indicates
that there were numerous third-order eustatic events prior to 36 Ma that cannot
be attributed to glaciation (Fig. 1). Cloetingh (1992) has proposed a mechanism
which could account for the creation of this high frequency accommodation. He has
argued that stress builds up within continental margins in response to mid-oceanic
ridge spreading and is released at third-order time scales along plate margins,
causing local and regional transgressions. He has suggested that subsidence at the
continental margin is induced by a combination of ridge spreading, thermal cooling,
and the flexural response to in place sediment loading. Towards the continental
interior, uplift occurs in response to pressure from ridge spreading and the flexural
response to down dip sediment loading. This leads to an offlapping sediment wedge
on the continental margin. Cloetingh (1992) then argued that regional release of
the stress along faults at the continental margin leads to the reduction of pressure
from oceanic ridge spreading. Subsequently, uplift of the immediate edge of the
continent occurs and the interior of the continental margin subsides. This produces
the characteristic onlapping transgressive systems tract that extends across the
outer edge of the continent.
The mechanism proposed by Cloetingh (1992) does not explain global third-
order sequences, but is capable of producing local or regional cycles. Even when
the induced release of stress is accompanied by a magnitude 7 to 9 earthquake
(1027 ergs), this effect is only expressed as a regional signal along plate margins
and lacks the capability to produce a global record. While the signal from large
The Enigma of 3rd-Order Sea Level Cycles: A Cosmic Connection? 369
STAGES
EUSTATIC CURVES
after Haq et 01., 1981 AGE
STAGES
~R~~Aif~~ ~l{~~~S1987 AGE
~50 150 50 OM Ma AGE OF IMPACTS 2&0 _ 150 5Q_..tJM Ma
o a/ler Grieve, 1991 ,...--- 6fi.S --~~~~~~~~~~f[-AGE
a{ler OF IMPACTS
................ .
Quaternary ............................ .. Grieve, 1.991
~
t
~
~ ---~----------
8;:~~!!ia!!:n~~~-----
;~.~~~~~~~~~~~~~:~ ~~~1f- :~:-"-~ ~.~.~.~.~.
Maastrichtian 70
----
5
----- -1- - --~--
<:=
~.............. 75
--
c '4 - -
t, '_'.".._.' .t: ......
r--
--
Tortonlan
"
.Q
- 1 0 " - 1 - .' _
_._. '_'_"+- "-::::0_'..."," -10 iii r)
Serravallian
............ , ..... . <:= 'u
.J!!
"~
Campanian
80
- 15 ~
Q)
c '---- 84 - --- F,----------
--
""@ 85
--
20 a Santonian
0
Aqullanlan
If' a ----ConECiaL- ::::- :::::::::::::::
'8 90
c- ".2---t:~:j:~:.~-:~:~:~:.~-./'.-~:--~-~-"2'~''~'ij' -25 .~
'---- .,
Turonian
----
....
:-:;.
............ .
--
CL
---------- --~-----
k ............. .
}~
ChaH'an
95
- --1f-'-''-''-'-''-f.'t"..-'-'-' '-'-1' ',,-'
1...,,:.-=. 30
Cenomanian
Rupellan -------((-----------
....... .......... .
f ..... .............. , 100 __ <:=
-'t-
~
*--
~ 35
l
---;:~~,a:n1~-8.8.8.~.~.~.~:~:~:i:~:'~.:B:8:8:~:~: <:= Albian
-----\---------
_______ ~----------- --
------- ---<,,---------
-~rt:l~
39A~i:~~~tJ:;:;~~coco,,.;:;.;:;.~.~. 40
- - - >-,-------
105
-- -;------------ ----- ....... t1 .....
..
---- IUS
---
Lutetian ...... /; ............. 45
<:=
Aptian 110 --
- 113 --t:=:::::=~~:::::==~
50
- - 116.5
Barremian _ J===t=====~
--\
....... ..... .
:55
<:= Hauterivian _~rf~~~3?~~~.~..~.~.~.~.~.~.~. 120 <:=
--
~ 121 --
~
Thanetlan
- - - - - - -~ - - - - - - - - - --
60
Valanginian K 125
_ .... _--l-----,/rL------j
Danian 1,"< - --------f--=-----i
_Ii l~ I 130
-----&a5-~~~~~~~~~~~~~~~~~~
1
<II
f-~3r - -------
------- 1'\------
-134 -
135
KEY
*
IMPACT CRATER DIMENSIONS
--
-J411 - 140 <:=
Diameter>100knu
Kimmeridgian
..... Diameter>10kms
~ Diameter<10kmll
TRANSGRESSIONS
Oxfordian
C~r::d 150
1);t::C"I'Sf.~Y:'
Error in Age + /. 5m.y.
Kendall, Ehrlich, Levine 19."15
Fig, 1. Timing of eustatic sea level variation (Haq and others, 1987) and extraterrestrial impacts of
different sizes (Grieve, 1991),
370 c.G.St.c. Kendall, P.A. Levine, R. Ehrlich
earthquakes can be detected on the other side of the globe, active movement of
the crust is not expressed much beyond the vicinity of the epicenter. For instance,
a magnitude 9 earthquake in Bolivia in 1994 caused buildings to sway in Toronto,
but interplate displacement was localized.
4. Third-Order Sequences Driven by Sediment Supply
Another mechanism which has been invoked to explain third-order sequence de-
velopment involves variations in sediment supply which modulate low frequency
second-order accommodation. Galloway (1989a,b) maintained that in the Tertiary
Gulf of Mexico section, changes in sediment supply were driven by tectonic uplift
of the continental interior and subsequent erosion. This uplift is a flexural response
to down dip sediment deposition at the shelf margin, in conjunction with plate col-
lision resulting from mid-oceanic ridge spreading. The accommodation also was
enhanced by subsidence related to crustal cooling. Following highstand prograda-
tion at the shelf margin, sediment loading led to further uplift of the interior shelf
edge, and the beginning of sediment offlapping and formation of a bypassing shelf
margin wedge.
As with Cloetingh's (1992) model, the Galloway (1989a,b) model also does
not adequately explain global third-order transgressions and sequences. The uplift
of the continental interior, variations in sediment supply, and thermally driven
subsidence are local or regional processes only.
5. Third-Order Sequences Driven by Geoidal Distortion
Changes in the Earth's gravitational field cause distortions in the crust in the form
of localized bulges and sags which can be detected with satellites (Morner, 1981,
1987). Morner has suggested tha~ these global distortions change their location
periodically, thereby inducing relative sea level changes which can locally produce
third-order sequences. He postulated that these sequences do not occur globally
and felt that these "relative" sea level changes might be misinterpreted as eustatic
signals because of inadequate stratigraphic correlation. The shortcoming of this
model is that the changes in the gravitational field which produce the crustal bulges
and sags will also be matched by sea level being raised or lowered. The net effect
is that there is no relative change in sea level. Furthermore, geoidal changes take
place on time scales that are much longer than third-order eustatic changes.
A similar argument can be applied to the Sabadini and others (1990) model,
which proposes that relative sea level changes are a response to variations in the
Earth's axis of rotation. Again, we expect that the elevation of the sea and the crust
would tend to change in tandem.
6. Hypothesis: Extraterrestrial Impacts Cause Third-Order Sequences
We propose an alternative model in which extraterrestrial collisions with the Earth

are considered to be responsible for third-order frequency accommodation (Kendall
and others, 1994). When asteroids and/or comets greater than 10 kilometers in
diameter strike the Earth, traveling at velocities exceeding 30 to 40 kilometers per
second, they can release 1032 to 1033 ergs of energy (Spudis, 1993). The amount
of energy released by hypervelocity impacts is supported by studies of crater and
basin dimensions on the Moon, Mercury, Mars, Venus, the Jovian moons, and
Earth (Spudis, 1993). This energy release is 106 times greater than the amount
of the energy released by magnitude 7 to 9 earthquakes. We propose that the
energy of the impact event induces the crust to undulate worldwide (Clube and
Napier, 1982), causing the global release of stress developed by mid-oceanic ridge
spreading along continental margin faults (Fig. 2).
Prior to the collision, mid-oceanic ridge spreading, coupled with crustal cooling
and sediment loading, causes subsidence to occur at the most seaward edge of the
continents, while uplift occurs towards the continental interiors. A relative sea
level fall is produced and sediment is deposited along the continental margins
as offlapping shelf margin wedges (Fig. 2). Following the collision, stress release
occurs and the crust moves into isostatic equilibrium. This results in the continental
margins being uplifted at their most seaward edge, while subsidence occurs towards
the continental interiors. A rapid transgression ensues, driving clastic sediment
sources shelfward (Fig. 2). This model is essentially Cloetingh's (1992) mechanism
for generating third-order sequences, however stress release is caused by an impact
event rather than conventional geophysical forces such as earthquakes.
Consequently, with an increase in water depth and accommodation space on
the shelf, the sediments of the transgressive systems tract are condensed and
carbonate production is diminished. The resulting sediment is deposited worldwide
and predominantly consists of black shales that are rich in organics and radioactive
minerals (Baum and Vail, 1988; Loutit and others, 1988). Movement of the oxygen
minimum zone above the shelf encourages the preservation of these organic rich
shales (Wignall, 1994). The airborne products of the hypervelocity collision may
cause the element iridium to reach anomalously high values (Alvarez and others,
1982; Alvarez, 1983) in the basal transgressive shales deposited at the onset of the
sea level rise.
We propose that following the collision, the crust assumes a position of iso-
static equilibrium. This occurs at rates that initially match those following glacial
unloading, which can last for tens of thousands of years (Revelle, 1990; Plint and
others, 1992). However, the Earth's isostatic response to the impact event is longer
term, when compared with isostatic rebound rates from glacial melting. The iso-
static response begins much the same as the response to glacial melting, and then
slows as the natural hysteresis of the Earth exerts its effect. The process of isostatic
crustal rebound following the impact event appears to last from 100 thousand to
2 million years, corresponding to the rates of third-order marine transgressions of
Haq and others (1987).
Following the initial rapid transgression, isostatic uplift of the seaward edge
of the continental margin slows and highstand systems tract sediments start to pro-
grade across the interior portions of the margins (Fig. 2). Eventually, mid-oceanic
ridge spreading is re-established as the dominant process, producing horizontal
372 C.O.St.c. Kendall, P.A. Levine, R. Ehrlich
MID-OCEAN RIDGE DRIVEN TECTONIC STRESS (BOTH EXTENSION

& COMPRESSION) INDUCES RELATIVE SEA LEVEL LOW
HYPER VELOCITY COLLISION INDUCES RELEASE OF TECTONIC

STRESS & RAPID GLOBAL TRANSGRESSION
/"ul 82
II
TECTONIC STRESS FROM RIDGE SPREADING RESUMES

WITH GLOBAL HIGHSTAND PROGRADATION
III
MID-OCEAN RIDGE DRIVEN TECTONIC STRESS (BOTH EXTENSION

& COMPRESSION) INDUCES RELATIVE SEA LEVEL LOW
IV
Fig. 2. Diagrammatic cross sections tracking the evolution of the tectonic behavior and sedimentary
fill of a continental margin as it responds to an extraterrestrial hypervelocity collision.
RIDGE SPREADING exerts a horizontal stress on the adjacent oceanic and continental crust.
SUBSIDENCE at the continental margin is induced by ridge spreading, crustal cooling, and flexural
response to in place sediment loading.
UPLIfT towards the continental interior is induced by ridge spreading and flexural response to down
dip sediment loading.
ISOSTATIC SUBSIDENCE is induced by the release of stress along shelf margin faults.
ISOSTATIC UPLIfT is induced by the release of stress along shelf margin faults.
I. Continental margin prior to collision when tectonic stress from mid-ocean ridge spreading (both
extension and compression) induces a relative sea level low. Note the offtapping shelf margin wedge
(insets A I and B I).
II. Continental margin immediately after hypervelocity impact when mid-ocean ridge stress is re-
leased along shelf margin faults. Isostatic subsidence occurs towards the continental interior, whereas
isostatic uplift occurs at the continental margin. Ridge spreading continues but loses relief. Note that
the transgression drives the clastic sediment source shelfward (insets A2 and B2) and leads to a
condensed section down dip. The condensed section sediments consist of organic rich shales which
may be anomalously rich in iridium.
III. Continental margin when effect of tectonic stress from mid-oceanic ridge spreading resumes,
causing uplift of the continental interior and subsidence of the margin. Ridge spreading begins to
recover relief and a rapidly prograding highstand systems tract develops (insets A3 and B3).
IV. Continental margin when tectonic stress once again induces relative sea level low. Note the
development of an offlapping sediment wedge that bypasses the shelf margin.
stress that causes uplift of the interior portions of continental margins, while of-
flapping sediment wedges are formed that bypass the shelf (Fig. 2).
The most compelling evidence for the relationship between extraterrestrial im-
pacts and third-order sequences can be seen when the age of the impacts compiled
by Grieve (1991) is compared with the eustatic events of the Haq and others (1987)
sea level chart (Fig. 1). Note that the frequency of impact events appears to coincide
with third-order marine transgressions. Also, several of the more extensive marine
transgressions (lasting a relatively longer time), and associated black shales and
condensed sequences, coincide with some of the larger collisions (Fig. 1). These
include the impact at 35 Ma and the impact near 65 Ma, each of which is followed
by long periods of highstand. From Fig. 1, it appears that third-order sea level
changes occur more frequently than impacts resulting from collisions with large
diameter asteroids and/or comets. For this reason, we suspect that asteroids and/or
comets less than 10 km in diameter, which strike the Earth more frequently, also
are involved in generating third-order sequences.
We propose that during the middle to late Cenozoic (35 Ma and younger), the
third-order eustatic signals were the product of a combination of glacial processes
and impact events. If an impact event coincided with a time oflarge ice volumes and
a corresponding low sea level, a transgression would follow the impact. However,
if an impact event coincided with low ice volumes and a corresponding sea level
high, the sea level would be coincidentally higher. A higher periodicity in the sea
level cycle would likely reflect that of the glacial events, while the longer term
perturbations would be driven by the impacts. The higher frequency glacial events
would tend to obscure the impact events, making it difficult to separate the two
processes.
7. Evidence for Impacts
If our hypothesis is proven correct, third-order stratigraphic sequences and trans-

gressions may be added to the list of effects attributable to extraterrestrial collisions.
Direct evident of these impact events are craters, shock metamorphism of sedi-
mentary grains, tektites, tsunamis, and ore deposits. Indirectly, they are thought
to cause major extinctions (Alvarez and others, 1982; Alvarez and others, 1983),
anomalous iridium concentrations in shales (Alvarez and others, 1982) (depending
on the composition of the impacting body), magnetic field reversals (Clube and
Napier, 1982), the development of magmatic plumes and flood basalts (Huffman,
1990); and additionally, extensive marine transgressions and organic rich shales.
Evidence for interplanetary impacts includes the recent comet fragment col-
lisions with Jupiter and the numerous large impact craters which occur on the
surface of the Moon, Mercury, Venus, Mars, the Jovian moons, and Earth. The
craters which are formed by impacts can be enormous and their effects can be
far reaching. Areas on the Moon exceeding 3,000 kilometers have been affected
by impacts, as for example, the Orientale rings. Recent comet fragment collisions
with Jupiter have left holes the size of the Earth in Jupiter's upper atmosphere.
Asteroid and/or comet impacts have occurred throughout Earth's history.
Grieve (1991) has catalogued more than 130 impact craters on the Earth pro-
374 c.G.St.c. Kendall, P.A. Levine, R. Ehrlich
duced during the past 2 billion years. The diameter of these craters varies from 10
to 200 kilometers. Several craters have been identified on the Earth greater than
100 kilometers in diameter. These sites include: Sudbury, Ontario (Grieve, 1991);
Chicxulub, Mexico (Hildebrand and others, 1991); and Veredfort, South Africa
(Spudis, 1993). Although many more impacts have probably occurred, craters on
the Earth are commonly unrecognized because craters smaller than 20 kilometers
in diameter are easily eroded and most extraterrestrial impacts are in the ocean
and would not be identified (Grieve, 1991; Spudis, 1993). Large craters also can
be removed as a result of tectonic processes and plate motion.
The potential for asteroid and/or comet collisions is quite high. Shoemaker
(1983) has presented data illustrating that more than 1000 objects greater than one
kilometer in diameter can be identified within Earth's orbit. He projects that objects
with a diameter of 0.5 kilometer will produce craters of 10 kilometers in size every
100 thousand years. A 10 kilometer diameter object will produce a crater with a
diameter of 200+ kilometers every 40 million years. He also proposes that a 20
kilometer diameter object will collide with the Earth every 3 billion years. The
actual rate of bombardment of the Earth appears to match that of the Moon and
Shoemaker's predictions. That asteroid and/or comet impacts occur frequently is
supported by the fact that between 1975 and 1992, the USA early warning system
detected 136 explosions resulting from extraterrestrial objects striking the Earth's
upper atmosphere (Economist, 1993).
8. Conclusions
The hypothesis is proposed that extraterrestrial hypervelocity impacts may be re-

sponsible for some third-order rapid global transgressions. Asteroid and/or comet
impacts could cause global relaxation of stress in the Earth's crust, leading to uplift
of the crust at the continental margins and the subsequent development of rapid
transgressions. Organic rich shales accumulate with the corresponding increase in
accommodation space on the shelf and movement of the oxygen minimum zone
to shallower depths that encourage the preservation of organic matter. Tektites
are wide spread and anomalously high iridium concentrations may occur in the
shales. With the exception of glaciation, other mechanisms cited as responsible
for the third-order transgressive cycles appear to be local or regional in extent. If
our hypothesis is proven correct, extraterrestrial impacts may explain many of the
wide spread enigmatic global events which are associated with organic rich, hy-
drocarbon source rocks found throughout the geological record. We also propose
that these impact events and associated rapid third-order transgressions form good
stratigraphic markers and can be used for global correlation and prediction of hy-
drocarbon source rocks. We conclude, therefore, that impact events may ultimately
be responsible for the global significance of the Haq and others (1987) eustatic sea
level chart.
Acknowledgments
We would like to express our appreciation for the editorial comments of Des. Gerry
Baum, James Iliffe, Greg Mountain, and Bilal U. Haq.
References
ALVAREZ, L. w., LUIS, W. A., FRANK A. and HELEN, V. M., 1982, Current status of the
impact theory for the terminal Cretaceous extinction, Geological Society of America,
Special Paper 190, 1982, p. 305-315.
ALVAREZ, L. w., 1983, Experimental evidence that an asteroid impact led to the extinction
of many species 65 million years ago, Proceedings of National Academic Science USA,
v. 80, p. 627-642.
BAUM, G. R. and VAIL, P. R., 1988, Sequence stratigraphic concepts applied to Paleocene
outcrops, Gulf and Atlantic Basins, in Wilgus, C. K., Hastings, B. S., Kendall, C. G. St.
C., Posamentier, H. W., Ross, C. A., and Van Wagoner, 1. c., eds., Sea-Level Changes -
publication, v. 42, p. 309-328.
CLOETINGH, S. P., 1992, Lithospheric dynamics and the tectonics of sedimentary basins,
reprinted from Proceedings of Koninklijke Nederlandse Akademie van Wetenschappen,
Amsterdam,95 (3), p. 349-369.
CLUBE, S. V. M. and NAPIER, W. M., 1982, The role of episodic bombardment in
geophysics, Earth and Planetary Science Letters, v. 57, p. 251-262.
ECONOMIST, 1993, Science and technology: The hard rain, The Economist, September
lIth-17th, 1993, p. 81-84.
GALLOWAY, w., 1989a, Genetic stratigraphic sequences in basin analysis I: Architecture
and genesis of flooding-surface bounded depositional units, The American Association
of Petroleum Geologists Bulletin, v. 73, no. 2, p. 125-142.
GALLOWAY, w., 1989b, Genetic stratigraphic sequences in basin analysis II: application
to Northwest Gulf of Mexico Cenozoic Basin, The American Association of Petroleum
Geologists Bulletin, v. 73, no. 2, p. 143-154.
GRIEVE, R. A. E, 1991, Terrestrial impact: The record in the rocks, Geological Survey of
Canada, Ottawa, Canada, Meteoritics 26, p. 175-194.
HAQ, B., HARDENBOL, J. and VAIL, P. R., 1987, Chronology of fluctuating sea levels
since the Triassic (250 million years to present), Science, no. 235, p. 1156-1167.
HILDEBRAND, A. R., GLEN, T. P., DAVID, A. K., MARK, P., ANTONIO, C. Z., STEIN,
B. J. and WILLIAM, V. B., 1991, Chicxulub Crater: A possible Cretaceousffertiary
boundary impact crater on the YucatOn Peninsula, Mexico, Geology, v. 19., p. 867-871.
HUFFMAN, A. R., 1990, An endogenous mechanism for extinctions, Geotimes, v. 35, no.
8, p.16-17.
KENDALL, C. G. ST. C., EHRLICH, R. and RIA, L., CHRISTENSEN, B., . LEVINE,
P. A., MWAKAHESYA, L., NESTEROFF, 1. L., PRUESER, R. L., THIBODEAU, P.,
SAFOURI, Y. L., SIREGAR, P. H. and WONG, M. L., 1994, Extraterrestrial collisions:
An explanation for high-frequency changes in accommodation misinterpreted as eustatic
sea level changes, Proceedings of the American Association of Petroleum Geology, 1994
Annual Convention, v. 3, p. 186.
LOUTIT, T. S., HARDENBOL, J., VAIL, P. R. and BAUM, G. R., 1988, Condensed
sections: The key to age determination and correlation of continental margin sequences,
in Wilgus, C. K., Hastings, B. S., Kendall, C. G. St. c., Posamentier, H. W., Ross, C.
376 c.G.St.C. Kendall, P.A. Levine, R. Ehrlich
A., and Van Wagoner, J. c., eds., Sea-Level Changes - an Integrated Approach, Society
of Economic Paleontologists and Mineralogists Special pUblication 42, p. 183- 216.
MILLER, K. G., KENT, D. V., BROWER, A. N., BYBELL, L. M., FEIGENSON, M. D.,
OLSSON, R. K. and POORE, R. Z., 1990, Eocene-Oligocene sea level changes on the
New Jersey coastal plain linked to the deep-sea record, Geological Society of America
Bulletin, v. 102, p. 331-339.
MORNER, N. A., 1981, The Northwest European "Sea-Level Laboratory" and regional
Holocene eustasy, Palaeogeography, Palaeoclimatology, Palaeoecology, v. 29, p. 281-
300.
MORNER, N. A., 1987, Pre-Quaternary long-term changes in sea level, in Devoy, R. J.
N., ed., Sea surface studies, a global view, New York, Croom Helm, p. 242-263.
PUNT, A. G., EYLES, N., EYLES, C. H. and WALKER, R. G., 1992, Control of sea level
change: in R.G. Walker and N.P. James, eds., Facies Models: Response to Sea Level
Change, Geological Association of Canada, p. 15-25.
REVELLE, R., ed., 1990, Sea level change, National Research Council, Studies in Geo-
physics, Washington, D.C., National Academy Press, p. 234.
SABADINI, R., DOGUONI, C. and YUEN, D.A., 1990, Eustatic sea level fluctuations
induced by polar wander, Nature, v. 345, p. 708-710.
SHOEMAKER, E.M., 1983, Asteroid and comet bombardment of the Earth, American
Review of Earth and Planetary Sciences, v. II, p. 461-494.
SPUDIS, P. D., 1993, The geology of multi-ring impact basins: The moon and other planets,
in Axford, W.I., Hunt, G.E., and Greeley, R., Cambridge Planetary Science, Cambridge
University Press, p. 263.
VAIL, P. R., AUDEMARD, F., BOWMAN, S. A., EISNER, P. N. and PEREZ-CRUZ,
c., 1991, The Stratigraphic signatures of Tectonics, Eustasy and Sedimentology - an
overview, in G. Einsele, W. Rieken and W. Seilacher, eds., Cycles and Events in Stratig-
raphy, Springer Verlag, Berlin, p. 617-659.
VAIL, P. R. and TODD, R. G., 1981, Northern North Sea Jurassic unconformities, chronos-
tratigraphy and sea level changes from seismic stratigraphy, in L. V. B1ing and G. D.
Hobson, eds., Proceeding of the Petroleum Geology of the Continental Shelf of NW
Europe Conf., March 4-6,1980, London, Heydon and Son Ltd., p. 216-235.
VAIL, P. R., MITCHUM JR., R. M., TODD, R. G., WIDMIER, J. M., THOMPSON III, S.,
SANGREE, J. B., BUBB, J. N. and HATLEUD, W. G., 1977, Seismic stratigraphy and
global changes of sea level, in Payton C. E., eds., Seismic Stratigraphy - Applications to
Hydrocarbon Exploration, American Association of Petroleum Geologists Memoir 26,
p.49-212.
WIGNALL, P.B., 1994, Black Shales, Clarendon Press, Oxford, p. 127.
WORSLEY, T. w., NANCE, D. and MOODY, J. B., 1984, Global tectonics and eustasy
for the past 2 billion years, Marine Geology, v. 58, p. 373 - 400.
List of Contributors
Roy D. Adams Douglas Hellmann

Utah Geological Survey Department of Computer Sciences
2363 South Foothill Drive University of South Carolina
Salt Lake City, UT 84109-1497 Columbia, SC 29208
U.S.A. U.S.A.
Thomas J. Algeo T. Don Hickey

Laboratory of Sedimentology Department of Geological Sciences
Department of Geology University of South Carolina
University of Cincinnati Columbia, SC 29208
Cincinnati, OH 45221-0013 U.S.A.
U.S.A.
Steven M. Holland
Robert Cannon Geology Department
Department of Computer Sciences University of Georgia
University of South Carolina Athens, GA 30602-2501
Columbia, SC 29208 U.S.A.
U.S.A.
David K. Jacobs
Jesus E. Caracuel Department of Biology
Departamento de Estratigraffa y Paleontologfa University of California Los Angeles
Universidad de Granada 405 Hilgard Avenue
18002 Granada Los Angeles, CA 90095-1606
Spain U.S.A.
John S. Compton Christopher G.St.C. Kendall

Department of Marine Science Department of Geological Sciences
University of South Florida University of South Carolina
St. Petersburg, FL 33701 Columbia, SC 29208
U.S.A. U.S.A.
Robert Ehrlich Phillip A. Levine

Department of Geological Sciences Department of Geological Sciences
University of South Carolina University of South Carolina
Columbia, SC 29208 Columbia, SC 29208
U.S.A. U.S.A.
Craig S. Fulthorpe David J. Mallinson

Institute for Geophysics Department of Marine Science
University of Texas at Austin University of South Florida
Austin, TX 78759 St. Petersburg, FL 33701
U.S.A. U.S.A.
377
378 List of Contributors
Phil Moore Thomas RUffer

Department of Geological Sciences Department of Geology and Paleontology
University of South Carolina Ruprecht-Karls University
Columbia, SC 29208 D-69120 Heidelberg
U.S.A. Germany
Federico Ol6riz Dork L. Sahagian

Departamento de Estratigraffa y Paleontologfa Institute for the Study of the Earth, Oceans
Universidad de Granada and Space
18002 Granada University of New Hampshire
Spain Durham, NH 03824
U.S.A.
David A. Osleger
Department of Earth Sciences Kirill B. Seslavinsky
University of California Institute of Physics of the Earth
Riverside, CA 92521 B. Gruzinskaya 10
U.S.A. 123810 Moscow
Russia
Luis Pomar
Departamento de Ciencies de la Terra William C. Ward
Universitat de les Illes Balcars Department of Geology and Geophysics
E-07071 Palma de Mallorca University of New Orleans
Spain New Orleans, LA 70148
U.S.A.
L. Bruce Railsback
Department of Geology Gregory L. Whittle
University of Georgia Department of Geological Sciences
Athens, GA 30602-250\ University of South Carolina
U.S.A. Columbia, SC 29208
U.S.A.
Francisco J. Rodriguez-Tovar
Departamento de Estratigraffa y Paleontologfa Rainer ZUhlke
Universidad de Granada Department of Geology and Paleontology
18002 Granada Ruprecht-Karls University
Spain D-69120 Heidelberg
Germany
Index
Accommodation sequence stratigraphy 306

and carbonate production 105 Caspian 335
and ecospace 65 Chad Basin 335
and stratal patterns 66 Cincinnati Arch 16
Late Cambrian history 256 Coal deposition 137
sedimentary response to - 264 climate and geography 140
Afghan-Zabol Basin 335 controls on deposition 140
Aggrading systems tract 101 global rates 138
Alcudia Basin 89 hypotheses 145
Alps, Early and Mid Triassic 161 land plant evolution 143
Berglental Section 187 numerical modeling 148
Dolomites 165 peat swamps 141
Elferschartenspitze Section 171 sea-level control 142
facies and sequences 168 sequence stratigraphy 265
geological setting 163 tcctonic control 142
HochgIckkar Section 184 Continental epeirogeny in Paleozoic 235
Laliderer Wnde Section 187 Armorica 237
northern Calcareous Alps 180 Baltica 237
Pufels Section 173 China 238
sedimentation controls 176 Chukotka 238
Selvapiana Section 171 Gondwana 236
Anthropogenic sea level change 353 Kazakhstania 238
contributions to rise 355 Siberia 238
dam building 356 Continental flooding in Paleozoic 213
water fluxes to sea 354 Continental hypsometry 215
in Paleozoic 217
Balkash-Zungar Basin 335 modern coastal 220
Baluchistan Basin 335
Basin floor topography and production 105 Death Valley 281
Basin volumes 335 Desert Range Limestone 304
Bonanza King Limestone 305 Dresbachian Fischer plots 258
California 277 Eagle Mountain Shale 299

Campos Basin 89 East Iran Basin 335
Canterbury Basin 121 Echo Shale 300
seismic section 123 Ecostratigraphic trends 60
Cap Blanc 80 averaged faunal spectra 66
Carbonate platforms ecosedimentary evolution 74
Florida 25 ecostratigraphy: zone level 75
L1ucmajor, Spain 88 tectono-eustatic interactions 79
Upper Cambrian of US 247 trends in abiotic components 77
Carbonate production rate 345 Espahan Basin 335
Carrara Formation 277 Eustasy and epeirogeny 209
grand cycles 298, 306 Eustasy and geochemical cycles 154
lithostratigraphy 289 Eustatic elevations in Paleozoic 229
379
380 Index
Extraterrestrial impacts and sea level 367, 370 Lithofacies descriptions 29

evidence of impacts 373 dominant facies around SB 41
Exxon model, test 117 Lithospheric stress release and sea level 368
Llucmajor Platform 88
Facies control on sequences 2 3D architecture 96
depth range 10 Lower-Middle Cambrian 277
depth tolerance 9 Low stillstand systems tract 10 1
extinction 10
first and last occurrences 10 Mallorca, Spain 87
origination 10 Marratxi Platform 89
peak abundance 9 Milankovitch cycles 329
preferred depth 9 Milankovitch monsoonal response 338
relation to sea level 32 Miocene of Florida Platform 25
sensitivity analysis 7 Modern coastal hypsometry 220
Florida Platform 25 Modern continental epeirogeny 231
geological setting 26 epeirogenic mechanisms 232
sea-level curves 52 Monsoon and sea level 332
seismic reflection 37 Monsoonal water storage 347
sequences 36 in Cambro-Ordovician 349
Sr isotopes 32 in Devonian 349
Fore-reef facies 91 in Early Cretaceous 351
in Jurassic 351
Geoidal anomalies 370 in Late Triassic 347
Global comparisons 197 in Permo-Triassic 350
Anisian 199
Ladinian-Carnian 200 Neogene 113
Scythian 199 Nevada 277
Gold Ace Limestone 300 New Zealand, offshore 113
Grand Cycles 277 Northern Calcareous Alps 180
complete and incomplete cycles 321 Ladinian-Carnian 183
sequence stratigraphic model 317 Scythian-Anisian 181
Groundwater and sea level 336 sequences and facies 184
sequence stratigraphy 189
High-frequency reef model 100 Northern Tethys 192
High stillstand systems tract 101 biostratigraphy 192
Holocene precipitation 333 sequence correlations 195
Iberian Upper Jurassic 67 Offlapping systems tract 101

depositional sequences 73 Oklahoma, subsidence curve 251
comparisons in other areas 76 Open-shelf reef facies 91
Ice and sea level 330, 368
Ichnologic features 69 Pahrump Hill Shale 302
Indiana 17 Paleozoic 209
Internal Drainage 340 epeirogenic trends 210, 222, 231
eustasy 210, 229
Jangle Limestone 303 continental flooding 209,211
continental hypsometry 215
Lagoonal facies in reefs 91 global sea-level elevations 227
Lakes and climate models 357 Palma Basin 87
environmental policy 358 Papoose Lake Member 305
Laurentia in Late Cambrian 249 Parasequence depth range 10
Index 381
Phanerozoic eustatic trends 230 Taphonomic information 63

Phosphogenesis during highstands 47 Tarim Basin 335
Pluvially-induced sea-level changes 346 Tennessee, subsidence curve 251
Pyramid Shale 30 I Texas, subsidence curve 251
Thimble Limestone 299
Quaternary ice volume 330 Third-order sea-level cycles, causes 368
Quaternary monsoonal proxy 332 Twentieth century sea-level change 356
Quidam Basin 335
Upper Cambrian 247
Red Pass Limestone 301 accommodation history 256
Reef-core facies 91 depositional sequences 251
Reef platform 88 Laurentian paleogeography 249
basic building blocks 91 meter-scale cycles 256
depositional sequences 96 paleobathymetric trends 253
larger accretional units 94 platforms in US 247
lithofacies 90 sea-level curves 262
systems tract boundaries 102 sedimentary response 264
Reef-slope facies 91 subsidence analysis 259
Upper Jurassic 61
Sampling bias in facies 2 western Tethys 70
Santanyi Platform 89 Upper Ordovician 16
Sediment supply and sea level 370 Urs Nor Basin 335
Sedpak simulations 116 Utah, subsidence curve 251
Sigmoid 92
Silurian-Miocene land plant evolution 144 Virginia, subsidence curve 251
Sr-isotopes 32
Subsidence in Paleozoic 251 Water storage potential of basins 342
Oklahoma 251
Tennessee 251 Zabriskie Quartzite 298
Texas 251
Virginia 251
Utah 251

Sequence Stratigraphy and Depositional Response To Eustatic, Tectonic and Climatic Forcing

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Sequence Stratigraphy and Depositional Response To Eustatic, Tectonic and Climatic Forcing

Загружено:

Авторское право:

Доступные форматы

SEQUENCE STRATIGRAPHY AND DEPOSITIONAL RESPONSE TO EUSTATIC,

TECTONIC AND CLIMATIC FORCING

Springer-Science+Business Media, B.V.

Sequence <tratigraphy and depositional response to eustatic, tectonic,

ISBN 978-90-481-4633-8 ISBN 978-94-015-8583-5 (eBook)

Printed on acid-free paper

All Rights Reserved

5.2. Ladinian to Early Carnian Depositional Environments . . . . . . . . . 183

6.2.Second Sequence/Grand Cycle: Echo Shale-Gold Ace Limestone . . . 308

Sequence stratigraphy has advanced considerably since the early applications of

The case studies appearing in this volume represent a variety of depositional

Depositional Sequences, Facies Control

ABSTRACT: Three increasingly realistic models of the stratigraphic distribution of

Sequence stratigraphy is forcing a reinterpretation and re-evaluation of many

2. The Problem of Facies Control

2.1. THE PERFECT STRATIGRAPHIC RECORD

The simplest approach to predicting the stratigraphic distribution of fossils would

3. The Effect of Sampling Bias

B - Sampling Effects Model Water Depth

3.1. THE EFFECTS OF FACIES AND DEPOSITIONAL SEQUENCES

A-50 % sampling B-10 % sampling

A - Same facies control Water Depth B - Different facies control

4. A Sensitivity Analysis of Facies Control

0.25 0.50 0.75 1.00

shallow Preferred Depth ----+ deep

0.58 0.36 0.20 0.10 0.06

eurytopic ...- - - Depth Tolerance----+' stenotopic

1.50 1.00 0.75 0.50 0.25

A single shallowing-upward parasequence was simulated (Fig. 5). Because the

4.1. PREFERRED DEPTH

4.2. DEPTH TOLERANCE

4.3. PEAK ABUNDANCE

4.4. EXTINCTION AND ORIGINATION

Origination or extinction within a parasequence will have the effect of truncating

4.5. DEPTH RANGE OF PARASEQUENCE

5. Sequence Control on First and Last Occurrences

5.1. A No-LOWSTAND SEQUENCE

First Occurrences Water Depth (m) Last Occurrences

5.2. A COMPLETE SEQUENCE

5.3.1. Sequence Boundary and Lowstand Systems Tract

5.3.2. Transgressive Systems Tract

5.3.3. Highstand Systems Tract

6. Preliminary Tests of the Model

6.1. THE UPPER ORDOVICIAN OF THE CINCINNATI ARCH

The Upper Ordovician of the Cincinnati Arch consists of a thick assemblage of

Upper Ordovician - Southeast Indiana

First Occurrences Last Occurrences

Some combined sequence boundaries/transgressive surfaces involve relatively

the disappearance of numerous shallow water taxa (e.g, brachiopods Platystrophia,

6.2. OTHER EVIDENCE

In his study of the evolution of the bryozoan Metrarabdotos, Cheetham (1986)

5) In the absence of high-amplitude parasequences, no clustering of first or

P(collection) = A e-(d-PD)2/ 2T2,

paleontology of the "type Cincinnatian": Ohio Di vision of Geological Survey Guidebook

David J. Mallinson and John S. Compton

ABSTRACT: Miocene depositional sequences are identified based on sequence strati-

carbonate depositional environment to a predominantly siliciclastic depositional

The Florida Platform is an extensive (2 x 105 km 2 ), tectonically stable carbonate

moldic dolostone intraclasts; bored and intermediate shelf ::!1

_/0 ~ phosphorite/dolomite gravel (LF1)

126.71 dolomite age (see Table 2) GIl diatoms @ echlnolds

truncated burrows, caliche-like horizons, and hiatuses recognized by Sr-isotope

4. Results and Interpretations