Disentangling Late Quaternary climatic and seismo-tectonic controls on Lake Mucubají sedimentation (Mérida Andes, Venezuela)

Available online at www.sciencedirect.com

alaeoecology 259 (2008) 284–300www.elsevier.com/locate/palaeo

Palaeogeography, Palaeoclimatology, P

Disentangling Late Quaternary climatic and seismo-tectonic controlson Lake Mucubají sedimentation (Mérida Andes, Venezuela)

Eduardo Carrillo a,b, Christian Beck a,⁎, Franck A. Audemard c,Eva Moreno d, Reinaldo Ollarves c

a Instituto de Ciencias de la Tierra, Universidad Central de Venezuela, Apdo. 3805, Caracas 1010-A, Venezuelab Laboratoire de Géodynamique des Chaînes Alpines, U.M.R. C.N.R.S. 5025, Université de Savoie, F-73376 Le Bourget du Lac, France

c Fundación Venezolana de Investigaciones Sismológicas, FUNVISIS, El Llanito, Caracas 1030, Venezuelad Département Histoire de la Terre - USM 203, Muséum National d’Histoire Naturelle de Paris, 43 rue Buffon, Case Postale 48, 57 Rue Cuvier,

75231 Paris Cedex 05, France

Received 3 October 2005; accepted 14 August 2007

Abstract

Lake Mucubají is a moraine-dammed lake, crossed by the Boconó Fault, in the Mérida Andes, Venezuela. Four long piston-cores (up to 8 m long) and 24 short gravity cores (0.6 to 0.8 m long) were collected to study the sedimentary fill of the lake.Lithostratigraphy, magnetic susceptibility, organic and inorganic contents, and 14C ages, permit lateral correlations between thecores, and with an emerged part of the former larger lacustrine basin fill. The cored succession spans the last 16,000 years. Severalabrupt changes in sediment texture and composition are ascribed to depth and surface modifications. Together with soft-sedimentdisturbances, they are related to the seismo-tectonic activity of the Boconό Fault: shock-induced phenomena (slumps, seicheeffects, liquefaction) and co-seismic scarps. Four major earthquakes indicate a mean minimum recurrence interval of 1200 years,which is consistent with trench data obtained from a neighbouring active trace of the Boconó Fault system. Despite the seismo-tectonic imprint, a general post-LGM climatic trend can be traced and correlated with northern-hemisphere global evolution, inparticular for the last 13,000 years.© 2007 Elsevier B.V. All rights reserved.

Keywords: Mérida Andes; Boconó Fault; Late Glacial; Holocene; Lacustrine sediments; Palaeo-earthquakes; Lake Mucubají

1. Introduction

Sedimentation in active tectonic environments andglacier-influenced watersheds is driven by a combina-tion of climatic factors and seismo-tectonic activity. Anexample of a lake affected by both tectonics and

⁎ Corresponding author.E-mail address: [email protected] (C. Beck).

0031-0182/$ - see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.palaeo.2007.10.012

glaciation/deglaciation is Lake Mucubají, located in theSierra Nevada National Park, in the Mérida Andes(northwestern Venezuela). Previous investigations bySchubert and Henneberg (1975), Giegengack et al.(1976), Giegengack (1984), Bradley et al. (1985),Henneberg and Schubert (1986), Weingarten et al.(1990a,b), Salgado-Labouriau et al. (1992) and Schubertand Vivas (1993) already dealt with these factors of LateQuaternary sedimentation and morphology in the centralpart of the Mérida Andes (Fig. 1). Lake Mucubají,

285E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

locally known as “Laguna de Mucubají” (Fig. 2), is3540 m high and situated at 9° N, midway between theeastern equatorial Pacific and the southeastern Caribbe-an, in particular the Cariaco Trough. This lake is directlycrossed by the Boconó Fault, an active part of thesouthern boundary of the Caribbean Plate (Fig. 1).

Representing an important earthquake source, theBoconó Fault has been surveyed as part of apalaeoseismic study undertaken by the VenezuelanFoundation for Seismological Investigation (FUNVI-SIS) (Audemard et al., 2000). Investigations of the faultsystem have included trench analysis and study oflacustrine sedimentary records; the present articleconcerns the latter (see also Audemard et al., 2002a,b;Carrillo et al., 2002; Beck et al., 2005; Carrillo et al.,2006). Our main objective was to test, for a commonarea, the addition of lacustrine palaeoseismic archives totrench data. This involved the search for specific shock-induced deposits and reworking features, as previouslyanalyzed by different authors in other lacustrine ormarine settings (Sims, 1975; Ben-Menahem, 1976;Hempton and Dewey, 1983; El-Isa and Mustafa, 1986;

Fig. 1. Geodynamic and morphological settings of the studied area. (A) simpliAudemard et al., 2000); (B) location of the study area. Relief map from HacklBF arrows.

Siegenthaler et al., 1987; Plaziat et al.; 1988; Doig,1991; Van Loon et al., 1995; Marco and Agnon, 1995;Beck et al., 1996; Cita and Rimoldi, 1997; Calvo et al.,1998; Lignier et al., 1998; Chapron et al., 1999; Morettiet al., 1999; Nakajima and Kanai, 2000; Gorsline et al.,2000; Föhlisch and Voigt, 2001; Arnaud et al., 2002;Rodríguez-Pascua et al., 2002). Other authors havediscussed the driving parameters of sub-aquatic sedi-ment failures (Field et al., 1982; Mulder and Cochonat,1996; Piper et al., 1992). In a study of present moraine-dammed lakes, Clague and Evans (2000) mentionedearthquakes as one of the causes of moraine failure andcatastrophic emptying.

We also assumed that co-seismic offsets can directlyaffect lake geometries and drainage (tributaries andoutlet), and induce significant changes in sedimentation.Lake Mucubají was chosen for these investigations; itdeveloped on the central section of the Boconό Faultduring the beginning of the last deglaciation. Here, wepresent and discuss the main results of the analysis ofthe in situ sedimentary fill of Lake Mucubají; inparticular, we attempt to decipher how the palaeoseismic

fied main present-day active tectonics of northern South America (fromey et al. (2005). Main active trace of the Boconò Fault between the two

Fig. 2. Lake Mucubají within its morphological and morphotectonic environment. Aerial photograph from Instituto Geográfico de Venezuela SimónBolívar. The white star indicates location of the lacustrine sediment outcrop studied by Salgado-Labouriau et al. (1977) (MCLO: Mucuñuque CreekLacustrine Outcrop, mentioned on Figs. 4 and 5).

286 E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

record is superimposed on the general Late Glacial-Holocene climatic evolution.

2. Geodynamic and climatic setting of LakeMucubají

2.1. Regional tectonic setting

The northern boundary of the South-Americancontinent is marked by a dextral transform system thataccommodates the relative displacement of the Carib-bean Plate with respect to the South-American Plate.Surface and subsurface geology, offshore and onshoregeophysical imagery, seismology and GPS kinematicsprovide evidence for the structures summarized on

Fig. 1 (see Audemard et al., 2000). The averagedisplacement on the bounding fault system is about2 cm/year (Pérez et al., 2001). Several strike-slip faultsaccommodate part of this movement within a broaddeformed zone (Fig. 1A). The northeast-strikingBoconó Fault takes up about half of this amount, andmuch of the remainder is accommodated on the west-east Oca/San Sebastian/El Pilar system (Rod, 1956;Audemard et al., 1999). The Boconó Fault follows theaxial part of the Mérida Andes (Fig. 1), whichculminates in Bolivar Peak (5002 m) near to the cityof Mérida. Although the main topography is associatedto Neogene deformation and uplift (Stéphan, 1985;Audemard and Audemard, 2002) the Mérida Andescomprises different Precambrian to Palaeogene igneous,

287E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

metamorphic, and sedimentary rocks. Its northern partwas partly overthrusted by the important allochtonouspile of the Caribbean Chain (Bellizzia et al., 1976;Stéphan, 1985).

2.2. Lake Mucubají area

In the study area, the Boconó Fault cuts Precambrianbasement as well as Late Quaternary glacial sedimentsof the two last major glaciations. Moraine systems and,often, associated lakes, developed at altitudes higherthan 2600 m. Deposits of the last Pleistocene glaciation,known here as the Mérida Glaciation (Schubert, 1974;Schubert and Vivas, 1993) are particularly abundantalong the central section of the Boconó Fault. Previousinvestigations have underlined the direct impact of theactive tectonics on the glacial (moraine) and post-glacial(lake) deposits (Giegengack et al., 1976; Schubert,1980, 1981, 1982; Giegengack, 1984). More recently,sediments of several palaeolakes, located on the trace ofthe Boconό Fault, were investigated to document localand regional palaeoseismicity (Audemard et al., 2001,2002b; Carrillo et al., 2002, 2006; Beck et al., 2005).Other researchers have studied glacial and post-glacialsediments to document changes in climate and vegeta-tion in the region, and compared them to global changes(Salgado-Labouriau and Schubert, 1976; Rull, 1996;Mahaney et al., 1997; Rull and Vegas-Vilarrubia, 1998;Rull, 1999; Mahaney et al., 2000); they consider theMérida Glaciation as equivalent of the global LateGlacial Maximum (LGM).

Fig. 3. Bathymetry of Lake Mucubají and location of long piston cores. B

Lake Mucubají has a maximum length of 780 m, amaximum width of 540 m, and an area of 0.26 km2

(Fig. 2). The maximum depth is about 16 m, and thevolume in the dry season 2.33×103 m3. The present-daylake level is 3540 m a.s.l. and relatively stable (less than1 m of annual variation) since the building of a smallconcrete dam at the outlet in 1967 (in Salgado-Labouriauet al., 1992).

Lake Mucubají is impounded by terminal and lateralmoraines constructed during the Mérida Glaciation(Schubert, 1974; Schubert and Vivas, 1993). Themoraines have been offset in a dextal strike-slip senseby the Boconό Fault. In addition, as underlined onFig. 2, the right-lateral moraine has been affected by alandslide, possibly polyphasic. The present-day lakeoutlet cuts the lowest part of this landslide. The southernshore of the lake is controlled by two active strands ofthe Boconó Fault. The scarps are also recognizable onthe bathymetric map (Fig. 3). An abandoned outlet westof the terminal moraine (Fig. 2) is 20 m higher than thepresent-day outlet. The elevation of this abandonedoutlet requires that the opposite shore at the presentoutlet was initially 20 m higher. Thus a previous higherlake surface is inferred, along with a southwardextension of the lake (to the white star in Fig. 2). Atthis location (mentioned MCLO in the following),outcropping lacustrine sediments have been dated to14,550 to 13,550 cal. yr BP by Salgado-Labouriau et al.(1977). Different processes may account for the high-level lake: 1) eastward tilting on the north side of theBoconó Fault and relative uplift of the south-south-

athymetry is based on a 575 points grid measured during this work.

Fig. 4. Summary of lithostratigraphy of cores MUCL-0201 to MUCL-0204 and outcrop MCLO. Illustration of main layering types. Core summits arerepresented with the actual depth differences; cores MUCL-0204 and MUCL-0202 are on opposite sides of the Boconó Fault last active trace.

288 E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

289E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

western side of the lake, 2) fast erosion or earthquake-induced collapse of the eastern lateral moraine near thelake, or 3) northeastward tilting of the entire Mucubajímoraine complex due to sliding from the basementrocks (Audemard et al., 2002b).

3. Logistics and methods

The bathymetric map shown on Fig. 3 was producedwith an echosounder coupled to a GPS, and is based on575 data points. Two sets of cores were retrieved inDecember 2002, using a portable UWITEC™ coringdevice (platform, tripod, winches, and corers). Twoperpendicular cables were strung across the lake and theplatform was fixed at their intersection. The wholesystem was moved for each piston-coring site; alldisplacements and coring operations were hand-pow-ered since the lake is located inside a National NaturalPark and the use of any thermal engine is forbidden onthe lake.

Twenty-four short gravity cores (mean length: 60 cm)were collected to sample the water-sediment interface.Six-cm-diameter piston cores (length: 3 to 8 m) werecollected at four sites (Fig. 3). At each site, 3-m-longsections were obtained with a 1.5 m depth offset toensure overlap at section ends. Thus, each log (Figs. 4and 5) represents the combination of successive sectionsand their overlapping parts. Magnetic susceptibility(MS) profiles were measured at 2 cm intervals on non-split cores, using a 72-cm-diameter BARTINGTON™loop, prior to storage in a refrigerated core repository.After core splitting in the laboratory, detailed MSprofiles were measured with a BARTINGTON MS2contact sensor (5 mm intervals). Laser grain-sizeanalyses were conducted with a MALVERN™ Mas-tersizer S apparatus (core MUCL-0202). X-ray diffrac-tion was performed on b2 μm non-carbonate fraction ofdifferent clay intervals, at the University of Lille(France). Anisotropy of magnetic susceptibility (AMS)on oriented 8 cm3 cubic boxes, with a 10 cm spacing(core MUCL-0203), were performed on a KLY-3Kappabridge using FUSAR software at the Institut dePhysique du Globe de Paris (France).

With respect to chronology, layers rich in organicmatter (particulate OM) could be used for dating. Twoseries of AMS 14C dates were produced at BetaAnalytics Radiocarbon Dating Laboratory (15 samples)and at the Poznan Radiocarbon Laboratory (2 samples).The obtained ages (on cores MUCL-0201 and MUCL-0202) were calibrated using OXCAL 3.3 software. Thedated material has different possible origins: in situaquatic production or reworking from the watershed

(peat bogs and organic soils); we did not findidentifiable terrestrial plant detritus. In case of lacustrinegenesis, a correction related to a possible reservoir effectshould have been applied, based on isotopic data on thelake water carbon content and estimation of theresidence time. Due to the uncertainty on the origin oforganic matter, we used the measured and calibratedvalues; nevertheless, these possible limitations have tobe taken in account when the chronological model isconsidered. The depth-age curve (Table 1 and Fig. 6) isbuilt using the 2σ (95%) interval and the highest modeof intersection between the calibration curve and 14Cconventional age.

4. Results

4.1. Core stratigraphy and sediment composition

Fig. 4 summarizes the main lithologies and texturaltypes, including close-ups of split cores illustrating themain sedimentary features. The main sedimentarycomponents are: siliciclastics (clay to coarse sand),and particulate organic matter. No carbonate mineralswere detected, nor are there calcareous rocks in thewatershed.

For each of the four cores, a synthetic log was ob-tained overlapping portions of successive sections,through visual correlation. The MS profiles improvedand strengthened the initial correlations. The four coresare subdivided into three main units (from base to top):

- unit I consists of laminated to thin-bedded sandy siltand clayey silt. Some layers have basal horizons withfine to medium sand (quartz, white mica, feldspars).The lowest part of the unit is a thick massive clay(H on Fig. 4) overlying coarse sand. It records a uniquedepositional event as discussed later. The proposedcorrelation of upstream Mucuñuque outcrop (MCLO,Figs. 2, 4, and 5) with unit I is based on lithology,magnetic susceptibility (Fig. 5), and 14C dating;

- unit II consists of sandy to silty clay with abundantparticulate organic matter; the unit is thin-bedded,with strata thicker than in unit I (E and I on Fig. 4).Radiocarbon ages obtained from core MUCL-0201show that the transition between unit I and unit IIdates to the beginning of the Late Glacial;

- unit III lacks visible layering and consists of silt, clay,and abundant particulate organic matter. The uppermeter of core MUCL-0203 is rich in fine aquatic-plant debris (labelled as authigenic organic matter onFig. 4). Diatom frustules were observed in smearslides throughout unit III, but not in underlying ones;

Fig. 5. Magnetic Susceptibility logs of cores MUCL-0201 to MUCL-0203 and outcrop MCLO (Bartington MS2 contact sensor, low frequency, 5 mm intervals).

290E.Carrillo

etal.

/Palaeogeography,

Palaeoclim

atology,Palaeoecology

259(2008)

284–300

Table 1Radiocarbon dating results

Sample CoreMUCL

Prof Conventional age 14C δ13C Calibrated ageδ2 intervals

Intersectionage(cm) (BP) (‰)

Beta-197598 0201 60 8180+/−60 −24.9 9290–9010 9120Beta-197602 0201 120 9620+/−50 −25.2 11,170–10,740 10,890Poz-5965 0201 176 10,490+/−50 – 12,900–12,050 12,500Beta-197599 0201 203 10,910+/−60 −22.5 13,140–12,820 12,960

12,730–12,660Beta-197600 0201 269 11,750+/−70 −25.4 14,050–13,460 13,810Beta-197601 0201 453 13,200+/−80 −23.0 16,340–15,390 15,870Beta-197603 0202 150 2630+/−50 −24.9 2800–2730 2780Poz-6092 0202 340 7890+/−50 – 8990–8580 8640Beta-197604 0202 590 10,800+/−90 −23.2 13,120–12,630 12,890

12,460–12,440Beta-197605 0202 800 13,200+/−120 −23.8 16,430–15,220 15,870

291E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

it is a minor component and variations in its contentdo not produce any distinct layering.

Four species of clay minerals were detected in theclay-rich layers of coreMUCL-0201 (location on Fig. 5):

Fig. 6. Age-depth curves of core MUCL-0201 and MUCL-0202, based on AMblack dot corresponds to higher probability value; depth positions refer to lake

chlorite (15 to 20%), illite (50 to 80%), smectite-richillite/smectite mixed-layered (5 to 30%), and kaolinite(5 to 10%). Magnetic susceptibility (Fig. 5) is generallylow (b30×10−6 SI) except in the lowest part of unit I.The general decrease in MS beginning in unit II (core

S 14C data (Oxcal Calibrated ages, horizontal error bars refer to 2σ /95%;floor, taking in account the top missing portion of Core MUCL-0201).

292 E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

MUCL-0203) is roughly parallel to the increase inorganic matter; as discussed later, other factors may ex-plain this evolution.

4.2. Chronology and lateral correlations

The radiocarbon ages are in correct stratigraphicorder and allow a detailed chronology to be establishedfor sedimentation in the lake (Table 1, Figs. 4 and 5).The boundary between units II and III coincides with thebeginning of the Holocene. The average sedimentationrate for the Holocene is low compared to that of the LateGlacial (Fig. 6), as previously measured by Weingartenet al. (1990a,b) and Yuretich et al. (1991). Detailed MSprofiles (Fig. 5) strengthen the subdivisions andcorrelations proposed for cores MUCL-0201 andMUCL-0203. Differences in the upper parts of thesetwo cores are due to coring retrieval problems(sediment-water interface not retrieved in core MUCL-0201; core MUCL-0203 stretched and ruptured), and tothe presence of a slumped interval in core MUCL-0203.

Correlations of MS profiles proposed for coresMUCL-0201 and MUCL-0202 (Fig. 5) are in goodagreement with the radiocarbon ages, and indicate ageneral thickening of sediments towards the maintributary. For the upper 6.5 m (ca. 13,000 cal. yr BP toPresent) of core MUCL-0202, the thickening isattributed to continuous uniform sedimentation withhigher regular siliciclastic input. For the lower 1.5 m ofthe core (ca. 16,000 to 13,000 cal. yr BP), major slumps(A on Fig. 4) account for the difference. If we eliminatereported slump intervals in the lower part of the section,sedimentation average rates at sites MUCL-0201 andMUCL-0202 are not significantly different.

This change in the character of sedimentationbetween Late Glacial and post-glacial times must alsobe considered in interpreting lacustrine sedimentsexposed in the upstream reaches of the moraine system(Fig. 2, white mark; MCLO log in Fig. 4). This outcropconsidered as a terrace by Salgado-Labouriau et al.(1977) is interpreted to be the product of co-seismicuplift. It consists of 4 m of faintly layered silty clayintercalated with organic-rich layers. The exposedsediment at this site section is identical to unit II andpart of unit I in cores MUCL-0201 and MUCL-0203.Salgado-Labouriau et al. (1977) give an age of 13,550–14,550 cal. yr BP for the top of these outcroppingsediments. This age is close to the one yielded at thebase of unit II. Furthermore, the top of the outcrop maybe synchronous with the major change at 6.5 m depth incore MUCL-0202 (Fig. 5). Taking only in account themorphology, the steep slope and the adjacent horizontal

shore (southeast of site MUCL-0202; see Fig. 3) couldrepresent a deltaic foreset/topset system. Nevertheless,above-mentioned remarks on lateral variations betweencores MUCL-0201 and MUCL-0202 do not favour thisinterpretation; this morphology rather represents part ofan alluvial plain cut by one of the Boconό Fault strands.

4.3. Textural characterization

In addition to direct observations, two parameters wereused to characterize sedimentary structures and deposi-tional processes: high-resolution grain-size distribution(1-cm sampling interval) and anisotropy of magneticsusceptibility (10-cm sampling interval). High-resolutiongrain-size trends for core MUCL-0201 are shown onFig. 7. The latter displays mean size and mode, which areproxies for current energy, and skewness and sorting,which provide information on depositional mechanisms.The MS profile has been added to show the non-systematic influence of grain-size on magnetic suscepti-bility, an issue discussed below. The boundary betweenunit I and unit II coincides with a change in particle size(Fig. 7). Most of the coarsest sediment is supplied prior to14,000 cal. yr BP. The boundary between unit II and unitIII corresponds with a major change in skewness (closerto 1) and mode (becoming rather constant), with no majorchange in grain size.

Three noteworthy types of sedimentary structureswereobserved: (1) slumps (close-up A, Fig. 4); (2) fractures(close-up F), and (3) massive coarse layers (close-ups Band F). The latter ones are presentmainly in coresMUCL-0201 and MUCL-0202 on, or at the toe of, the maintributary delta (Figs. 2 and 3). They comprise a non-layered coarse base, sharply overlain by a fine-grainedmassive layer. The 40-cm-thick set, labelled E on close-upB, includes a third sub-unit: 2-cm-thick cross-stratifiedsand and mud layers (flaser-bedding type) intercalatedbetween the coarse base and the fine upper part. In coresMUCL-0201 and MUCL-0203, the lower half of unit I,consisting of a 1.6 m of homogeneous silty clay, sharplyoverlies weakly stratified coarse sand; we tentativelyconsider the whole as similar to the 40-cm-thick setpreviously mentioned. The corresponding part of the MSprofiles (Fig. 5) underlines the lack of transition betweenthe coarse base and the homogeneous fine upper part.Grain-size parameters display a similar contrast: the upperpart (below 450 cm) is homogeneous and the base showssuccessive sharp high-amplitude changes.

Post-depositional, brittle or plastic, soft-sedimentdeformations are observed in unit I from cores MUCL-0201,MUCL-0203, and also in unit II from coreMUCL-0202 (close-ups A, D, and F, Fig. 4). The strongest

Fig. 7. Detailed evolution of grain-size distribution in core MUCL-0201. Magnetic susceptibility curve is added to test relationships between textureand mineralogy.

293E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

disturbances are in core MUCL-0203 which is close tothe steep southern margin of the lake and to the BoconόFault. MS responses (Fig. 5) appear to coincide: there is alack of variation in the slumped part (upper half of unit I).

Anisotropy of magnetic susceptibility (AMS) has beenused as an indicator of sedimentary fabric (Elwood, 1980;Joseph et al., 1998; Thouveny et al., 2000, 2004). Shapeand size of magnetic particles, and hydrodynamicconditions during settling represent the primary factorsfor AMS. Diagenesis (compaction, chemical and miner-alogical reactions) and bioturbation can modify theoriginal fabric. Re-suspension, slumping, liquefaction,and other mechanical processes may also have a strongimprint (in Tarling andHrouda, 1993). Some authors havesuggested that the anisotropy of magnetic susceptibilitycan be used as an indicator of shock-induced disturbance(Beck et al., 1992; Chapron, 1999). In a lake in thenorthwestern Alps, a homogeneous mud layer 30 cm

below the sediment-water interface displays a strongmagnetic foliation. The latter could not be explained bycompaction, and was attributed by Chapron et al. (1999)to the particular settling of suspended sediment affectedby a seiche-induced oscillation. Similarly, we use AMSdata to argue for changes in depositional, or post-depositional, processes caused by earthquakes. CoreMUCL-0203was sampled for AMS analysis immediatelyafter splitting. Intensity of magnetic susceptibility ismeasured along different axes and its 3D distribution isrepresented by an ellipsoid; K1, K2, and K3 are themaximum, intermediate, and minimum susceptibilityvalues along orthogonal axes. They are combined tocharacterize the shape of the ellipsoid. Fig. 8 shows two ofthem: the anisotropy degree P′ (higher values representstronger degree of anisotropy), and the shape parameterT (Jelinek, 1981) indicating a disk-shape (positive values)or a rod-shape (negative value). Values of inclination

Fig. 8. Parameters of anisotropy of magnetic susceptibility (AMS) measured in core MUCL-0203. T parameter=general shape of MS ellipsoid.P′=degree of anisotropy. K3 Index= inclination ellipsoid short axis K3. Volume magnetic susceptibility curve is added to check theindependence of AMS from magnetic content.

294 E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

of K3 axis (IK3), with respect to the horizontal plane,are added.

The oriented sediment cubes were taken from theaxis of the core to avoid coring disturbances developedalongside the PVC liners. These disturbances are clearlyvisible in units I and II as affecting the initial layering.The 50–130 cm section of core MUCL-0203 (Fig. 8)could not be correctly sampled due to the consistency ofthe sediment. In the upper 50 cm, lacking visiblelayering, coring disturbance could not be visuallydetected; nevertheless, considering the high organicmatter and water contents of this section, we assumed itcould have been strongly disturbed. Thus we attributethe strong rod-shape, in parallel with a quite horizontalK3 axis, to coring problems. The rest of the logs can bedivided as follow:

- from600 cm to 470 cm, a disk-shape iswell developed,except at the top of the homogeneous fine-grainedlayer. A strong anisotropy (foliation) and an inclinationof the ellipsoid also characterize this interval.

- from 470 cm to 270 cm, significant modification ofanisotropy (disk-shape to almost spherical) appears inparallel with IK3 fluctuations; MS is unchanging,

indicating a predominant influence of texture onAMS;

- the interval between 270 cm and 130 cm has aconstant T parameter while P′ decreases; the volumemagnetic susceptibility (MS) decreases indicatingparallel change in mineralogical content. The disk-shape ellipsoid stays horizontal as indicated by IK3values close to 90°.

5. Interpretation and discussion

5.1. Depositional processes

Apart from the “event” deposits discussed in the nextparagraph, the analyzed part of the sedimentary fill ofLake Mucubaji consists of:

- siliciclastic terrigenous input: silt and clay, and, inminor proportion, fine sand;

- primary production: diatoms.

A third component, organic matter, may either belongto lake production (sub-aquatic vegetation) or to bereworked from the catchment (soils and peat bogs).

295E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

Mineral detritus constitute the bulk of units I and II.Grain-size parameters (mean-size and distributionspectra) indicate deposition from suspended-load andfrom bed-load driven by slow currents. The millimeter-to centimeter-scale laminations – mainly in units I andII – are related to slight variations in the relative roles ofthese two processes. These rhythmic changes are thusrelated to fluctuations in the run-off from the watershed.The progressive decrease of magnetic susceptibility,starting in the upper half of unit II, can be associatedwith an increase in organic content; the parallel decreaseof anisotropy of magnetic foliation can thus beexplained by a relative decrease in phyllosilicateparticles. In the upper part of unit III, lower sedimen-tation rate and higher organic content (MS null) favourthe hypothesis of a production within the lake.

Among the detected clay minerals, kaolinite occur-rence has to be discussed. Weingarten et al. (1990a,b)considered unlikely that kaolinite is currently beingformed in the Lake Mucubají watershed. It may bereworked from previous palaeosoil alteration. We addtwo other possibilities: eolian transport from lowaltitude (undergoing tropical wet/dry alternation), or ahydrothermal alteration along faults.

In the four cores, we did not find any indication ofproglacial environments, such as dropstones, and not evencoarse sand grains dispersed in mud, which could haveindicated episodes of lake-surface freezing.We comparedthe cored succession with the contemporaneous infill ofthe neighbouring Los Zerpa palaeolake (Beck et al., 2005;Carrillo et al., 2006). At this site (10 km NE of LakeMucubají), we did not observe any gravel or coarse clastsimbedded in the outcropping fine-grained lacustrinesediments. Based on these observations, we dismiss anyglacigenic/glacitectonic process to explain some of thedisturbances discussed hereafter.

5.2. “Event” deposits and abrupt changes of LakeMucubají’s geometry

Two types of depositional “events” are intercalatedwithin the “normal” continuous sedimentation: coarsedeposits and slumps (three major ones in unit I andlower unit II). Together with several abrupt changes insedimentation (Figs. 4 and 7) they will be emphasized aspossible indicators of regional tectonic activity.

The occurrence of coarse deposits, mainly interca-lated in units I and II, requires conditions favouring boththe genesis and the storage of terrigenous components inthe watershed, as well as appropriate energy fortransport and spreading on the lake floor. Thick sandlayers from the base of unit I (cores MUCL-0201 and

MUCL-0203, Fig. 4) lack any clear type of stratifica-tion, which could have argued for a long polyphasicaccumulation. Textural parameters and lack of system-atic MS changes in these layers (Fig. 7) reinforce theirinterpretation as grain flows. We assume the sameinterpretation for thinner sand layers from units II and III(Fig. 4), especially in core MUCL-0202 which is closeto the steep southeastern slope and shows a highersedimentation rate (see correlations on Fig. 5). Althoughthese grain flows may be explained by flooding events,we underline their frequent association with overlyinghomogeneous fine-grained layers (unit I of coresMUCL-0201 and MUCL-0203). The association oflayers, labelled E on close-up C (Fig. 4), is similar towhat has been ascribed to a unique gravity deposit bydifferent authors (Pickering and Hiscott, 1985; Chapronet al., 1999; Lignier, 2001; Beck et al., 2007). They haveunderlined the specificity of turbidites displaying asharp separation between a highly homogeneous fine-grained upper part and a coarse lower layer. Anintermediate thin composite layer may occur, which isinterpreted as an indicator of “to-and-fro” particledisplacement (“seiche effect”) following an earthquakeand mass-wasting. These inferred oscillations arethought to be responsible for the strong segregation offine-grained components (suspended-load) from thecoarser fraction (bed-load). As a hypothesis, weconsider the basal thick homogenite (cores MUCL-0201 and MUCL-0203) as associated with underlyingsand in a unique depositional event similar to a seismo-turbidite. In core MUCL-0203, AMS parameters (strongfoliation, parameter P′, Fig. 8) confirm a specific arrayof phyllosilicate particles. IK values of upper unit I andlower unit II (450 cm to 300 cm) can be related withslumps observed in fine-grained laminated sediments.Both types of sedimentary events (sand plus homo-genite, slump) are considered as related to earthquake-induced reworking. We also underline the stratigraphicposition of the radiocarbon-dated samples: fine-grainedOM at the top of major reworking events, possiblyconcentrated during final settling of lighter suspendedparticles. Thus, most of the obtained ages could directlyrepresent those of these events, taking in account theimprecision due to the reworking of the OM itself andthe possible mixing of different sources of organicdetritus. In cores MUCL-0201, MUCL-0202, andMUCL-0203, coarse deposits and slumps are developedin units I and lower unit II, before approximately13,000 cal. yr BP. This is similar to what has beenreported from other post-glacial lakes or lacustrinebasins (Beck et al., 1996; Van Rensbergen et al., 1999;Beck et al., 2003) in which immediate post-LGM

296 E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

millennia are characterised by high terrigenous input.Thus, high sedimentation rates and outbuilding of deltasare thought to enhance the triggering of mass-wasting.In parallel, above-mentioned authors underline thepossible additional effect of a higher seismic activityrelated to the rapid disappearance of the LGM ice load.

Apart from these sedimentary “events”, the coredsuccession shows some abrupt changes shown on Fig. 7for core MUCL-0201: at 370 cm, at 305 cm (limit unit I/unit II), at 170 cm (limit unit II/unit III), and at 80 cm. Thetwo first modifications do not show equivalent shifts onthe MS curve; thus, these textural changes correspond tomodifications of depositional dynamics rather than to

Fig. 9. Schematic post-LGM evolution of Lake Mu

changes in the production of the sediment components (inthe catchment or within the lake). We link these changesto abrupt modifications of the lake geometry (surface and/or depth). Reconstructed successive longitudinal profilesof Fig. 9 display this interpretation. The disappearance oflacustrine sedimentation in the upstream part of Mucu-ñuque Creek (already discussed above) could correspondto the second and major change around 14,000–14,500 cal. yr BP. According to historical magnitudesand co-seismic fault offsets measured in trenches in thesame area (Audemard et al., 1999, 2000), each of thedifferent modifications here proposed for Lake Mucubajícould represent the consequence of a major earthquake.

cubají geometry along a longitudinal profile.

297E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

The 2700–2800 cal. yr BP (core MUCL-0203) eventcorresponds to a gravity reworking deposit but was notnecessarily followed by significant change of the lake’sgeometry. The overall cored sediments in Lake Mucubajípossibly recorded at least four major co-seismic dis-turbances and/or geometrical modifications. One of thepalaeo-earthquakes detected both in the neighbouring LosZerpa lacustrine sediments and in theMorro de los Hoyostrench (Audemard et al., 1999; Carrillo et al., 2006) alsofits with the 9000–9300 cal. yr BP event recorded in LakeMucubají.

5.3. Post-LGM palaeoclimatic record

Base on radiocarbon ages and lateral correla-tions (Figs. 5 and 6, Table 1), a curve of magneticsusceptibility versus time has been constructed for coreMUCL-0202 (Fig. 10). This curve was established afterremoving the coarse deposits considered as instanta-neous events, and thus concerns only the permanentsedimentation. Slumped parts, before 13,000 cal. yr BP,

Fig. 10. Tentative correlations of the record of core MUCL-0202 with region(2003). Geenland ice-core data from Brook et al. (1996). Dotted line: lower

were not removed and will not be discussed. MS waschosen as a palaeoenvironmental proxy often linked tothe importance and characteristics of terrigenous input(in Evans and Heller, 2003). On our different profiles(Fig. 7), higher values do not appear systematicallyrelated to coarser sediments. The watershed is entirelycomposed of the same granitic and gneissic association,ruling out a possible impact of changes in source area.Thus, we consider that MS is here mainly linked to theintensity of alteration. On Fig. 10, the obtained curve iscompared with two published post-LGM palaeoclimatecurves of the northern-hemisphere: Greenland ice-coredata from Brook et al. (1996) and data from Cariacobasin from Lea et al. (2003). For the last 13,000 years,several correlations can be suggested. The YoungerDryas (YD on Fig. 10) cooling appears here as a low inMS, and the abrupt warming at the onset of theHolocene (EH) as a shift to the highest measuredvalues. Although it is not possible to assume synchro-nism due to chronological imprecision, several generalvariations or peaks can be tentatively correlated.

al and global climatic fluctuations. Cariaco Basin data from Lea et al.limit of possible correlations.

298 E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

6. Conclusions

During the analyzed last 16,500 years of sedimen-tation in Lake Mucubají, a few abrupt changes andseveral specific events occurred. Co-seismic displace-ments along the Boconó Fault are inferred to beresponsible for these major sedimentary events. Theirrecurrence time interval – about 1200 years – iscompatible with the results obtained from neighbouringLos Zerpa palaeolake and from trench data (Audemardet al., 1999; Carrillo et al., 2006). One event registeredin Lake Mucubají can be correlated between threedifferent palaeoseismic recording sites.

Despite these changes and disturbances, a generalpost-LGM global climatic evolution could also bederived from the sedimentary record. Except in unit I(homogenite and slumps), undisturbed laminated por-tions recorded – as in the synchronous Los Zerpadeposits – decadal to centennial climatic fluctuations.Thus, our study shows that it is possible to discriminatethe respective impacts of Late Quaternary climatechanges and local seismo-tectonic activity. To disentan-gle these two factors, Lake Mucubají’s sedimentary fillwas examined at three scales: 1) at first-order, theoverall cored succession, which displays a completeLate Glacial and Holocene general palaeoclimaticsequence (Figs. 7 and 10); 2) at second-order,decimeter-scale subdivisions based on textural changes,which are attributed to major seismic events andassociated fault offsets; 3) at third-order, centimeter-scale changes in layering or lamination, which mayrepresent high-frequency fluctuations in precipitation.

Acknowledgements

The authors thank the Venezuelan Foundation forSeismological Research, the Venezuelan NationalFoundation for Science and Technology (FONACITgrants 20001002492 and 2003000090), and the FrenchNational Council for Scientific Research (UMR 5025),for funding and logistic support. The French Ministry ofForeign Affairs, through EGIDE and the FrenchEmbassy in Caracas (Technological and CulturalCooperation), supported our joint Venezuelan-Frenchscientific program; we specifically thank Yves Yard andEstrella Marciano. This work is also supported by the2001002492 and 2003000090 FONACIT grants. Mrs.Maura Elena Remiro (Instituto Nacional de Parques),and Mrs. Ana Elisa Osorio Granado, Minister ofEnvironment and Natural Resources, provided permis-sion to core Lake Mucubají. E. Carrillo’s research andstay in the Laboratoire de Géodynamique des Chaînes

Alpines were supported by the Scientific and Human-istic Development Centre of the Central University ofVenezuela. Bernard Henry and Maxime Legoff (Institutde Physique du Globe de Paris) assisted with themeasurements of the anisotropy of magnetic suscepti-bility. Engineers Raymi Castilla, Luis Melo, and VictorCano helped with coring, and María AuxiliadoraChacón assisted for customs clearance. Philippe Recourtperformed the X-ray diffraction measurements. We arealso especially grateful to the manager, his wife, and thestaff, of the Hotel Santo Domingo for allowing us theuse of the hotel as our home, office, core repository andlaboratory, during field work. J. Clague and D. Ariztéguigreatly helped to clarify the discussion and to improvethe manuscript.

References

Arnaud, F., Lignier, V., Revel, M., Desmet, M., Beck, C., Pourchet,M., Charlet, F., Trentesaux, A., Tribovillard, N., 2002. Flood andearthquake disturbance of 210Pb geochronology (Lake Anterne,NWAlps). Terra Nova 14, 225–232.

Audemard, F.E., Audemard, F.A., 2002. Structure of the MéridaAndes, Venezuela: relations with the South America-Caribbeangeodynamic interaction. Tectonophysics 345, 299–327.

Audemard, F.A., Pantosti, D., Machette, M., Costa, C., Okumura, K.,Cowan, H., Diederix, H., Sawop Participants, 1999. Trenchinvestigation along the Merida section of the Boconó fault (centralVenezuelan Andes). Tectonophysics 308, 1–21.

Audemard, F.A.,Machette,M.,Cox, J., Hart, R., Haller, K., 2000.MapofQuaternary faults of Venezuela. Scale 1:2,000,000; jointly publishedwith database of Quaternary faults in Venezuela and offshore regions(USGSOpen-File Report 00-18; 78 p). A project of the InternationalLithosphere Program Task Group II-2: major active faults of theworld (Regional Coord.: Carlos Costa, Univ. San Luis-Argentina;ILP II-2 co-chairman Western Hemisphere: Michael Machette,USGS-Colorado).Also available fromUSGSwebpage. Reprinted asa XXX FUNVISIS Anniversary special edition.

Audemard, F.A., Beck, C., Carrillo, E., Cousin, M., Paterne, M., 2001.Sedimentary record of late-Pleistocene seismic activity alongthe Boconó Fault (Mérida Andes): preliminary observations on theLos Zerpas Moraine-dammed paleo-Lake. 8ème Congrès Associ-ation des Sédimentologues Français. Publications ASF, vol. 36,p. 19.

Audemard, F.A., Beck, C., Carrillo, E., Cousin, M., 2002a. LatePleistocene–Holocene major earthquakes along the Boconó Fault(Mérida Andes, Venezuela): sedimentary record in the Los Zerpamoraine-dammed paleo-lake. Abstract (Oral. Pres.) EnvironmentalCatastrophes and Recovery in the Holocene, INQUA. BrunelUniversity, London, pp. 14–15. 28/08-2/09 September 2002.

Audemard, F.A., Beck, C., Cano,V., Carrillo, E., Castilla, R., Cousin,M.,Jouanne, F., Melo, L., Villemin, T., 2002b. Stability of Quaternarymorainic complexes cut by the Boconó Fault, Mérida Andes,Western Venezuela. 5th International Symposium on AndeanGeodynamics, Toulouse, 16–18 September 2002, pp. 49–52.

Beck, C., Rochette, P., Tardy, M., 1992. Interprétation en termes depaléosismicité de niveaux déstructurés intercalés dans des ryth-mites lacustres quaternaires des AlpesNord-Occidentales. ComptesRendus de l’Académie des Sciences, Paris, t. 315, s. II, 1525–1532.

299E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

Beck, C., Manalt, F., Chapron, E., Van Rensbergen, P., De Batist, M.,1996. Enhanced seismicity in the early post-glacial period:evidence from the post-Würm sediments of Lake Annecy,northwestern Alps. Journal of Geodynamics 22, 155–171.

Beck, C., Schneider, J.-L., Cremer, M., Mercier de Lépinay, B.,Çagatay, N., Labeyrie, L., Turon, J.-L., Wendenbaum, E.,Boutareau, S., Ménot-Combes, G., Hadjas, I., Cortijo, E.,MARMACORE Leg Shipboard Scientific Party, 2003. LatePleistocene major sedimentary reworking event (homogenite) inthe Marmara Sea Central Basin: a combination of Late Glacial highterrigenous supply with a major earthquake ? Preliminary results ofgiant piston-coring and high-resolution seismic reflexion. Amer-ican Geophysical Union-European Geophysical Society-EuropeanUnion of Geoscientists, joint meeting, Nice (April 2003).

Beck, C., Carrillo, E., Audemard, F.A., Ollarves, R., Diaz, G., Cano,V., 2005. Assessing paleoseismic sedimentary records reliability:compared case studies in Venezuelan Andes and NorthwesternAlps. Geophysical Research Abstracts 7 (AGU05-A-08059).

Beck, C., Mercier de Lépinay, B., Schneider, J.-L., Cremer, M., Çağatay,N., Wendenbaum, E., Boutareaud, S., Ménot-Combes, G., Schmidt,S.,Weber, O., Eris, K., Armijo, R.,Meyer, B., Pondard, N., Labeyrie,L., Cortijo, E., Turon, J.-L., Malavielle, J., Dominguez, S., Gutscher,M.-A., 2007. Late Quaternary co-seismic sedimentation in theSea ofMarmara's deep basins. In: Bourrouilh-Le Jan, F.G., Beck, C.,Gorsline, D. (Eds.), Sedimentary Records of Catastrophic Events.Special Issue. Sedimentary Geology, vol. 199 (1/2), pp. 65–89.

Bellizzia, A., Pimentel, N., Bajo de Osuna, R., 1976. Mapa geológico-estructural de Venezuela. Scale 1:500,000. Ministerio de Minas eHidrocarburos, Ed. Foninves, Caracas.

Ben-Menahem, A., 1976. Dating historical earthquakes by mud pro-files of lake-bottom sediments. Nature 262, 200–202.

Bradley, R.S., Yuretich, R., Salgado-Labouriau, M.L., Weingarten, B.,1985. Late Quaternary paleoenvironmental reconstruction usinglake sediments from the Venezuelan Andes: preliminary results.Zeitschrift für Gletscherkunde und Glazialgeologie 21, 97–106.

Brook, E.J., Sowers, T., Orchardo, J., 1996. Rapid variations inmethane atmospheric concentration during the past 110000 yrs.Science 273, 1087–1091.

Calvo, J.P., Rodríguez-Pascua, M., Martín-Velásquez, S., Jiménez, S.,De Vicente, G., 1998. Microdeformation of lacustrine laminitesequences from Late Miocene formations of SE Spain: aninterpretation of loop bedding. Sedimentology 45, 279–292.

Carrillo, E., Audemard, F.A., Beck, C., Cousin, M., 2002. Sedimentarydisturbances in Late Quaternary fluvio-lacustrine deposits of theLos Zerpa moraine, induced by the Boconó Fault, Mérida Andes,Venezuela. 5th International Symposium on Andean Geody-namics. Toulouse, 16-18 September 2002, pp. 125–128.

Carrillo, E., Audemard, F., Beck, C., Cousin, M., Cano, V., Castilla,R., Melo, L., Villemin, T., 2006. A late Pleistocene-Holocenenatural seismograph along the Boconó Fault (Merida Andes,Venezuela): the moraine-dammed Los Zerpa paleo-lake. Bulletinof the French Geological Society 177, 13–17.

Chapron, E., 1999. Contrôles climatiques et sismo-tectonique de lasédimentation lacustre dans l’Avant-Pays Alpin (Lac du Bourget)durant le Quaternaire récent. Géologie Alpine, Mémoire H.S. no. 30,Université J. Fourier, 261 pp.

Chapron, E., Beck, C., Pourchet, M., Deconinck, J.-F., 1999. 1822 ADearthquake-triggered homogenite in Lake Le Bourget (NWAlps).Terra Nova 11, 86–92.

Cita, M.B., Rimoldi, B., 1997. Geological and geophysical evidencefor a Holocene tsunami deposit in the eastern Mediterranean deep-sea record. Journal of Geodynamics 24, 293–304.

Clague, J.J., Evans, S.G., 2000. A review of catastrophic drainage ofmoraine-dammed lakes in British Columbia. Quaternary ScienceReviews 19, 1763–1783.

Doig, R., 1991. Effects of strong seismic shaking in lake sediments,and earthquake recurrence interval, Temiscaming, Québec.Canadian Journal of Earth Sciences 28, 1349–1352.

El-Isa, Z.H., Mustafa, H., 1986. Earthquake deformations in the Lisandeposits and seismotectonic implications. Geophysical Journal ofthe Royal Astrophysical Society 86, 413–424.

Elwood, B.B., 1980. Application of the anisotropy of magneticsusceptibility method as an indicator of bottom water flowdirection. Marine Geology 24, 83–90.

Evans, M.E., Heller, F. (Eds.), 2003. Environmental Magnetism.Academic Press, Elsevier Science, Amsterdam, 299 pp.

Field, M.E., Gardner, J.V., Jennings, A.E., Edwards, B.D., 1982.Earthquake-induced sediment failure on a 0.25° slope, KlamathRiver delta, California. Geology 10, 542–546.

Föhlisch, K., Voigt, T., 2001. Synsedimentary deformation in theLower Muschelkalk of the Germanic Basin. In: McCaffrey, W.D.,Kneller, B.C., Peakall, J. (Eds.), Particulate Gravity Currents.International Association of Sedimentologists Special Publication,vol. 31. Blackwell Science, pp. 279–297.

Giegengack, R., 1984. Late Cenozoic tectonic environments of thecentral Venezuelan Andes. Geological Society of America Memoir162, 343–364.

Giegengack, R., Grauch, R.I., Sahagam, R., 1976. Geometry of LateCenozoic displacement along the Boconó Fault, VenezuelanAndes. Boletín de Geología Publicación Especial 7/2, 1201–1223.

Gorsline, D.S., De Diego, T., Nava-Sanchez, E.H., 2000. Seismicallytriggered turbidites in small margin basins: Alfonso Basin, WesternGulf of California and Santa Monica Basin, California Borderland.Sedimentary Geology 135, 21–35.

Hackley, P.C., Urbani, F., Karlsen, A.W., Garrity, C.P., 2005. Geologicshaded relief map of Venezuela. Open-File Report — U. S.Geological Survey. U.S. Geological Survey, Reston, VA, UnitedStates (2 sheets).

Hempton, M.R., Dewey, J.F., 1983. Earthquake-induced deformation-al structures in young lacustrine sediments, East Anatolian Fault,southwest Turkey. Tectonophysics 98, 7–14.

Henneberg, H.G., Schubert, C., 1986. Geodetic networks along theCaribbean–South-American plate boundary. Tectonophysics 130,77–94.

Jelinek, V., 1981. Characterization of the magnetic fabric of rocks.Tectonophysics 79, 63–73.

Joseph, L.H., Rea, D.K., van der Pluijm, B.A., 1998. Use of grain sizeand magnetic fabric analyses to distinguish among depositionalenvironments. Paleoceanography 13, 491–501.

Lea, D.W., Pak, D.K., Peterson, L.C., Hughen, K.A., 2003.Synchroneity of tropical and high-latitude Atlantic temperaturesover the last glacial termination. Science 301, 1361–1364.

Lignier, V., 2001. Les sédiments lacustres et l’enregistrement de lapaleoséismicité. Etude comparative de différents cas dans leQuaternaire des Alpes Nord-Occidentales et du Tien-ShanKyrghize. Ph.D. thesis, Université de Savoie, France.

Lignier, V., Beck, C., Chapron, E., 1998. Caractérisation géométriqueet texturale de perturbations synsédimentaires attribuées à desséismes, dans une formation quaternaire glaciolacustre des Alpes(les “Argiles du Trièves”). Comptes Rendus de l’Académie desSciences. Paris 327, 645–652.

Mahaney, W., Kalm, V., Bezada, M., 1997. Estratigrafía delCuaternario tardío de un ambiente proglacial en el area deMucubají, Mucuchache, El Pedregal. Andes Venezolanos.

300 E. Carrillo et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 259 (2008) 284–300

Memorias Ier Congreso Latinoamericano de Sedimentología,Sociedad Venezolana de Geólogos, I, pp. 417–424.

Mahaney, W., Milner, M.W., Voros, J., Kalm, V., Hütt, G., Bezada, M.,Hancock, M.G.V., Aufreiter, S., 2000. Stratotype for the MéridaGlaciation at Pueblo Llano in the northern Venezuelan Andes.Journal of South American Earth Sciences 13, 761–774.

Marco, S., Agnon, A., 1995. Prehistoric earthquake deformations nearMasada, Dead Sea Graben. Geology 23, 695–698.

Moretti, M., Alfaro, P., Caselles, O., Canas, J.A., 1999. Modellingseismites with a digital shaking table. Tectonophysics 304,369–383.

Mulder, T., Cochonat, P., 1996. Classification of offshore massmovements. Journal of Sedimentary Research 66, 43–57.

Nakajima, T., Kanai, Y., 2000. Sedimentary features of seismoturbi-dites triggered by the 1983 and older historical earthquakes in theeastern margin of the Japan Sea. Sedimentary Geology 135, 1–19.

Pérez, O., Bilham, R., Bendick, R., Hernández, N., Hoyer, M.,Velandia, J., Moncayo, C., Kozuch, M., 2001. Velocidad relativaentre las placas del Caribe y Sudamérica a partir de observacionesdentro del sistema de posicionamiento global (GPS) en el norte deVenezuela. Interciencia 26, 69–74.

Pickering, K.T., Hiscott, R.N., 1985. Contained (reflected) turbiditycurrents in the Middle Ordovician Cloridorme Formation, Québec,Canada: an alternative to the antidune hypothesis. Sedimentology32, 373–394.

Piper, D.J.W., Cochonat, P., Ollier, G., Le Drezen, E., Morrison, M.,Baltzer, A., 1992. Evolution progressive d’un glissement rota-tionnel en un courant de turbidité: cas du séisme de 1929 desGrands Bancs (Terre Neuve). Comptes Rendus de l’Académie desSciences. Paris 314, 1057–1064.

Plaziat, J.-C., Purser, B.H., Philobbos, E., 1988. Diversity of Neogeneseismites of the NW Red Sea (Egypt): a characteristic sedimentaryexpression of rifting. Tectonophysics 153, 281–295.

Rod, E., 1956. Strike-slip faults of northern Venezuela. Bulletin of theAmerican Association of Petroleum Geologists 40, 457–476.

Rodríguez-Pascua, M.A., Calvo, J.P., De Vicente, G., Gómez-Gras, D.,2002. Soft-sediment deformation structures interpreted as seis-mites in lacustrine sediments of the Prebetic Zone, SE Spain, andtheir potential use as indicators of earthquake magnitudes duringthe Late Miocene. Sedimentary Geology 135, 117–135.

Rull, V., 1996. Late Pleistocene and Holocene climates of Venezuela.Quaternary International 31, 85–94.

Rull, V., 1999. Palaeoclimatology and se-level history in Venezuela.New data, land-sea correlations and proposals for future studies inthe frame of the IGBP-PAGES Project. Interciencia 24/2, 92–101.

Rull, V., Vegas-Vilarrubia, T., 1998. Palynomorphs and other thanPollen and Fern Spores in Glacial Sediments from the VenezuelanAndes. Preliminary assessment on their potential paleoecologicalvalue. Boletín de la Sociedad Venezolana de Geólogos 23, 5–27.

Salgado-Labouriau, M.L., Schubert, C., 1976. Palynology of Holo-cene peat bogs from the central Venezuelan Andes. Palaeogeo-graphy, Palaeoclimatology, Palaeoecology 19, 147–156.

Salgado-Labouriau, M.L., Schubert, C., Valastro Jr., S., 1977.Paleoecologic analysis of a Late-Quaternary terrace from Mucu-bají, Venezuelan Andes. Journal of Biogeography 4, 313–325.

Salgado-Labouriau, M.L., Bradley, R.S., Yuretich, R., Weingarten, B.,1992. Paleoecological analysis of the sediments of Lake Mucubaji,Venezuelan Andes. Journal of Biogeography 4, 313–325.

Schubert, C., 1974. Late Pleistocene Mérida Glaciation, VenezuelanAndes. Boreas 3, 147–152.

Schubert, C., 1980. Late Cenozoic pull-apart basins, Boconó fault zone,Venezuelan Andes. Journal of Structural Geology 2, 463–468.

Schubert, C., 1981. Evolución post-glacial de un valle morrénico,Andes Merideños. Acta Científica Venezolana 32, 151–158.

Schubert, C., 1982. Neotectonics of Boconó Fault, western Venezuela.Tectonophysics 85, 205–220.

Schubert, C., Henneberg, H.G., 1975. Geological and geodetic investi-gations on themovement along the Boconό fault, VenezuelanAndes.Tectonophysics 29/1–4, 199–207.

Schubert, C., Vivas, L., 1993. El Cuaternario de la Cordillera deMérida, Andes Venezolanos. Universidad de Los Andes, Funda-ción POLAR, Mérida, 345 pp.

Siegenthaler, C., Finger, W., Kelts, K., Wang, S., 1987. Earthquakeand seiche deposits in Lake Lucerne, Switzerland. EclogaeGeologicae Helveticae 80, 241–260.

Sims, J., 1975. Determining earthquake recurrence intervals fromdeformational structures in young lacustrine sediments. Tectono-physics 29, 141–152.

Stéphan, J.-F., 1985. Andes et Chaîne Caraïbe sur la transversale deBarquisimeto (Vénézuela). Evolution géodynamique. Symposiumon Caribbean Geodynamics, Ed. Technip, Paris, pp. 505–529.

Tarling, D.H., Hrouda, F., 1993. The Magnetic Anisotropy of Rocks.Chapman and Hall, London. 217 pp.

Thouveny, N., Moreno, E., Delanghe, D., Candon, L., Lancelot, Y.,Shackleton, N.J., 2000. Rock-magnetic detection of distal Ice RaftedDebries: clue for the identification of Heinrich layers on thePortuguesesMargin. Earth andPlanetary Science Letters 182, 61–75.

Thouveny, N., Carcaillet, J., Moreno, E., Leduc, G., Nérini, D., 2004.Geomagnetic moment variation and paleomagnetic excursionssince 400 kyrs BP: a stacked record from sedimentary sequences ofthe Portuguese margin. Earth and Planetary Science Letters 219/3–4,377–396.

Van Loon, A.J., Brodzikowski, K., Zielinski, T., 1995. Shock-inducedresuspension deposits from a Pleistocene proglacial lake (Kleszczowgraben, central Poland). Journal of Sedimentary Research 65/2,417–422.

Van Rensbergen, P., De Batist, M., Beck, C., Chapron, E., 1999. High-resolution seismic stratigraphy of glacial to interglacial fill of adeep glacigenic lake: Lake Le Bourget, Northwestern Alps,France. Sedimentary Geology 128, 99–129.

Weingarten, B., Yuretich, R.F., Bradley, R.S., Salgado-Labouriau,M.L., 1990a. Environmentally-controlled variations in claymineral assemblages from lake sediments in the VenezuelanAndes. In: Farmer, V.C., Tardy, Y. (Eds.), Proc. 9th InternationalClay Conference, Strasbourg, Sciences Géologiques. Mémoire,vol. 88, pp. 53–61.

Weingarten, B., Yuretich, R.F., Bradley, R.S., Salgado-Labouriau,M.-L., 1990b. Characteristics of sediments in an altitudinalsequence of lakes in the Venezuelan Andes: climatic implications.Journal of South American Earth Sciences 3, 113–124.

Yuretich, R., Weingarten, B., Bradley, R., Salgado-Labouriau, M.L.,1991. Late Quaternary climatic evolution of the VenezuelanAndes. Final Report NSF Grant ATM83-03171. 158 pp.