Quaternary International xxx (2013) 1e14

Contents lists available at SciVerse ScienceDirect

Quaternary International
journal homepage: www.elsevier.com/locate/quaint

Historical tsunami deposits in Peru: Sedimentology, inverse modeling and

optically stimulated luminescence dating
Michaela Spiske a, *, Jens Piepenbreier a, b, Carlos Benavente c, Alexander Kunz d, e, Heinrich Bahlburg a,
Jens Steffahn f
a
Westfälische Wilhelms-Universität, Institut für Geologie und Paläontologie, Corrensstrasse 24, 48149 Münster, Germany
b
Dornier Consulting, P.O. Box 2730, Riyadh 11461, Saudi Arabia
c
Instituto Geológico, Minero y Metalúrgico INGEMMET, Av. Canadá 1470, Lima, Peru
d
Leibniz-Institut für Angewandte Geophysik e LIAG, Stilleweg 2, 30655 Hannover, Germany
e
Department of Geosciences, National Taiwan University, No. 1, Sec. 4, Roosevelt Road, Taipei 10617, Taiwan, ROC
f
GeoDienste GmbH, Leinestraße 33, 30827 Garbsen, Germany

a r t i c l e i n f o a b s t r a c t

Article history: The entire coast of Perú was surveyed to document deposits of historical tsunami. Evidence of four tsunami
Available online xxx was found. At Puerto Casma, in northern Perú, a graded, heavy mineral-rich layer with shell- and rock
fragments was detected. OSL dating revealed an age of 0.37
0.03 ka (1615e1667 AD). Inverse modelling
of tsunami onshore flow parameters indicates a flow depth of w6 m and a flow speed of w8 m/s at a
distance of w60 m from the shoreline. For the Chimbote earthquake of 1619 AD, a tsunami was not
listed in the historical tsunami catalogues, although it is seen as a predecessor of the 1996 Chimbote
earthquake and tsunami. Hence, this study may provide the first evidence of a local tsunami triggered
by this event. A graded, shell-rich event layer that contains cobbles with attached marine organisms
was found at Vila Vila in southern Perú. This layer was dated to 0.17
0.04 ka (1797e1871 AD) and is most
probably the result of the 1868 Arica tsunami. Two additional event layers found in Boca del Río (southern
Perú) were dated 2.26
0.37 ka (615 BCe119 AD) and 1.98
0.23 ka (207 BCe255 AD). Thus, the layers
record for the first time tsunami much older than the events listed in tsunami catalogues. These two events
exhibit similar parameters with flow speeds of 7e8 m/s and depths of 6 m in a distance of 460 m from the
present shoreline. The time that passed between the events is similar to the recurrence interval of the two
Arica tsunami of 1604 and 1868 AD, hence giving evidence of another prehistoric earthquake and tsunami
couplet.
Ó 2013 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction Additionally, radiocarbon dating cannot be used for very young

Research on ancient wave events such as tsunami and severe
storms can provide a long-term record of the history of a coastline
and help to estimate recurrence intervals for risk management. For
age determination of these historical events, a reliable method is
necessary. The most common methods for dating young sediments
are optically stimulated luminescence (OSL) and radiocarbon
dating.
Despite some shell accumulations, high-energetic wave deposits
are frequently composed of siliciclastic material (e.g., Dawson and
Shi, 2000; Nanayama et al., 2000; Goff et al., 2004). Plant material
is sparse. Hence, radiocarbon dating is not always possible.
* Corresponding author.
E-mail address: spiske@uni-muenster.de (M. Spiske).
samples of <350 years because of the complexity of the calibration
curve which can lead to non-unique solutions and multiple ages
(Stuiver, 1978; Aitken, 1990, 1998; Huntely and Clague, 1996). OSL
dating provides the advantage of using potassium feldspar or
quartz, which are abundant in most coastal settings, and in contrast
to radiocarbon dating there is no reservoir effect involved. More-
over, OSL covers a wider time span than radiocarbon dating and
can be used for very recent sediments (e.g., Kunz et al., 2010a;
Alappat et al., 2011), as well as for deposits of up to 500 ka (e.g.,
Aitken, 1990, 1998).
OSL dating is a well-established method for dating long-term
processes, including deposition ages of glacial or fluvial sedi-
ments and dune development or migration (e.g., Wood, 1994;
Wallinga et al., 2001; Bateman, 2008; Fuchs and Owen, 2008;
Lancaster, 2008; Rittenour, 2008; He et al., 2010). Nevertheless,
dating event sediments which have been deposited during a very

1040-6182/$ e see front matter Ó 2013 Elsevier Ltd and INQUA. All rights reserved.
http://dx.doi.org/10.1016/j.quaint.2013.02.010

Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
2 M. Spiske et al. / Quaternary International xxx (2013) 1e14

short time is still a challenge. First approaches of dating storm surge

deposits were made, amongst others by Sommerville et al. (2003),
van Heteren et al. (2008), Madsen et al. (2009) and Davids et al.
(2010). Tsunami sediments have been dated by Huntely and
Clague (1996), Banerjee et al. (2001), Ollerhead et al. (2001),
Bishop et al. (2005), Murari et al. (2007), Cunha et al. (2009), Kunz
et al. (2010a), Brill et al. (2012) and Prendergast et al. (2012).
This study uses OSL dating to determine depositional ages of
suspect event beds along the coast of Perú and give evidence of
their tsunamigenic origin. By dating the respective event deposits, a
tsunami history that reaches beyond the limit of the tsunami cat-
alogues (Appendix A) for Perú, may be obtained.

2. Regional setting

2.1. Seismic hazard

The Perú-Chile Trench is one of the seismically most active areas

in the world (Kulikov et al., 2005). The subduction of the Nazca
Plate under the South American Plate causes earthquakes with
magnitudes (Mw) greater than 8.2 every w25 years (Lockridge,
1985). Consequently, the risk for destructive regional tsunami
along the coast of Perú is high. There is also an additional risk of
ocean wide tsunami generated along the margins of the Pacific
plate and propagating across the Pacific, affecting the coasts of Perú
(Lockridge, 1985; Kulikov et al., 2005). Submarine slides along the
continental margin only add a minor component to the generation
of tsunami (Kulikov et al., 2005) because of the sediment starved
nature of the marine forearc (Thornburg and Kulm, 1987; von
Huene and Scholl, 1991). Berninghausen (1962) notes that a
tsunami affects the west coast of South America approximately
every three years.
The list of historical tsunami starts some tens of years after the
Spanish conquest of Perú at ca. 1532e1535 AD (Dorbath et al., 1990;
Appendix A). Berninghausen (1962) lists 49 tsunami along the west
coast of South America from 1562 to 1960 AD. Twenty of these
events occurred in the period from 1900 to 1960 AD. A compilation
by Lockridge (1985) records 34 tsunami between 1586 and 1974 AD.
The greatest historical tsunami events along the Perú-Chile Trench
were the two Arica tsunami in 1604 and 1868 AD (Okal et al., 2006),
and the Chile tsunami in 1960 AD (Cisternas et al., 2005). The most
recent tsunami were the Chimbote tsunami in 1996 AD (Petroff
et al., 1996; Bourgeois et al., 1999), the Camana tsunami in 2001
AD (EERI, 2001; JSCE, 2001; USGS, 2001; Okal et al., 2002; Dengler
et al., 2003; Jaffe et al., 2003) and the Pisco-Paracas tsunami in 2007
AD (Astroza, 2007; Sanchez, 2007; Fritz et al., 2008; Yauri et al.,
2008; Spiske et al., 2010).
2.2. Study area
This study surveyed the w2400 km long coast of Perú in order to
find traces of historical tsunami events. Drilling or trenching was
conducted at 90 sites Fig. 1; Appendix B).
Most of the coast of Perú is a semi-desert with a hyperarid
climate, primarily shaped by aeolian processes. The annual pre-
cipitation is below 10 mm/y (NCDC, 2011). Storms are generally
absent and storm surges do not exceed 1 m height (Bourgeois et al.,
1999). Consequently, marine inundations along the Peruvian coast
most likely were caused by tsunami.
The coastal environments of Perú display various morphologies
with different preservation potentials of event deposits (Spiske
et al., 2010). Coastal environments include steep coastal cliffs,
sandy beaches, broad river plains and marshes. Steep coasts and
regions with a high rate of aeolian erosion are not suitable for the
preservation of event deposits since deposits will be modified or
Fig. 1. Map of Perú depicting the main field sites and plate tectonic elements.
reworked. Fluvial muddy and potentially subsiding settings like
marshes provide a better preservation potential as long as the event
deposits are not affected by floodwaters of rivers that may be
connected either to the annual snowmelt in the Andes or to El Niño
events. Sandy event deposits are easiest to identify within wetland
deposits due to their differences in grain-size, sediment composi-
tion and sediment color. The recognition of a sandy event bed
within beach sand may not be as easy.
3. Methods
3.1. Field work and sediment analysis
Sediments were studied in trenches and boreholes down to
depths of 3 m. Drilling was carried out using a window sampler and
a drop hammer. If drill cores contained evidence of event deposits
and if digging was possible, trenches were additionally excavated to
study the spatial extent of possible sedimentary structures. In order
to trace the inland extent of the deposits, several trenches or
boreholes were made with even spacing (25 m) at 90 to the coast
line. Both normal beach or marsh sediments and intercalated event
layers were sampled. If possible, samples were taken in different
levels of an event layer in order to detect possible faintly developed
grading.
For the study of the grain-size distribution (Figs. 2e4), a rotary
sample splitter was used to separate a representative amount of
30e60 g (dependent on grain-size) from the dried bulk sample. The
grain-size analysis was conducted using a digital particle counter
(PartAn 2001L by AnaTec) which determines the size of the parti-
cles that fall in front of a high resolution camera. The effective range
of the particle counter lies between 5.0 and þ6.0 phi. Samples
that included coarser particles were measured by sieving or with a
caliper.
The magnitude of historical tsunami can be estimated by the
application of inverse models, such as TsuSedMod by Jaffe and
Gelfenbaum (2007). This model uses the thickness and grain-size

Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
 M. Spiske et al. / Quaternary International xxx (2013) 1e14 3

Fig. 2. Historical tsunami deposit and grain-size distributions at Puerto Casma dated at 0.37
0.03 ka (1615e1667 AD). The outcrop is located 60 m from the present coastline.

distribution of a tsunami inflow layer to calculate the flow speeds
and depths necessary to entrain and transport the given sediment
volume and grain-size. The model assumes a steady and uniform
tsunami flow, and an equilibrium distribution of sediment in the
water column. Additionally, the model postulates that at least 90%
of the sediment is transported in suspension and that typical
tsunami sediments are hence normally graded. As sediment
thickness is one of the input parameters, it is worth noting that
processes, such as erosion by successive waves or by the backwash,
and post-depositional processes including surface erosion or
compaction can reduce the original thickness of the event layer
(Spiske et al., 2010). Consequently, the observed thickness of a
tsunami deposit is a minimum value, as are the computed Tsu-
SedMod results (Spiske et al., 2010). Taking these assumptions into
account, Apotsos et al. (2012) argue that the model is most
appropriate for gently sloping broad coastal plains where the

Fig. 3. Historical tsunami deposits and grain-size distributions at Boca del Río dated at 1.98
0.23 ka and 2.25
0.37 ka. The outcrop is located 460 m from the present coastline.

Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
4 M. Spiske et al. / Quaternary International xxx (2013) 1e14

Fig. 4. Historical tsunami deposit and grain-size distributions at Vila Vila dated at 0.17
0.04 ka. The outcrop is located 80 m from the present coastline.

backwash is either not strong, or restricted to morphological de-
pressions by channeling. For a more detailed description on the
application of the model, refer to Spiske et al. (2010) and Witter
et al. (2012).
All samples were examined for foraminifera using a reflected-
light microscope and separated foraminifera were photographed
using a JEOL 840 scanning electron microscope (SEM). The fora-
minifera were described regarding their preservation condition and
genetically classified (Fig. 5d).
Cathodoluminescence (CL) microscopy was applied to explain
differences in OSL-signal properties of quartz from southern and
northern Perú in regard to their original source area. CL microscopy
can be used to discriminate between quartz of different rock types,
such as metamorphic, plutonic and volcanic rocks (e.g., Zinkernagel,
1978; Augustsson et al., 2011; Augustsson and Reker, 2012). CL of
quartz takes advantage of the geochemical composition and lattice
defects formed due to variations in temperature and pressure and
by substitution of ions during crystal growth and subsequent
alteration events (e.g., Zinkernagel, 1978; Götze et al., 2001).
CL measurements were performed with a hot-cathode lumi-
nescence microscope (HC-1LM), coupled with a water-cooled
(63  C) ANDOR OE-CCD detector. Polished thin sections were
coated with carbon and single grains were exposed to electron
bombardment for 50 s. The obtained light spectra are in the range
of 375e890 nm and were corrected for background signal. The
spectra were evaluated using the method of Augustsson and Reker
(2012) in which a comparison of the positions of the two main
intensity peaks and the trough position in between is used to define
the origin of each quartz grain.
3.2. OSL dating
OSL dating dates the depositional age itself and not the death of
incorporated organic material, as being the case for radiocarbon
dating (Aitken, 1998). OSL determines the time elapsed since a
mineral grain was exposed to sunlight prior to deposition. The
method is based on electrons being activated by natural radioac-
tivity and trapped within defects in the crystal lattice, such as
cosmic radiation, as well as a-, b-, and g-radiation emitted by the
sediments that surround the mineral grain. The status of trapped
electrons will be undone when energy is applied to the minerals in
the form of sunlight, artificial light, and heat. The stimulated elec-
trons will leave the traps and jump to lower energetic levels and
recombine with luminescence centers, causing the emission of light
as OSL signal (Aitken, 1998). The emitted light is proportional to the
number of formerly trapped electrons. This means that the OSL
signal gets more intensive the longer natural radiation activates the
electrons to jump into traps and the more electrons are captured in
these holes (Aitken, 1998).
The OSL signal which represents the time since the last exposure
of a mineral grain to sunlight, can only be preserved if the sample is
taken in light proof tubes protecting them from bleaching by recent
sunlight. For this study, sampling included both the event layer and
the under- and overlying layers (Figs. 2e4). The samples were
processed under red light conditions at the OSL laboratory of the
Leibniz Institute for Applied Geophysics in Hannover, Germany.
Preparation of coarse grained samples from 100 to 150 mm included
drying, sieving (grain-sizes are listed in Appendix C), dissolution of
organic and carbonate matter, heavy liquid separation of quartz and
potassium feldspar and etching of quartz with hydrofluoric acid for
60 min to remove the outer rim. For a more detailed description of
the sample preparation procedure, refer to Aitken (1998) and Kunz
et al. (2010a,b). The quartz and feldspar grains were mounted using
silicon oil (Rüsch Silkospray) on stainless steel discs. The grains
were placed in the center of the aliquot covering an area of 2e3 mm
in diameter containing w1000 grains. The fine grained samples
were prepared according to Frechen et al. (1996). Basically this
method follows the same steps as for coarse grain samples. Sepa-
ration of the 4e11 mm fraction was done by settling (based on
Stokes law) using a centrifuge. The prepared polymineral fine grain

Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
 M. Spiske et al. / Quaternary International xxx (2013) 1e14 5

Fig. 5. Detailed sedimentological composition of the tsunami unit at Vila Vila. a) Imbricated shell fragments with landward inclination angles of 30e45 , b) rock fragments, some of
them with attached barnacles or vermetids, and c) shells and shell fragments incorporated into the event layer. d) Foraminifera of the event layer and the surrounding beach sand.

samples were mounted on aluminum discs by settling in acetone.
Each aliquot contained approximately 2 mg of sample material
covering the complete disc.
First test measurements showed unsatisfactory OSL properties
for the quartz samples. The natural signal was low and in general
the quartz appeared dim (Fig. 6). Additionally a dominant slow
component precluded the proper use of OSL. Therefore, in this
study, potassium feldspar was used for dating.
The use of feldspar entails further advantages because feldspar
emits an up to 100 times more intense luminescence signal than
quartz, which entails that they can be used to date younger samples
and that the obtained precision is higher than for quartz (Wagner,
Fig. 6. OSL properties of ideally behaving quartz from dunes in southeast India (after
Kunz et al., 2010b) compared to quartz from coastal Perú. The signal from the Peruvian
quartz is the average of natural signals from the samples from Puerto Casma and Boca
del Río.
1998; Lian and Roberts, 2006; Li et al., 2007). An analytical disad-
vantage of feldspar is the possible presence of anomalous fading
(Wintle, 1973). Anomalous fading seems to be connected to electron
leakage which entails a signal loss, i.e. electrons can escape from
traps within the mineral lattice without being lifted into the con-
duction band (Wintle, 1973; Aitken, 1998). Therefore, anomalous
fading may lead to an age underestimation and consequently, ages
need to be corrected for this effect (e.g., Huntley and Lamothe, 2001;
Lamothe et al., 2003). The g-value defines the rate of anomalous
fading in percent per decade. More recent studies try to avoid the
fading correction for Pleistocene deposits by using higher optical
stimulation temperatures to measure signals from non-fading
affected traps (e.g., Thomsen et al., 2008; Buylaert et al., 2009).
Measurements were conducted with a Risø Reader TL/OSL-
DA-15. Feldspar grains were stimulated with infrared light
(l ¼ 875 nm) and the IRSL-signal was detected through a combi-
nation of a Schott BG 39 and Corning 7-59 filters. Irradiation was
done with a 90Sr/90Y beta source with a dose rate of 0.175 Gy/s. The
measurement was done using a single aliquot regenerative (SAR)
dose protocol based on Kunz et al. (2010a) and Urban et al. (2011).
The settings for the SAR protocol derived from preheat tests made
on one sample from each location. The results are shown in
Appendix D. All three samples show a preheat plateau in the
temperature range from 160  C to 200  C. For preheat temperatures
higher than 220  C and 240  C for sample SVV3a the equivalent
doses increases significantly. Based on these results for the samples
in this study preheat and cutheat temperatures were set to 180  C
and held for 10 s. Optical stimulation was done for 100 s at 50  C
with IR-diodes. The test dose was 1.75 Gy and the regenerative
doses ranged from 0.32 Gy to 10.5 Gy depending on the expected
age of the sample. After each SAR-cycle a hotbleach step was
included with 40 s of IR-stimulation at 290  C. Before measuring
the samples, the SAR protocol was checked with a dose recovery

Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
6 M. Spiske et al. / Quaternary International xxx (2013) 1e14

test on all samples if an artificially administered dose can be
recovered using this protocol. For all samples the dose could be
recovered with dose recovery ratios of unity within 10%. For each
sample 24 aliquots were measured with the SAR protocol to
determine the equivalent dose. Aliquots have been accepted for
further analysis when the recycling ratio was unity within 10% and
the recuperation did not exceed 10%. In Appendix C the number of
accepted aliquots for each sample is listed. The feldspar from
sample NPCA1a showed exceptional poor luminescence behavior
and high fading rates that made this sample unusable for an age
determination. The final equivalent dose (De) for each sample was
obtained from the dataset of single De using the central age model
after Galbraith et al. (1999). Ages were calculated by dividing the
equivalent dose (De) with the dose rate of the sediment. The ob-
tained feldspar ages were then corrected for anomalous fading
using the method from Huntley and Lamothe (2001) and Auclair
et al. (2003).
The dose rate of the natural radiation of the samples was
measured with a HPGe N-type detector (see Kunz et al., 2010a) at the
Leibniz Institute for Applied Geophysics in Hannover (Germany).
Samples for dose rate analysis were dried, homogenized and filled
into beakers (50 g material). The beakers were sealed and stored for
at least four weeks so that the radon disequilibrium could readjust.
The results of the gamma-ray spectrometry and dose rate calculation
are shown in Table 1.
southern Perú (Fig. 1). In Puerto Casma and Vila Vila a single event
layer was documented. At Boce del Río, two successive layers were
found.
4.1.1. Puerto Casma (northern Perú)
The beach of Puerto Casma (Fig. 1) is located at the southern end
of a w3 km wide bay, 45 km south of Chimbote. At a depth of 0.6 m, a
conspicuous 3e6 cm thick layer of sand was found (Fig. 2). It can
clearly be distinguished from the normal beach sand (mean grain-
size þ1.8 to þ2.3 phi) by its coarser grain-size (þ1.6 phi) and
different colour. The layer exhibits an undulating, erosive contact to
the underlying sand and contains some shells and rock fragments
(1e2 cm in size), as well as heavy mineral accumulations at its base.
Two subsamples were collected from the layer. Grain-size analysis
documents the presence of normal grading which indicates depo-
sition of the layer from suspension. The variation in colour between
the event layer and the surrounding sands is due to the different
mineral composition, i.e. higher content of heavy minerals and
coarse plagioclase and quartz grains in the event layer. Grain-size
distributions additionally show that the event layer contains
reworked beach sand, whereas coarser components and heavy
minerals indicate a probable shallow marine origin. The presence of
heavy minerals within the event layer reflects that the possible
tsunami possessed higher energetic flow conditions in comparison
to normal wave conditions, hence being able to transport and erode

Table 1
Results of IRSL-measurements and age calculation for historical tsunami deposits and the related surrounding sediments at Puerto Casma (northern Perú), Boca de Río and Vila
Vila (both southern Perú). Equivalent doses were calculated using the central age model (Galbraith et al., 1999). Fading corrected ages were calculated using the approach from
Huntley and Lamothe (2001). If the KeS test value is smaller than the critical KeS value (in brackets) the De distribution follows a normal distribution.

Location Sample ID N KeS test results Total dose rate Equivalent g-Value [% per Fading uncorrected Fading-corrected Historical age
[Gy/ka] dose De [Gy] decade] OSL age [yrs]b OSL age [yrs]b range
Puerto Casma NPCA1a 7 0.171 (0.483) 3.671
0.980 0.577
0.114 12.1
1.4c 0.17
0.03 kac 0.48
0.21 kac 1315e1739 ADc
09 27.060 S NPCA1ba 24 0.166 (0.269) 4.764
0.300 1.272
0.040 3.8
0.2 0.28
0.02 ka 0.37
0.03 ka 1615e1667 AD
078 22.50 W NPCA1c 23 0.140 (0.275) 3.647
0.201 0.981
0.025 4.0
0.5 0.28
0.02 ka 0.38
0.03 ka 1599e1655 AD

Boca del Río SBDR1a 20 0.129 (0.294) 3.845
0.500 7.133
0.110 3.4
0.4 1.95
0.26 ka 2.58
0.33 ka 902e240 BC
18 09.290 S SBDR1ba 22 0.161 (0.275) 3.507
0.470 4.505
0.337 5.9
0.6 1.35
0.16 ka 2.26
0.37 ka 615 BCe119 AD
070 40.190 W SBDR1c 20 0.071 (0.294) 3.909
0.500 5.789
0.093 3.4
0.6 1.55
0.21 ka 2.05
0.27 ka 316 BCe230 AD
SBDR1da 21 0.100 (0.287) 3.421
0.350 3.836
0.151 6.0
0.4 1.19
0.12 ka 1.98
0.23 ka 207 BCe255 AD
SBDR1e 24 0.113 (0.269) 4.002
0.510 2.914
0.050 5.1
0.8 0.76
0.10 ka 1.16
0.17 ka 685e1019 AD

Vila Vila SVV3a 24 0.183 (0.269) 2.775
0.270 0.240
0.016 6.7
0.5 90
7 a 145
18 a 1844e1880 AD
18 06.260 S SVV3ba 23 0.178 (0.275) 3.560
0.600 0.319
0.038 8.3
0.8 94
12 a 173
37 a 1797e1871 AD
070 43.350 W SVV3c 24 0.098 (0.269) 3.116
0.261 0.168
0.009 7.4
0.9 57
4 a 95
12 a 1900e1924 AD

N ¼ number of aliquots used for age calculation.

a
Event layer.
b
OSL ages related to the year of sampling in 2007 AD.
c
Fading rate is very high so that obtained age may not be used.

Furthermore, it was tested if the analyzed samples were
completely bleached during transport, i.e. prior to deposition. This
was done with the distribution of the equivalent doses (De) in a
sample that can give evidence about the amount of bleaching of the
sediment prior deposition. Fully bleached sediments should show a
Gaussian distribution of De values. Whereas partly or incomplete
bleached sediments show a skewed distribution of De values with a
peak shifted to lower values (Rittenour, 2008). The distribution of De
values from all measured samples here was checked for normal
distribution using the KolmogoroveSmirnov test (KeS test; Table 1).
4. Results
4.1. Historical tsunami deposits
Evidence of historical tsunami deposits was found at Puerto
Casma in northern Perú, and at Vila Vila and Boca del Río in
i) material from environments below the storm wave base and ii)
material that was too heavy (too large or too dense) for being
transported by the normal wave regime. Tracing the event layer for
about 100 m landward direction gives evidence of a slight fining
trend. OSL dating gives an age of 0.37
0.03 ka (1615e1667 AD) for
the event layer. The event layer and the overlying beach sands with
an OSL age of 0.38
0.03 ka (1599e1655 AD) are in stratigraphic
order within the range of error. The OSL age of the underlying
beach sand has a relative error of 44% and an exceptional high
fading rate with a g-value of 12.1
1.4%. Thus, the unsatisfactory
luminescence properties of this sample give doubt to the reliability
of the dating result, so that the obtained age may not be used.
Flow parameters necessary to generate the possible tsunami-
genic layer at Puerto Casma were calculated using a deposit thick-
ness of 2 cm and a mean grain-size of þ1.6 phi. The inverse
modeling resulted in a flow speed of 7.8 m/s and a flow depth of
6.2 m within w60 m from the present shore line.

Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
 M. Spiske et al. / Quaternary International xxx (2013) 1e14
4.1.2. Boca del Río (southern Perú)
At Boca del Río in southernmost Perú (Fig. 1), dunes separate the
beach from a wide coastal plain which is crossed by an ephemeral
river. In a distance of w460 m from the sea, two distinct sand layers
(samples SBDR 1b and 1d) were found within muddy marsh sedi-
ments (Fig. 3). Both sand layers have sharp and undulating, most
probably erosional contacts to the underlying marsh deposits. The
lower deposit is found 0.4 m below the current surface, the upper
layer 0.3 m below the surface. These layers are interpreted as the
result of two historical tsunami events flooding the dunes and
transporting sediments onto the coastal plain. The two layers are
composed of coarse sand, including some shell fragments. Both
layers have a thickness of w4 cm. The event layers exhibit a much
wider grain-size range from 3.0 to þ5.0 phi, compared to the
marsh sediments (þ1.9 to þ5.0 phi) over- and underlying them. The
coarse tail (3.0 to 2.0 phi) of the tsunami deposits is determined
by the contained platy shells and shell fragments. Furthermore, a
weak trend of normal grading is present within the two tsunami
layers.
OSL dating documents a concordant stratigraphic succession
with each layer being reasonably younger than the underlying
strata and similar time passing during the steady deposition of the
marsh sediments (Fig. 3; Table 1). The lower event layer was dated
to 2.26
0.37 ka (615 BC to 119 AD) and the upper event layer has
an OSL age of 1.98
0.23 ka (207 BC to 255 AD).
Inverse modeling for the lower layer provided a possible
tsunami flow speed of 7.1 m/s and a flow depth of 5.1 m in 460 m
distance to the present coastline. The upper layer reveals a flow
speed of 7.8 m/s and flow depth of 6.1 m.
4.1.3. Vila Vila (southern Perú)
Vila Vila (Fig. 1) is the only location where a distinctive shell
layer was found within fine to medium grained beach sand (Figs. 4
and 5). The shell-rich layer is 11e13 cm thick and was detected in a
depth of 1.1 m below the present surface. The contained shells
entail a polymodal composition of the event layer with a dominant
coarse grain fraction around 5.0 to 3.0 phi. Subsamples docu-
ment normal grading. The grading can further be defined as coarse
tail grading because the decrease in grain-size mainly appears for
the shell fragments, but not for the incorporated siliciclastic ma-
terial. Furthermore, some shell fragments show a landward
imbrication with angles of 30e45 (Fig. 5a) and hence give evi-
dence of bed load transport of components during landward
transport. Moreover, the event layer contains subrounded to
rounded rock fragments of up to 12  11  5 cm in size (about
1.75 kg), some of them with attached barnacles or vermetids,
indicating a marine origin (Fig. 5b). Largest shell fragments are
9  7  4 cm and partly broken (Fig. 5c). Vila Vila is the only location
where foraminifera were found both within the beach sands and
the event layer (Fig. 5d). Foraminifera were exclusively benthic and
can be found in all analyzed layers from Vila Vila. The most frequent
species are Anomalinoides ssp. (Fig. 5d 1e3) and Cribrorotalia ssp.
(among them Cribrorotalia meridionalis; Fig. 5d 4e6). According to
Malumián and Scarpa (2005) this association is typical for middle
to inner shelf regions with water depths of less than 100 m along
the Pacific coast. The complete absence of planktic foraminifera
appears to support a neritic provenance. Even if some species of
Anomalinoides possess a wider depth range as far as the continental
slope (e.g., Heinz et al., 2008), the bad preservation of the forami-
nifera in this study prevents a more detailed assignation. On the
other hand Cribrorotalia, especially C. meridionalis unquestionably
and exclusively inhabits infralittoral environments (Hayward and
Brook, 1994; Zapata and Gutiérrez, 1995; Páez and Zuñiga, 2001).
While fragile elongated and delicate forms are completely absent
the discovered more compact trochospiral tests are often broken,
Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
7
abraded, partially dissolved or show secondary crystallization and
seem to resemble reworked material from the Pleistocene cliffs.
This material was later transported onshore both during normal
wave action and by tsunami. Thus, the foraminifera assemblage
represents post-mortem transportation and cannot be used as
tsunami indicator or for the inference of water depths of the
sediment entrainment.
OSL dating of the sedimentary succession at Vila Vila resulted in
an age of 0.15
0.02 ka (1844e1880 AD) for the underlying beach
sand, 0.17
0.04 ka (1797e1871 AD) for the event layer and
0.10
0.01 ka (1900e1924 AD) for the overlying beach sand
(Fig. 4; Table 1). Whereas the event layer and the overlying sand
layer are in chronological order, the ages of the lower sand layer
and the event layer overlap significantly.
Inverse modeling of tsunami parameters of the event layer ob-
tained flow speeds of 21.4 m/s and flow depths of 47 m necessary to
transport the given amount of sediment and the respective grain-
sizes. Compared to recent tsunami events in Perú and to the his-
torical deposits studied at Boca del Río and Puerto Casma, these
values are unrealistically high and very likely inaccurate. The un-
realistic results of flow parameter calculations imply that the
portion of bed load may have exceeded 10% because the Tsu-
SedMod inverse model can only simulate deposition out of a flow
with <10% bed load (e.g., Jaffe and Gelfenbaum, 2007; Spiske
et al., 2010).
4.2. OSL properties
Many studies use quartz for OSL measurements of very young
sediments. However, in some cases quartz grains do not emit any
measurable luminescence or only signals of low intensity. Compared
to ideal quartz, these samples emitted much slower signals and in
association more background noise (Fig. 6). In Fig. 6, the OSL signals
from Holocene Peruvian and Indian quartz are compared. The Indian
quartz shows an ideal behavior with a fast decaying and intense OSL
signal. The Peruvian quartz shows a very slow decay and weak signal
intensity with a high background. This implies that the Peruvian
quartz lacks a fast component which is sensitive to light exposure
required for registering short-term irradiation during transport or in
case of tsunami turbulent transport processes (e.g., Duller et al.,
1995; Gemmell, 1997). Dose recovery tests with different preheat
temperatures showed that it was not possible to measure an artifi-
cially applied dose. Other parameters which are indicating the
quality and reliability of a SAR measurement (Wintle and Murray,
2006) were not satisfying. The recycling ratios in the SAR measure-
ments are from 0.55 to 0.82 and hence lower than the range of
acceptance of 1.0
0.1. Furthermore, the recuperation values ranging
from 30% to 173% are outside of the range of acceptance of 5%. Based
on these poor OSL properties of quartz, in this study the use of po-
tassium feldspar was more appropriate. For the sampled Peruvian
feldspars, the g-values for anomalous fading are 3.4
0.4% to
8.3
0.8% with only one outlier (12.1
1.4% for the lower sand layer
at Puerto Casma; Table 1). The errors of the OSL ages from the
investigated sections range from 7% to 21%.
Tests of the amount of bleaching, using the KeS test, indicate
that the De distribution of all samples is normal (Table 1). The
skewness as parameter for the shift of the peak in a distribution
shows no significant shift to lower values for all measured samples.
Based on the distribution of the De values, no evidence was found
for incomplete bleaching of the samples.
Several studies have shown that the quality of minerals in OSL
often relies on the age of the minerals and the number of recycling
cycles these grains underwent during which the number of lattice
defects could increase (Aitken, 1998; Sawakuchi et al., 2012).
Therefore, CL was performed to determine the possible sources of
8 M. Spiske et al. / Quaternary International xxx (2013) 1e14
the beach and event sediments and to obtain information of the
quality of the quartz grains regarding their use for OSL dating.
Fig. 7 displays the CL-based origin of quartz grains of the
southern (Vila Vila and Boca del Río) and northern Peruvian (Puerto
Casma) samples. Using the luminescence spectral scheme of
Augustsson and Reker (2012), in southern Perú 95% of the quartz
have CL spectra typical of a low-temperature metamorphic origin
and 5% are volcanic. For northern Perú, 51% of the quartz is low-
temperature metamorphic, 45% is felsic plutonic, pegmatitic or
high-temperature metamorphic, and 4% is of volcanic origin.
5. Discussion
5.1. Challenges in age determination
Generally, if abundant within the sample, quartz is used for the
dating procedure because it is much more sensitive to irradiation
and bleaching (e.g., by sunlight) than feldspar. Whereas a quartz
grain that exhibits the fast component can be completely bleached
and consequently the luminescence signal be zeroed within an
exposure time of 10 s, feldspar grains need to be exposed to irra-
diation for at least 9 min (Godfrey-Smith et al., 1988). In this study,
the sampled quartz does not exhibit a fast component and pre-tests
show a significant background noise compared to quartzes ideal for
their use in OSL dating (Fig. 6). Consequently, OSL ages needed to be
determined using the contained potassium feldspar.
The deficiency of Peruvian quartz grains, especially those from
coastal and Coastal Cordilleran southern Perú, for the application in
OSL dating was also observed by Steffen et al. (2009a,b, 2010). Steffen
et al. (2009a,b) noted that quartz from Quaternary alluvial terraces in
central Perú does not exhibit a fast component, but instead contains
unstable components that cannot be removed, e.g., by preheating,
and consequently providing underestimated ages. Instead, they used
feldspar for the dating procedure because the rate of anomalous
fading was insignificant. Other studies, such as from Argentina, have
shown that Andean quartz in general seems to offer unsuitable
characteristics for OSL dating (e.g., Robinson et al., 2005; Spencer and
Robinson, 2008). The main reasons seem to be the formation age of
the quartz and its provenance. Steffen et al. (2009b) used CL mi-
croscopy to analyze the origin of the quartz grains. The sources of the
magmatic quartzes were Cretaceous to Tertiary rock formations of
the Peruvian Coastal Cordillera (Steffen et al., 2009b). These rela-
tively young quartz grains did not experience long transport and
Fig. 7. Cathodoluminescence of quartz grains from southern (Boca del Río and Vila
Vila; n ¼ 43) and northern (Puerto Casma; n ¼ 51) Perú.
Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
have not undergone multiple sedimentary cycles. Hence, they do not
provide ideal properties for storing OSL signals (e.g., Rhodes and
Pownall, 1994; Preusser et al., 2006; Pietsch et al., 2008; Thrasher
et al., 2009; Zheng et al., 2009). Such multiple sedimentary cycles
with repeated irradiation, bleaching and heating during transport
are supposed to enhance the sensitivity of quartz grains (e.g.,
Sawakuchi et al., 2012).
Additionally, Sawakuchi et al. (2012) note that the origin of the
quartz grains used for OSL dating has a significant influence on their
OSL sensitivity. Therefore, CL analysis was performed to determine
why the quartz grains do not supply a satisfactory OSL sensitivity and
connected weak OSL signal. The CL analysis of the sampled Peruvian
quartz resembles the regional geology (Fig. 7). Puerto Casma lies
close to the mouth of the Río Sechín. The river cuts through Meso- to
Cenozoic plutonic rocks of the coastal batholith (INGEMMET, 1995).
The minor content of volcanic quartz may originate from mainly
Cretaceous (Casma Group) to Tertiary (Calipuy Group) andesites,
basalts or rhyolites (Soler, 1991; INGEMMET, 1995; Noble et al.,
2005). The Cretaceous sand- and mudstones of the Casma Group
were partly metamorphosed during the intrusion of the coastal
batholith and very likely contribute the metamorphic quartz com-
ponents (Pitcher et al., 1985; INGEMMET, 1995).
Boca del Río is located at the mouth of Rio Sama. Since Vila Vila
is only located 8 km to the northwest, Rio Sama will also act there
as main sediment source, e.g., by coast parallel currents trans-
porting the siliciclastic sediments northwards. The samples from
Boca del Río and Vila Vila resemble a dominant component of
metamorphic quartz most likely originating from the Arequipa
Massif (INGEMMET, 1995; Mamani et al., 2010). Some volcanic
quartz that may originate from Tertiary to Cretaceous rocks of the
Coastal Cordillera is also represented (Shackleton et al., 1979;
INGEMMET, 1995; Mamani et al., 2010). With regard to the active
volcanism in the Arequipa region, it is surprising that there is no
higher amount of young volcanic quartz represented in the coastal
sediments from southern Perú. This is probably due to the Coastal
Cordillera acting as a barrier, preventing the westward transport of
clastic material. Furthermore, the absence of young volcanic quartz
might also underline the slow erosion and transport processes that
are most possibly a function of the arid climate, low amounts of
annual rain fall that normally do not exceed 0e200 mm along the
coast (SENAMHI, 2012), and the connected low transport capacity
of the seasonal rivers. This observation underlines the poor OSL
properties because of the relatively short transport paths (max.
100e150 km) and the fact that these sediments were not recycled.
In addition to the provenance of the minerals, the proposed
tsunamigenic origin of the event layers will have influenced the
OSL properties, too. High-energetic, short-term event sedimenta-
tion may not provide enough exposure time to the sunlight and
hence not all grains may be bleached and single grains may not be
completely zeroed. Especially in a turbulent, sediment laden flow,
such as the onshore flow of a tsunami, grains may not reach the
surface of the water column (e.g., Ditlefsen, 1992; Rendell et al.,
1994; Rhodes and Bailey, 1997; Murari et al., 2007). Additionally,
if tsunami transport takes place at night, no natural irradiation will
be available for bleaching. Consequently, the tsunami sediments
may inherit age signals of sediments from different sources and
thus mixed ages. Several studies show that tsunami sediments
exhibit incomplete bleaching and that consequently residual OSL
signals entail an overestimation of the measured age (e.g., Huntely
and Clague, 1996; Ollerhead et al., 2001; Prendergast et al., 2012). It
was shown that some tsunami sediments had residual ages of tens
to hundreds of years (Bishop et al., 2005; Murari et al., 2007; Cunha
et al., 2009; Brill et al., 2012). Only rarely, a complete zeroing of the
OSL signal during event sedimentation is documented (Banerjee
et al., 2001).
 M. Spiske et al. / Quaternary International xxx (2013) 1e14
Putting these influences together, zeroed minerals within
tsunami deposits occur most likely in i) already bleached onshore
surface sediments that were entrained and re-deposited by the
tsunami and ii) in surface sediments which were directly exposed
to sunlight in the morning after a tsunami hit at night or iii) in
sediments that constitute the surface of the tsunami layer and were
bleached after the water flew back into the sea. These criteria would
imply that the sampling of the top of a tsunami deposit may pro-
vide a greater amount of zeroed minerals for OSL dating, whereas
within the middle and lower part of the layer a greater amount of
unbleached, partially bleached and zeroed grains are mixed,
resulting in a tampered age signal. Furthermore, finer grains that
constitute an event deposit and that can be found in the upper part
of a graded layer, are preferable for OSL dating since they were most
likely transported in the upper part of the water column during
tsunami inundation and will be bleached more rapidly compared to
coarser grains.
Reasonable procedures to check the reliability of an event layer’s
age are to i) “sandwich” the layer, i.e. to date the under- and
overlying background sediments and ii) if available, to use plant or
shell material found within the event layer for comparative radio-
carbon dating. Nevertheless, one must be precautious because
erosion of the pre-tsunami surface by the tsunami itself can remove
the upper horizon that stored an age just before the tsunami and
hence, the age of the sediments that are in direct contact to the base
of the tsunami layer may give much older ages. Therefore, the time
spanned by the ages of the under- and overlying sediments may
appear longer than it actually was. In some cases where organic
material was not removed from its natural surroundings by the
tsunami, but by earlier processes, double checking the age of the
event layer with a second method of age determination may also be
imprecise because the age of the organics may be older than the
tsunami deposits. Considering radiocarbon or AMS dating applied
to calcareous marine material, such as shells, a reservoir effect
which may span some hundred years of inaccuracy has to be
considered. Therefore, irrespective of the dating method, age
determination of event layers that were deposited within a very
short time, can only give age estimates. Despite of the need of
further research on procedures for dating the depositional age of
sediments, with special attention to event deposits, current studies
can at least help to establish approximate ages and recurrence in-
tervals of events.
5.2. Modeling of tsunami parameters
The TsuSedMod inversion model makes several assumptions
entailing that it can only be applied for calculating flow speeds and
depths of tsunami deposits that were deposited during runup,
where >90% of the material settled from suspension, and backwash
did not modify the thickness (Jaffe and Gelfenbaum, 2007).
At Puerto Casma and Boca del Río, the presence of erosive
contacts at the base and of marine shells within the tsunami layers
most likely implies a deposition during landward transport. Addi-
tionally, the grading of the layers underlines a deposition from
suspension. At Puerto Casma a slight trend of landward fining was
observed that underlines the deposition during the inland flow.
Along the wide and flat coastal plains of Puerto Casma and Boca del
Río he tsunami was free to inundate the coastal area until its energy
completely dissipated and the backwash was weak, probably not
being able to erode and redistribute the runup sediments. There-
fore the coastal setting and sedimentological features of the
tsunami deposits at Puerto Casma and Boca del Río seem to match
the requirements for the application of the inversion model.
The minimum flow speeds and depths calculated using Tsu-
SedMod provided realistic values for the event layer at Puerto
Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
9
Casma (7.80 m/s and 6.2 m) and the two event layers at Boca del
Río (7.06 m/s and 5.1 m; 7.76 m/s and 6.1 m). These values are in
good agreement with the inundation parameters observed during
recent events. Dengler et al. (2003) report flow depths of at least
3.6e7.3 m for the 2001 AD Camana tsunami. In open coastal set-
tings, such as the sites visited during this study, Dengler et al.
(2003) reported flow depths of at least 4.3e5.1 m in a distance of
750 m from the coast. Flow speeds of 2.80e6.60 m/s were modeled
for the 1996 Chimbote tsunami (Bourgeois et al., 1999).
At Vila Vila, the imbrication of the shells indicates a deposition
during the landward flow (Fig. 5a). Nevertheless, the imbrication
itself depicts a significant portion of bed load transport (Harms
et al., 1975; Johansson, 1976). Consequently, the model cannot be
applied there. The latter fact is underlined by the unrealistically
high values of 21.37 m/s and 47 m calculated by the model. The
normal grading of the tsunami layer (Fig. 4) also indicates that part
of the sediment settled from suspension. Nevertheless, the portion
of bed load seems to have exceeded 10% violating the assumptions
of the model.
5.3. Interpretation of the event sediments
Taking into account that the Perú-Chile-Trench is one of the most
active seismic areas in the world (Kulikov et al., 2005) with a high-
frequency recurrence of large, possibly tsunamigenic earthquakes,
it was surprising to find traces of historical tsunami at only three
sites within coastal Perú. According to Kulikov et al. (2005) a tsunami
with average wave heights of at least 11 m is generated by earth-
quakes within the Perú-Chile Trench every 50 years, with a wave
height of at least 14 m every 100 years and with one of at least 16 m
height every 200 years. Dorbath et al. (1990) indicate that the return
period of great earthquakes within the Perú-Chile Trench that may
cause Pacific-wide tsunami is about one century. Considering events
listed in tsunami catalogues (Appendix A) the return period of
smaller, local tsunami with average wave heights <10 m is about one
decade in Perú. This number is more or less consistent with the most
recent Peruvian events of 1996, 2001 and 2007. Tsunami catalogues
also document that the number of historic tsunami observations in
general and the number of severe tsunami is higher for central and
southern Perú than for northern Perú. This might be a result of the
different tectonic stress conditions along the plate boundaries north
and south of the Nazca Ridge (Cahill and Isacks, 1992; Machare and
Ortlieb, 1992; Gutscher, 2002; Bilek, 2010) or due to the fact that the
coasts of central and southern Perú are more densely populated than
northern Perú and thus more documented historical accounts are
available for these regions.
The small number of detected tsunami layers in Perú may be
due to i) the fact that large portions of the coastline are composed
of cliffs, ii) that coastal deposits and tsunami deposits may have the
same sources and therefore may have a similar appearance, and iii)
the low preservation potential of onshore event layers. The latter
case was documented by Spiske et al. (2010) who studied the
sediments deposited by the most recent tsunami in Perú (Chim-
bote, 1996 AD, Camana, 2001 AD, Pisco-Paracas, 2007 AD) and
found that none of the sediments deposited in Chimbote was left
eleven years after the event and only half to the deposits where left
six years after the Camana event. The low preservation potential
seems to be mainly related to the arid climate and rapid aeolian
erosion of the onshore tsunami sediments. Despite of this, coastal
Perú is an appropriate location for studying historical tsunami
deposits because the region does not experience seasonal storms
and storm surges (NCDC, 2011). Therefore, storm surges can be
excluded as depositing process for coastal event layers.
Torrential rains during El Niño-years are capable of eroding and
transporting huge amounts of sediments. El Niño induced floods
10 M. Spiske et al. / Quaternary International xxx (2013) 1e14

deposit sediments that display a range of features from mud to
sheet and debris flows (e.g., Wells, 1987; Keefer et al., 1998;
Fontugne et al., 1999; Wells and Noller, 1999; Vargas et al., 2000;
Magilligan and Goldstein, 2001; Ortlieb and Vargas, 2003; Keefer
and Moseley, 2004; Vargas et al., 2006). The flood sediment
thickness reaches from a few decimeters to several meters (Fig. 8).
The sediments consist of a mixture of mud, silt, sand and cobble to
boulder sized, often angular, components and can exhibit parallel
lamination, cross-bedding and grading (e.g., Wells, 1987; Wells and
Noller, 1999; Ortlieb and Vargas, 2003). Thus, El Niño induced flood
deposits should exhibit thicker and coarser layers in landward di-
rection, whereas for storm surges or tsunami it would be vice versa
(e.g., Wells, 1987; Witter et al., 2001; Ortlieb and Vargas, 2003;
Fig. 8). Furthermore, flood deposits should be much thicker and
exhibit more sedimentary structures compared to tsunami sedi-
ments (Fig. 8). The event layers documented in this study contain
marine components, have only a small thickness of a few centi-
meters and exhibited landward sorting. Consequently, El Niño
induced floods may be excluded as a cause for the studied coarse
grained event layers in coastal Perú.
In Puerto Casma, there is one event layer that was dated
0.37
0.03 ka (1615e1667 AD; Fig. 2). Around this period the
historical records list seven major earthquakes in the years 1604,
1615, 1619, 1647, 1664, 1678 and 1687 AD (Table 2; Berninghausen,
1962; Iida et al., 1967; Lomnitz, 1970; Lockridge, 1985; Dorbath
et al., 1990; Carpio and Tavera, 2002; Kulikov et al., 2005). The
1615, 1647, 1664 and 1678 AD event and related tsunami affected
only southern Perú. The 1604 AD event was also reported from
northern Perú, but wave heights were small and are not in agree-
ment with calculated flow depths of 6.2 m for Puerto Casma.
Furthermore, the 1604 AD event does not fit within the calculated
range of 1615e1667 AD. This also applies for the 1687 earthquake
with an epicentre close to Lima that caused a Pacific-wide
tsunami that also affected northern Perú. For the two events of
1619 and 1663 AD no tsunami was reported. Excluding the earth-
quakes that affected mainly southern Perú, and also the 1604 and
1687 AD events that lie outside the calculated age range, the 1619
AD event or an unrecorded local event remain as possible triggers
for the deposition of the event layer at Puerto Casma.
Dorbath et al. (1990) noted that it is surprising that for the 1619
AD event no tsunami was reported. However, the quality of the
historical data is in many cases questionable because even today
the Peruvian coastline is in many parts very remote and sparsely
populated (Dorbath et al., 1990; Kulikov et al., 2005). This is also
reflected in the tsunami catalogues that mainly list Lima (Callao)
and Arica, as well as some smaller coastal towns. Supporting
Dorbath et al. (1990), Wells (1996) proposes a tsunami event
depositing organic debris within a beach ridge complex at Santa
around 1618 AD. Despite the possibility of a local event which did
not affect major coastal towns and hence was not registered in
historical tsunami catalogues, the 1619 event with an assumed
magnitude of 7.7e8.0 (Table 2) and a rupture length of 100e150 km
seems to be the most reasonable trigger. The earthquake may be
seen as predecessor of the 1996 AD Chimbote event that had a
magnitude of 7.5, a rupture length of about 100 km and caused a
regional tsunami with wave heights of 3e5 m and a maximum
inundation of 455 m (Petroff et al., 1996; Bourgeois et al., 1999). The
1619 AD event could have triggered a tsunami of similar magnitude.
This would be in good agreement with the flow calculations at
Puerto Casma and the only local occurrence of the related event
layer. Hence, this study may provide the first evidence of a local
tsunami that was triggered by the 14th February 1619 Chimbote
earthquake and can support the assumption of Wells (1996).
The second historical event layer was found at Vila Vila in
southern Perú (Fig. 4) and OSL dated at 0.17
0.04 ka (1797e1871
AD). Tsunami catalogues (Appendix A) list seven major earthquakes
around the obtained time span, namely the 1833, 1835, 1847, 1868,
1869, 1871, and 1877 AD events (Table 2). For the 1833 AD event no
tsunami was reported. All other earthquakes triggered a tsunami that
affected southern Perú, whereas the 1847, 1869 and 1871 AD tsunami
were probably too small to deposit the event layer at Vila Vila. Three
events (1835,1868, 1877 AD) caused a Pacific-wide tsunami. The 1835
AD tsunami was generated in Central Chile, w2100 km south of Vila
Vila, and the main impact was observed in Concepcion, but no severe
damage occurred in southern Perú (Iida et al., 1967; Lomnitz, 1970;
Lockridge, 1985). Consequently, the 1868 and 1877 AD events remain
as possible causes for the event deposit at Vila Vila. The earthquake
of 13th August 1868 was the second major earthquake generated

Table 2
List of possible tsunamis around 1641
26 yr AD (1615e1667 AD) for the event layer in Puerto Casma and 1862
18 yr AD (1797e1871 AD) for Vila Vila. Note that the location
Lima is mainly standing for Callao, the main port of Lima and that Arica belonged to Perú until 1883 AD and was then annexed to Chile.

Date [A.D.] Location of Earthquake Locations of tsunami Rupture Max. tsunami Tsunami Comments References
epicenter magnitude observations length [km] wave height [m] magnitude
24.11.1604 Arica 8.2e9.0 Talcahuano (Chile) 450 10e15 (26) 4 Tsunami affected 1400e1900 km of 2, 3, 4, 5, 6
to Trujillo coastline; max. reported inundation was
10 km in Arica; Arica destroyed by tsunami
16.09.1615 Arica 7.5e7.9 Arica 4 1.5 1, 3, 4, 5, 6
14.02.1619 Chimbote 7.7e8.0 e 100e150 Trujillo and Santa severely destroyed by 2, 5
earthquake; no tsunami observations
07.05.1647 Pisco? 8.5 Arica to Lima 2.8 2 1, 3, 5
12.05.1664 South of Lima 7.3e7.5 e 75 No tsunami observations 2, 5
17.06.1678 North of Lima 7.7e8.0 Lima 2 1, 6
20.10.1687 Lima 8.0e8.5 Pisco to Trujillo >300 5e10 3e3.6 Pacific-wide tsunami 1, 3, 2, 5

18.09.1833 Arica 7.2e7.7 e 50e100 No tsunami observations 2, 5

20.02.1835 Concepcion 7.0e8.2 Puerto Montt (Chile) 7.5e24 3 Pacific-wide tsunami 1, 3, 4, 6
(Chile) to Lima
23.05.1847 Lima Lima 2 1.5 3
13.08.1868 Arica 8.5e9.0 Puerto Montt (Chile) 500 14e21 4 Max. reported inundation was 3 km about 1, 2, 3, 4, 5, 6
to Talara 100 km south of Arica; Arica destroyed by
tsunami; Pacific-wide tsunami
24.08.1869 Arica 6.8e7.4 Arica, Lima 2 1 1, 3, 6
21.08.1871 Lima 6.8e7.0 Lima 1e2 1 1, 3, 6
09.05.1877 Tarapaca 8.0e8.5 Puerto Montt (Chile) 450e500 16e24 4e4.5 Pacific-wide tsunami; 20 m tsunami 2, 3, 4, 5
(Chile) to Talara amplitude in Arica

1. Iida et al. (1967), 2. Dorbath et al. (1990), 3. Lockridge (1985), 4. Lomnitz (1970), 5. Carpio and Tavera (2002) and 6. Berninghausen (1962).

Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
 M. Spiske et al. / Quaternary International xxx (2013) 1e14
Fig. 8. Schematic overview of sedimentological differences of sediments deposited by
tsunami inundation and (e.g., El Niño induced) flooding (no to scale).
offshore Arica after the 1604 AD earthquake. Reported maximum
wave heights in Arica were 15e21 m and w12 m further north close
to Islay (e.g., Berninghausen, 1962; Iida et al., 1967; Lockridge, 1985).
Thus, wave height of at least 10 m may also be assumed for Vila Vila,
w180 km south of Islay and 65 km north of Arica (Fig. 1). Additionally,
this event is closest to the obtained age of 1797e1871 AD. The 1877
Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
11
AD event is slightly out of the obtained age range and was
generated by an earthquake offshore Tarapaca, w200e250 km
south of Vila Vila. Nevertheless, the reported w20 m tsunami
height in Arica document that it also severely affected the coastline of
southern Perú (Berninghausen, 1962; Iida et al., 1967; Lomnitz, 1970).
Compared to the 1868 event, the tsunami height seems to have
dissipated significantly north of Arica because Islay only reports
tsunami heights of 3 m (Lockridge, 1985). The 1877 AD event cannot
be excluded, but the most reasonable event to deposit the event
sediments at Vila Vila is the 1868 AD event which is in historical
sources also called “The great Peruvian earthquake and tsunami”
(Iida et al., 1967).
The two event layers at Boca del Río, dated to 2.26
0.37 ka
(615 BC to 119 AD) and 1.98
0.23 ka (207 BC to 255 AD), are beyond
the limit of the tsunami catalogues that do not list Peruvian events
older than 1513 AD (Carpio and Tavera, 2002). The time that passed
between both events in Boca del Río is ca.272 years. As discussed for
the case of Vila Vila and as depicted in Table 2, southern Perú is a
seismically very active region. The most severe Peruvian earthquakes
and tsunami were generated in the Arica region in the years 1604
and 1868 AD. Interestingly, the time span passing between these two
major events is 264 years and hence very close to the ca. 272 year
interval at Boca del Río. Furthermore, it is documented that the 1604
and 1868 AD earthquakes had identical ruptures and the triggered
tsunami possessed similar inundation parameters (Table 2; Dorbath
et al., 1990). This also applies for the two layers at Boca del Río where
flow parameters are 7.1 m/s and 5.1 m depth for the older and 7.8 m/s
and 6.1 m for the younger event. These observations imply that the
observed layers at Boca del Río represent an earthquake and a
resulting tsunami couplet similar to the two “Great Arica earth-
quakes and tsunami” of 1604 and 1868 AD.
6. Conclusions
The number of tsunamigenic earthquakes along the Perú-Chile
trench implies that such events occur frequently and that the
geological record should be full of tsunami deposits.
This study i) for the first time documents sedimentological evi-
dence of past Peruvian tsunami beyond the limit of the tsunami
catalogues, and ii) confirms that the preservation potential of onshore
event deposits is low since only four event layers were found.
 An event layer from northern Perú is supposed to be the first
documentation of a local tsunami caused by the 1619 AD
Chimbote earthquake. A tsunami was not reported for this
event, but suggested because the earthquake is assumed to be a
predecessor of the tsunamigenic 1996 Chimbote event.
 In southern Perú, a shell-rich layer can be related to the 1868
AD Arica earthquake and tsunami that was one of the most
severe events affecting Perú.
 The two event layers found in southern Perú and dated at
2.26
0.37 ka and 1.98
0.23 ka seem to resemble an earth-
quake and tsunami couplet similar to the major events of 1604
and 1868 AD.
Acknowledgements
Fieldwork on was funded by DFG grant (Deutsche For-
schungsgemeinschaft) BA 1011/32-1. INGEMMET (Instituto Geo-
lógico Minero y Metallurgico Geological de Perú) provided help in
organizing the field trip and financially supported CB. Thanks to M.
Frechen (LIAG Hannover) for helpful discussions on OSL ages and
commenting on the manuscript. S. Riemenschneider (LIAG Hann-
over) supported us during sample preparation. We acknowledge B.
Jaffe (USGS) who supplied us with the latest version of the
12 M. Spiske et al. / Quaternary International xxx (2013) 1e14
TsuSedMod inversion model and R. Weiss (Virginia Tech) for his
support during modeling. We also thank C. Augustsson (Jena)
for her help with CL analysis and for critically reading the manu-
script. We appreciate the support from N. Catto, L. Borrero and the
anonymous referees.
Appendix A. Supplementary material
Supplementary material related to this article can be found at
http://dx.doi.org/10.1016/j.quaint.2013.02.010.
References
Adamiec, G., Aitken, M., 1998. Dose-rate conversion factors: update. Ancient TL 16,
37e50.
Aitken, M.J., 1990. Science-based Dating in Archaeology. Longman Archaeology
Series, London, 274 pp.
Aitken, M.J., 1998. An Introduction to Optical Dating e The Dating of Quaternary
Sediments by the Use of Photon-stimulated Luminescence. Oxford University
Press, Oxford, 280 pp.
Alappat, L., Frechen, M., Ramesh, R., Tsukamoto, S., Srinivasalu, S., 2011. Evolution of
late Holocene coastal dunes in the Cauvery delta region of Tamil Nadu, India.
Journal of Asian Earth Sciences 42, 381e397.
Apotsos, A., Gelfenbaum, G., Jaffe, B., 2012. Time-dependent onshore tsunami
response. Coastal Engineering 64, 73e86.
Astroza, M.I., 2007. Estudio de la zona afectada por el terremoto de Pisco, Agosto 15
del 2007-Intensidades y daños del terremoto. Centro Regional de Sismologia
para America del Sur, Lima, Peru. http://www.ceresis.org/informacion/Informe_
M_Astroza.pdf (last accessed Jan. 2013).
Auclair, M., Lamothe, M., Huot, S., 2003. Measurement of anomalous fading for
feldspar IRSL using SAR. Radiation Measurements 37, 487e492.
Augustsson, C., Rüsing, T., Adams, C.J., Chmiel, H., Kocabayoglu, M., Buld, M.,
Zimmermann, U., Berndt, J., Kooijman, E., 2011. Detrital quartz and zircon
combined: the production of mature sand with short transportation paths
along the Cambrian West Gondwana margin, Northwestern Argentina. Journal
of Sedimentary Research 81, 284e298.
Augustsson, C., Reker, A., 2012. Cathodoluminescence spectra of quartz as prove-
nance indicators revisited. Journal of Sedimentary Research 82 (8), 559e570.
Balescu, S., Lamothe, M., 1994. Comparison of TL and IRSL age estimates of feldspar
coarse grains from waterlain sediments. Quaternary Geochronology 13, 437e444.
Banerjee, D., Murray, A.S., Foster, I.D.L., 2001. Scilly Isles, UK: optical dating of a
possible tsunami deposit from the 1755 Lisbon earthquake. Quaternary Science
Reviews 20, 715e718.
Bateman, M.D., 2008. Luminescence dating of periglacial sediments and structures.
Boreas 37, 574e588.
Berninghausen, W.H., 1962. Tsunamis reported from the west coast of South
America from 1562e1960. Bulletin of the Seismological Society of America 52
(4), 915e921.
Bilek, S.L., 2010. Seismicity along the South American subduction zone: review of
large earthquakes, tsunamis, and subduction zone complexity. Tectonophysics
495, 2e14.
Bishop, P., Sanderson, D., Hansom, J., Chaimanee, N., 2005. Age-dating of tsunami
deposits: lessons from the 26 December 2004 tsunami in Thailand.
Geographical Journal 171, 379e384.
Bourgeois, J., Petroff, C., Yeh, H., Titov, V., Synolakis, C.E., Benson, B., Kuroiwa, J.,
Lander, J., Norabuena, E., 1999. Geologic setting, field survey and modeling of
the Chimbote, Northern Peru, tsunami of 21 February 1996. Pure and Applied
Geophysics 154, 513e540.
Brill, D., Klasen, N., Brückner, H., Jankaew, K., Scheffers, A., Kelletat, D., Scheffers, S.,
2012. OSL dating of tsunami deposits from Phra Thong Island, Thailand. Qua-
ternary Geochronology 10, 224e229.
Buylaert, J.P., Murray, A.S., Thomsen, K.J., Jain, M., 2009. Testing the potential of an
elevated temperature IRSL signal from K-feldspar. Radiation Measurements 44,
560e565.
Cahill, T., Isacks, B.L., 1992. Seismicity and shape of the subducted Nazca Plate.
Journal of Geophysical Research 97, 17503e17529.
Carpio, J., Tavera, H., 2002. Estructura de un catalogo de tsunamis para el Peru
basado en el catalogo de Gusiakov. Boletin de la Sociedad Geologica del Peru 94,
45e59.
Cisternas, M., Atwater, B.F., Torrejon, F., Sawai, Y., Machuca, G., Lagos, M., Eipert, A.,
Youlton, C., Salgado, I., Kamataki, T., Shishikura, M., Rajendran, C.P., Malik, J.K.,
Rizal, Y., Husni, M., 2005. Predecessors of the giant 1960 Chile earthquake.
Nature 437, 404e407.
Cunha, P.P., Buylaert, J.P., Murray, A.S., Andrade, C., Freitas, M.C., Fatela, F.,
Munhá, J.M., Martins, A.A., Sugisaki, S., 2009. Optical dating of clastic deposits
generated by an extreme marine coastal flood: the 1755 tsunami deposits in the
Algarve (Portugal). Quaternary Geochronology 5 (2e3), 329e335.
Davids, F., Duller, G.A.T., Roberts, H.M., 2010. Testing the use of feldspars for optical dating
of hurricane overwash deposits. Quaternary Geochronology 5 (2e3), 125e130.
Dawson, A.G., Shi, S., 2000. Tsunami deposits. Pure and Applied Geophysics 157,
875e897.
Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
Dengler, L., Borrero, J., Gelfenbaum, G., Jaffe, B., Okal, E., Ortiz, M., Titov, V., 2003.
Tsunami: Ch. 7 in Southern Peru earthquake of 23 June 2001 reconnaissance
report. Earthquake Spectra Supplement 19, 115e144.
Ditlefsen, C., 1992. Bleaching of K-feldspars in turbid water suspensions: a com-
parison of photo- and thermoluminescence signals. Quaternary Science Re-
views 11, 33e38.
Dorbath, L., Cisternas, A., Dorbath, C., 1990. Assessment of the size of large and great
historical earthquakes of Peru. Bulletin of the Seismological Society of America
80 (3), 551e576.
Duller, G.A.T., Wintle, A.G., Hall, A.M., 1995. Luminescence dating and its application
to key pre-Late Devensian sites in Scotland. Quaternary Science Reviews 14 (5),
495e519.
EERI, 2001. Preliminary Observations on the Southern Peru Earthquake of June 23, 2001.
Earthquake Engineering Research Institute. Special Earthquake Report. www.
cismid.uni.edu.pe/descargas/redacis/redacis53_a.pdf (last accessed Jan. 2013).
Fontugne, M., Usselmann, P., Lavallee, D.l., Julien, M.l., Hatte, C., 1999. El Niño
variability in the coastal desert of southern Peru during the Mid-Holocene.
Quaternary Research 52 (2), 171e179.
Fritz, H.M., Kalligeris, N., Borrero, J.C., Broncano, P., Ortega, E., 2008. The 15 August
2007 Peru tsunami runup observations and modeling. Geophysical Research
Letters 35, L10604.
Frechen, M., Schweitzer, U., Zander, A., 1996. Improvements in sample preparation
for the fine grain technique. Ancient TL 14, 15e17.
Fuchs, M., Owen, L.A., 2008. Luminescence dating of glacial and associated sedi-
ments: review, recommendations and future directions. Boreas 37, 636e659.
Galbraith, R.F., Roberts, R.G., Laslett, G.M., Yoshida, H., Olley, J.M., 1999. Optical
dating of single and multiple grains of quartz from Jinmium Rock Shelter,
northern Australia e Part 1, experimental design and statistical models.
Archaeometry 41, 339e364.
Gemmell, A.M.D., 1997. Fluctuations in the thermoluminescence signal of sus-
pended sediment in an alpine glacial meltwater stream. Quaternary Science
Reviews 16 (3e5), 281e290.
Godfrey-Smith, D.I., Huntley, D.J., Chen, W.H., 1988. Optical dating studies of
quartz and feldspar sediment extracts. Quaternary Science Reviews 7 (3e4),
373e380.
Götze, J., Plötze, M., Habermann, D., 2001. Origin, spectral characteristics and
practical applications of the cathodoluminescence (CL) of quartz-a review.
Mineralogy and Petrology 71, 225e250.
Goff, J., McFadgen, B.G., Chague-Goff, C., 2004. Sedimentary differences between the
2002 Easter storm and the 15th-century Okoropunga tsunami, southeastern
North Island, New Zealand. Marine Geology 204, 235e250.
Gutscher, M.A., 2002. Andean subduction styles and their effect on thermal structure
and interplate coupling. Journal of South American Earth Sciences 15, 3e10.
Harms, J.C., Southard, J.B., Spearing, D.R., Walker, R.G., 1975. Depositional Environ-
ments as Interpreted from Primary Sedimentary Structures and Stratification
Sequences. Society of Economic Paleontologists and Mineralogists, Short Course
no. 2, 161 pp.
Hayward, B.W., Brook, F.J., 1994. Foraminiferal paleoecology and initial subsidence
of the early Miocene Waitemata basin, Waiheke Island, Auckland. New Zealand.
Journal of Geology and Geophysics 37, 11e24.
He, Z., Zhou, J., Lai, Z., Yang, L., Liang, J., Long, H., Ou, X., 2010. Quartz OSL dating of
sand dunes of Late Pleistocene in the Mu Us Desert in northern China. Qua-
ternary Geochronology 5 (2e3), 102e106.
Heinz, P., Ruschmeier, W., Hemleben, C., 2008. Live benthic foraminiferal assem-
blages at the Pacific continental margin of Costa Rica and Nicaragua. Journal of
Foraminiferal Research 38, 215e227.
Huntley, D.J., Baril, M.R., 1997. The K content of the K-feldspars being measured in
optical dating or in thermoluminescence dating. Ancient TL 15, 11e13.
Huntely, D.J., Clague, J.J., 1996. Optical dating of tsunami-laid sands. Quaternary
Research 46, 127e140.
Huntley, D.J., Lamothe, M., 2001. Ubiquity of anomalous fading in K-feldspars and
the measurement and correction for it in optical dating. Canadian Journal of
Earth Sciences 38 (7), 1093e1106.
Iida, K., Cox, D.C., Pararas-Carayannis, G., 1967. Preliminary Catalog of Tsunamis
Occurring in the Pacific Ocean. Data Report 5. University of Hawaii, Honolulu.
http://www.soest.hawaii.edu/Library/Tsunami%20Reports/Iida_et_al.pdf, 274
pp. (last accessed Jan. 2013).
INGEMMET, 1995. Geológica del Perú. In: Instituto Geológico Minero y Metallurgico
Boletin Serie A, vol. 55. Carta Geológica Nacional, Lima, 177 pp.
Jaffe, B., Gelfenbaum, G., Rubin, D., Peters, R., Anima, R., Swensson, M., Olcese, D.,
Anticona, L.B., Gomez, J.C., Riega, P.C., 2003. Identification and interpretation of
tsunami deposits from the June 23, 2001 Peru tsunami. http://walrus.wr.usgs.
gov/tsunami/srilanka05/CS03_PeruTsuDep.pdf (last accessed Jan. 2013).
Jaffe, B.E., Gelfenbaum, G., 2007. A simple model for calculating tsunami flow speed
from tsunami deposits. Sedimentary Geology 200 (3e4), 347e361.
Johansson, C.E., 1976. Structural studies of frictional sediments. Geografiska Annaler
Serie A: Physical Geography 58, 201e301.
JSCE, 2001. Provisional Report on the June 23, 2001 Atico Earthquake. Japan Society
of Civil Engineers, Peru. http://www.jsce.or.jp/report/14/01/provisional_n.pdf
(last accessed Jan. 2013).
Keefer, D.K., deFrance, S.D., Moseley, M.E., Richardson, J.B., Satterlee, D.R., Day-
Lewis, A., 1998. Early maritime economy and El Niño events at Quebrada
Tacahuay, Peru. Science 281 (5384), 1833e1835.
Keefer, D.K., Moseley, M.E., 2004. Southern Peru desert shattered by the great
2001 earthquake: implications for paleoseismic and paleo-El NinÌf o-Southern
 M. Spiske et al. / Quaternary International xxx (2013) 1e14
Oscillation records. Proceedings of the National Academy of Sciences of the
United States of America 101 (30), 10878e10883.
Kulikov, E.A., Rabinovich, A.B., Thomson, R.E., 2005. Estimation of tsunami risk for
the coasts of Peru and Northern Chile. Natural Hazards 35 (2), 185e209.
Kunz, A., Frechen, M., Ramesh, R., Urban, B., 2010a. Revealing the coastal event-
history of the Andaman Islands (Bay of Bengal) during the Holocene using
radiocarbon and OSL dating. International Journal of Earth Sciences 99 (8),
1741e1761.
Kunz, A., Frechen, M., Ramesh, R., Urban, B., 2010b. Periods of recent dune sand
mobilisation in Cuddalore, Southeast India. Zeitschrift der Deutschen Gesell-
schaft für Geowissenschaften 161, 353e368.
Lamothe, M., Auclair, M., Hamzaoui, C., Huot, S., 2003. Towards a prediction of longterm
anomalous fading of feldspar IRSL. Radiation Measurements 37, 493e498.
Lancaster, N., 2008. Desert dune dynamics and development: insights from lumi-
nescence dating. Boreas 37, 559e573.
Li, B., Li, S.H., Wintle, A.G., Zhao, H., 2007. Isochron measurements of naturally
irradiated K-feldspar grains. Radiation Measurements 42 (8), 1315e1327.
Lian, O.B., Roberts, R.G., 2006. Dating the Quaternary: progress in luminescence
dating of sediments. Quaternary Science Reviews 25, 2449e2468.
Lockridge, P.A., 1985. Tsunamis in PerueChile. World Data Center A for Solid Earth
Geophysics. http://www.ngdc.noaa.gov/hazard/data/publications/Wdcse-39.
pdf. Report SE-39 (last accessed Jan. 2013).
Lomnitz, C., 1970. Major earthquakes and tsunamis in Chile during the period 1535
to 1955. International Journal of Earth Sciences 59 (3), 938e960.
Machare, J., Ortlieb, L., 1992. Plioquaternary vertical motions and the subduction of
the Nazca Ridge, central coast of Peru. Tectonophysics 205, 97e108.
Madsen, A.T., Duller, G.A.T., Donnelly, J.P., Roberts, H.M., Wintle, A.G., 2009.
A chronology of hurricane landfalls at Little Sippewissett Marsh, Massachusetts,
USA, using optical dating. Geomorphology 109 (1e2), 36e45.
Magilligan, F.J., Goldstein, P.S., 2001. El Niño floods and culture change: a late
Holocene flood history for the Rio Moquegua, southern Peru. Geology 29 (5),
431e434.
Malumián, N., Scarpa, R., 2005. Foraminíferos de la Formación Irigoyen, Neogeno,
Tierra del Fuego, Argentina: su significado paleobiogeográfico. Ameghiniana 42
(2), 363e376.
Mamani, M., Wörner, G., Sempere, T., 2010. Geochemical variations in igneous rocks of
the Andes orocline (13 to 18 ): tracing crustal thickening and magma generation
through time and space. Geological Society of America Bulletin 122, 162e182.
Mejdahl, V., 1979. Thermoluminescence dating: beta-dose attenuation in quartz
grains. Archaeometry 21, 61e72.
Murari, M.K., Achyuthan, H., Singhvi, A.K., 2007. Luminescence studies on the
sediments laid down by the December 2004 tsunami event: prospects for the
dating of palaeo tsunamis and for the estimation of sediment fluxes. Current
Science 92 (3), 367e371.
Murray, A.S., Olley, J.M., 2002. Precision and accuracy in the optically stimulated
luminescence dating of sedimentary quartz: a status review. Geochronometria
21, 1e16.
Nanayama, F., Shigeno, K., Satake, K., Shimokawa, K., Koitabashi, S., Miyasaka, S.,
Ishii, M., 2000. Sedimentary differences between the 1993 Hokkaido-nansei-oki
tsunami and the 1959 Miyakojima typhoon at Taisei, southwestern Hokkaido,
northern Japan. Sedimentary Geology 135, 255e264.
NCDC, 2011. Climate Data Archive. National Climatic Data Center. NOAA Satellite
and Information Service. http://www.ncdc.noaa.gov/oa/ncdc.html (last
accessed Jan. 2013).
Noble, D., Rios, A., Vidal, C., Spell, T., ZanettiI, K., Angeles, C., Ochoa, J.Y., Cruz, S.,
2005. Late Cretaceous basalt in the Rio Mala valley, central Peru: evidence for
extension and mafic magmatism prior to Latest CretaceousePaleocene plu-
tonism and silicic volcanism. Sociedad Geológica del Perú 6, 141e148 (special
volume).
Okal, E., Dengler, L., Araya, S., Borrero, J., Gomer, B., Koshimura, S., Laos, G., Olcese, D.,
Ortiz, M., Swensson, M., Titov, V., Vegas, F., 2002. Field survey of the Camana,
Peru tsunami of June 23, 2001. Seismological Research Letters 73 (6), 907e920.
Okal, E.A., Borrero, J.C., Synolakis, C.E., 2006. Evaluation of tsunami risk from
regional earthquakes at Pisco, Peru. Bulletin of the Seismological Society of
America 96 (5), 1634e1648.
Ollerhead, J., Huntley, D.J., Nelson, A.R., Kelsey, H.M., 2001. Optical dating of tsunami-
laid sand from an Oregon coastal lake. Quaternary Science Reviews 20,1915e1926.
Olley, J.M., Murray, A., Roberts, R.G., 1996. The effects of disequilibria in the uranium
and thorium decay chains on burial dose rates in fluvial sediments. Quaternary
Science Reviews 15, 751e760.
Ortlieb, L., Vargas, G., 2003. Debris flow deposits and El Niño impacts along the
hyperarid southern Peru coast. In: Haas, J., Dillon, M. (Eds.), El Niño in Peru:
Biology and Culture over 10,000 Years. Fieldiana Botany, vol. 43, pp. 24e51.
Páez, M., Zúñiga, O., 2001. Foraminíferos recientes en el área del río Loa (21250 S;
70 040 W), Chile. Estudios Oceanológicos 20, 3e12.
Petroff, C., Bourgeois, J., Yeh, H., May 1996. The February 21, 1996 Chimbote Tsu-
namis in Peru. EERI Special Earthquake Report. http://www.eeri.org/lfe/pdf/
peru_chimbote_eeri_preliminary_report.pdf (last accessed Jan. 2013).
Pietsch, T.J., Olley, J.M., Nanson, G.C., 2008. Fluvial transport as a natural lumines-
cence sensitiser of quartz. Quaternary Geochronology 3, 365e376.
Pitcher, W., Atherton, M., Cobbing, E., Beckinsale, R., 1985. Magmatism at a Plate
Edge e the Peruvian Andes. John Wiley and Sons, New York, 328 pp.
Prendergast, A.L., Cupper, M.L., Jankaew, K., Sawai, Y., 2012. Indian Ocean tsunami
recurrence from optical dating of tsunami sand sheets in Thailand. Marine
Geology 295e298, 20e27.
Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010
13
Preusser, F., Ramseyer, K., Schlüchter, C., 2006. Characterisation of low OSL intensity
quartz from the New Zealand Alps. Radiation Measurements 41, 871e877.
Rees-Jones, J., Tite, M.S., 1997. Optical dating results for british archaeological sed-
iments. Archaeometry 39, 177e187.
Rendell, H.M., Webster, S.E., Sheffer, N.L., 1994. Underwater bleaching of signals
from sediment grains: new experimental data. Quaternary Science Reviews 13
(5e7), 433e435.
Rhodes, E.J., Pownall, L., 1994. Zeroing of the OSL signal in quartz from young gla-
ciofluvial sediments. Radiation Measurements 23 (2e3), 581e585.
Rhodes, E.J., Bailey, R.M., 1997. The effect of thermal transfer on the zeroing of the
luminescence of quartz from recent glaciofluvial sediments. Quaternary Science
Reviews 16, 291e298.
Rittenour, T.M., 2008. Luminescence dating of fluvial deposits: applications
to geomorphic, palaeoseismic and archaeological research. Boreas 37, 613e
635.
Robinson, R.A.J., Spencer, J.Q.G., Strecker, M.R., Richter, A., Alonso, R.N., 2005.
Luminescence dating of alluvial fans in intramontane basins of NW Argentina.
In: Dynamics. Special Publications, vol. 251. Geological Society, London,
pp. 152e168.
Sanchez, J.C., 2007. Daños por el terremoto del 15 de Agosto del 2007. Instituto de
Ingenieria. Universidad Nacional Autonoma de Mexico. http://www.igp.gob.pe/
sismologia/libro/trabajo_12.pdf (last accessed Jan. 2013).
Sawakuchi, A.O., Guedes, C.C.F., DeWitt, R., Giannini, P.C.F., Blair, M.W.,
Nascimento Jr., D.R., Faleiros, F.M., 2012. Quartz OSL sensitivity as a proxy for
storm activity on the southern Brazilian coast during the Late Holocene.
Quaternary Geochronology 13, 92e102.
SENAMHI, 2012. Informacion del tiempo, clima y agua. Servicio Nacional de
Meteorologia e Hidrologia del Peru. http://www.senamhi.gob.pe/ (last accessed
Jan. 2013).
Shackleton, R.M., Ries, A.C., Coward, M.P., Cobbold, P.R., 1979. Structure, meta-
morphism and geochronology of the Arequipa Massif of coastal Peru. Journal of
the Geological Society 136, 195e214.
Soler, P., 1991. Contribution a l’étude du magamatisme associe aux marges actives
pétrographie, géoquimmie et geochimie isotopique du magmatisme crétacé a
Pliocène le long d’une transversale des andes du Pérou central implications
géodynamiques et metallogeneques. Unpublished PhD thesis, Université Pierre
et Marie Curie, Paris, 832 pp.
Sommerville, A.A., Hansom, J.D., Sanderson, D.C.W., Housley, R.A., 2003. Optically
stimulated luminescence dating of large storm events in Northern Scotland.
Quaternary Science Reviews 22, 1085e1092.
Spencer, J.Q.G., Robinson, R.A.J., 2008. Dating intramontane alluvial deposits from
NW Argentina using luminescence techniques: problems and potential. Geo-
morphology 93, 144e155.
Spiske, M., Weiss, R., Bahlburg, H., Roskosch, J., Amijaya, H., 2010. The TsuSedMod
inversion model applied to the deposits of the 2004 Sumatra and 2006 Java
tsunami and implications for estimating flow parameters of palaeo-tsunami.
Sedimentary Geology 224, 29e37.
Spiske, M., Piepenbreier, J., Bahlburg, H., Benavente, C., 2010. A note on the
onshore preservation potential of siliciclastic tsunami depositseexamples
from coastal Peru. European Geosciences Union General Assembly, Wien,
Austria, 02e07 May 2010. http://meetingorganizer.copernicus.org/EGU2010/
EGU2010-8927-1.pdf.
Steffen, D., Schlunegger, F., Preusser, F., 2009a. Drainage basin response to climate
change in the Pisco valley, Peru. Geology 37 (6), 491e494.
Steffen, D., Preusser, F., Schlunegger, F., 2009b. OSL quartz age underestimation due
to unstable signal components. Quaternary Geochronology 4 (5), 353e362.
Steffen, D., Schlunegger, F., Preusser, F., 2010. Late Pleistocene fans and terraces in
the Majes valley, southern Peru, and their relation to climatic variations. In-
ternational Journal of Earth Sciences 99 (8), 1975e1989.
Stuiver, M., 1978. Radiocarbon timescale tested against magnetic and other dating
methods. Nature 273, 271e274.
Thomsen, K.J., Murray, A.S., Jain, M., Bøtter-Jensen, L., 2008. Laboratory fading rates
of various luminescence signals from feldspar-rich sediment extracts. Radiation
Measurements 43, 1474e1486.
Thornburg, T.M., Kulm, L.D., 1987. Sedimentation in the Chile Trench: depositional
morphologies, lithofacies, and stratigraphy. Geological Society of America
Bulletin 98, 33e52.
Thrasher, I.M., Mauz, B., Chiverrell, R.C., Lang, A., 2009. Luminescence dating of
glaciofluvial deposits: a review. Earth-Science Reviews 97, 133e146.
Urban, B., Kunz, A., Gehrt, E., 2011. Genesis and dating of Late Pleistocene-Holocene
soil sediment sequences from the Lüneburg Heath, Northern Germany. Qua-
ternary Science Journal (Eiszeitalter und Gegenwart) 60, 164e184.
USGS, 2001. Preliminary Analysis of Sedimentary Deposits from the June 23, 2001
Peru Tsunami. United States Geological Survey. http://walrus.wr.usgs.gov/
peru2/ (last accessed Jan. 2013).
van Heteren, S., Bakker, M.A.J., Oost, A.P., van der Spek, A.J.F., van der Valk, L., 2008.
Sedimentary signature of a storm-surge unit in the western Netherlands coastal
dunes. Geophysical Research Abstracts 10. EGU2008-A-10264.
Vargas, G., Ortlieb, L., Rutllant, J., 2000. Aluviones historicos en Antofagasta y su
relacion con eventos El Niño/Oscilacion del Sur. Revista Geologica de Chile 27,
157e176.
Vargas, G., Rutllant, J., Ortlieb, L., 2006. ENSO tropical-extratropical climate tele-
connections and mechanisms for Holocene debris flows along the hyperarid
coast of western South America (17 e24 S). Earth and Planetary Science Letters
249, 467e483.
14 M. Spiske et al. / Quaternary International xxx (2013) 1e14

von Huene, R., Scholl, D.W., 1991. Observations at convergent margins concerning
sediment subduction, subduction erosion, and growth of continental crust.
Reviews of Geophysics 29, 279e316.
Wagner, G.A., 1998. Age Determination of Young Rocks and Artifacts e Physical
and Chemical Clocks in Quaternary Geology and Archaeology. Springer, Berlin,
466 pp.
Wallinga, J., Murray, A.S., Duller, G.A.T., Tornqvist, T.E., 2001. Testing optically
stimulated luminescence dating of sand-sized quartz and feldspar from fluvial
deposits. Earth and Planetary Science Letters 193, 617e630.
Wells, L.E., 1987. An alluvial record of El Niño events from northern coastal Peru.
Journal of Geophysical Research 92 (C13), 14463e14470.
Wells, L.E., 1996. The Santa beach ridge complex: sea-level and progradational history
of an open gravel coast in Central Peru. Journal of Coastal Research 12, 1e17.
Wells, L.E., Noller, J.S., 1999. Holocene coevolution of the physical landscape and
human settlement in northern coastal Peru. Geoarchaeology 14 (8), 755e789.
Wintle, A.G., 1973. Anomalous fading of thermoluminescence in mineral samples.
Nature 245, 143e144.
Wintle, A.G., Murray, A.S., 2006. A review of quartz optically stimulated lumines-
cence characteristics and their relevance in single-aliquot regeneration dating
protocols. Radiation Measurements 41, 369e391.
Witter, R.C., Kelsey, H.M., Hemphill-Haley, E., 2001. Pacific storms, El Niño and
tsunamis: competing mechanisms for sand deposition in a coastal marsh,
Euchre Creek, Oregon. Journal of Coastal Research 17 (3), 563e583.
Witter, R., Jaffe, B., Zhang, Y., Priest, G., 2012. Reconstructing hydrodynamic flow
parameters of the 1700 tsunami at Cannon Beach, Oregon, USA. Natural Hazards
63, 223e240.
Wood, P.B., 1994. Optically stimulated luminescence dating of a late quaternary
shoreline deposit, Tunisia. Quaternary Science Reviews 13, 513e516.
Yauri, S., Tavera, H., Moncca, G., Herrera, B., 2008. Características Generales del
Tsunami del 15 de Agosto de 2007 in: El Terremoto de Pisco del 15 de Agosto de
2007. Dirección de Sismología - CNDG/Instituto Geofísico del Perú. Volumen
Especial IGP (Versión Preliminar). http://www.igp.gob.pe/sismologia/libro/
trabajo_21.pdf, 17 pp. (last accessed Jan. 2013).
Zapata, M.J., Gutiérrez, M.A., 1995. Foraminíferos litorales recientes de Tocopilla
(22 060 S, 70 130 W), Chile. Estudios Oceanológicos 14, 49e59.
Zheng, C.X., Zhou, L.P., Qin, J.T., 2009. Difference in luminescence sensitivity of
coarse-grained quartz from deserts of northern China. Radiation Measurements
44, 534e537.
Zinkernagel, U., 1978. Cathodoluminescence of quartz and its application to sand-
stone petrology. Contributions to Sedimentology 8, 1e69.

Please cite this article in press as: Spiske, M., et al., Historical tsunami deposits in Peru: Sedimentology, inverse modeling and optically
stimulated luminescence dating, Quaternary International (2013), http://dx.doi.org/10.1016/j.quaint.2013.02.010