www.elsevier.com/locate/tecto

Tectonophysics 399 (

Paleoseismic inferences from a high-resolution marine sedimentary

record in northern Chile (238S)

Gabriel Vargasa,b,c,T, Luc Ortliebc, Emmanuel Chaprond,1,

Jorge Valdesc,e, Carlos Marquardtf

aDepartamento de Geologıa, Universidad de Chile, Plaza Ercilla 803, Santiago, ChilebUMR-CNRS 5805 EPOC Universite Bordeaux I Avenue des Facultes, 33405, Talence Cedex, France

cIRD, UR055-PALEOTROPIQUE, 32 Avenue Henri Varagnat, 93143, Bondy Cedex, FrancedI.S.T.O. Universite d’Orleans, UMR CNRS 6113, BP 6759, 45067 Orleans Cedex, France

eFacultad de Recursos del Mar, Universidad de Antofagasta, Casilla 170, Antofagasta, ChilefSERNAGEOMIN, Avenida Santa Marıa 0104, Providencia, Santiago, Chile

Received 15 November 2002; received in revised form 10 November 2003; accepted 23 December 2004

Available online 12 February 2005

Abstract

The active margin of northern Chile is characterized by strong seismic events which induce gravity instability for

sedimentary sequences located along the outer forearc of the Andean range. Even though the narrow continental platform of this

area limits the accumulation of marine sediments, a detailed paleoceanographic reconstruction, using high-resolution

sedimentological and geochronological techniques, of a Holocene sedimentary sequence in Mejillones bay (238S) providesimportant tools to infer the occurrence of past great subduction earthquakes and/or associated tsunamis.

From the analyses of short cores retrieved in this shallow marine basin, we infer the occurrence of two strong seismic events

during historical times. The first one, dated between the years 1409 and 1449 AD, produced an angular unconformity and

associated lenticular coarse-grained deposits generated after local reworking of shallower material. The second one, dated

between the years 1754 and 1789 AD, caused slumping, erosive processes and local reworking of material. The analysis of

seismic profile data acquired in this bay allowed the recognition of a major slump deposit within the upper sedimentary

sequence, tentatively dated as of Middle Holocene age. This event might record the greatest earthquake (and tsunami) affecting

the basin within the last few thousand years.

We interpret that destabilization of sediments occurs most probably in response to local reactivation of faults within the

Mejillones peninsula during strong earthquakes, especially the Mejillones fault.

D 2005 Elsevier B.V. All rights reserved.

Keywords: Paleoseismicity; Paleoceanography; Laminated sediments; Northern Chile; Seismic gap; Mejillones bay

T Corresponding author. Departamento de Geologıa, Universidad de Chile, Plaza Ercilla 803, Santiago, Chile.

0040-1951/$ - s

doi:10.1016/j.tec

E-mail addre

jvaldes@uantof.1 Present addr

2005) 381–398

ee front matter D 2005 Elsevier B.V. All rights reserved.

to.2004.12.031

sses: gvargas@ing.uchile.cl (G. Vargas)8 Luc.Ortlieb@bondy.ird.fr (L. Ortlieb)8 chapron@erdw.ethz.ch (E. Chapron)8

cl (J. Valdes)8 cmarquar@sernageomin.cl (C. Marquardt).

ess: Geological Institute, ETH Zentrum, Zqrich, Switzerland.

G. Vargas et al. / Tectonophysics 399 (2005) 381–398382

1. Introduction

1.1. Regional geodynamic context in the outer forearc

of northern Chile: the Mejillones peninsula

The continental margin of northern Chile is

characterized by active tectonics and great historical

seismic events related to the subduction of the Nazca

plate under the South American plate. This tectonic

contact controls the geometry of the margin and the

disposition of the main geological features. One major

structure in this region is the N–S Atacama fault

system (Fig. 1), which was originated during the

Mesozoic period, but exhibits superficial manifesta-

tions of recent reactivations (Armijo and Thiele, 1990;

Niemeyer et al., 1996; Delouis et al., 1997; 1998;

Gonzalez et al., 2003). The Mejillones peninsula

constitutes an anomalous geomorphological feature

Fig. 1. Right: Regional geodynamical context of the forearc of northern C

coastal cordillera, indicating faults showing evidence of recent activity (Ar

b: central depression, c: pre-cordilleran basins and d: main cordillera showi

with a mean velocity around 8 cm y�1 in a N778E direction (Somoza, 199

of northern Chile, indicating the inferred rupture lengths (modified from

Pritchard et al., 2002).

which interrupts the N–S trending coastline of north-

ern Chile (Fig. 2). Uplifted Plio-Quaternary marine

terraces and coastal deposits affected by extensional

fault scarps disposed in N–S, SW–NE and SE–NW

directions, characterize this peninsula (Fig. 3). Both

vertical uplift and brittle deformation result from

strong tectonic activity in a dominant extensional

stress regime during the Late Cenozoic (Armijo and

Thiele, 1990; Ortlieb et al., 1996b; 1996c; Delouis et

al., 1998; Gonzalez et al., 2003).

Major seismic activity in this region has been

associated with the seismogenic interplate contact

zone between both tectonic plates (Comte and Pardo,

1991; Comte et al., 1992; Comte et al., 1994). The last

great earthquake, in July 30, 1995 (Mw=8.1, hypo-

center at 23.438S and 70.488W, 36 km depth; Monfret

et al., 1995), occurred at the southern limit of the area

considered as a seismic gap region, which encom-

hile showing the main morphostructural units. a: Coastal fringe and

mijo and Thiele, 1990; Delouis et al., 1998; Marquardt et al., 2002),

ng Quaternary volcanoes. The Nazca plate moves towards the trench

8). Left: Historical seismic events of magnitude MN7 along the coast

Comte and Pardo, 1991; Delouis et al., 1997; Comte et al., 2002;

Fig. 2. 3D view of the regional bathymetry in front of the Mejillones peninsula, based on a Numerical Model of Terrain (NMT) using Kriging-

method of interpolation. Original data of bathymetry from Zapata (2001). Subdivision of the continental margin in Upper Slope (US: 1000–

3000 m), Middle Slope (MS: 3000–6000 m) and Lower Slope (LS:N6000 m) is indicated. The US is composed by coherent crust affected by

normal faults. The MS is characterized by frequent terraces containing blocks that disintegrate, forming sediments and debris that constitutes the

LS (von Huene et al., 1999).

G. Vargas et al. / Tectonophysics 399 (2005) 381–398 383

passes the southern Peru and northern Chile (Comte

and Pardo, 1991). Important coseismic vertical and

horizontal deformation (coastal uplift up to 40 cm in

the southern Mejillones peninsula and 70 cm of

continental horizontal displacement toward the trench;

Ortlieb et al., 1996a; Ruegg et al., 1996), as well as

localized superficial manifestations along some por-

tions of the Atacama fault system (Delouis et al.,

1998; Gonzalez et al., 2003), illustrates how active

processes control the present day geomorphological

configuration of this region. Block-tectonic mecha-

nisms explain localized impact of co-seismic defor-

mation during strong earthquakes, as the limited

impact of the 1995 seismic event in the northern part

of the Mejillones peninsula, at a short distance north

of the epicenter (Monfret et al., 1995; Delouis et al.,

1997; 1998). Considering that the previous large

earthquake that struck the coast of northern Chile

occurred in 1877 (Comte and Pardo, 1991), the 1995

earthquake may be viewed as a precursor of the end of

the seismic quiescence in the southern portion of the

seismic gap region (Delouis et al., 1997; Fig. 1).

This important seismic activity is a source of

instability for sedimentary systems located in the

narrow continental platform of northern Chile (Fig. 2).

The interpretation of processes yielding structural

anomalous features affecting recent sedimentary

sequences in this area can provide relevant informa-

tion related to the occurrence of strong earthquakes in

the past. The hearth of the Atacama Desert, between

198S and 248S hosted very scarce inhabitants and no

written reports are available prior to the 18th century.

Ongoing high-resolution paleoceanographic recon-

struction from the sedimentary sequence in Mejillones

bay (238S) offers a potential for the determination of

great seismic events through the recognition and

dating of anomalous sedimentological and structural

features, as discontinuities and slump deposits.

1.2. The high-resolution sedimentary sequence in

Mejillones bay

While the continental margin of northern Chile is

characterized by a steep slope, a deep trench which

Fig. 3. Bathymetric chart of the Mejillones Bay with S–N and W–E bottom profiles and location of the sedimentary cores mentioned in this

paper. Onshore geomorphological and structural indicated features include uplifted marine terraces, Quaternary shorelines assigned to the

Marine Isotopic Stages (MIS) 5, 7 and 9, as well as normal Quaternary faults (dipping block indicated; from Ortlieb et al., 1996a,b,c and

Marquardt et al., 2002).

G. Vargas et al. / Tectonophysics 399 (2005) 381–398384

reaches 8000 m and a narrow (near absent) continental

platform (in the classical sense, b200 m water depth),

two small platforms are preserved north and south of

the Mejillones peninsula (Fig. 2). The Mejillones bay

is located at the northern extremity of this peninsula

and exhibits a maximum diameter of 15 km, with 125

m of maximum depth. The western margin of the bay,

which exhibits an abrupt morphology, coincides with

the northward projection of the Mejillones fault

(Ferraris and Di Biase, 1978; Fig. 3). Gentle slopes

are observed on the eastern and southern margins,

where the basin is limited by a coastal cliff exposing

Neogene marine sequences (Ferraris and Di Biase,

1978) and uplifted Quaternary marine terraces

(Ortlieb et al., 1996c). In the center of the basin, at

depths greater than 50 m, the morphological slopes

dip less than 2%.

The sedimentation within the basin is strongly

influenced by an important upwelling cell at Punta

Angamos (Rodrıguez et al., 1991), which drives high

levels of primary productivity within the bay (Marın

et al., 1993; Marın and Olivares, 1999). As the bay is

protected from the direct influence of the dominant

SW winds, pelagic sedimentation of biogenic rest

(mainly diatom-skeletons) and organic matter is

favoured (Vargas, 1998; Valdes, 1998; Ortlieb et al.,

2000; Valdes et al., 2000; Vargas, 2002). The

preservation of these sediments is in part due to

persistent bottom hypoxia within the bay (Escribano,

1998). The accumulation of organic-rich material of

marine algal origin at a short distance from the shore

is only possible because of the lack of alluvial input in

the basin, which results from limited alluvial pro-

cesses in this region of the coastal Atacama Desert

(Vargas et al., 2000). Therefore, northward eolian

transport of particles from the lowlands of Mejillones

Peninsula is the main factor controlling the input of

lithic material to the bay (Vargas, 2002).

Through downcore 14C and 210Pb analyses, mass

accumulation rates about 0.034–0.032 g cm�2 y�1

(sedimentation rate of 160–130 cm ky�1) have been

calculated in gravity cores from the central part of the

G. Vargas et al. / Tectonophysics 399 (2005) 381–398 385

basin (Vargas, 2002), while lower (than 100 cm ky�1)

values were inferred (from 14C downcore data) in

cores from the margins (Vargas, 2002). Given the high

values of mass accumulation rates in the central area

of the bay, reconstructions of ocean-climate variability

from decadal to millennial time-scales based on the

analysis of sediment cores are possible (Ortlieb et al.,

2000; Valdes et al., 2000; Valdes and Ortlieb, 2001;

Vargas, 2002; Vargas et al., 2004). This type of studies

obviously requires that a continuous record be

preserved and that the studied sequence be not

interrupted by anomalous sedimentary features. Pre-

vious works reported at least one angular uncon-

formity in several cores retrieved from this shallow

marine basin and, in one case, evidence for a slump

deposit (Valdes, 1998; Ortlieb et al., 2000, Vargas,

2002; Vargas et al., 2002; Vargas et al., 2004).

Considering that these features could be linked to

the occurrence of major seismic events in the

southern part of the seismic gap region of northern

Chile, their identification and dating are of major

importance, both for paleoceanographic reconstruc-

tions and for paleoseismological studies. The paper

thus focuses on the relationship that can be

established between sedimentary unconformities

within the Holocene marine sequence of Mejillones

bay and the former seismic activity in northern

Chile.

2. Materials and methods

2.1. Coring

Short cores were retrieved using a 3-in. diameter

gravity core aboard the boat Puri and the R/V

Purihaalar of the University of Antofagasta (Fig. 3).

Cores 23 and 24 were retrieved in 1993, while cores

32B, 32C and 33C were obtained in 1996. Cores

F981A and 41C were obtained in 1998, using a gravity

box-core of 20 cm width and 50 cm length, and a 3-in.

diameter gravity core, respectively. All these cores

were preserved at 48C until the moment of analyses.

2.2. Sampling and sedimentology of short cores

X-ray images were taken using conventional

medical equipment with mean parameters of 50 kV,

0.16 s and 1 m distance. Contact prints (positive

phases) of radiographs were produced and treated

with specialised software to characterize grey-level

variations between adjacent lamina. In the case of core

F981A, the X-ray image was obtained from the

SCOPIX equipment at the University of Bordeaux 1

(Migeon et al., 1999).

Samples for sedimentological and mineralogical

analyses were taken every 1 cm in the case of core

33C, 0.5 cm for cores F981A and 41C. Sampling

according the limits of individual lamina, thus with

variable thickness, was done in the case of core 32B.

The water content (expressed as weight percentage)

and the bulk dry density were measured after drying

up the sediment at 40–50 8C using a F10�4 g

precision balance. In the case of core F981A, repeated

measurements provided a final accuracy of F0.01 g/

cc for dry density values. Samples of core 33C were

washed and sieved at 500 Am, 160 Am and 63 Am for

grain-size analysis. Grain size fractions are expressed

as weight percentage.

The identification and quantification (weight per-

centages) of lithic minerals and amorphous silica were

performed following the FTIR (Fourier Transformed

Infrared Spectrometry) method used at IRD laborato-

ries in Bondy. This technique described in Bertaux et

al. (1998) permits to quantify the percentage of

mineralogical species in ground sediments with a

precision of F1%.

Thin sections every 10 cm from cores 33C, F981A

and 41C, were cut from resin-impregnated samples

after a replacement of water by acetone. A detailed

petrographic study on the structure, texture, sedimen-

tology and mineralogy of these sediments was

performed using a polarised-light microscope with

magnifications of �20, �80 and �320. The high-

resolution variability (samples 1 mm thick) of

minerals in core 33C was quantified using a magni-

fication �80. The minor precision of this quantifica-

tion method is estimated as 2r of the total variance.

2.3. Chronological models for sedimentary cores

The chronological model for core F981A was

elaborated through a comparison between detailed14C and 210Pb downcore analyses (Fig. 4). The novel

methodology and the obtained results (Vargas, 2002)

will be published elsewhere. The constant mass

Fig. 4. Comparison between chronological models derived from 14C

and 210Pb analyses on samples of core F981A. The respective

values of the calculated constant mass accumulation rates are

indicated. The maximum difference between both chronological

models is 14.6 years, at 42 cm depth. Original data and method-

ology used can be found in Vargas (2002).

G. Vargas et al. / Tectonophysics 399 (2005) 381–398386

accumulation rates calculated from 14C and 210Pb

results are 34F5 g cm�2 ky�1 and 32F2 g cm�2

ky�1, respectively. A maximum difference of

14.6F0.5 years calculated between both chronolog-

ical models at the base of this core (42 cm length)

indicates that the chronological framework for this

sequence is set with a high degree of accuracy, and

can be relied upon.

A lateral correlation between core F981A and the

topmost part of cores 41C, 32B and 33C was

performed comparing the variation of the concen-

tration in lithic grains, the X-ray and photography

images. We used this correlation to calculate the

corresponding mass accumulation rates for each core.

After the corroboration for the absence of any

anomalous feature within the respective sedimentary

sequences, chronological models were derived for

the top sequence of each core, through the extrap-

olation of the corresponding mass accumulation

rates.

Taking into account the two independent methods

used for the determination of each chronology, Vargas

(2002) calculated a mean value for the Local

Reservoir Effect: DR=262F13 years, valid for radio-

carbon ages obtained from the organic fraction formed

during the last few centuries in the Mejillones bay

area. This correction is necessary to determine precise

chronological models from sediments which include

CO2 with depleted activity of 14C, such as those

produced in upwelling areas (e.g. Southon et al.,

1995). We used this value of DR to calibrate the

conventional radiocarbon ages from cores 32C and

24, which have been reported in Valdes (1998) and

Ortlieb et al. (2000), as well as for the radiocarbon

ages at the base of core 33C. For calibrations we used

Calib4.3 (Stuiver and Reimer, 1993).

2.4. Seismic reflection profiling

High-resolution seismic reflection profiles data

were obtained during November 2002 in Mejillones

bay using an ODEC Bathy 2000 acquisition device.

This equipment operates on a CHIRP mode at a main

frequency of 12 kHz, modulated over 4 kHz.

Navigation on the R/V Purihaalar, controlled by

GPS, resulted in a dense grid of profiles. Data

processing using Bathy 2000 W software provided a

vertical resolution of ~20 cm and an acoustic

penetration up to 45 m below the sea floor. Two-

way travel times were converted in depths during the

processing using an acoustic constant velocity of 1500

m/s. Based on the preliminary analysis and interpre-

tation of this data set, it was possible to map changes

in the nature of the acoustic substratum of the sea

floor in the bay, and to recognize changes in

sedimentary facies and structural features within the

overlying sediments.

3. Results

3.1. Mineralogy and high-resolution sedimentology of

short cores

The retrieved sediments correspond to a diatoma-

ceous organic-rich mud, characterized by a dominant

grain size fraction b63 Am (Fig. 5). Mean concen-

trations values of amorphous silica around 46%, 55%,

52% and 50% in the upper 40–50 cm of cores F981A,

41C, 32B and 33C, respectively, reflect the high

abundance of skeletons of siliceous phytoplankton.

The mean content of lithic grains (mainly quartz,

albite and oligoclase) diminishes with distance from

the southern shoreline, from 10% (core F981A) to 4%

(core 33C).

X-ray image and microscopic observations from

thin sections confirm the preservation of the

original structure. The thickness of the individual

laminae varies from a few millimetres to a

centimetre (or slightly more), denoting a relatively

Fig. 5. X-ray image and high-resolution variability of minerals grains content of core 33C. A: Variability (every centimetre) of grain size

fractions (in Am). B: Variability of felsic minerals grains (quartz, albite, oligoclase) per centimetre, as determined trough the FTIR method

(F1%). C: Variability of total minerals grains per centimetre, as estimated from microscope observations (F2r of total variance). Correlation

between B and C is r=0.5 at pb0.05, which increases the confidence of these estimations based on two totally different techniques. D: Laminae

limits with a millimetre precision. E: Variability of total minerals grains, as counted through microscope observations (F2r of total variance)

every millimetre, on a 7 mm2 surface. Grey thick lines figure two levels anomalously enriched in lithic grains; the older one indicated by a black

arrow, and the younger one by a white arrow.

G. Vargas et al. / Tectonophysics 399 (2005) 381–398 387

rhythmic sedimentation pattern. Microscope obser-

vations revealed that dark laminae are systemati-

cally enriched in lithic particles and coarser grain

size fractions (mean value around 80–100 Am) with

respect to the adjacent light laminae (mean value

around 40–60 Am) (Fig. 5). Dark lamina are

interpreted to be formed during periods with

intensified SW winds, enhanced upwelling and

primary productivity processes, driving higher

fluxes of biogenic remains and organic matter

toward the sea floor, with respect to adjacent light

facies (Vargas, 2002; Vargas et al., 2004).

Two unconformities, observed in core 33C, are

associated with lenticular deposits anomalously

G. Vargas et al. / Tectonophysics 399 (2005) 381–398388

enriched in lithic grains (Fig. 5). The older one is

an angular unconformity, located at a 47.5 cm

depth, which is associated with a layer exhibiting a

maximum content in lithic grains (20%, from FTIR

measurements) and a high percentage of coarse

grains. The younger one, located at a 45.5 cm

depth, is associated to a local peak in lithic grain

abundance (4%, from FTIR measurements), as

observed through the high-resolution microscope

study (every 1 mm, Fig. 5). These features are

interpreted as manifestations of at least two excep-

tional geodynamic events within the more or less

rhythmic pelagic sedimentation prevailing in the

bay.

Fig. 6. Detailed sedimentological characterization of the thin section 5

variability of the concentration in mineral grains (Nb/A is number of mine

lenses which are anomalously enriched in coarse lithic grains. The grain si

unconformity situated at 47.5 cm depth in core 33C is shown also.

The detailed characterisation of the thin section 5,

which encompasses the sequence between 40.5 and

50.5 cm depth in core 33C, revealed detailed aspects

of the deposits associated to both unconformities (Fig.

6). Grain size distributions of lithic grains contained in

these levels, show mean values around 110–115 Am,

significantly greater than those of the rest of the

sequence. As this sandy grain size fraction is

dominant in shallower areas, at 25–50 m depth, we

interpret that these deposits result from sporadic local

reworking of material within the basin. This hypoth-

esis is supported by the similar mineralogical compo-

sition of the sediments from these two lenses with

respect to that of the rest of the sequence.

(40.5–50.5 cm depth) in core 33C, showing the high-resolution

ral grains per 7 mm2). The two discontinuity levels are associated to

ze distribution of the anomalous deposit associated with the angular

G. Vargas et al. / Tectonophysics 399 (2005) 381–398 389

Ortlieb et al. (2000) reported a slump deposit

between 46 and 57.5 cm depth as well as an angular

unconformity at 60 cm depth in core 23. A similar

angular unconformity was observed at 70 cm in core

24 (Ortlieb et al., 2000) and at 74 cm depth in core

32C (Valdes, 1998). The X-ray image of this core

suggests the presence of a slump deposit between 56

and 62 cm depth (Fig. 7). These anomalous structures,

tentatively correlated with those described in core

33C, can be dated using the available geochronolog-

ical data reported in this and previous works.

Fig. 7. Photography and X-ray images of cores showing anomalous struct

(black arrow) or a discontinuity/slump deposit (white arrow) observed with

the sedimentary sections analysed by the radiocarbon method. Convention

(see Table 1). A lateral correlation of anomalous structures between all th

possibly related to sudden variations of the Local Reservoir Effect (DR).

3.2. Dating anomalous structural features observed in

short cores

Conventional and calibrated radiocarbon data for

cores 24, 32C and 33C are compiled in Table 1. The

calibrated radiocarbon data suggest that the entire

cores 24 and 32C represent roughly the last millen-

nium and not the last 2000 years as interpreted earlier

(Valdes, 1998; Ortlieb et al., 2000). Two anomalously

old dates in core 32C (1717 AD and 256 AD, Table 1)

may represent sediments affected by anomalously

ural features corresponding to a discontinuity/angular unconformity

in the sequences. Grey rectangles indicate the length and position of

al radiocarbon data and corresponding calibrated ages are indicated

ese cores is proposed. Asterisks (*) indicate anomalously old dates

Table 1

Conventional and calibrated radiocarbon ages from cores 24, 32C and 33C

Core Section depth (cm) Lab. no. Conventional age

(years BP) F1rMean calibrated

age (AD)

2r interval for

calibrated ages (AD)

24 16–25 OBDY-1428 738F70 1874 1678–1950

24 36–45 OBDY-1427 848F150 1693 1446–1950

24 81–85 OBDY-1421 1428F50 1255 1153–1313

24 91–95 OBDY-1403 1928F110 716 534–981

32C 16–18 Beta-100117 810F80T 1717T 1623–1950

32C 71–74 Beta-110118 1144F100 1449 1303–1647

32C 80–82 Beta-100119 1225F60 1408 1302–1474

32C 82–87 Pa-1620 1225F80 1408 12841498

32C 8797 Pa-1612 2375F100T 256T 28–489

33C 49–50 Beta-107629 1400F40 1278 1196–1319

33C 65–66 Beta-107630 1620F40 1038 981–1151

Conventional ages for cores 24 and 32C were from Valdes (1998) and Ortlieb et al. (2000). Calibrated ages were computed with Calib4.3

(Stuiver and Reimer, 1993) and a DR value of 262F13 years (Vargas, 2002; Vargas et al., 2002). Asterisks (T) figure anomalously old results

(see text).

G. Vargas et al. / Tectonophysics 399 (2005) 381–398390

increased DR, resulting from the inclusion of boldQorganic carbon during periods with intensified upwell-

ing processes (Vargas, 2002; Vargas et al., 2002).

Taking into account the middle points of each section

dated in core 24, assuming a date of 1993 AD (the

collection year) for the top of the sequence, and

considering the calibrated ages shown in Table 1, we

calculated sedimentation rates values between 97 and

171 cm ky�1 (Fig. 8). These values seem reasonable

with respect to the mean sedimentation rates of

131F7 cm ky�1, calculated from 210Pb data, and

167F22 cm ky�1, from 14C data, for the upper 10 cm

Core 24: Calibrated ages andSedimentation rates

2000

1750

1500

1250

1000

750

5000 10 20 30 40 50 60 70 80 90

Depth (cm)

Dat

e (A

.D.)

171 cm/ky

111 cm/ky

19 cm/ky

97 cm/ky

Fig. 8. Sedimentation rates estimated for core 24. Bars indicate the

2r range for the calibrated ages (see Table 1). Conventional

radiocarbon data from Ortlieb et al. (2000).

and the entire core F981A, respectively (Vargas, 2002;

Vargas et al., 2002).

Radiocarbon data available from sediments located

below the angular unconformity in cores 24, 32C and

33C support the lateral correlation proposed for this

feature (Fig. 7). The calibrated radiocarbon ages

suggest that the generation of the discontinuity

occurred after 1255 AD (core24), 1278 AD (core

33C) and 1408 AD (core 32C). A single radiocarbon

result from above the discontinuity in core 32C

provides a minimal calibrated age of 1449 AD. Thus,

the event responsible for the generation of this angular

unconformity might have occurred between the years

1408 and 1449 AD (1302–1647 AD considering the

2r range).

The subsequent anomalous feature, observed in

cores 23 and 32C, is tentatively correlated with the

younger discontinuity in core 33C (Fig. 7). If

correct, this lateral correlation would imply that

several tens of centimetres of the sedimentary

sequence had probably been eroded prior to this

second event. Considering the lack of radiochrono-

logical data for the upper segment of this and other

cores, it is difficult to estimate the age of this event.

An attempt to meet this aim is proposed through

lateral correlations between cores F981A, 41C, 32B

and 33C. Between cores F981A and 32B, a detailed

statistically significant (r=0.60, pb0.05) lateral cor-

relation is proposed (Fig. 9). This correlation based

upon a comparison of the downcore content in lithic

Fig. 9. Detailed lateral correlation between cores F981A and 32B, using the comparison between the X-ray images and the computed grey level,

as well as the variability of the content in felsic minerals and carbonates. An acceptable correlation coefficient of 0.60 (at pb0.05) supports the

lateral correlation using the mineral grains-content.

G. Vargas et al. / Tectonophysics 399 (2005) 381–398 391

minerals (quartz, albite and oligoclase) supports the

confidence of this approach.

The lateral correlation of two marker horizons

representing the years 1907 and 1988 AD in core

F981A, with the upper sections of cores 41C, 32B and

33C, allowed the determination of mass accumulation

rates for each sequence (Fig. 10). A chronological

model for the section located below the angular

unconformity (47.5 cm depth) in core 33C was

deduced from two calibrated radiocarbon measure-

ments (Table 1). This chronological framework

suggests that cores F981A and 32B and the upper

section of core 33C (above the younger discontinuity)

represent roughly the last 250 years. In the case of

core 41C, the extrapolation of the upper mass

accumulation rate led to interpret that the former

encompasses roughly the last 550 years (Fig. 10).

Thus, the maximum content in felsic minerals (21%)

observed at 29.5 cm depth (6.95 g cm�2 of cumulated

mass depth) in this core, could be correlated with the

peak of felsic minerals (13%) observed at 43.4 cm

depth (6.93 g cm�2) in core 32B. Since the micro-

scopic observation of thin sections of core 41C

indicates that no other anomalous structure is pre-

served within the sequence; we correlate confidently

these anomalous levels with the youngest disconti-

nuity and local maximum of felsic minerals in core

33C (at 6.81 g cm�2). This proposed lateral correla-

tion led to estimate the date of occurrence of the

youngest exceptional event between the years 1754

and 1789 AD.

3.3. Acoustic profiles description and interpretation

Near 400 km of seismic reflection profiles were

acquired in the Mejillones bay area (Fig. 11). We

recognized two major seismic units: an acoustic

substratum and overlying well stratified unit (Fig.

12). Two distinct facies observed in the acoustic

substratum are probably related to the presence of

granitic basement in the nearby area (substratum-1;

Ferraris and Di Biase, 1978) and to Neogene platform

marine deposits commonly observed along the south-

ern and eastern coastline of the bay (substratum-2;

Ferraris and Di Biase, 1978). The overlying stratified

unit has a maximum thickness where the water depth

is comprised between 50 and 100 m and shows a

progressive thinning toward the NE. Three strong

Fig. 10. Lateral correlation between the upper section of cores F981A, 41C 32B and 33C, based upon the content variation in felsic minerals.

For cores 41C, 32B and the upper part of core 33C, indicated dates (to the left of each profile) are deduced from the extrapolation of the mass

accumulation rate calculated from the correlation of two marker levels (1907 and 1988 AD) in core F981A. Dates for the base of core 33C are

from Table 1. Black and white arrows designate the same anomalous structural features as in Figs. 5, 6 and 7.

G. Vargas et al. / Tectonophysics 399 (2005) 381–398392

amplitude reflections named A, B and C (Fig. 12)

characterize this unit and appear to be either regional

(A and C) or localized (B) reflections. As C seems to

Fig. 11. A: High-resolution bathymetry of Mejillones bay (isobaths every 1

in dashed line. Profiles P1123-04 and P1125-01 are indicated. The �120 m

shoreline during the LGM, is indicated. B: 3D view of bathymetry of the

denote an important change in acoustic facies within

the stratified unit, we suggest that the origin of this

reflection be associated with the onset of the dominant

0 m). Seismic profiles lines recently acquired in this basin are shown

isobath, which could correspond to the approximate situation of the

Mejillones bay. Isobaths every 10 m. Grey levels every 100 m.

Fig. 12. Interpretation of two seismic profiles along W–E (A) and SW–NE (B) directions, illustrating the sedimentary infill of Mejillones Bay.

The upper seismic unit is characterized by two regional strong reflections (A and C) and a locally strong and discontinuous reflection (B) that is

stratigraphically associated with a major slump (contorted and chaotic reflections in the centre of the bay; C and D). The substratum shows two

seismic facies. Substratum-1 is characterized by total absorption of energy (A), while substratum-2 is characterized by strong dips towards the

NE (B). The base of the sedimentary infill develops offshore downlaps above a strong erosion surface (reflection E) on top of the acoustic

basement (parallel reflections dipping toward the NE; substratum-2; B). The uppermost seismic unit (above reflection C) forms coastal onlaps,

develops a main depocentre in the centre of the bay and thins toward the NE where reflections are parallel to reflection E.

G. Vargas et al. / Tectonophysics 399 (2005) 381–398 393

G. Vargas et al. / Tectonophysics 399 (2005) 381–398394

pelagic sedimentation within the basin. This may have

occurred in the Middle Holocene, at the end of the last

marine transgression following the Last Glacial

Maximum (LGM). During the LGM (around 20 ka),

the global sea level was roughly 120 m below its

present day position (Lambeck et al., 2002), which

means that it was below the largest part of the present

bottom of Mejillones bay (Fig. 11).

The upper seismic unit is affected by a large

chaotic body in some areas of the bay (around 75 m of

water depth), between C and A (Fig. 12). B is locally

associated with the top of this chaotic acoustic facies.

This body occurs at water depths greater than 60 m

and 100 m in the central part and NE area of this

basin, respectively. In this last area, B is clearly

associated to the development of contorted reflections,

which include C. The occurrence of these chaotic

acoustic facies along a steeper slope of the bay

suggests soft-sediment deformation related to gravity

reworking phenomena. According a preliminary

interpretation of the Bathy 2000 data, no slide or

slump scar could be recognized and the shape of this

chaotic body does not indicate sediment displacement

over a long distance. Thus, the loss of internal

stratification in this chaotic body probably resulted

from limited slumping and development of sediment

liquefaction or fluidization (Lowe, 1975; Allen, 1982;

Mulder and Cochonat, 1996). Long coring is needed

to confirm the precise nature of this major event.

4. Discussion

4.1. Sedimentary instability and earthquakes

The analysis of short cores suggests that the

presence of anomalous sedimentological and struc-

tural features is associated to exceptional events that

produce gravity instability in the sedimentary

sequence, slumping and local reworking of material

from shallower areas within the basin. In situ

liquefaction of sediments due to cyclic loading and

development of soft-sediment deformation have been

described as related to cyclic gravity acceleration

during strong earthquakes (Sims, 1975; Seilacher,

1991; Moretti et al., 1999). A similar phenomenon has

been associated to pore-pressure fluctuations caused

by fluid escape, induced by cyclic and residual stress

of storm or tsunami waves (Lowe, 1975; Allen, 1982;

Molina et al., 1998, Rossetti, 1999). While it is

accepted that storm surges result in the deposition of

discrete sedimentary units, tsunami waves generally

result in deposition of sediment sheets over relatively

wide areas and considerable distances inland (Daw-

son, 1999). Moreover, the successive run-up and

backwash of tsunami waves can produce strong

seaward currents causing additional erosion and sedi-

ment re-deposition, depending on the nature of the

local coastal topography and bathymetry (Bondevik et

al., 1997; Shi et al., 1998; Dawson, 1999).

4.2. Chronological interpretation of unconformities

and slump deposits, and their correlation with

regional seismic events

From the analysis of historical reports and

chronicles from southern Peru and northern Chile,

Comte and Pardo (1991) inferred seismic parameters

(rupture length and magnitude) for strong historical

earthquakes (MS(I)N7, where MS(I) is proportional to

Mw) since the 16th century. Return periods of

118F44 (based on four events: 1513, 1604, 1784

and 1868) and 111F33 years (based on four events:

1543, 1615, before 1768 and in 1877) were estimated

for these two areas, respectively. For the two last great

tsunamigenic earthquakes, in 1868 in southern Peru

(epicenter at 17.708S and 71.608W) and in 1877 in

northern Chile (epicenter at 21.008S and 70.258W),

magnitudes of MS(I)=8.8 and MS(I)=8.8 and rupture

lengths of 500 km and 420 km were respectively

inferred (Fig. 1).

Considering the morphological configuration of

the north-western extremity of the Mejillones pen-

insula, which protects the basin from the direct

influence of the dominant S–SW winds and associated

storms (Fig. 3), we interpret that altogether the

anomalous sedimentological and structural features

observed in the short cores and the great slump

deposit detected in the seismic profiles, most probably

reflect the effects of strong earthquakes and/or

tsunamis in Mejillones bay. We suggest that a seismic

event occurred between the years 1408 and 1449 AD,

at a time during which no written archives existed.

Such an earthquake would have produced: (1) the

angular unconformity described in cores 24, 32C, 33C

and 23, and (2) local reworking of material yielding to

G. Vargas et al. / Tectonophysics 399 (2005) 381–398 395

lens-shaped deposits enriched in lithic and coarse

grains. This event would have occurred about 94–135

years before the 1543 earthquake reported in the

Tarapaca area (epicenter at 19–208S, rupture length

LN100 km; Comte and Pardo, 1991). Such interpre-

tation would be consistent with the estimated return

period (111 years) of great seismic episodes in the

region. The second event recorded in the sediment

cores, which would have occurred between the years

1754 and 1789 AD, might correspond to the 1768

historical event documented in northern Chile (Fig. 1).

The scarcity of historical reports on this event

explains that its epicenter has been imprecisely

located around 20.58S, with an inferred MS(I)N7.7

and a rupture length LN100 km (Comte and Pardo,

1991). We surmise that this earthquake produced

liquefaction and slumping within the bay, thus

generating the slump deposit seen in core 23 (and

core 32C?). Local reworking of material from shallow

areas within the basin and its later deposition in

deeper zones possibly produced the mineral-rich

deposit observed in cores 41C, 32B and 33C. We

hypothesize that the 1768 event was also responsible

for the loss of several tens of centimetres of

sedimentary sequence at the place where core 33C

was retrieved.

4.3. On the lack of evidence for local effects of the

major 1877 earthquake

It is puzzling that we did not observe, in the studied

sedimentary sequences, any anomalous structural

feature susceptible to be associated with the major

1877 earthquake. We did not observe either anom-

alous deposits associated with the 15–20 m high

tsunami waves generated during this seismic event,

known to have destroyed the Mejillones village. It

must be noted, though, that the epicenter of this

earthquake was located several hundred kilometres

north of Mejillones bay, so that the magnitude of this

event was lower than Mw=8 in the Mejillones area.

The same inference can be made to explain the

absence of any anomalous structure susceptible to be

associated to the earthquakes of 1543 and 1615

(epicenter at 19.58S and 70.58W; MS(I)=7.9; L=137

km; Comte and Pardo, 1991). Thus, we propose that

the lack of anomalous features related with these

historical earthquakes within the studied cores, may

reflect either one, or a combination, of the following

considerations:

1. the energy associated with the propagation of these

events was significantly reduced, or even stopped,

near the Mejillones peninsula, hypothesis which is

consistent with the role of tectonic-barrier proposed

for this peninsula (Armijo and Thiele, 1990;

Delouis et al., 1997; 1998);

2. the run-up and backwash fluxes of the tsunami

waves produced during the 1877 major earthquake,

were of reduced amplitude and did not allow that

the sediment reworked in the coastal area reach the

coring sites in the centre of the bay;

3. these anomalous features were associated to local

reactivations of faults that cut the Mejillones

peninsula, which generated gravity instability

yielding anomalous structures in limited areas

within the basin.

4.4. Possible effects of reactivation of the Mejillones

fault

The last hypothesis could be relevant to explain the

generation of the great slump deposit observed

through acoustic data in the sedimentary infill of the

bay. The Mejillones peninsula displays an E–W

extensional deformation during Plio-Quaternary times

and is affected by numerous normal faults, tension

gashes, tilted blocks and vertical motions (Okada,

1971; Armijo and Thiele, 1990; Hartley and Jolley,

1995; Niemeyer et al., 1996; Delouis et al., 1998).

The N–S Mejillones fault, a main feature located at

the north-western end of this peninsula (Fig. 3),

separates two blocks which were submitted to differ-

ent long-term uplift motions since the Early Pleisto-

cene, estimated at 0.4 m/ka (west of the fault) and 0.2

m/ka (on the eastern compartment) (Ortlieb et al.,

1996a; 1996b; 1996c). New chronological data, based

on cosmogenic isotopes measurements obtained from

alluvial fan’s surfaces affected by this fault, suggest a

recent vertical normal offset around 13F1 m during

the last 43F3 ka and 5.5F0.5 m during the last 20F3

ka (Marquardt et al., 2003). A detailed interpretation

of the acoustic reflection data is still pending, but

observations of profiles obtained on the western

margin of the basin already show an important

morphological scarp displacing 10–15 m the acoustic

G. Vargas et al. / Tectonophysics 399 (2005) 381–398396

substratum-1, which is systematically disposed along

the northward projection of the Mejillones fault (Fig.

12A). Considering the 5.5F0.5 m vertical offset

measured along this fault for the post-LGM period,

it is possible that local re-activations of this fault

during strong regional earthquakes occurred coevally

with the deposition of the Holocene sedimentary infill

within the basin. Actually, Delouis et al. (1998)

previously discussed the possible link between local

reactivations of this fault during some strong sub-

duction seismic events, in the light of cumulative

Coulomb’s stress in a regional E–W extensional

regime. Similar phenomena were envisioned along

some specific portions of the Atacama fault system

during the Antofagasta 1995 seismic event (Delouis et

al., 1998; Gonzalez et al., 2003).

Taking into account the mass accumulation rates

estimated from the analysis of short cores and

considering the Late Quaternary age inferred for the

whole upper seismic unit, we suggest that an

important seismic event occurred during or shortly

after the Middle Holocene. This event may have

generated a reactivation of the Mejillones fault and

induced gravity instability and slumping within the

Mejillones bay, as evidenced by acoustic data. The

wide extension of reflection B may reflect the effects

of backwash flows along the sandy coastal sediments

which succeeded to tsunami waves. We furthermore

surmise that this major tectonic event might have

played a role in the carving of the several-metre-high

cliffs which now border the southern and eastern

shores of Mejillones bay.

4.5. Final considerations

The study shows how a detailed chronological

analysis of a laminated marine sequence may yield

relevant information for both paleoceanographic and

paleoseismic reconstructions. The Holocene sedi-

mentary record preserved in Mejillones bay is quite

favourable to develop high-resolution chronological

models encompassing the last few centuries.

Ongoing and future works will thus be concentrated

on the detailed 3D interpretation of the acoustic data

and the recuperation of longer sediment cores in the

same bay that encompass longer and older time

periods. Such studies, which will involve the upper

reflection horizons identified in the seismic profiles,

should provide a more complete picture of the effect

of paleoseismic events during the last few millennia

on the recent sedimentary infill in this basin.

From the point of view of the impacts of both the

regional seismic and local faulting activities, we still

need to test hypotheses and work on the propagation

modelling of the energy released during seismic

events all along the rupture zones, the deformation

mechanisms proper to the Mejillones fault and the

propagation of tsunami waves within the Mejillones

bay.

Acknowledgements

Several aspects of this work was supported by

the program Paleotropique (Paleo-environnements

tropicaux et variabilite climatique), of IRD, France.

Acquisition of seismic profile data was partly

supported by the PCCI project bEstudio sısmico

de alta resolucion del relleno sedimentario holoceno

de la bahıa Mejillones del Sur: implicancias neo-

tectonicas y oceano-climaticasQ, of CONICYT,

Chile, and the PRODAC (Programa de Dinamica

de la Atmosfera y del Clima) of the University of

Chile.

The Department of Geology and Oceanography

of the University of Bordeaux I, the laboratory of

IRD at Bondy, the Department of Geology of the

University of Chile and the FAREMAR of the

University of Antofagasta, Chile, greatly helped for

the acquisition of cores, treatment of samples and

analyses, as well as for the acquisition of seismic

profile data.

We thank J.F. Saliege (LODYC, Univ. Paris-

CNRS-IRD) for radiocarbon data on several cores,

and Laurent Perderau (ISTO, Universite d’Orleans)

for his useful help during the acquisition of seismic

profiles. The preparation of the seismic survey

benefited largely from Emilio Vera’s (University of

Chile) experience in marine geophysics of the area.

Two anonymous reviewers provided useful

comments.

Jacques Bertaux (from IRD, Bondy) provided

important support for FTIR mineralogical analyses

of the short cores. Jacques Bertaux died quite

prematurely in August 2002, and we all miss him.

This paper is dedicated to his memory.

G. Vargas et al. / Tectonophysics 399 (2005) 381–398 397

References

Allen, J.R.L., 1982. Sedimentary Structures. Their Character and

Physical Basis. Elsevier, Amsterdam.

Armijo, R., Thiele, R., 1990. Active faulting in northern Chile:

ramp stacking and lateral decoupling along a subduction plate

boundary? Earth and Planetary Science Letters 98, 40–61.

Bertaux, J., Frfhlich, F., Ildefonse, Ph., 1998. Multicomponent

analysis of FTIR spectra: quantification of amorphous and

crystallized mineral phases in soils and sediments. Journal of

Sedimentary Research 68 (3), 440–447.

Bondevik, S., Svendsen, J.I., Mangerud, J., 1997. Tsunami

sedimentary facies deposited by the Storrega tsunami in shallow

marine basins and coastal lakes, western Norway. Sedimentol-

ogy 44, 1115–1131.

Comte, D., Pardo, M., 1991. Reappraisal of great historical

earthquakes in the Northern Chile and Southern Peru seismic

gaps. Natural Hazards 4, 23–44.

Comte, D., Pardo, M., Dorbath, L., Dorbath, C., Haessler, H.,

Rivera, L., Cisternas, A., Ponce, L., 1992. Tectonophysics 205,

13–22.

Comte, D., Pardo, M., Dorbath, L., Dorbath, C., Haessler, H.,

Rivera, L., Cisternas, A., Ponce, L., 1994. Geophys. J. Int. 116,

553–561.

Comte, D., Haessler, H., Dorbath, L., Pardo, M., Monfret, T.,

Lavenu, A., Pontoise, B., Hello, Y., 2002. Seismicity and stress

distribution in the Copiapo, northern Chile subduction zone

using combined on- and off-shore seismic observations. Physics

of the Earth and Planetary Interiors 132, 197–217.

Dawson, A.G., 1999. Linking tsunami deposits, submarine

slides and offshore earthquakes. Quaternary International 60,

119–126.

Delouis, B., Monfret, T., Dorbath, L., Pardo, M., Rivera, L., Comte,

D., Haessler, H., Caminade, J.P., Ponce, L., Kausel, E.,

Cisternas, A., 1997. The Mw=8.0 Antofagasta (northern Chile)

earthquake of 30 July 1995: a precursor to the end of the large

1877 gap. Bulletin of the Seismological Society of America 87

(2), 427–445.

Delouis, B., Philip, H., Dorbath, L., Cisternas, A., 1998. Recent

crustal deformation in the Antofagasta region (northern Chile)

and the subduction process. Geophysical Journal International

132, 302–338.

Escribano, R., 1998. Population dynamics of Calanus chilensis in

Chilean eastern boundary Humboldt current. Fisheries Ocean-

ography 7, 245–251.

Ferraris F., Di Biase F., 1978. Hoja Antofagasta, region Antofa-

gasta. Carta Geol. Chile no 30, Instituto de Investigaciones

Geologicas Chile, 48p. 1 map 1:250.000.

Gonzalez, G., Cembrano, J., Carrizo, D., Macci, A., Schneider,

H., 2003. The link between forearc tectonics and Pliocene–

Quaternary deformation of the Coastal Cordillera, northern

Chile. Journal of South American Earth Sciences 16,

321–342.

Hartley, A.J., Jolley, J., 1995. Tectonics implications of Late

Cenozoic sedimentation from the Coastal Cordillera of northern

Chile (22–248S). Journal of the Geological Society of London

152, 51–63.

Lambeck, K., Esat, T.M., Potter, E.-K., 2002. Links between

climate and sea levels for the past three million years. Nature

419, 199–206.

Lowe, D.R., 1975. Water escape structures in coarse-grained

sediments. Sedimentology 22, 157–204.

Marın, V., Olivares, G., 1999. Estacionalidad de la productividad

primaria en Bahıa Mejillones del Sur (Chile): una aproximacion

proceso-funcional. Revista Chilena de Historia Natural 72,

629–641.

Marın, V., Rodrıguez, L., Vallejos, L., Fuenteseca, J., Oyarce, E.,

1993. Efectos de la surgencia costera sobre ela productividad

primaria primaveral de Bahıa Mejillones del Sur (Antofagasta,

Chile). Rev. Chilena de Historia Natural 66, 479–491.

Marquardt, C., Braucher, R., Ritz, J.F., Philip, H., Bourles, D.,

Lavenu, A., Delouis, B., Ortlieb, L., 2002. Late Quaternary slip

rates of the Mejillones Fault, northern Chile (238S), using 10Be

dates. 5th ISAG, Toulouse, France, abs. vol., pp. 409–412.

Marquardt, C., Ritz, J.-F., Philip, H., Bourles, D., Braucher, R.,

Lavenu, A., Ortlieb, L., Delouis, B., 2003. Slip rates of a Late

Quaternary normal fault in Mejillones peninsula (238S),estimated using 10Be exposures ages. 108 Congreso Geologico

Chileno, Abs. Volume, Concepcion, 6 pp.

Migeon, S., Weber, O., Faugeres, J.C., Saint-Paul, J., 1999.

SCOPIX: a new X-ray imaging system for core analysis. Geo-

Marine Letters 18, 251–255.

Molina, J.M., Alfaro, P., Moretti, M., Soria, J.M., 1998. Soft-

sediment deformation structures induced by cyclic stress of

storm waves in tempestites (Miocene, Guadalquivir Basin,

Spain). Terra Nova 10 (3), 145–150.

Monfret, T., Dorbath, L., Caminade, J.P., Pardo, M., Comte, D.,

Ponce, L., Cisternas, A., Delouis, B., Rivera, L., 1995. The July

30, 1995 Antofagasta earthquake: a bhypocriticalQ seismic event.

EOS Abstract of the Fall Meeting of the American Geophysical

Union 76 (46), F427 (supplement).

Moretti, M., Alfaro, P., Caselles, O., Canas, J.A., 1999. Modelling

seismites with a digital shaking table. Tectonophysics 304,

369–383.

Mulder, T., Cochonat, P., 1996. Classification of offshore mass

movements. Journal of Sedimentary Research 66 (1), 43–57.

Niemeyer, H., Gonzalez, G., Martınez De los Rıos, E., 1996.

Evolucion tectonica cenozoica del margen continental activo de

Antofagasta, norte de Chile. Revista Geologica de Chile 23 (2),

165–186.

Okada, A., 1971. On the neotectonics of the Atacama fault zone

region. Preliminary notes on Late Cenozoic faulting and

geomorphic development of the Coastal Range of Northern

Chile. Bulletin of the Department of Geography, Kyoto

University 3, 47–65.

Ortlieb, L., Barrientos, S., Guzman, N., 1996a. Coseismic coastal

uplift and coralline algae record in northern Chile: the 1995

Antofagasta earthquake case. Quaternary Science Reviews 15,

949–960.

Ortlieb, L., Diaz, A., Guzman, N., 1996b. A warm interglacial

episode during oxygen isotope stage 11 in northern Chile.

Quaternary Science Reviews 15, 857–871.

Ortlieb, L., Zazo, C., Goy, J.L., Hillaire-Marcel, C., Ghaleb, B.,

Cournoyer, L., 1996c. Coastal deformation and sea-level

G. Vargas et al. / Tectonophysics 399 (2005) 381–398398

changes in the Northern Chile subduction area (238S) during thelast 330 ky. Quaternary Science Reviews 15, 819–831.

Ortlieb, L., Escribano, R., Follegati, R., Zuniga, O., Kong, I.,

Rodrıguez, L., Valdes, J., Guzman, N., Iratchet, P., 2000.

Recording ocean-climate changes during the last 2,000 years in

a hypoxic marine environment off northern Chile (238S).Revista Chilena de Historia Natural 73, 221–242.

Pritchard, M.E., Simons, M., Rosen, P.A., Hensley, S., Webb, F.H.,

2002. Co-seismic slip from the 1995 July 30 Mw=8.1 Antofa-

gasta, Chile, earthquake as constrained by InSAR and GPS

observations. Geophysical Journal International, 150, 362–376.

Rodrıguez, L., Marin, V., Farias, M., Oyarce, E., 1991. Identifica-

tion of an upwelling zone by remote sensing and in situ

measurement, Mejillones del Sur Bay (Antofagasta, Chile).

Scientia Marina 55, 467–473.

Rossetti, D.F., 1999. Soft-sediment deformation structures in Late

Albian to Cenomanian deposits, Sao Luis Basin, northern Brazil:

evidence for palaeoseismology. Sedimentology 46, 1065–1081.

Ruegg, J.C., Campos, J., Armijo, R., Barrientos, S., Briole, P.,

Thiele, R., Arancibia, M., Canuta, J., Duquesnoy, T., Chang, M.,

Lazo, D., Lyon-Caen, H., Ortlieb, L., Rossignol, J.C., Serrurier,

L., 1996. The MW=8.1 Antofagasta (north Chile) earthquake of

July 30, 1995: first results from teleseismic and geodetic data.

Geophysical Research Letters 23 (9), 917–920.

Seilacher, A., 1991. Events and their signatures—an overview. In:

Einsele, G., Ricken, W., Seilacher, A. (Eds.), Cycles and Events

in Stratigraphy. Springer-Verlag, pp. 221–226.

Shi, S., Dawson, A.G., Smith, D.E., 1998. Coastal tsunami

sedimentation processes. In: CEA/DASE (Ed.), Proceedings of

the International Conference on Tsunamis, Paris, France, May

26–28, 1998, pp. 43–55.

Sims, J.D., 1975. Determining earthquake recurrence intervals from

deformational structures in young lacustrine sediments. Tecto-

nophysics 29, 141–152.

Somoza, R., 1998. Updated Nazca (Farallon)–South America

relative motions during the last 40 My: implications for

mountain building in the Central Andean region. Journal of

South American Earth Sciences 11, 211–215.

Southon, J.R., Oakland Rodman, A., True, D., 1995. A comparison

of marine and terrestrial radiocarbon ages from Northern Chile.

Radiocarbon 37 (2), 389–393.

Stuiver, M., Reimer, P.J., 1993. Extended 14C data base and revised

Calib 3.0 14C age calibration program. Radiocarbon 35 (1),

215–230.

Valdes, J., 1998. Evolucion oceanografica reciente de la Bahıa

Mejillones del Sur (238S). Evidencia geoquımica en sedimentos

marinos. PhD tesis, Centro EULA-Chile, Universidad de

Concepcion, 114 p.

Valdes, J., Ortlieb, L., 2001. Paleoxigenacion subsuperficial de la

columna de agua en la bahıa Mejillones del Sur (238S):Indicadores geoquımicos en testigos de sedimento marino.

Investigaciones Marinas 29 (1), 25–35.

Valdes, J., Lopez, L., Lo Monaco, S., Ortlieb, L., 2000. Condiciones

paleoambientales de sedimentacion y preservacion de la materia

organica en bahıa Mejillones del Sur (238S), Chile. Revista de

Biologıa Marina y Oceanografıa 35 (2), 169–180.

Vargas, G., 1998. Approches methodologiques en paleo-ocean-

ographie realisees a partir de carottes de la Baie de Mejillones,

Chile (238S). Memoire DEA, Departement de Geologie et

Oceanographie. Universite Bordeaux I, France. 32 pp.

Vargas, G., 2002. Iteractions ocean–atmosphere au tours des

derniers siecles sur la cote du Desert d’Atacama : analyse

multi-proxies des sediments lamines de la Baie de

Mejillones (238S). These de Doctorat, Universite Bordeaux

I, 290 p.

Vargas, G., Ortlieb, L., Rutllant, J., 2000. Aluviones historicos en

Antofagasta y su relacion con eventos El Nino/Oscilacion del

Sur. Revista Geologica de Chile 27 (2), 155–174.

Vargas G., Ortlieb L., Valdes J., 2002. Radiochronology in a marine

sequence of northern Chile, for high resolution paleosismolog-

ical and paleoclimatic reconstructions. 5th ISAG, Toulouse,

France, abs. vol., pp. 673–676.

Vargas, G., Ortlieb, L., Pichon, J.J., Bertaux, J., Pujos, M., 2004.

Sedimentary facies and high resolution primary production

inferences from laminated diatomaceous sediments off northern

Chile (238S). Marine Geology 211, 79–99.

von Huene, R., Weinrebe, W., Heeren, F., 1999. Subduction erosion

along the North Chile margin. Geodynamics 27, 345–358.

Zapata, 2001. Estudio batimetrico del margen continental chileno.

Tesis Ms. Cs. Departamento de Geofısica, Universidad de Chile,

113 pp.