Pile dwellers occupation gap in Lake Geneva (France-Switzerland) possibly explained by an

Transcription

1 Pile dwellers occupation gap in Lake Geneva (France-Switzerland) possibly explained by an earthquake mass movement tsunami event during Early Bronze Age Katrina Kremer a,b*, François Marillier c, Michael Hilbe d,e, Guy Simpson a, David Dupuy c, Ble J.F. Yrro c,1, Anne-Marie Rachoud-Schneider f, Pierre Corboud g, Benjamin Bellwald e, Walter Wildi h, Stéphanie Girardclos a,b a Department of Earth Sciences, University of Geneva, Rue des Maraîchers 13, 1205 Geneva, Switzerland b Institute for Environmental Sciences, University of Geneva, Site de Batelle, Carouge, Switzerland c Research Centre in Terrestrial Environment, University of Lausanne, Geopolis Bldg., 1015 Lausanne, Switzerland d Eawag, Swiss Federal Institute of Aquatic Science and Technology, Ueberlandstrasse 133, 8600 Dübendorf, Switzerland e Institute of Geological Sciences, University of Bern, Baltzerstrasse 1+3, 3012 Berne, Switzerland f Musée et Jardins botaniques cantonaux, Avenue de Cour 14bis, 1007 Lausanne, Switzerland g Archéologie préhistorique et anthropologie, Institute Forel, University of Geneva, Route des Acacias 18, 1211 Geneva, Switzerland h Institute F.-A. Forel, University of Geneva, Route de Suisse 10, 1290 Versoix, Switzerland *Corresponding author: Postal address: Katrina Kremer Department of Earth Sciences Rue des Maraichers Geneva Tel: katrina.kremer@unige.ch 1 Present address: Département des génies civil, géologique et des mines, Polytechnique Montreal, C.P. 6079, Montréal, Quebec, Canada

2 Appendices: Appendix A: Supplementary Figures Figure A.1: Earthquakes (Mw > 2) distribution in Lake Geneva region (Earthquake catalogue of Switzerland ECOS09 from ; Fäh et al., 2011; Earthquakes (Mw > 4.5) occurred mostly in the Prealpine region (green letters a to h refer to Table A1). During the period of time covered by the catalogue, the northern side of Lake Geneva was seismically less active, with only Mw 2-4 earthquakes, mostly located near fault zones of the Molasse units.

3 Figure A.2: Watergun seismic reflection profiles showing the location of multiple failure scars between 80 and 220 m water depth (points and dotted black lines, inferred from multibeam data, see also Fig. 4). Watergun profiles a to e are positioned on the bathymetric map (f) with red lines. Delimitation of the failure scars (black lines on bathymetry map) are inferred from multibeam data (Fig. 4).

4 Figure A.3: Thickness map showing the thickness distribution and volumes of MTDs A, B and C. MTD A has two depocentres (A1 and A2). Thickness contour line interval is 1 m from 3 to 10 m and 10 m from 10 to 20 m. Detected failure scars (Figs. 4 and A2) and arrows indicate the source area of MTD A.

5 Figure A.4: Photographs detail the different sediment facies cored in MTD A. The main cored facies consists of ~2 m of deformed and folded laminated mud (1). Muddy matrix with

6 mudclasts follows over 0.5 m (2). MTD A ends with a 0.1 m thin layer of homogeneous, darker, faintly laminated mud topped with a thin, fine-grained white layer (3). Same legend as in Fig.5.

7 Figure A.5: Calculated tsunami wave-height in metres as a function of time in minutes simulated over 110 minutes for the main cities and the archaeological sites on the shores of

8 Lake Geneva. These wave spectra show that, for most locations, the first wave is not the highest except for Lausanne and Evian that are located in immediate vicinity of the failure scar. Table A1: Recorded Earthquake events in Lake Geneva region with maximum magnitudes Mw larger than 4.5 from 1500 to 2009 (ECOS09; Date Mw Intensity Area of Epicentre Reference number (Fig.A1) VIII Aigle a VII Aigle b VII Chablais (St. Jean d'aulps) c VI Aigle d V-VI Lausanne e V St. Prex f VI-VII Chablais (Abondance) g VII Chablais (Abondance) h

9 Appendix B: Supplementary Methods Supplementary Methods 1. Calculation of MTD volumes: The volumes of the mass transport deposits (MTDs) were calculated from both the pinger and watergun data using the SMT Kingdom Suite software. The first isopach map was established by subtracting the picked base horizon from the picked top horizons depth values. The MTD s volumes were then calculated with the volumetric algorithm of SMT Kingdom Suite using the Single Isopach Type Grid function. 2. Numerical simulation: Simulations were carried out by numerically solving the shallow water equations in two dimensions. In our model, surface water waves were generated by kinematically displacing a sediment volume of 0.13 km 3 from a failure scar located on the northern lake slope, just south of the city of Lausanne (Figs. 7 and A5). This is based on the assumption that the mass movement originated as slide on the slope and evolved then into a mass flow and was deposited as a slide-evolved mass flow, as suggested by seismic and coring data. Areas corresponding to the slide origin and eventual depocenter were estimated on the basis of bathymetric data and seismic data. Using the estimated volume, these areas were then converted to mean heights, respectively. The variation in height with time was obtained by linearly interpolating between the initial and final locations. The variation in the slide velocity through time was computed using a simple empirical relationship (Ward and Day, 2002), using a slope of 4 and a total runout distance of 2.5 km. The change in slide height with time changes the bathymetry and thus induces disturbances in the overlying water column. We have not accounted for the transformation of the slide into a mass flow, for which we have no

10 sedimentological data. We also didn t include erosion, sediment transport or sedimentation or any other interaction between the moving mass and the water column that may introduce momentum dissipation, something that would reduce wave heights compared to the results presented. The entire modern lake border was assumed to be a no flow (closed) boundary, so we do not compute run-up distances relative to the lake shore. This will tend to overestimate the height of reflected waves.