Вы находитесь на странице: 1из 28

Tectonophysics 406 (2005) 39 66

www.elsevier.com/locate/tecto

Interplay between tectonic, fluvial and erosional


processes along the Western Border Fault of the
northern Upper Rhine Graben, Germany
Gwendolyn Peters a,b,*, Thies J. Buchmann a,b, Peter Connolly c, Ronald T. van Balen b,
Friedemann Wenzel a, Sierd A.P.L. Cloetingh b
a
Geophysical Institute, University of Karlsruhe, Hertzstrasse 16, D-76187 Karlsruhe, Germany
b
Faculty of Earth and Life Sciences, vrije Universiteit, De Boelelaan 1085, NL-1081 HV Amsterdam, The Netherlands
c
Rock Mechanics Team, Chevron Texaco ETC-Drilling Solutions Unit, 1500 Louisiana St., Houston, Texas, 77002-7308, USA
Received 23 October 2004; received in revised form 3 May 2005; accepted 24 May 2005
Available online 21 July 2005

Abstract

The northern Upper Rhine Graben, situated in the central part of the European Cenozoic rift system, is currently
characterized by low intra-plate seismicity. Historical earthquakes have not been large enough to produce surface rupturing.
Moreover, the records of Quaternary surface processes and human modifications are presumably better preserved than the
record of the relatively slow tectonic deformation.
In order to gain information on the neotectonic activity and paleoseismicity in this setting, the geological and geomor-
phological records of fault movements along a segment of the Western Border Fault (WBF) were studied using an integration
of techniques in paleoseismology, structural analysis and shallow geophysics. The WBF segment investigated follows a 20 km
long linear scarp of unclear origin. A series of geophysical measurements were performed and the results suggested that near-
surface deformation structures are present at the segments southern end. Several trenches opened at this location revealed fault
structures with consistent extensional style and a maximum vertical displacement of 0.5 m. In one trench, the deformation
structures were dated between 19 and 8 ka. Assuming the deformation has been caused by an earthquake, a M w ~ 6.5
earthquake would be implied. Aseismic deformation would point to a fault creep rate z 0.04 mm/yr.
A reconstruction of the sequence of events at the trench site, from Middle Pleistocene to Present, demonstrates that the
morphology at the base of the scarp is the result of interplay between tectonic activity and fluvial and erosional processes. At the
regional scale, a mixed origin for the WBF scarp is proposed, combining the effects of fluvial dynamics, erosion, regional uplift
and localized tectonic activity on the WBF.
D 2005 Elsevier B.V. All rights reserved.

Keywords: Active tectonics; Surface processes; Paleoseismology; Shallow geophysics; Upper Rhine Graben

* Corresponding author. Tel.: +49 721 608 4596; fax: +49 721 71173.
E-mail address: Gwendolyn.Peters@gpi.uni-karlsruhe.de (G. Peters).

0040-1951/$ - see front matter D 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.tecto.2005.05.028
40 G. Peters et al. / Tectonophysics 406 (2005) 3966

1. Introduction show contemporary movements of 0.41 mm/yr


(Schwarz, 1974; Demoulin et al., 1995). Several
This paper presents results of a study on the neo- authors use these high present-day movements as
tectonic activity of a segment of the Western Border evidence for recent tectonic activity (Illies, 1974;
Fault (WBF) in the northern part of the Upper Rhine Schwarz, 1974; Demoulin et al., 1995; Franke, 2001).
Graben (URG; Fig. 1B, D). This area is characterized The most obvious topographic feature of the gra-
by low seismicity, although damaging historical earth- ben is the significant height difference between the
quakes have occurred (e.g. Leydecker, 2003; Fig. 1C). graben shoulders and the River Rhine plain (up to
In order to provide greater constraint on the neotec- ~1,000 m), representing a clear morphological signa-
tonic activity and paleoseismicity, the geological and ture of the border faults. This suggests that tectonic
geomorphological records of fault movements were movements have occurred on the latter during the
studied by integrating techniques in paleoseismology, Quaternary. The River Rhine Fault south of Freiburg,
structural analysis and shallow geophysics. the AchenheimHangenbieten Fault near Strasbourg
The study area is part of the 300 km long, 3040 and the Omega Fault system near Edenkoben/Landau
km wide URG, i.e. the central part of the European are documented faults in the URG that exhibit topo-
Cenozoic rift system (Fig. 1A). Rifting of the URG graphic scarps and that may be the expression of
started in Early to Middle Eocene, approximately recent surface deformation (Fig. 1B; Illies and Grei-
contemporaneous with an important phase of the Al- ner, 1976; Cushing et al., 2000; Brustle, 2002). Since
pine orogeny (e.g. Illies, 1977; Sissingh, 1998). a Pleistocene fluvial terrace of the River Rhine coin-
Changes in the stress regime during the evolution of cides with the location of the scarp Brustle (2002)
the graben have resulted in a complex evolution with argues that the River Rhine Fault scarp has both a
different subsidence and uplift phases (e.g. Illies, tectonic and an erosional origin.
1975; Villemin et al., 1986; Ziegler, 1992; Sissingh, The recent tectonics of the URG have been studied
1998; Schumacher, 2002; De`zes et al., 2004). by several authors between the 1960s and early 1980s
The seismicity of the URG is characterized by low (Semmel, 1968; Illies and Greiner, 1976; Monninger,
to moderate intra-plate seismicity (e.g. Ahorner, 1983; 1985). More recently European network projects such
Bonjer, 1997; Fig. 1C). A wide distribution of small as PALEOSIS (PALEOSIS, 2000), EUCOR-UR-
earthquakes occurs over the whole graben, with the GENT (EUCOR-URGENT, 2005) and ENTEC (Cloe-
southern part of the URG exhibiting higher activity tingh et al., 2003; Cloetingh and Cornu, 2005) have
than the northern part (e.g. Leydecker, 2003). No focused on the interaction between recent surface
prominent active faults have been identified from processes and crustal deformation. However, the Qua-
the seismicity distribution alone and no recent earth- ternary tectonic activity of the graben is only under-
quake in the URG was large enough to produce a stood for local areas in its southern part.
surface rupture. Important seismic events in the north- In this study, deformation observed in trenches
ern URG were a magnitude 5.1 earthquake (pers. across a segment of the WBF in the northern URG
comm. Leydecker, 2004) located close to the WBF is used to investigate evidence of surface deformation
(Worms 24.02.1952), three intensity VII earthquakes in young sediments. The WBF segment chosen is part
at Mainz (858 AD, 14.03.1733 and 18.05.1733; Kai- of a 20 km long, 50100 m high linear scarp (Fig. 2).
ser, 1999) and an earthquake swarm in the 19th The scarp has been highly modified by human activ-
century located in the central part of the northern ity, especially viniculture, dating back at least to
graben, which produced a maximum intensity of VII Roman times. At several accessible sites along the
(near Gross-Gerau, 18691871 swarm; Fig. 1C). scarp pre-trenching surveys were undertaken. Based
The large thickness variations of Pliocene and on these surveys a trench site at the southern end of
Quaternary sediments in the graben imply syndeposi- the scarp was chosen (Figs. 2, 3). This paper presents
tional tectonic movements and suggest an average the results of 3 trenches and high-resolution geophysi-
subsidence rate for the Quaternary of 0.10.2 mm/yr cal measurements (ground-penetrating radar, geoelec-
(e.g. Pflug, 1982; Monninger, 1985; Fig. 1E). Precise trical tomography, reflection seismics) and discusses
levellings across major faults in the northern URG their implications for recent URG tectonics.
G. Peters et al. / Tectonophysics 406 (2005) 3966 41

nish
North
1692 - 2001
A Sea Rhessif
Mainz
Gro-Gerau
a 50
50 Amsterdam M ML
Mainz

B.
1-2

d
HG 1733

inz

w al
LRG

rn URG
NPH 2-3
Worms

Ma

n
3-4

Ode
1952
G
UR 4-5

northe
LG Basel
BG
5 - 6.4

Wa lzer
Alpes Heidel-

ld
RG berg

Pf

OF
au
hg
Marseille Landau r aic
Mediterranean K
RG 49
49 Karlsruhe
Rastatt
U

1933
le
dd
mi

Straburg Alb
AHF
Albstadt
ian

1978
s
ain

ab
UR
unt

Sw
rn
Mo

t
the

res

Freiburg
48 48
es

sou

Fo
g

basement
F
Vos

ck
RR

Bla

border and
main faults
Basel 0 50 km
B C
7 8 9 7 8 9

0 20 km 0 20 km
Frankfurt a.M. Frankfurt a.M.
if
ass Hofheim n
ni sh M a.T.
M
ai
Rhe n
inge Mainz ai
n
Rh

Rh

50 i nz B en M 50 Mainz
i ne

a
i ne

M b
Gra Rsselsheim
WBF

Bingen
sin
z Ba Darm-
Main Upper stadt Darm-
stadt
Oppenheim Oppenheim
Rhine
Pa
le
o

Graben
Rh

Mettenheim
ge lze -

Odenwald
H pf ord

in

824m
lla r
nd
e
N

Osthofen Osthofen
EBF
F
OG

Worms 0m
F

20 m
WB

Rhi n

Rhi n

40 m
WF

Grnstadt 60 m
e

122m 80 m
1035m
Nec k 100 m
ar
120 m Ne
Pflzer
Kraichgau

49.5 Bad Drk- Mann- 49.5 150 m ck


ar
Wald heim heim 175 m
Heidel- 200 m Heidel-
berg 250 m berg
Landau 300 m
D E > 350 m

8 8.5 8 8.5

Fig. 1. A: The European Cenozoic Rift system with the centrally located Upper Rhine Graben (URG). LRG = Lower Rhine Graben,
HG = Hessian Grabens, BG = Bresse Graben, LG = Limagne Graben, RG = Rhone Graben. B: Structural map of the URG. NPH = Nordpfalzer
Hugelland, AHF = AchenheimHangenbieten Fault, OF = Omega Fault, RRF = Rhine River Fault. C: Distribution of earthquakes with magnitude
M L N 1 recorded in the area between 1692 and 2001. The Swabian Alb is characterized by the highest recent seismicity, with the Albstadt 1978
earthquake being the largest (M L 5.6). The Worms 1952 (ML 5.1) and Rastatt 1933 (ML 5.4) earthquakes were substantial twentieth century
earthquakes located inside the URG. D: The location of the Western Border Fault (WBF), the Worms Fault (WF) and the OppenheimGrunstadt
Fault (OGF) in the northern URG. Offsets along the WBF and WF are measured for the Lower Miocene Hydrobia Beds (Stahmer, 1979). The
course of the River Rhine between Upper Miocene and Early Pliocene is indicated with a dashed line. E: Isopach map of Quaternary deposits
(modified after Bartz, 1974).
42 G. Peters et al. / Tectonophysics 406 (2005) 3966

2. Geology of the Mainz Basin and the northern


Upper Rhine Graben

The study area is situated in the northern URG,


along the contact of the western graben border with
the Mainz Basin. A review of the Mainz Basin and
northern URG, with emphasis given to the WBF and
the investigated part of the 20 km long WBF segment,
is presented in the following section.

2.1. The Mainz Basin

The Mainz Basin is bounded by three structural


blocks (Fig. 1D): (1) In the North, the Mainz Basin is
separated from the Rhenish Massif by the ENE-trend-
ing HunsruckTaunus Border fault, (2) to the south-
west of the Mainz Basin, separated by a NWSE
trending fault contact, is the Nordpfalzer Hugelland
and (3) in the east, bounded by the NNESSW trend-
ing WBF, the Mainz Basin borders the URG.
From Late Eocene to the Early Oligocene the
Mainz Basin and the URG developed as a single
sedimentary system. Since Late Oligocene the
Mainz Basin has subsided less than the URG and
tilted southeastward as it took part in the episodic
uplift of the Rhenish Massif (Meyer et al., 1983).
Remnants of the Upper MioceneLower Pliocene
Dinotherium sands in the Mainz Basin are assigned
to a paleo River Rhine. Its course was across the
Mainz Basin from Worms to Bingen (Fig. 1D;
Bartz, 1936). Between Early Pliocene and Early Pleis-
tocene the relative uplift of the Mainz Basin forced the
River Rhine to migrate northeastward to its present
course along the Mainz Bingen Graben (Wagner,
1930; Kandler, 1970; Abele, 1977).
During the Quaternary solifluction, fluvial erosion
and loess deposition influenced the landscape evolu-
tion in the Mainz Basin. At present, the Mainz Basin
has a tableland landscape formed by resistant Miocene
limestones overlying Oligocene sands and marls. The
landscape is covered by Pliocene to Pleistocene flu-
Fig. 2. Shaded relief map of the 20 km long scarp investigated. The vial sands and Upper Pleistocene loess and slope wash
surface trace of the WBF (dashed line) is mapped by morphology deposits.
(Steuer, 1911; Franke, 2001). Two ~EW striking faults subsidiary
to the WBF (solid lines; after Steuer, 1911) south of Oppenheim
2.2. The northern Upper Rhine Graben
bound a landslide. Lateral resolution of the DEM is 20 m and
vertical resolution is 0.5 m. Coordinates are in GaussKruger
coordinate system. (Data provided by Landesamt fur Vermessung The Cenozoic graben fill in the northern URG
und Geobasisinformation, 2002). consists of a sequence of fluvial and lacustrine sedi-
G. Peters et al. / Tectonophysics 406 (2005) 3966 43

Alsheim
180

160
160
5514000 140

0
10
160

180
0
19

90

120

WB F
160
140
5512000

180

Mettenheim 90

Wo r m
180
0
14

160

s
Fau l t
0
10

Bechtheim 90

trench site
120

5510000
160
profile Fig. 11 (Fig. 4A)
0
14

F
WB

140
120
0
5508000 12

100 Osthofen 90
90

140 120 0 1 km
12

3448000 3450000 3452000 3454000

Local deposits Holocene Rhine meander faults


(Atlantic), oxbow
Holocene Rhine meander Western Border
Loess (mainly Wuermian) Fault (WBF)
(Atlantic)
Pliocene - Pleistocene Younger Dryas aolian sands contour interval
sands 10 m
Miocene limestones Pleistocene terrace, modified after: Rosenberger
lower level et al., 1996, Franke 2001

Fig. 3. Geological and tectonic map of the southern end of the scarp investigated between Alsheim and Osthofen. At the trench site shallow
geophysical measurements and corings were undertaken. The pre-survey results indicated this site as the most favorable for trenching.
Additionally, the location of a west to east topographic profile given as Fig. 11 is shown.

ments, interrupted by marine sediments. The total berger Loch) to west (20 m along the WBF segment
thickness varies in this part of the graben from 400 investigated; Straub, 1962; Doebl, 1967; Bartz, 1974),
m at Darmstadt to 3200 m at Mannheim (Doebl and reflecting the asymmetry of the graben structure (Fig.
Olbrecht, 1974). The fluviolacustrine Quaternary 1E). This thickness variation is the result of continu-
deposits of the northern URG show large thickness ous subsidence on the eastern side, localized in the
variation from east (~380 m in the so-called Heidel- Heidelberger Loch. In contrast, several fault branches
44 G. Peters et al. / Tectonophysics 406 (2005) 3966

of the WBF accommodate the relatively low subsi- activity along this structure. In contrast, the amount of
dence on the western side of the graben, leading to a Quaternary uplift of the adjacent Mainz Basin (170 m;
more widespread distribution of deformation. Bruning, 1977) and the formation of fluvial terraces
Several fluvial terraces of the River Rhine system on the western margin imply a significant vertical
developed in the northern URG during the Quaterna- displacement along the WBF.
ry. The distribution of these terraces is different in
distinct areas. Only small remnants of Early Pleisto- 2.4. The WBF scarp
cene terraces are preserved on the eastern side. In
contrast to this, widespread Early to Late Pleistocene The 20 km long scarp along the WBF segment
terraces exist on the western side at the foot of the (Figs. 2, 3) is a striking morphological feature. The
Pfalzer Wald (Stablein, 1968; Monninger, 1985), and scarp has a height of 100 m in the north and 50 m in
a complete sequence of Pleistocene terraces exists on the south relative to the recent River Rhine flood
both sides of the Mainz Bingen Graben (Kandler, plain. The time of formation and origin of this scarp
1970; Abele, 1977). The Lower Terrace, which was is unclear. It could be fluvial, tectonic, or a combina-
formed during the last glacial period (Wurm), is well tion of both. Since Early Pleistocene the River Rhine
defined and extends from the eastern to western bor- migrated from the uplifted Mainz Basin to the Mainz
der of the northern URG. During the Holocene, the Bingen Graben (Wagner, 1930; Kandler, 1970; Abele,
River Rhine eroded into the Lower Terrace; periods of 1977; Fig. 1D). It is likely that scarp formation was
more or less pronounced meandering left three distinct initiated at this time but no evidence exists to prove
meander generations (Fetzer et al., 1995; Dambeck this hypothesis.
and Thiemeyer, 2002). Based on enhanced seismic activity and precise
levelling data, Illies (1974) suggests that the WBF
2.3. The Western Border Fault (WBF) between Hofheim and Oppenheim, north of the seg-
ment investigated, is an active fault zone (Fig. 1C, D).
The WBF strikes in a NNE-direction from Hof- In contrast, precise levelling data from Schwarz
heim am Taunus at the northern end of the graben (1976) show enhanced relative vertical movements
southward across Russelsheim, Oppenheim, Osthofen occurring along the WBF segment investigated.
towards Bad Durkheim (Fig. 1D). Along this section, These measurements indicate increased vertical move-
it separates the Mainz Basin from the URG. Several ment rates from 0.24 mm/yr in the northern part to
sub-parallel faults reflect the complex deformation 1.26 mm/yr in the southern part of the scarp.
history of the western margin of the URG. The verti- In the northern part (Oppenheim to Alsheim) the
cal offset along the WBF is highly variable (Fig. 1D). scarp exhibits several indentations caused by landsli-
In the central part of the scarp segment investigated ding and fluvial erosion of the River Rhine (Fig. 2).
(Mettenheim), Lower Miocene Hydrobia Beds are At Oppenheim, landslides have been frequent in pre-
offset by 824 m (Franke, 2001). Southwards, the historical and historical times (e.g. Steuer, 1911;
vertical offset of Hydrobia Beds on the WBF gradu- Wagner, 1941; Krauter and Steingotter, 1983). These
ally decreases to 122 m with the extension taken up on landslides are related to the lithological conditions at
the Worms Fault, which accommodates offsets of this part of the scarp, where the Lower Oligocene
Hydrobia Beds to a maximum of 1035 m (Stahmer, sands tend to flow once they become water saturated.
1979). The vertical movements on both faults oc- Here, the largest landslide is located between two ~E
curred mainly during the Miocene, contemporaneous W trending faults (Steuer, 1911) indicating that fault
with the major depositional phase (2000 m of the structure has some influence on landform evolution.
maximum 3200 m sediment fill was deposited during During the Middle Wurm, the River Rhine eroded into
the Miocene; Doebl and Olbrecht, 1974). In the Qua- the scarp near Guntersblum and triggered a river bank
ternary, tectonic activity remained similar to its Late collapse (Semmel, 1986; Fig. 2).
Tertiary level along the eastern border fault. The The southern end of the scarp (Alsheim to Ostho-
relatively thin accumulation of Quaternary sediments fen) is the most linear (Figs. 2, 3). Here, landslides are
(20100 m; Fig. 1E) along the WBF suggests minor less likely since the scarp is lower and less steep.
G. Peters et al. / Tectonophysics 406 (2005) 3966 45

Furthermore, the scarp is mainly formed by resistant A


Miocene limestones covered with PliocenePleisto- Me8 GPR
cene sands, alluvial loess and slope wash (Fig. 3). 1
Land consolidation has been less intense than in the
5510000

120
north and no recent mass movements have been

39
Me3
reported. In addition, this part of the scarp has the

L4
most natural morphology. For these reasons it was

F
expected to be the best location along the WBF with

WB
near-surface deformation preserved.

0
10
Me1 4
5509800 GPR
3. Pre-trenching surveys
Me10
GPR5

120
For trenching, the trace of the fault at surface needs seismic line 1
Me2 GPR6
to be known to within a few meters. In the case of the
GPR7
WBF segment, its location is only known at depth
from industry seismics and cores. The faults surface
5509600
B

90
trace has been mapped assuming that it coincides with
the base of the scarp (Steuer, 1911; Franke, 2001; Me7

F
WB
Figs. 2, 3). High-resolution geophysical imaging tech-
0
11
GPR8
niques (reflection seismics, geoelectrical tomography 3450700 3450900
and ground-penetrating radar) were used to map the
0 200 m
fault at surface accurately. In addition, shallow coring
B
was performed at several locations.
The site chosen for intense geophysical measure-
ments and trenching was situated north of Osthofen
,5

and covered an area from the base of the scarp to the


2
11

first vineyard terrace 10 m above (anthropogenic V1, 5509700


TR1
Figs. 3, 4). The base of the scarp corresponds to the section
TR2 (Fig. 10 A
)
Lower Terrace from the Wurm glacial period (Franke, 52

2001; Fig. 4A). Presence of older fluvial terraces on


110

the scarp at this location has not been previously


95

vinyard terrace
93

92
investigated. It is therefore possible that V1 also V1
105

base of the scarp


corresponds to a fluvial terrace. Geophysical measure-
94

ments and drilling were undertaken on fallow vine-


yards and paths. They provide many indications of TR3
near-surface structures, which are presented in the
section
L 439

following section and on Fig. 5. 5509600 56 B (Fig.


10)

3450700 3450800
3.1. Description of the geophysical profiles
0 100 m

3.1.1. High resolution reflection seismics Fig. 4. A: Geological map of the trench site chosen, with location of
A drop weight of 50 kg and a geophone spacing of all geophysical profiles performed and inferred surface trace of the
12 m were used for a profile with a penetration depth WBF. Grey solid lines: geoelectrical profiles Me1, 2, 3, 7, 8, 10.
of about 200 m (seismic line 1 on Figs. 4A, 5A). Stippled dark grey lines: GPR profiles GPR 48. Black dashed line:
seismic line 1. Solid and dashed black line: WBF surface trace,
Unfortunately, measurements on the steep slope of the known and inferred (legend as Fig. 3). B: Detailed topographic map
scarp were not possible. The processed seismic sec- of the trench site showing the location of cores (open circles),
tion is shown in Fig. 5A. In the eastern part of the trenches and restored sections given as Fig. 10.
46
G. Peters et al. / Tectonophysics 406 (2005) 3966
Fig. 5. A: Processed results and interpretation of line 1 seismic reflection profile. B: 2D-geoelectrical tomography profiles showing resistivity contrasts along a subvertical to east
dipping surface. C: Two representative GPR profiles showing disturbed and steeply dipping reflectors in a 30 m wide zone extending eastwards from the base of the scarp. Dashed
line: maximum penetration depth. The box in GPR 7 indicates the location of TR1.
G. Peters et al. / Tectonophysics 406 (2005) 3966 47

section, reflectors are continuous and horizontal (Fig. 5C). The eastern parts of the profiles show
(meter 240400). In the middle and west of the sec- undisturbed, subhorizontal reflectors.
tion (meter 100240), reflectors are interrupted at four
zones, which are interpreted to be high angle exten- 3.1.4. Percussion cores
sional faults (F1F4). Fault F3 extends almost to the In total 34 percussion cores (63 mm diameter) were
surface, whereas the other three faults probably ter- taken in order to obtain a three dimensional picture of
minate ~4050 m (0.0250.06 s TWT) below surface. the various lithological units at the trench site. These
The uppermost reflectors in the western part form a cores were located both on V1 and at the base of the
wedge shaped structure (meter 100200) indicating scarp where they reach a maximum depth of 8 m (Fig.
westward thickening of the strata towards the graben 4B). Cores obtained at the base of the scarp in general
border fault. This structure is interpreted as the growth reveal a uniform stratigraphy. Fine sand and silt
sequence associated with a half graben, of which the deposits with a thickness of 45 m are overlying a
bounding, east dipping, synthetic fault (F5; Fig. 5A) is sequence of cohesionless sands and clay-rich sands
thought to be located to the west of the profile. This that continue to the maximum coring depth. No clear
interpretation is supported by the increase in fault evidence for faults could be derived from the cores at
structures and decrease in spacing between the faults the base of the scarp because of the heterogeneity of
in the western part of the section. the lower sequence with the various layers thinning
out and/or forming lenses. However, it is interesting to
3.1.2. Geoelectrical tomography profiles note that the largest variations in thickness of the sand
Several 2D-geoelectrical tomography profiles layers occur near the base of the scarp, i.e. where the
using a Wenner array have been carried out. The geoelectrical and GPR measurements indicate resis-
profiles have a resolution of 2  2 m and reach depths tivity contrasts and deformed structures, respectively.
of 2030 m. The resistivity of the ground is dependant In contrast to the sequence found at the base of
to several geological parameters including porosity, scarp, cores taken on V1 show a different stratigraphy.
mineral content and fluid saturation (e.g. Friedel, On V1 a thin unit of cohesionless sand overlies 11 m
1997; Loke, 2001). thick clay-rich sand, which forms this part of the
Five geoelectrical profiles (Me1, 2, 3, 8, 10) cross scarp. This unit overlies the sequence of cohesionless
the scarp and one profile (Me7) is located on V1 (Fig. sands and clay-rich sands found in cores at the base of
4A). Low resistivities (b145 Vm) suggest that only the scarp.
soft sediments occur in the upper 30 m (Fig. 5B). On
all profiles a distinct inclined boundary between re- 3.2. Trenching locations
latively high and low resistivities occurs (Fig. 5B). The
surface projection of this distinct change in resistivity Integration of all geophysical measurements and
is mapped in Fig. 4A. In the north this zone is parallel core data suggest a zone of sediment deformation
to the scarp whereas in the south it crosses the scarp. located along the strike of the scarp (Figs. 4A, 5).
The origin of these deformations (i.e. tectonic or
3.1.3. Ground-penetrating radar (GPR) sedimentary) cannot be unambiguously derived
The reflection pattern of high frequency electro- from the surveys. For this reason, trenched profiles
magnetic waves detected by GPR can represent lith- were made in order to obtain further data on these
ological boundaries, moisture content or fault structures. The choice of trench locations was deter-
structures in the shallow subsurface. A 200 MHz mined jointly by the results of the survey and the
antenna was used to investigate up to 3 m deep at site conditions.
0.2 m resolution. A series of parallel profiles crossing The first trench, TR1, was located at the base of the
the base of the scarp were made (Fig. 4A). Six of scarp (Fig. 4B) along the two most promising geo-
these profiles exhibit similar reflection patterns: an physical profiles (Me2 and GPR7, Fig. 5B,C) and
approximately 30 m wide zone extends eastwards where thickness variations in cores were largest.
from the base of the scarp in which reflectors are Here, a fault zone was discovered. A second 30 m
steeply eastward dipping and frequently disturbed long trench, TR2, was situated on V1 in the westward
48
Table 1

G. Peters et al. / Tectonophysics 406 (2005) 3966


Description of lithostratigraphic units identified in the trenches, their estimated and absolute ages and interpretation of their formation environment
Unit Interpretation Lithology Sedimentary Base Remarks Color (Munsell code) Period Estimated TL and 14C dating (ka),
structures age (ka) heavy mineral analysis
samples

5 B Soil Silts, sands, organic Bioturbation Carbonate-rich, average Greyish brown 10YR 5/2, Holocene, MIS 1 10Recent MET14: 5.83 F 0.76;
material slope of 88 in TR1, 148 dark greyish brown 2.5Y 4/2 MET33: 10.64 F 0.95;
in TR2 and TR3 MET36: 8.48 F 0.93

A Colluvium Sands, silts, gravels, Erosive Partly carbonate-rich, detrital Dark yellowish brown 10YR Late Holocene ~0.75Recent MET115: 24.3 F 2.5; MET118:
carbonate concretions charcoal fragments, slag, 4/4, yellowish brown 10YR (since Middle 3.98 F 0.51; MET131:
plastic, weakly developed soils, 5/4, brownish yellow 10YR 6/6 AgeRecent) 23.3 F 3.8; Metten-Tr.1-2-SW/
Middle Age and Roman C14: 44.46 + 1.17/ 1.02 BP;
artifacts, slope 12198 Metten-Tr.1-6-SW/C14:
2.64 F 0.025 BP;
Metten-Tr.1-11- SW/C14 :
0.207 F 0.021 BP

4 B Alluvial loess, Silts, fine sands, Parallel lamination Erosive ? Quartz and mica-rich, Pale yellow 2.5Y 7/4 Late Wurm 2314 MET32: 18.9 F 2.4; MET34:
slope wash deposit partly medium sands (mmcm scale) mass movement features, 14.6 F 1.5; MET35: 14.2 F 1.3;
and course sands liquefaction features, MET119: 23.4 F 2.5; MET120:
slope 11288 22.7 F 2.0; MET133: 18.2 F 1.2;
MET134: 17.9 F 1.4; MET135:
19.4 F 1.4; MET136: 19.9 F 1.8

A Slope wash deposit Medium sands, silt Lamination Erosive Iron coating on sand Yellowish brown 10YR 5/8 Middle Late 3122 MET31: 30.8 + 4.1/ 3.6;
(cm scale) grains, slope 5 248 Wurm MET81: 22.2 F 2.7

3 B Slope wash deposit Fine sands, silt Parallel lamination Sharp and Quartz and mica-rich, rounded Light yellowish brown 2.5Y 6/4 EarlyMiddle ?100~37 MET12:N 100; MET13:
(mmcm scale), erosive limestone pebbles and one Wurm?, MIS 5b-3 19.4 F 1.9; MET132:
sandy fan deposits flattened (20 cm  5 cm) silt 37.3 + 4.6/ 3.9
with fining upward clast at base, average slope 208

A Basal conglomerate Gravel, limestone Lenticular bedding Erosive Miocene limestone fragments Strong brown 7.5YR 5/8 Early Wurm?, ?100
clasts in sandy (cm scale) with sharp edges, rounded MIS 5b
matrix (upper part) quartz pebbles, sandstone
and clayey matrix pebbles, manganese and iron
(lower part), nodules, carbonate concretions,
clay layers slope 9128
2 B Paleocambisol Sands in clayey Bioturbation Gradual Manganese nodules and layer Yellowish brown 10YR 5/6 Intra and post? ?130?Recent MET111:N100; MET112:N
matrix at the base, no slope Eemian, MIS 5e 100; MET114:N100

A Braided river deposit, Fine-medium sands, Crossbedding Erosive Cohesionless, carbonate-free, Pale yellow 2.5Y 8/2 brownish Riss terrace?, ?450?130 MET113:N 100; heavy mineral
terrace body course sand, gravel, and parallel quartz pebbles, 10 cm thick yellow sand layers: 10YR 6/8 MIS 6 analysis, sample no.:
gravel layer at the lamination gravel layer with pebbles, MET-T4-01; MET-T4-02;
base (cm scale) hematite nodules, wood pieces METT4- 03; MET-T3-03;
at base, Buntsandstein MET-T3-02; MET-T3-01
provenance (Pfalzer Wald),
no slope

1 B Fluvial or lacustrine Clayey, fine-medium Sharp Pseudogley formation, Yellowish brown parts : 10YR Pliocene? ?N 3000 Heavy mineral analysis, sample
still water deposit sands manganese layers and nodules, 5/8; light grey parts: 2.5Y 7/1 no. and depth: MET-BK56-03:
mottled, Buntsandstein provenance 3.603.93 m; METBK56-05:
(Pfalzer Wald) with zirkon, 5.505.92 m; MET-BK56-10:
tourmaline, rutile, anatas, no slope 78 m; MET-BK56-11: 89 m,
depths counting from top of unit

G. Peters et al. / Tectonophysics 406 (2005) 3966


1A

A Braided river deposit Fine-medium sands, Lenticular bedding ? Carbonate concretions, upper Pale yellow sands: 2.5Y 8/2; Pliocene? ?N 3000 MET11: N100
clayey sands, clays (dmm scale) part carbonate-free, lower part light grey clays: 2.5Y 7/2
carbonate-rich, rounded quartz
grains, liquefaction features,
slope 188

49
50 G. Peters et al. / Tectonophysics 406 (2005) 3966

continuation of TR1, leaving a gap between the 4.1.2. Faulting


trenches (Fig. 4B). This trench was opened in order The fault zone consists of a large number of mainly
to gain information on the sedimentary evolution of extensional faults with offsets of up to 0.60 m (Fig.
the scarp and on the westward extension of the fault 6D,E). A minor fault set of strike-slip faults with
zone uncovered in TR1. Since geophysical profiles offsets of a few centimeters has been observed on
and the structural analysis of the fault zone indicated the west wall (Fig. 6C). On the footwall side of the
that one fault splay crossed the scarp, a third 20 m fault zone, both synthetic and antithetic faults occur.
long trench, TR3, was opened ~100 m south of TR2 The synthetic faults dip steeply (~608) to the east. The
on V1 (Fig. 4B). antithetic faults are generally shorter, especially on the
north wall. They dip steeply to the west (~658) and
abut the synthetic faults. The core of the fault zone is
4. Trench descriptions defined as the synthetic fault with the largest vertical
throw (0.6 m) and is termed the main fault (thick solid
The stratigraphy of the trenching area has been line on Figs. 6D,E). The hangingwall consists of a 1 m
determined by detailed mapping of the trench walls wide zone of mainly synthetic faults. Note that the
(1 : 10 mapping), percussion coring, absolute dating hangingwall and footwall exhibit a different fault
and lithostratigraphic correlation. In total 10 strati- style. Furthermore, the fault core is located in different
graphic units have been identified and are presented stratigraphical positions on the north and south walls.
in detail in the Appendix and Table 1. This section On the south wall, it is located mainly at the erosional
describes the individual trenches. It is followed by a contact between unit 3B with 4A (Fig. 6D) suggesting
structural analysis and interpretation of the faults un- that faulting has used the erosional contact. On the
covered in the trenches. Finally, a reconstruction of north wall the fault core is entirely situated within unit
the site is presented. 3B (Fig. 6E). The fault density is related to lithology.
A high fault density exists in the sandy and silty units
4.1. Trench 1 (1A, 3B, 4A, 4B) while a lower density occurs in the
clay-rich conglomerate (3A). On the south wall a
4.1.1. Overview horst structure is situated above the lower, inclined
TR1 is located at the base of the scarp (Fig. 4B) part of the reactivated erosional surface that indicates
along parts of the geophysical profiles Me2 and GPR7 slip on an irregular surface (Fig. 6D).
(Fig. 5B,C) and where thickness variations in cores
were largest. TR1 exhibits deformation caused by 4.1.3. Liquefaction features
faulting (Fig. 6D,E), liquefaction features (Fig. 7) Liquefaction features, indicating injection of water-
and soft sediment deformation related to mass move- saturated material under overpressure, exist at several
ments (Fig. 6A, inset IV). Most of the tectonic defor- locations in TR1. They are restricted to the sandy and
mation is located in a 6 m wide fault zone in the silty sediments of units 1A and 4B. On the base of the
western part of the trench. west wall cohesionless sand has been injected into
TR1 exhibits 6 of the stratigraphic units identified overlying clay-rich sand, both of unit 1A, vertically
(Fig. 6). Unit 1A (sand with clay lenses) occurs in the and laterally (Fig. 6C, marked dIT, Fig. 7). A 5 cm
westernmost basal part of the trench. Above unit 1A thick clay layer overlying these structures acted as a
lays unit 3A, a conglomerate, and unit 3B of inter- seal. Upwardly injected silt occurs as bubbles within
bedded sand and silts. Both unit 3A and 3B dip to the sand-rich silt (Fig. 6A, marked dIIT). On the north
east. Unit 4A (iron coated sand and silt) has a steeply, wall, a 1 m long silt dyke exists (Fig. 6E, marked
eastward dipping erosional discontinuity on the lower dIIIT). Wavy bedding, micro-flame structures and
units (1A, 3A, 3B). In the middle part of the trench, stretched sand layers in the alluvial loess are frequent
unit 4A is horizontally bedded and lies on unit 1A. on both the north and south walls. Such features have
Unit 4B (alluvial loess) conformably overlays unit been described as liquefaction features induced by
4A. Recent colluvium (unit 5A) and a thin Holocene seismic events (e.g. Obermeier, 1996a,b, in particular
horizon (unit 5B) truncate units 3B and 4B. Fig. 7.23; Tuttle, 2001).
G. Peters et al. / Tectonophysics 406 (2005) 3966 51

4.1.4. Mass movement features Late Wurm could have created conditions favorable
From the central to the eastern part of TR1 (Fig. 6A for gravitational movements on the existing slope. It is
meter 1226, inset marked dIVT), decollements with a also likely that fault activity may have triggered the
slight westward dip occur in the alluvial loess. These mass movement observed. Extensional movements on
decollements are rooted in a fine sand layer and sand the nearby faults could have possibly had an influence
has been injected along them. Bedding is crosscut by on the stability of the slope.
these structures, indicating they formed after sediment
deposition (i.e. post 14 ka; MET34; Tab. 1, Fig. 6A). 4.2. Trench 2
The offset along these structures is extensional in the
lower part and compressional (east vergent folding) in 4.2.1. Overview
the upper part. All these features strongly indicate a TR2 is situated west of TR1. It was made in order
mass movement. However, the origin of this mass to investigate the westward extension of the fault zone
movement is ambiguous. The alternating process of discovered in TR1 and to gain information on the
thawing and freezing of the top of the water-saturated sedimentary evolution of the scarp. TR2 exhibits
soil located above impermeable permafrost during the only one single extensional fault structure in the east-

W C14: 2.64+/-0.025 ka BP
0 C14: 0.205+/-0.025 ka BP
C14: 44.46+1.17/-1.02 ka BP
MET 14: 5.83+/-0.76 ka MET 33: 10.64+/-0.95 ka

-2 MET 32: 18.9+/-2.4 ka

'IV'
-4 normal fault
E
MET 13:
19.4+/-1.19 ka
-6
MET 12: >100 ka

MET 11: >100 ka 'II'


-8 D
MET 36: 8.48+/-0.93 ka
MET 31: 30.8+4.1/-3.6 ka ground water table MET35: 14.2+/-1.3 ka
-10 m MET 34: 14.6+/-1.5 ka
A south wall
0 2 4 6 8 10 12 14 16 18 20 22 24 26 m
0
W north wall E S strike-slip
N 5B - recent soil

faults 5A - colluvium
MET 81: 22.2+/-2.7 ka -2
4B - alluvial loess
-2
4A - slope wash
3B - slope wash
3A - conglomerate
-4 -4
brown silt 2B - paleosol
marker
horizon 2A - braided river deposit
see Fig. 8
-6 1B - still water deposit
'I', see Fig. 7 1A - braided river deposit
-6 m
B C west wall fault
decollement
-8 m
2 4 6 8 10 m 0 2 4m

Fig. 6. TR1 trench log (based on 1 : 10 mapping) of the south (A), north (B) and west (C) wall showing the lithostratigraphic units, faults,
decollements and absolute ages from dating. Insets D and E show the fault zone in detail and define the footwall, main fault (fault core) and
hangingwall. Legend is for Figs. 6, 8 and 10.
52 G. Peters et al. / Tectonophysics 406 (2005) 3966

core 1m

brown silt
layer
marker hangingwall
horizon

co
re
horst
footwall

0.6 m offset
on the main
fault

D - south wall core

core 1m

hangingwall
brown silt layer
co

marker horizon
re

with 0.5 m offset


on the main fault

footwall

'III'
dyke
lateral
injection
of silt
E - north wall core

Fig. 6 (continued).

ern part of the trench (Fig. 8A). In this part the within units 1B, 3B and 4B (Fig. 8A, inset marked
westward continuation of the units 3B and 4B from dIT). All fault strands terminate within unit 4B (alluvial
TR1 is exposed (Figs. 6A, 8A). These sloping units loess). Adjacent to the faults, a few folds and decolle-
overlie the 11 m thick unit 1B (clay-rich sand), which ments, showing the same deformation style as in TR1,
forms this part of the scarp. are seen (Fig. 8A, inset marked dIT).

4.2.2. Faulting 4.2.3. Sedimentary structures


A steeply eastward dipping (628) extensional struc- Units 1B and 2A were deposited on a subhori-
ture, consisting of a few fault strands, is situated zontal surface and at a later stage eroded from the
G. Peters et al. / Tectonophysics 406 (2005) 3966 53

20 cm 20 cm

glo- -
nd

con rate glo


co n rate
& s ed sa

m e
r me
laye
ilt
at

clay clay la debris

ult
-co

yer
fa
clay- cohesion
iron

le ai n
rich ss san clay layer
d
m
fault

clay-rich sand
d
main

h san clay-r
clay-rich
sand
y -ric ich sa
nd
cla

injection cohes injection


ionles
s sa nd cohesionless sand

A 20 cm B 20 cm

Fig. 7. Photos of liquefaction features at the base of TR1. Photo B was taken at a later stage of excavation, when the wall was located
approximately 20 cm further west than at the time of photo A.

eastern part of TR2. On the erosional and eastward trench log shows that TR3 exposes a 5 m wide
sloping surface of unit 1B units 3B and 4B were fault zone in the central part of the trench (Fig.
deposited (Fig. 8A). During historical times the 8B,C).
slope was modified, most likely for a vineyard ter-
race, and was covered by an almost 2 m thick 4.3.2. Faulting
colluvium (unit 5A). The fault zone deforms the units 1B, 2A and 2B
(Fig. 8B,C). Similar to the fault zone in TR1, this
4.3. Trench 3 zone consists of numerous steeply dipping, exten-
sional synthetic and antithetic fault strands. In the
4.3.1. Overview west, the footwall exhibits multiple large synthetic
TR3 is also situated on V1 about 100 m south of faults (Fig. 8C). Fewer shorter antithetic fault strands
TR2. The stratigraphic position of TR2 and TR3 is abut the synthetic faults. The fault core is situated
therefore very similar (Fig. 8). The results of the along a synthetic fault structure consisting of multi-
structural analysis of the fault zone discovered in ple, closely spaced fault strands (thick solid lines on
the first trench indicated that one part of the fault Fig. 8C). The strands merge into a single fault
zone crossed the scarp and extended on V1 in a towards the top of the fault zone where a remnant
NNESSW orientation (Fig. 4A). The geoelectrical of a paleosol, unit 2B, is faulted against the sands of
profile Me7 showed a distinct resistivity contrast at unit 2A. The hangingwall deformation consists main-
the position of the extrapolated fault zone (Figs. 4A, ly of antithetic faults and a few synthetic faults.
5B). Based on this information, TR3 was opened in Here, the antithetic faults are less steep than in the
order to verify the location of a fault zone. The footwall. Similarly to TR1, fault density is clearly
54 G. Peters et al. / Tectonophysics 406 (2005) 3966

W E
0 A - south wall of Trench 2
-2
Late Middle Age artifact

-4 MET136: 19.9 +/-1.8ka

MET131: 23.3 +/-3.8ka


-6
MET132: 37.3 +4.6/-3.9ka
1m
MET134: 17.9 +/-1.4ka
-8 m 'I' MET133: 18.2 +/-1.2ka 'I'
MET135:
19.4 +/-1.4ka
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 m

WNW MET112: >100ka ESE


B - north wall of Trench 3
0
MET115: 24.3 +/-2.5 ka

-2 Roman artifact Late Middle Age


MET111: >100ka artifact

MET113: >100ka
-4

MET114: >100ka MET 118: 3.98 +/-0.51 ka


-6 m
MET 119: 23.4 +/-2.5 ka MET 120: 22.7 +/-2.0 ka
0 2 4 6 8 10 12 14 16 m
footwall
core
0.5 m

hangingwall

marker
horizon
0.4 m

0.7 m

C ~ 13% lateral extension

Fig. 8. Trench logs of TR2 (A) and TR3 (B). The inset of the fault zone in TR3 (C) highlights the 0.7 m vertical displacements and the net 0.4 m
displacement of the top of unit 1B. Unit 2B is offset a minimum of 0.5 m. Due to lateral variations, the marker horizon (red) cannot be correlated
across the main fault and itsT fault strands (fault core, thick black lines).

related to the lithology. A high fault density exists in slip faults that offset the top surface by a few
the sandy unit, 2A, whereas a lower density occurs centimeters was exhumed during excavation.
in the clay-rich unit 1B. In order to investigate the
faulted surface in 3 dimensions the base of the trench 4.3.3. Sedimentary structures
was carefully excavated to expose the top surface of The sedimentary pattern in TR3 is very similar to
unit 1B. In addition to the extensional faults known TR2. Units 1B and 2A are subhorizontally bedded
from the trench walls, a minor set of sinistral strike- while unit 4B deposited on a sloping surface. The
G. Peters et al. / Tectonophysics 406 (2005) 3966 55

thickness (maximum 2.5 m) of unit 5A, a colluvium conservative 19 ka estimate is chosen. The minimum
of historical age, suggests active man-made modifica- possible age is obtained from dates of the soil hori-
tions at this site. zon (unit 5A) truncating the faults in TR1 (Fig. 6).
TL dating of the soil yields a median age (3 samples)
of 8 ka (Tab. 1).
5. Structural analysis and timing of the faulting
5.3. Structural data
5.1. Extensional displacements
Structural analyses of the fault structures in TR1,
The majority of extensional, vertical displacements TR2 and TR3 have been performed in order to
on single fault strands in TR1 and TR3 are in the order compare the style of deformation amongst the
of a few centimeters. A few faults (main faults or fault trenches and to relate the fault system at the trench
core) show vertical offsets of ~0.5 m. In TR1, these site to active regional fault trends. If the fault struc-
large offsets are on the south wall within unit 1A and tures in the trenches are part of the same fault
on the north wall within unit 3B (Fig. 6D,E). Due to system at a larger scale, a similar style of deforma-
poor correlation of units across the erosional discon- tion with similar orientation of fault sets is to be
tinuity, which reactivated as a fault surface, these expected.
offsets are the maximum offsets that can reliably be The data of TR1 show a dominant set of conju-
determined. In TR2 the maximum vertical offset is gate, extensional faults with a NNESSW strike
only 0.17 m (Fig. 8A, marked dIT). In TR3 the max- direction (maxima at 0158; Fig. 9A). Synthetic faults
imum vertical offset is the accumulated offset of are more frequent than antithetic faults. Subsets are a
several synthetic faults of 0.7 m (Fig. 8C). Due to set of NWSE (1208) striking sinistral strike-slip
large offsets on antithetic faults a graben structure is faults (great circle 3) and an ENEWSW (0758)
developed. The net vertical offset of this graben struc- striking, oblique to dextral strike-slip set (great circle
ture, determined by line length restoration of units 1B 4). Measurements of the fault strands in TR2 yield a
and 2A, is 0.4 m. In comparison, the fault zones in 0058 striking set of synthetic faults and a few anti-
TR1 and TR3 show both a maximum vertical offset in thetic faults (from the north wall; Fig. 9B). The
the order of 0.5 m. None of the offsets in the trenches structural data from TR3 reveal a set of conjugate
give evidence for synsedimentary faulting, which NNESSW (0158) striking extensional faults (Fig.
implies that the deformation occurred entirely post 9C, great circle to 1 and 2). In the cohesive clay-
depositional. rich sand (unit 1B) these faults have only minor
variations in orientation. Conversely, in the cohesion-
5.2. Timing of the faulting event less sand (unit 2A) the faults dfan outT and have
strike orientations between 0158 and 0658 (Fig. 9C,
The timing of the faulting event can only be dIIT). Subsets are 0658-striking extensional faults
constrained by determining a maximum and a mini- (great circle 4) and 1158-striking, sinistral strike-
mum age of the sediments affected by faulting. The slip faults (great circle 3). The latter set was identi-
youngest datable sediments affected by faults are unit fied on the surface of unit 1B.
4B in TR1 and TR2 (Figs. 6, 8). Thermolumines- The structural analysis shows identical fault trends
cence (TL) dating of samples located close to the in all trenches. When plotted together (Fig. 9D) the
fault structures yields an age of 19 ka (Tab. 1; data yield a dominant, conjugate fault set of NNE
MET32, median age of MET133, 134, 135, 136). If SSW striking extensional faults (great circle 1 and 2)
the single extensional fault, located at the eastern end and three minor subsets (great circles 3, 4 and 5). The
of TR1 (Fig. 6A, marked dIVT), was active simulta- average fault dip is 708. In addition to the similarities
neously with the main fault zone at the western end in fault trends, the maximum vertical offsets are con-
of the trench, a maximum age would be 14 ka sistent with ~0.5 m. These observations strongly sug-
(derived from sample MET35; Tab. 1). Unfortunately, gest that the fault zones in TR1 and TR3 plus the
simultaneous activity cannot be proven so the more single fault in TR2 belong to the same fault system.
56 G. Peters et al. / Tectonophysics 406 (2005) 3966

TR1: N, S & W-wall TR2: N & S-wall, top of unit 1B


A B
2) 58/279 2) 59/280
Best fit great circle to 39 data 1) 58/096 Best fit great circle to
559 data 1) 46/104
1) synthetic, extensional 1) synthetic, extensional
2) antithetic, extensional 2) antithetic, extensional
3) sinistral strike-slip
4) oblique to dextral
strike-slip
4) 62/165

3) 68/032
'I' 'I'
0 15 0 0 515 0
'II' 345 'III' 'II' 355 'III'

55 55
120 75
270 90 90 270 90 90

TR3: N & S-wall, top of unit 1B combined TR1, TR2 & TR3 data
C D 2) 62/285
2) 59/291 1) 51/106 Best fit great circle to
232 data 1) 59/113 830 data 3) 74/330 1) synthetic, extensional
4) 75/334 2) antithetic, extensional
3) extensional
Best fit great circle to
4) oblique to dextral
1) synthetic, extensional
strike-slip
2) antithetic, extensional
5) sinistral strike-slip
3) sinistral strike-slip
4) extensional 4) 62/165

3) 77/026 5) 70/029
'I' 'I'
0 15 0 0 15 0
25
'II' 35 'III' 'II' 'III'

115 65
115 65 55 75 65
270 90 90 270 90 90 75

Fig. 9. Fault data from TR1 (A), TR2 (B) and TR3 (C) using data of all the trench walls and from the floors of TR2 and TR3: I) k poles to faults
plotted on lower hemisphere, equal area projection and best fit circles through dominant fault sets. II) Rose plot of fault data 18 of maxima. III)
Inclination plot of faults. D: Summary plot of all data. In order to ensure as high a quality data set as possible, only fault segments are included
in the analysis on which two or more consistent measurements where recorded.

The NNESSW (0158) trend of the fault zones is the and high dips (Dezayes et al., 2004). The good corre-
so-called Rhenish trend (Suess, 1885), which is repre- lation in fault orientation and faulting mechanism of the
sented by the border faults of the URG. Studies of trench site with the neotectonic faults and the basement
outcrops and reflection seismic profiles give indica- fractures of the region imply that the structures at the
tions of two principle trends of neotectonic faults in trench site are part of an active tectonic fault system.
the northern URG: (1) NS trend (Straub, 1962;
Anderle, 1974; Schneider and Schneider, 1975; Illies 5.4. Surface trace of the fault system
and Greiner, 1976), and (2) NESW to EW trend
(Stablein, 1968; Illies and Greiner, 1976). The kine- Analyses of structural data strongly suggest that
matics of these neotectonic faults is mainly extensional. the fault zones observed in the trenches are part of the
Studies of exploration wells at the Soultz-sous-Forets same fault system. In plan view, a line drawn between
geothermal reservoir site, located near the WBF in the the fault zones in TR1 and TR3 has a trend of 0208
middle URG, revealed an active fracture set in the (Fig. 4A). The geophysical measurements from the
basement rocks with conjugate, NS striking fractures trench site give additional information on the trace of
G. Peters et al. / Tectonophysics 406 (2005) 3966 57

the fault system. Geophysical profiles taken parallel to as a complex fault system consisting of several seg-
TR1 and TR3 exhibit deformed reflection patterns and ments and extending from the base of the scarp to V1
a distinct inclined resistivity contrast at the same over an area several tens of meters wide (Fig. 4A).
location as the fault zones in the trenches. It is there-
fore concluded that the structures in the measurements
(Fig. 5) represent the fault zone. A single straight line 6. Reconstruction of trenches 1, 2 and 3
passing through all known fault zones determined by
geophysical measurements and trenching cannot be The excavation of the 3 trenches provided an insight
constructed. It is therefore concluded that several into the near-surface expression of an active fault sys-
overlapping and merging fault zone segments exist, tem and the sedimentary conditions at the scarp. In
yielding a general strike for the fault system of 0158 order to obtain information on the chronological
(NNESSW). This strike is in good agreement with order of tectonic, erosional and depositional events at
the Rhenish trend of the WBF. Based on the similarity the site, a reconstruction has been built by backstrip-
in strike and on the location of the trench site, the fault ping along two sections: section A with TR1 and TR2
system seen in the trenches and the measurements is and section B with TR3 (Fig. 10, see also Fig. 4B).
regarded as a near-surface expression of the WBF. At Eight evolutionary steps are included. The timing of
the trench site this near-surface expression is defined these steps has been determined using absolute ages

Section A, TR1/TR2 Section B, TR3


WNW ESE WNW ESE
~ 12 ka - Present-day

Trench 3 5A+B
Trench 2
2B
2A 5A+B 2A 4B
4B
3B
28 m

1B
24 m

1B
Trench 1 3B
4A
step1 1A 4A step 1 1A
? ? ?
70 m 70 m

2B
2A 2A
19 - 8 ka

1B 1B
3B 3B 4B
4B
4A
step 2 1A 4A step 2 1A
? ? ?

23 ka
18 ka 2B
2A 2A
23 - 14 ka

19 ka
1B 4B 14 ka 1B
3B 3B 4B
4A
step 3 1A 4A step 3 1A
? ?

2B
2A 4B 2A 3B
31 - 22 ka

1B 22 ka
1B 4A
3B 31 ka
4A
step 4 1A step 4 1A
? ?

Fig. 10. Reconstruction of the sequence of events at the trench site along two sections shown on Fig. 4B. Section A follows the trace of TR1 and
TR2. Section B includes TR3 and the location of GPR 8 profile. Based on the GPR 8 profile (Fig. 5C) section B suggests another fault zone at
the base of the scarp (dashed line with question mark).
58 G. Peters et al. / Tectonophysics 406 (2005) 3966

Section A, TR1/TR2 Section B, TR3

2B
2A 37 ka 2A
37 - 31 ka

1B 1B
3B 3B

step 5 1A step 5 1A
? ?
loess
loess
2B
2A 2A
~ 100 - 37 ka

3A
1B > 100 ka 1B
3B
3B
step 6 1A step 6 1A
? ?

2B
~ 130 - ~ 100 ka

2A 2A

1B 1B

step 7 1A step 7 1A
? ?

2B crossbedded fluvial sand


2B crossbedded fluvial sand
~ 450 - ~ 130 ka

Riss Riss

2A 2A

1B
1B 1B
1B

step 8 1A step 8 1A
? ?

Fig. 10 (continued).

from TL and relative ages from correlation with marine accumulated to more than 2 m on parts of the trench
isotope stages (MIS; ages after Shackleton, 1987). site and smoothed the surface. Preceding human ac-
Since only few absolute ages are available, the corre- tivity caused partly removal of units 1B, 3B and 4B
lation with MIS is incorporated in order to support and and created a steep slope near the scarp base. Step 1
present as complete a reconstruction as possible. The includes the historic and prehistoric period. Late Mid-
correlation assumes the general simplifying concept dle Age artifacts in the lowest colluvium horizons
that interglacial periods coincide with river incision give a maximum age for the colluvium accumulation
whereas glacial periods coincide with fluvial accumu- (Tab. 1). TL dating yields a Late Wurm age for the
lation (e.g. Penck, 1910; Budel, 1977). It should be deposition of unit 4B.
noted that recent insights show that the fluvial response
to climate change is complex in detail (van Balen et al., 6.2. Step 2 faulting activity between Late Wurm and
2003). Early Holocene (198 ka)

6.1. Step 1 erosion and deposition from Late Wurm Step 2 represents the faulting event (Fig. 10, step 2).
to present (~12 kapresent) The fault zone in section A dconcentratesT at the base of
the scarp. Since TR1 uncovers only the eastern part of
The present-day situation is documented by 3 the fault zone, it is suggested that the fault zone extends
trenches, numerous cores and geophysical measure- further west into the scarp (Fig. 10, step 2, section A,
ments (Fig. 10, step 1). A soil and colluvium cover dashed lines with question marks). In TR3, a well
G. Peters et al. / Tectonophysics 406 (2005) 3966 59

defined, 5 m wide fault zone occurs, which is shown as seems unlikely that this lateral termination is coinci-
a narrow zone in section B. The deformation of unit 4B dentally located at the present-day fault zone. It is
in this section is an interpretation. It is concluded from possible that a fractured material influenced the ero-
section A that this unit also covered the slope at section sion, thus suggesting penultimate faulting during step
B during step 2 and has been eroded between step 1 and 5. The faults excavated in TR1 do not show offsets of
2. A GPR profile from the base of the scarp in section B this event but other faults that have not been excavat-
(GPR7) shows a similar reflection pattern as the profile ed could exist and be related to the event (Fig. 10, step
at TR1 (GPR8) and suggests that another fault zone 5, dashed lines with question marks).
may exist at this location (Figs. 5C, 10, step 2 dashed
line with question mark). The faulting event occurred 6.6. Step 6 deposition on the slope from early to
after the deposition of unit 4B and before the formation Middle Wurm (~10037 ka)
of the Holocene soil, unit 5B. TL dating from TR1
yields a maximum age of 19 ka and a minimum age of At step 6, units 3A and 3B covered the scarp and
8 ka for the event (Tab. 1). created a smooth surface (Fig. 10, step 6). Dating of
units 3A and 3B is difficult since a sample from the
6.3. Step 3 deposition of unit 4B in the Late Wurm base of unit 3B delivers an age of N 100 ka (Fig. 10,
(2314 ka) step 6; Tab. 1) suggesting that unit 3A is older than
100 ka. The correlation of sedimentary events with
Prior to the faulting, unit 4B accumulated on the MIS suggests a cold period for the deposition of the
slope and smoothed the scarp (Fig. 10, step 3). East- units 3A and 3B. From MIS 5b (~100 ka) the climate
ward dipping layers and strata diverging towards the was gradually cooling and reached a minimum during
east point to a downslope transport of the material. TL the MIS 4 (7459 ka). This would suggest that the
dates of 2314 ka give a Late Wurm age for the most likely rapid deposition of unit 3A initiated at
deposition of unit 4B (Fig. 10, step 3; Tab. 1). ~100 ka and was followed by the deposition of unit
3B, which lasted until 37 ka (step 5).
6.4. Step 4 deposition of unit 4A in the Middle to
Late Wurm (3122 ka) 6.7. Step 7 erosion and initial formation of the
scarp during Eemian (~130 ~100 ka)
It is inferred from TR1 in section A that unit 4A
was deposited on both the slope and at the base of the The base of units 3A and 3B show the limit of
scarp (Fig. 10, step 4). For section B, no information erosion into the older deposits (Fig. 10, step 7). This
from the base of the scarp exists, but a similar situa- lateral and vertical erosion caused the initial formation
tion to section A has been assumed for the reconstruc- of the scarp at the trench site. Similar to step 5, it is
tion. TL dates of 3122 ka give a Middle to Late possible that faulting activity influenced the position
Wurm age for the deposition of unit 4A. of this initial scarp. The correlation with MIS infers a
warm period for erosion, probably the MIS 5e
6.5. Step 5 erosion and retreat of the scarp in the (Eemian interglacial). Considering that the deposition
Middle Wurm (3731 ka) of unit 3A started in the MIS 5b (step 6) the erosional
event at step 7 may have occurred between MIS 5e
The onlap of unit 4A onto unit 3B in TR1 indicates until MIS 5b (~130~100 ka).
that unit 3B is older (Fig. 10, step 4). Furthermore, it
suggests that unit 3B originally extended eastwards 6.8. Step 8 deposition of a fluvial terrace and soil
and that parts of the unit have been eroded. Step 5 formation during Riss and Eemian (~450 ~130 ka)
represents this erosional phase. TL dating of units 3B
and 4A suggests that the erosion took place between The final step of the reconstruction suggests that a
3731 ka (Fig. 10, step 5; Tab. 1). The erosion caused flat surface existed and that a soil, unit 2B, developed
a retreat of the scarp and terminated at the location of on the cross-bedded sands of unit 2A, (Fig. 10, step
the present-day fault zone in TR1 (Fig. 10, step 1). It 8). The large thickness (2 m) of this soil observed in
60 G. Peters et al. / Tectonophysics 406 (2005) 3966

TR3, section B, implies that it extended over the to consider is the position of the fault zone in the
whole trench site area. It is concluded that the 2B trenches relative to the lithological units. The fault
soil has been eroded at section A, most likely during zones are both located at the base of the scarp within
step 7. The underlying units (1A, 1B, 2A) were slope wash and sands (units 1A, 3B, 4A, 4B in TR1)
deposited on a plain. Unit 2B and the sands of unit and on top of the slope within subhorizontally bedded
2A are regarded as forming a terrace body. This units (units 1B, 2A; TR3). A slide generating such
terrace is located about 20 m above the Wurm lower faults would be initiated on the sloping lithological
terrace, located at the base of the scarp, and could boundary between the impermeable unit 1B and the
therefore belong to the Middle Terraces from the Riss overlying units. However, along this particular contact
glacial period (~350130 ka). The thickness of unit no features of sliding were observed.
2B implies a long soil formation process, which could Due to the setting of the trench site, e.g. slope and
have initiated during the last interglacial (Eemian, Late Wurm permafrost conditions, a non-tectonic ori-
MIS 5e, ~130 ka). gin cannot be entirely excluded. However, based on
all observations, i.e. the correlation with regional fault
trends, the consistency in deformation style across the
7. Discussion trenches and the lack of mass movement features at
the most prone position; faults in the trenches have to
7.1. Discussion on the origin of faults be of a mainly tectonic origin.
In order to gather constraints about the tectonic
The results of trenching and geophysical measure- strain rate, i.e. seismic or aseismic deformation, the
ments show that the region exhibits near-surface fault present deformation pattern has to reveal temporal
structures. However, the presence of near-surface fault information about the faulting process. Since nume-
structures does not reveal information of their origin as rous analogue models show that the generation of
different processes result in surface faulting similar in similar fractures in sandy material both takes place
style to the structures observed in the trenches. The at rapid and creeping rates (e.g. McClay and Ellis,
trench site is situated along the border fault of an active 1987; Vendeville et al., 1987; Davy and Cobbold,
rift system. Thus, the faults observed may have a 1991), the brittle deformation of the sandy units ex-
tectonic origin. However, both surface processes and posed in the trenches cannot reveal temporal informa-
tectonic faulting can create slopes, which could be tion. In contrast, the clay-rich material of the 11 m
large enough to generate mass movements and could thick unit 1B observed in TR3 is considered to have a
also result in near-surface fault structures. This leads to time-dependent rheological behavior with the brittle
the following questions: are the faults at the trench site deformation of this unit likely to be related to relative
of tectonic or gravitational origin, and if the origin was high strain rates. To prove whether the deformation
tectonic, was the movement aseismic or seismic (i.e. occurred at high or low strain rates, the exact compo-
creeping or rapid)? sition and conditions under which the clay-rich mate-
The structural analysis of the faults in the trenches rial deformed (e.g. degree of water saturation) would
revealed a consistent style and orientation, which have to be known. Unfortunately, this information is
correlates well with independent data from regional unknown and cannot be retrieved. However, the li-
trends of active faults (Straub, 1962; Stablein, 1968; quefaction features observed in TR1 could be an
Anderle, 1974; Schneider and Schneider, 1975; Illies indication of co-seismic deformation. Obermeier
and Greiner, 1976; Dezayes et al., 2004). This sup- (1996a) developed five criteria to prove earthquake-
ports a tectonic origin. If however the faults in the induced liquefaction: (1) Rapid injection of material
trenches have a gravitational origin, one or more large under overpressure, (2) the existence of several types
slides would be required. This is considered unlikely of features, (3) ground water settings, (4) occurrence
since no morphological indication of large slides at multiple locations and (5) formation in short, dis-
exists at the trench site. Nevertheless small mass crete episodes. Criteria 1, 2 and 3 clearly apply to the
movements did occur at the site, as evidenced by trench site. No data is available for criteria 4 since the
the folds and decollements in TR1. Another aspect intensive land use of the study area precludes a re-
G. Peters et al. / Tectonophysics 406 (2005) 3966 61

gional study of liquefaction features. Furthermore, dynamics have played a major role in the evolution
criteria 5 cannot be proven because the features in of this part of the scarp (Fig. 10). Step 5 of the
TR1 cannot be dated with sufficient precision. In reconstruction of the trenches shows that lateral ero-
addition to the 5 criteria, gravitational sliding has to sion terminates within the sandy and silty unit 3B and
be ruled out for inducing liquefaction (Obermeier, at the location of the present-day fault zone (Fig. 10).
1996a). The most prominent liquefaction features This coincidence cannot be explained by the mechan-
occur in the footwall of TR1 within clay-rich and ical behavior of the material and there is no reason
cohesionless sand (Fig. 7). Since the lithological why the erosion would not have extended westward
units affected remained in situ and show no sign of into the more resistant clay-rich unit 1B. One possible
mass movement, these features are regarded as indi- explanation for this limit of erosion could be a frac-
cating a co-seismic event. For the liquefaction features tured, and thus weakened, material (unit 3B) whereas
observed in the hangingwall (Fig. 6A,E) a relation to the material further west was intact, and thus stronger.
mass movements cannot be excluded. Based on these Conversely, the existence of an erosional scarp
arguments, the liquefaction features observed in TR1 (formed during steps 5 and 7) could also have influ-
are interpreted to be earthquake-induced. enced the location of the surface termination of the
fault zone during a later faulting event. It remains an
7.2. Paleoseismological implications of this study open question for the trench site whether an erosional
topography influenced faulting or pre-existing faults
Since it remains unclear whether the tectonic defor- influenced erosion. At the northern end of the scarp
mation observed in the trenches was related to contin- the influence of fracturing on erosion can be inferred
uous creep or a single event, two scenarios for the for historic times. Here, it has been suggested that pre-
formation of the maximum vertical displacement ob- existing faults subsidiary to the WBF have aided large
served are possible. If the fault displacement is the landslides (Steuer, 1911; Fig. 2) by strongly influenc-
result of creep, the minimum average fault displace- ing the permeability on the slope.
ment rate is 0.04 mm/yr (0.5 m / 198 ka). If the fault The interplay of fluvial dynamics and tectonics is
displacement was caused by a single seismic event with also supported by the larger scale geomorphology of
a minimum vertical displacement of 0.5 m, calculation the area. The margins of the Mainz Bingen Graben are
of an earthquake magnitude yields a moment magni- formed by a sequence of fluvial terraces, i.e. wide
tude of M w = 6.5 (Wells and Coppersmith, 1994). horizontal surfaces on top of the plateau (Early Pleis-
A creep rate of 0.04 mm/yr is relatively low, though tocene terraces), multiple small terraces along the
it is of the same order of magnitude as average dis- scarp fronts (Middle to Late Pleistocene terraces)
placement rates of the region. The Quaternary uplift and the wide surface of the Late Pleistocene Lower
rate of the Mainz Basin obtained from the northern part terrace at the base (Kandler, 1970; Abele, 1977). The
of the WBF scarp (170 m uplift at Oppenheim; Brun- morphology of the WBF scarp closely resembles these
ing, 1977) is ~0.09 mm/yr. A paleoseismic event of M w margins (Figs. 2, 11). It is thus suggested that the wide
6.5 is larger than the low-magnitude historical earth- surfaces on top of the plateau and the smaller surfaces
quakes observed in the northern URG. Unfortunately, on the WBF scarp front are fluvial terraces of the
detailed information on multiple seismic events and River Rhine or a tributary. The arrangement of the
their timing cannot be derived from the trench site terraces implies an eastward migration of the river.
and the recurrence interval for large earthquakes cannot The suggestion that fluvial terraces form the scarp
be constrained. Other data from paleoseismological front is supported by the trench site interpretation
trenching would be necessary to support the observa- (Fig. 10, step 8).
tions of a possible M N 6 event of this study. The terraces of the Mainz Bingen Graben are
interpreted to have formed under regional tectonic
7.3. Discussion on the origin of the scarp uplift, which also caused the reactivation of faults
along the Mainz Bingen Graben and the migration
The sequential reconstruction of the trench site of the River Rhine away from the Mainz Basin
shows that, in addition to tectonic activity, fluvial (Wagner, 1930; Kandler, 1970; Abele, 1977). The
62 G. Peters et al. / Tectonophysics 406 (2005) 3966

W E
m a.s.l
10x exag.
wide surfaces small wide surface of
surfaces Lower Terrace
140 (Wrm)

100 trench site

0 1000 2000 3000 4000 m

Fig. 11. West to east topographic profile crossing the scarp at the trench site shown on Fig. 3. The profile is based on digital topography data and
field observations and shows a succession of wide and small horizontal surfaces interpreted as fluvial terraces.

terraces along the WBF scarp may have also been terplay of fluvial dynamics, erosion, regional uplift
influenced by this regional uplift. The formation of and localized tectonic activity on the WBF, has a
the scarp would then be the result of fluvial erosion mixed origin.
under increased regional tectonic activity. In addition,
the concentration of multiple small terraces along the
scarp front coinciding with the position of the WBF Acknowledgements
suggests a strong relationship between terrace and
scarp formation and localized fault activity. A similar We would like to thank M. Meghraoui, M. Wei-
relationship is suggested by Brustle (2002) for the denfeller, G. Greiner, K. Reicherter and J. Reinecker
Rhine River Fault scarp at the southern end of the for extremely useful discussions on the regional ge-
URG (Fig. 1B). Recent studies on the Frankenthaler ology, tectonics and the sedimentary environment of
Terrace scarp, 20 km south of the study area, also the northern URG. S. Husges, S. Reiss and W. Frei are
suggest a tectonic control on the scarp formation, thanked for their geophysical investigations and use-
since the scarp coincides with faults mapped in the ful discussions. Thanks are extended to N. Debenham,
subsurface (Weidenfeller and Karcher, 2004). P. Grootes, W. Boenigk and E. Hagedorn for dating
and analyzing the trench sediments. The project de-
veloped in the context of the EUCOR-URGENT proj-
8. Conclusions ect and the EU project of Environmental Tectonics
(ENTEC). It was funded by the Gesellschaft fur An-
In summary, the integration of shallow geophysics, lagen-und Reaktorsicherheit (GRS) mbH. Additional
paleoseismological trenching and structural analysis financial support for G. Peters was provided by ISES
enabled the location and characterization of near-sur- (Netherlands Research Centre for Integrated Solid
face deformation structures of the WBF at the base of Earth Science) and DAAD (Deutscher Akademischer
the southern end of the scarp. The results of 3 trenches Austauschdienst, PhD grant) and for P. Connolly by
point to extensional faulting producing a consistent, the World Stress Map Project. The manuscript benefit-
conjugate fault set of 0158-striking faults, parallel to ed substantially from the reviews of two anonymous
the WBF strike, and maximum vertical displacements reviewers.
in the order of 0.5 m. TL dating of deformed sedi-
ments shows that the faulting event occurred between
19 and 8 ka. The faulting may have been caused by a Appendix A. Stratigraphy of the trenches
single seismic event with a moment magnitude of 6.5
and/or creep movements. Reconstruction of the se- A.1. Unit 1A sand and clay lenses
quence of events at the trench site shows the local
scale interplay between tectonic activity on the WBF Unit 1A consists of sand with lenses of clay-rich
and fluvial and erosional processes. This interplay is sand and clay and is interpreted as a braided river
also important at the regional scale. Thus, the 20 km deposit. The unit is exposed at the bottom of TR1
long WBF scarp, which was associated with the in- (Fig. 6) and has been excavated in a number of cores
G. Peters et al. / Tectonophysics 406 (2005) 3966 63

near the trench. Core samples adjacent to this trench Schachtschabel, 1998). The 1 m thickness of the B-
display strong lateral changes associated with the horizon in TR3 is exceptional and implies a long soil
lenticular bedding. No absolute dating of this unit formation process. It is likely that this process started
could be performed, however comparable fluvial in the Eemian interglacial and continued into the Late
quartz-rich sands with interbedded clay have been Holocene.
documented as Pliocene in exposures south of the
trenching area (Franke, 2001). A.5. Unit 3A conglomerate

A.2. Unit 1B mottled clay-rich sand Unit 3A is a gravel layer with a matrix of material
from unit 1B, unconformably overlying unit 1A in
Unit 1B consists of a mixture of clay and sand. TR1 (Fig. 6). Unit 3A is interpreted as a basal con-
Reduction and oxidation patches and abundant man- glomerate of the overlying clastic succession. Parts of
ganese nodules throughout the whole unit give it a unit 3A have also been found in cores ~10 m west of
mottled appearance, typical for post depositional TR1 on top of the scarp (borehole 52, Figs. 4B, 10). A
pseudogley formation. The clay-rich sand is inter- TL sample from the base of the overlying unit 3B
preted as a still water deposit, possibly an aban- (MET12) is of N 100 ka.
doned meander or a lake. Unit 1B is exposed at the
bases of TR2 and TR3 (Fig. 8). The high kaolin A.6. Unit 3B interbedded sand and silt
content in the samples of unit 1B indicates Pliocene
weathering conditions, so a Pliocene age for unit Unit 3B consists of interbedded sand and silt. It is
1B is most likely (pers. comm. W. Boenigk, E. unconformably overlying both unit 3A in TR1 and
Hagedorn, 2003). unit 1B in TR2 (Figs. 6, 8). The lower 1 m exhibits
several deformed zones. The upper 1 m shows parallel
A.3. Unit 2A cross-bedded sand lamination. In TR2, unit 3B shows reworking of the
underlying unit 1B with embedded clayey sand lenses.
Unit 2A consists of cross-bedded, cohesionless Here, diverging strata and sandy deposits dip eastward
sands with gravel layers and is overlying unit 1B and indicate high-energy conditions. In this trench, a
with an erosional discontinuity (Fig. 8). Unit 2A is gradational fining upward transition to unit 4B exists,
interpreted to be a braided river deposit. It is exposed so that no clear boundary could be determined
at the western end of TR2 and in the middle and (boundary in Fig. 8A is arbitrary). Unit 3B is inter-
western parts of TR3 (Fig. 8). The sands are beyond preted as a slope wash deposit. TL dating of a sample
the dating range of TL, so that no absolute age for unit from the bottom (MET12) yields N100 ka and two
2A could be obtained. Correlation of the sands with samples from the top of the unit (MET13, MET132)
fluvial terrace levels at the trench site suggests that yield 19 and 37 ka (Table 1, Fig. 8).
they form, together with unit 2B, one of the Middle
terraces from the Riss glacial period at ~450130 ka. A.7. Unit 4A iron coated sand and silt

A.4. Unit 2B paleosol Unit 4A consists of iron coated interbedded sand


and silt. The unit is interpreted as the basal part of a
The fluvial sands of unit 2A form the substratum of second slope wash deposit. It forms an easterly dip-
a soil, unit 2B, consisting of 2 horizons (Fig. 8). In the ping erosional discontinuity with the underlying unit
lower C-horizon the clay content increases upward 3B in the western part of TR1 (Fig. 6). It has a wedge
and the original layering is obliterated. The 1 m shape and is located within the fault zone of TR1 (Fig.
thick B-horizon is characterized by higher clay con- 6D,E). Unit 4A occurs approximately 6 m east of the
tent and frequent manganese-oxide nodules. Unit 2B fault zone at the base of TR1 and in cores exhibiting
is interpreted as a paleocambisol similar, but with a almost no slope (Fig. 6A). TL dating from the western
thicker B-horizon, to Pleistocene cambisols seen else- part yields 22 ka (MET81) and 31 ka (MET31; Table
where (Scheer and Wourtsakis, 1986; Scheffer and 1, Fig. 6A) ages.
64 G. Peters et al. / Tectonophysics 406 (2005) 3966

A.8. Unit 4B parallel laminated silt and fine sand Bartz, J., 1974. Die Machtigkeit des Quartars im Oberrheingraben.
In: Illies, H., Fuchs, K. (Eds.), Approaches to Taphrogenesis,
Stuttgart, pp. 78 87.
Unit 4B consists of parallel laminated silts and fine Bonjer, K.-P., 1997. Seismicity pattern and style of seismic faulting
sands. The unit is interpreted as a slope wash deposit at the eastern borderfault of the Southern Rhine Graben. Tecto-
consisting of fluvially reworked loess and local sands. nophysics 275, 41 69.
Due to the high content of loess it is named dalluvial Bruning, H., 1977. Zur Oberflachengenese im zentralen Mainzer
Becken. Mainzer geographische Studien 11, 227 243.
loessT. The unit is present in all 3 trenches (Figs. 6,
Brustle, A., 2002. Morphostructural analyses of the region of Frei-
8). In TR1, unit 4B occurs in the whole 26 m long burg i. Br. (Upper Rhine Graben). Diploma thesis, Geological
trench overlying unit 4A (Fig. 6). In TR2 and TR3 Institute, University of Cologne, BRGM Orleans. Geological
unit 4B is exposed in the eastern parts of the trenches Institute, Albert-Ludwigs University Freiburg i.Br., 83 pp.
overlying units 1B and 3B, respectively (Fig. 8). TL Budel, J., 1977. Klima-Geomorphologie. Berlin, 304 pp.
dates from all 3 trenches have consistent ages for unit Cloetingh, S., Cornu, T. (Eds.), 2005. Neotectonics and Quaternary
Fault-reactivation in Europes Intraplate Lithosphere, Quaterna-
4B ranging from 23 to 14 ka and pointing to a Late ry Science Reviews, vol. 24 (34), pp. 235 508.
Wurm age (Table 1, Figs. 6A, 8). Due to reliable Cloetingh, S., Ziegler, P., Cornu, T., ENTEC Working group, 2003.
dating of this unit a reconstruction of the fault timing Investigating environmental tectonics in northern Alpine fore-
has been possible. land of Europe. EOS, Transactions, American Geophysical
Union 84 (36), 356 357 (349).
Cushing, M., Lemeille, F., Cotton, F., Grellet, B., Audru, J.-C.,
A.9. Unit 5A colluvium, unit 5B - recent soil Renardy, F., 2000. Paleo-earthquakes investigations in the
Upper Rhine Graben in the framework of the PALEOSIS proj-
A recent soil, unit 5B, is present in all 3 trenches ect. PALEOSIS-Project, Final Report, pp. 39 43.
consisting of silty and fine sandy material (Figs. 6, Dambeck, R., Thiemeyer, H., 2002. Fluvial history of the Northern
8). In TR2 and TR3 the soil overlies a Late Middle Upper Rhine River (Southwestern Germany) during the Late-
glacial and Holocene times. Quaternary International 9394,
Age colluvium (unit 5A), dated with Late Middle 53 63.
Age pieces of pottery (pers. comm. J. Schutz, 2003), Davy, P., Cobbold, P.R., 1991. Experiments on shortening of a 4-
and is therefore younger than this (Fig. 8). Converse- layer model of the continental lithosphere. Tectonophysics 188,
ly, TL dating from the base of the soil in TR1 gives 1 25.
ages from 510 ka (Fig. 6A). It is important to note Demoulin, A., Pissart, A., Zippelt, K., 1995. Neotectonic activity in
and around the southwestern Rhenish shield (West Germany):
that the TL ages of unit 5A between 4 and 24 ka indications of a levelling comparison. Tectonophysics 249,
(MET115, MET118, MET131; Table 1) indicate only 203 216.
the age of the reworked material and not the age of Dezayes, C., Genter, A., Gentier, S., 2004. Fracture network of the
the colluvium. EGS geothermal reservoir at Soultz-sous-Forets (Rhine Graben,
France). Geothermal Energy The Reliable Renewable, Indian
Wells, California, USA, August 29September 1, 2004, GRC
Transactions, vol. 28, pp. 213 218.
References De`zes, P., Schmid, S., Ziegler, P.A., 2004. Evolution of the Euro-
pean Cenozoic rift system: interaction of the Alpine and Pyr-
Abele, G., 1977. Morphologie und Entwicklung des Rheinsystems enean orogens with their foreland lithosphere. Tectonophysics
aus der Sicht des Mainzer Raumes. Mainzer geographische 389, 1 33.
Studien 11, 245 258. Doebl, F., 1967. The Tertiary and Pleistocene sediments of the
Ahorner, L., 1983. Historical seismicity and present-day microearth- northern and central part of the Upper Rhinegraben. Abhandlun-
quake activity of the Rhenish Massif, Central Europe. In: Illies, gen Geologisches Landesamt Baden-Wurttemberg 6, 48 54.
H., Fuchs, K., von Gehlen, K., Malzer, H., Murawski, H., Doebl, F., Olbrecht, W., 1974. An isobath of the tertiary base in the
Semmel, A. (Eds.), Plateau Uplift: the Rhenish Massif a Rhinegraben. In: Illies, H., Fuchs, K. (Eds.), Approaches to
Case History. Heidelberg, pp. 198 221. Taphrogenesis, Stuttgart, pp. 71 72.
Anderle, H.J., 1974. Block tectonic interrelations between Northern EUCOR-URGENT, 2005. Homepage: http://comp1.geol.unibas.ch/.
Upper Rhine Graben and Southern Taunus Mountains. In: Illies, Fetzer, K.D., Larres, K., Sabel, K.-J., Spies, E., Weidenfeller, M.,
H., Fuchs, K. (Eds.), Approaches to Taphrogenesis, Stuttgart, 1995. Hessen, Rheinland-Pfalz, Saarland. In: Benda, L. (Ed.),
pp. 243 253. Das Quartar Deutschlands, Berlin, pp. 220 254.
Bartz, J., 1936. Das Unterpliocan in Rheinhessen. Jahresberichte Franke, W.R., 2001. Geologische Karte von Rheinland-Pfalz
und Mitteilungen Oberrheinischer geologischer Verein, N.F. 25, 1:25.000, Erlauterungen, Blatt 6215 Gau-Odernheim. Geolo-
121 228. gisches Landesamt Rheinland-Pfalz, Mainz.
G. Peters et al. / Tectonophysics 406 (2005) 3966 65

Friedel, S., 1997. Hochauflosende Geoelektrik- Geoelektrische Final Report, Project No. ENV4-CT97-0578. Directorate-
Tomographie. In: Beblo, H. (Ed.), Umweltgeophysik, Berlin, General XII for Science, Research and Development (Brus-
pp. 131 151. sels): 135 pp.
Illies, H., 1974. Taphrogenesis and plate tectonics. In: Illies, Penck, A., 1910. Versuch einer Klimaklassifikation auf physiogra-
H., Fuchs, K. (Eds.), Approaches to Taphrogenesis, Stuttgart, phischer Grundlage. Preussische Akademie der Wissenschaften,
pp. 433 460. Sitz der Phys.-Math. 12, 236 246.
Illies, H., 1975. Intraplate tectonics in stable Europe as related to Pflug, R., 1982. Bau und Entwicklung des Oberrheingrabens,
plate tectonics in the Alpine system. Geologische Rundschau Darmstadt. 132 pp.
64, 677 699. Scheer, H.D., Wourtsakis, A., 1986. Bodenkarte von Rheinland-
Illies, H., 1977. Ancient and recent rifting in the Rhinegraben. Pfalz 1 : 25.000, Erlauterungen Blatt 6415 Grunstadt-Ost. Geo-
Geologie en Mijnbouw 56, 329 350. logisches Landesamt Rheinland-Pfalz, Mainz.
Illies, H., Greiner, G., 1976. Regionales stress-Feld und Neotekto- Scheffer, F., Schachtschabel, P., 1998. Lehrbuch der Bodenkunde,
nik in Mitteleuropa. Oberrheinische geologische Abhandlungen Stuttgart. 401 pp.
25, 1 40. Schneider, E.F., Schneider, H., 1975. Synsedimentare Bruchtektonik
Kaiser, D., 1999. Re-evaluation of the Largest Earthquakes in the im Pleistozan des Oberrheintal-Grabens zwischen Speyer,
Northern Upper Rhinegraben (Germany) Since the Year 858. Worms, Hardt und Odenwald. Munstersche Forschungen zur
International Union of Geodesy and Geophysics, Birmingham. Geologie und Palaontologie 36, 81 126.
(XXII General Assembly). Schumacher, M.E., 2002. Upper Rhine Graben: Role of preexisting
Kandler, O., 1970. Untersuchungen zur quartaren Entwicklung des structures during rift evolution. Tectonics 21, 1. doi:10.1029/
Rheintales zwischen Mainz/Wiesbaden und Bingen/Rudesheim. 2001TC900022.
Mainzer geographische Studien 3, 9 90. Schwarz, E., 1974. Leveling results at the northern end of the
Krauter, E., Steingotter, K., 1983. Die Hangstabilitatskarte des Rhinegraben. In: Illies, H., Fuchs, K. (Eds.), Approaches to
linksrheinischen Mainzer Beckens. Geologisches Jahrbuch Taphrogenesis, Stuttgart, pp. 261 268.
C34, 2 33. Schwarz, E., 1976. Prazisionsnivellement und rezente Krustenbe-
Landesamt fur Vermessung und Geobasisinformation. Rheinland- wegung dargestellt am nordlichen Oberrheingraben. Zeitschrift
Pfalz, Koblenz. fur Vermessungswesen, 14 25.
Leydecker, G., 2003. Erdbebenkatalog fur die Bundesrepublik Semmel, A., 1968. Studien uber den Verlauf jungpleistoza-
Deutschland mit Randgebieten fur die Jahre 800- 2001 ner Formung in Hessen. Frankfurter geographische Hefte 45
(www.bgr.de/quakecat). Federal Institute for Geosciences and (133 pp.).
Natural Resources, Hannover, Germany. Semmel, A., 1986. Angewandte konventionelle Geomorphologie-
Loke, M.H., 2001. Electrical imaging surveys for environmental Beispiele aus Mitteleuropa und Afrika. In: Bar, W.-F. (Ed.),
and engineering studies. A practical guide to 2-D and 3-D Frankfurter Geowissenschaftliche Arbeiten, Frankfurt a.M.
surveys. www.geoelectrical.com. Serie D, Band 6. 114 pp.
McClay, K.R., Ellis, P.G., 1987. Geometries of extensional Shackleton, N.J., 1987. Oxygen isotopes, ice volume and sea level.
fault systems developed in model experiments. Geology 15, Quaternary Science Reviews 6, 183 190.
341 344. Sissingh, W., 1998. Comparative Tertiary stratigraphy of the
Meyer, W., Albers, H.J., Berners, H.P., von Gehlen, K., Glatthaar, Rhine Graben, Bresse Graben and Molasse Basin: corre-
D., Lohnertz, W., Pfeffer, K.H., Schnutgen, A., Wienecke, K., lation of Alpine Foreland events. Tectonophysics 300,
Zakosek, H., 1983. Pre-quaternary uplift in the central part of 249 284.
the Rhenish Massif. In: Fuchs, K., von Gehlen, K., Malzer, H., Stablein, G. 1968. Reliefgenerationen der Vorderpfalz- Geomorpho-
Murawski, H. (Eds.), Plateau Uplift, the Rhenish Shield a logische Untersuchungen im Oberrheingraben zwischen Rhein
Case History. Heidelberg, pp. 39 46. und Pfalzer Wald. PhD thesis, Wurzburger geographische Arbei-
Monninger, R. 1985. Neotektonische Bewegungsmechanismen im ten. 23, 183 pp.
mittleren Oberrheingraben. PhD thesis. Geological Institute, Stahmer, G. 1979. Geologie und Stratigraphie des Raumes Grun-
University of Karlsruhe, 219 pp. stadt- Monsheim / Pfalz. Diploma thesis. Geological Institute,
Obermeier, S.F., 1996a. Use of liquefaction-induced features for University of Heidelberg, 92 pp.
paleoseismic analysis an overview of how seismic liquefac- Steuer, A., 1911. Geologische Karte des Groherzogtums Hessen
tion features can be distinguished from other features and how im Mastabe 1: 25.000, Erlauterungen, Blatt Oppenheim. Hes-
their regional distribution and properties of source sediment can sisches Landesamt fur Bodenforschung, Wiesbaden.
be used to infer the location and strength of Holocene paleo- Straub, E.W., 1962. Die Erdol-und Erdgaslagerstatten in Hessen und
earthquakes. Engineering Geology 44, 1 76. Rheinhessen. Abhandlungen geologisches Landesamt Baden-
Obermeier, S.F., 1996b. Using liquefaction-induced features for Wurttemberg 4, 123 136.
paleoseismic analysis. In: McCalpin, J.P. (Ed.), Paleoseismo- Suess, E., 1885. Das Antlitz der Erde. Tempsky, Prague. 778 pp.
logy, San Diego, pp. 331 397. Tuttle, M.P., 2001. The use of liquefaction features in paleo-
PALEOSIS, Commission of the European Communities (Ed.), seismology: lessons learned in the New Madrid seismic
2000. Evaluation of the potential for large earthquakes in zone, central United States. Journal of Seismology 5 (3),
regions of present-day low seismicity activity in Europe. 361 380.
66 G. Peters et al. / Tectonophysics 406 (2005) 3966

van Balen, R.T., Vandenberghe, J., Kasse, K. (Eds.), 2003. Quater- Wagner, W., 1941. Bodenversetzungen und Bergrutsche im Mainzer
nary Science Reviews, Fluvial Response to Rapid Environmen- Becken. Geologie und Bauwesen 13, 17 33.
tal Change, vol. 22 (20), pp. 2053 2235. Weidenfeller, M., Karcher, T., 2004. Terrassen, Kieslager und
Vendeville, B., Cobbold, P.R., Davy, P., Brun, J.P., Choukroune, P., Zwischenhorizonte - Neue Aspekte zur geologisch-hydrolo-
1987. Physical models of extensional tectonics at various scales. gischen Gliederung der quartaren Sedimente im linksrhei-
In: Coward, M.P., Dewey, J., Hancock, P. (Eds.), Continental nischen Teil des nordlichen Oberrheingrabens. From Source to
Extensional Tectonics, London, Geological Society Special Pub- Delta, DEUQUA Meeting, 30 August3 September 2004, Nij-
lication, vol. 28, pp. 95 107. megen, The Netherlands.
Villemin, T., Alvarez, F., Angelier, J., 1986. The Rhinegraben: Wells, D.L., Coppersmith, K.J., 1994. New empirical relationships
extension, subsidence and shoulder uplift. Tectonophysics 128, among magnitude, rupture length, rupture width, rupture area,
47 59. and surface displacement. Bulletin of the Seismological Society
Wagner, W., 1930. Bemerkungen zur Tektonischen Skizze des wes- of America 84 (4), 974 1002.
tlichen Mainzer Beckens. Notizblatt der Vereinigung Erd- Ziegler, P.A., 1992. European Cenozoic rift system. Tectonophysics
kunde und hessisches geologisches Landesamt V (12), 208, 91 111.
185 188.