7000 years of earthquake record in (Lake , ) William Rapuc, Pierre Sabatier, Maja Andric, Christian Crouzet, Fabien Arnaud, Emmanuel Chapron, Andrej Smuc, Anne-Lise Develle, Bruno Wilhelm, François Demory, et al.

To cite this version:

William Rapuc, Pierre Sabatier, Maja Andric, Christian Crouzet, Fabien Arnaud, et al.. 7000 years of earthquake record in Julian ALps (, Slovenia). International Meeting of Sedimentology, Oct 2017, Toulouse, France. 2017. ￿hal-01773779￿

HAL Id: hal-01773779 https://hal-sde.archives-ouvertes.fr/hal-01773779 Submitted on 25 Mar 2020

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés. 7000 years of earthquake record in Julian ALps (Lake Bohinj, Slovenia) William RAPUC1, Pierre SABATIER1, Maja ANDRIC2, Christian CROUZET3, Fabien ARNAUD1, Emmanuel CHAPRON4,5, Andrej SMUC6, Anne-Lise DEVELLE1, Bruno WILHELM7, François DEMORY8, Edouard REGNIER9, Jean-Louis REYSS1, Gerhard DAUT10 & Ulrich VON GRAFENSTEIN9

1 : EDYTEM, Université Savoie Mont Blanc, CNRS, Le Bourget du Lac, [email protected] ; 2 : ZRC SAZU, Institute of Archaeology, Ljubljana, Slovenia ; 3 : ISTerre, Université de Savoie, CNRS, bat. Belledonne, Le Bourget du Lac ; 4 : GEODE, Université Jean Jaurés, Toulouse ; 5 : ISTO, Université d’Orléans, Orléans; 6 : Department of Geology, University of Ljubljana, Slovenia ; 7 : LTHE, Université Grenoble, Grenoble ; 8 : CEREGE, Université Aix-Marseille, CNRS, IRD, Aix-en-Provence ; 9 : LSCE, Université de Versailles Saint-Quentin, CNRS, Gif-sur-Yvette ; 10 : Institut für Geographie, Jena, Germany

INTRODUCTION FIG. 1 RESULTS A B Surface Geology: 1690 Scree 46°30’ N MSK=VIII-IX • Julian Alps are characterised by a moderate to high 1348 Glacio-lacustrine clay • 4 generations of MWDs are identied across eastern sub- MSK=IX-X Unconsolidated Moraine seismic activity that has included several destructive earth- Mt basin (Fig. 1D) at the edges of the subaqueous slopes. MWDs 2000 Amonitico rosso Limestone Bovec Mid Jurassic-Low. Cretaceous quakes, such as the 1511 AD “Idrija” earthquake (MSK = X). 46°20’ N Fault evolved laterally into transparent acoustic facies developing Upper Triassic Limestone lity Accurate assessment of the seismic hazard is required to 1976 Upper Triassic Massive Limestone onlaps, typical signature for muddy turbidites or homogenites. MSK=IX-X Idrija Fault 1942 Kranj 1750 MSK=VI Watershed shape reduce the high vulnerability of this region. 46°10’ N Tolmin Ravne Fault Main and secondary Stream, Waterfall Thrust Fault, Fault 1511 1500 Udine MSK=X • Lake Bohinj sediment core is subdivided into 7 di erent Ljubljana • Earthquake shaking may destabilise previously deposited 46°N Idrija 1000 sedimentary units. Most of the sedimentary record is Unit IV Rasa Fault Lake Bohinj sediment, which slides to the bottom of the lake and formed 525m (443-998 cm, Fig. 2) composed of a coarse sandy thick erosive 1500 Sveti Duh (525.89 m asl) seismically induced mass wasting deposits (MWD), such as 45°50’ N base overlaid by a homogeneous silty facies and a clayey cap. Montfalcone turbidites or homogenites. 1750

45°40’ N 500m • The upper part of the sedimentary record (0-443 cm) is

13°E 13°30’E 14°E 14°30’E interbedded by two types of graded-beds: N S MATERIALS & METHODS Major Strike-slip Fault Lake Bohinj 20 km D o Type I is composed of (i) a coarse sandy base (Fig. 3), -35 m • This study focused on Lake Bohinj (Fig. 1), in the Julian Secondary Strike-slip Fault Major Historical Earthquakes (ii) a central part with homogeneous silt, (iii) a thin light clayey Alps (Slovenia). During the Quaternary, this area was a ec- cap. Typical characteristics of Homogenite-type deposits, C MWD-A -40 m MWD-D ted by major NW-SE striking faults (Camassi et al., 2011). MWD-B MWD-C conrmed by higher values of Foliation (Chapron et al., 2016). BO12 -45 m o Type II is characterized by a gray graded bed with a inlet coarse and erosive silty base becoming slightly ner until a • To reconstruct the earthquake chronicle, Lake Bohinj top Depth blf (m) MWD-D sediments were mapped with seismic reection survey and 1000 m outlet base -50 m white clayey cap (Fig. 4). Characteristic of turbidite-type 100 m cored in 2012 to provide a 12-m-long sedimentary record. >10 m > 20 m > 30 m >35 m >40 m >45 m deposits (Sturm and Matter, 1978) -55 m • A multiproxy analysis associating sedimentological and Fig. 1- General presentation of the study site. (A) Location of the study area with the major strike-slip faults and major • The age depth model was created associating short-lived ra- historical earthquakes, modied from Moulin et al., 2014. (B) Geological map of the Lake Bohinj watershed. 14 geochemical measurements was performed to reconstruct (C) Bathymetric map of Lake Bohinj. (D) Seismic prole oriented across the eastern basin showing the coring site location dionuclides proles and 21 calibrated C ages. 7 outliers were the long-term earthquake record. and four generations of mass wasting deposits (MWD). unused because sampled in instantaneous deposits (Fig. 5).

FIG. 2 FIG. 3 FIG. 5 Ti (cps) LOI 550° (%) LOI 950° (%) A B Age (cal. BP) Inc K3 Foliation Sorting Cl

St Pb 0102030 BO12 Sd 020000 40000 010203040 0 0 30 60 90 1 1,02 2 2.5 343.5 % BO12 0 4000 8000 12000 I 8 0 10 6 Short-lived radionuclides 1 4 II (210Pb,137Cs) 20 2 IIIb 0 1 2 B 14C Dates 30 C III MAG: 520 x HV: 20.0 kV WD: 26.1mm 100 µm Rejected 14C Dates 40 C 2

3 Depth (cm)

Base 30.5 50 I T1 Coarse 4 II 3 I 60

30.7 Depth (cm) 5 70 4 Continuous 1 1,02 0 10000 20000 010203040 10-1 100 101 102 103 Sedimentation Lineation Ti (cps) Median (µm) Grain Size (µm) Type 1 deposits MAG: 59 x HV: 20.0 kV WD: 26.0mm 700 µm 6 Depth (m ) Fig. 3- Detailed results for T1 deposits. (A) AMS and grain size. EDX cartography in (B) T1 deposit and (C) continuous 5 IV sedimentation. 7 6 FIG. 4 8 A Inc K3 Foliation Sorting Grain Size (µm) B 30 60 90 1 1,02 1,04 1,06 22.5 3 151.71 Depth (m ) 7 140 10-1 100 101 102 103 % 9 8 6 151.73 4 8 145 2 10 V 0 151.75 T2 Coarse Base 9

11 Depth (cm) VI 151.77 150 Depth (cm)

B 10 Hiatus 12 VII 151.79

155

Continuous 151.81 11 11,02 0 5000 10000 0102030 Sedimentation 02000400060008000 0 1000 2000 Lineation Ti (cps) Median (µm) 3 MAG: 160 x HV: 20.0 kV WD: 26.0mm Br (cps) Ca (cps x10 ) 70 µm Fig. 2- Main sedimentological and geochemical results. Lithological and geochemical results (Ti, Br and Type 2 deposits 12 Ca contents) associated with the BO12 sequence Lost On Ignition (LOI) 550° (dark-gray dots) and LOI 950° Fig. 4- Detailed results for T2 deposits. (A) AMS results, Ti content and grain size. (B) EDX cartography at the boundary (light-gray dots). of a T2 deposit and the continuous sedimentation. Fig. 5- Age-depth models from radiocarbon and short-lived radionuclide dates.

FIG. 7 DISCUSSION Age (k. yr cal BP.) 0 1 23456 7 • Several mechanisms can trigger MWDs: e.g., earthquakes, spontaneous delta collapses, and considerable lake-level changes. The 15-m-high moraine ridge (Fig. 1) prevents sediment contribution of hyperpycnal ow from the delta and delta collapses to 6000 + reach the coring location. 4000 Dynamics of organic inputs 2000 Br (cps ) • Earthquakes record by MWDs depends on the sensitivity of a lake to record a seismic event. To characterize the sensitivity of 0 - Lake Bohinj, earthquake-sensitivity threshold Index (ESTI) method was applied (Fig. 6; Wilhelm et al., 2016). 12

6 • Seismic trigger hypothesis for the three distinct types of deposits: (nb/30yrs) T2 Deposit s o Three most recent T1 correspond to three strong historic earthquakes (Fig. 6, 1348, 1511 and 1690 AD earthquake) 0 1511 AD 1348 AD o Between 3500 and 2000 yr cal BP major destabilizations occur in the watershed by human activities (Fig. 7) that led to 1

increases in sedimentation rate and thus increase ESTI (more earthquakes should have been recorded). High T2 deposits number AD Earthquake identied could be the consequence of sensitivity increase. 1690 Chronicle

o Unit IV occurred coevally with geomorphological changes in the area (Bovec Terrace, Fig. 1 , Marjanac et al., 2001) T1 Deposit s 0 + )

-1 3

• The Bohinj earthquake chronicle presents several time intervals with no local seismic activity recorded by homogenite-type 2 deposits. These periods could be linked to the lack of major seismic events or to ESTI decrease Sedimentation 1 rate (mm.yr Sedimentation 0 - FIG. 6 Age (k. yr cal BP.) CONCLUSION 0 1 23456 7 Units ABLIL I II III I IIIbIV Sedimentation X Earthquakes that 1976 1348 • This is the rst earthquake chronicle Low sedimentation rate Erosion Low sedimentation rate Good record of seismic activity Good record of seismic activity generated 1511 over the last 7000 years in the Julian Alps Evolution of mass movements with 29 homogenite-type deposits. Progressive rise of Opening A forest of spruce, beech and pine watershed 1690 BOHT2 VIII Grasses and Pines surround Lake Bohinj Anthropogenic impact The regional seismic activity recorded Opening BOH • Fig. 7- Lake Bohinj earthquake chronicle. Comparison between the organic input dynamic (Br), T2 and T1 1942 by a lake is highly connected to frequency with major historical earthquakes and the mean sedimentation rate. Gray shadings correspond to

VI ESTI Index intervals with potential higher ESTI. Earthquakes that earthquake settings, distance and the not generated lake’s sensitivity to recording a seismic mass movements Epicentral MSK Intensit y IV event (ESTI), which is related to the REFERENCES 1 10 100 sedimentation rate. • Camassi, R., et al. (2011) Journal of Seismology, 15, 191–213. • Rovida, A., et al. (2016) Istituto Nazionale di Geo sica e Vulcanolo- Distance lake-epicentre (km) Sedimentation rate (mm.yr-1) • Chapron, E., et al. (2016) In: Submarine Mass Movements and gia (INGV). • When sedimentary sequence presents their Consequences, pp. 341–349. Springer. • Sturm, M. and Matter, A. (1978) Wiley Online Library. Fig. 6- ESTI estimation for Lake Bohinj. (A) Lake Bohinj sequence results associated with CPTI15 data. Green line represents the limit of lake Bohinj • Marjanac, T., et al. (2001) Geologija, 44, 341–350. • Wilhelm, B., et al. (2016) Journal of Geophysical Research: Earth sensitivity to earthquakes calculated for the historical period. The red dotted line represents the hypothetical limit of lake Bohinj sensitivity with an ESTI of a variable ESTI, it should be prohibited to 0.18 calculated for the mean sedimentation rate over the period of 3500 to 2000 yr cal BP, unique period with T2 deposits. (B) The ESTI versus sedimentation • Moulin, A., et al. (2014) Tectonophysics, 628, 188–205. Surface, 121, 2–16. rate for Lake Bohinj (green) over the last centuries with previous results (black) from Wilhelm et al. (2016). The red cross corresponds to the ESTI and present a mean return period for seismic sedimentation rate for the period of T2 deposits (3500 to 2000 yr cal BP). This diagram suggests that a signicant increase (decrease) of the sedimentation Thanks to the LSM, ARSO, Robert Jensterle, D. Valoh, J. Dirjec, S. von Grafenstein, S. Kuharič, rate appears to be the dominant factor resulting in an increase (decrease) of earthquake deposits in the lake (Wilhelm et al., 2017). activity recorded by the lake system. M. Zaplatil and to ARRS Research program.

Laboratoire EDYTEM Environnements, Dynamiques et Territoires de la Montagne