Lake : Causes, Consequences and inves�gated in a mul�disciplinary project

Katrina Kremer*, Flavio S. Anselme�, Paola Bacigaluppi, Robert M. Boes, Frederic M. Evers, Donat Fäh, Helge Fuchs, Michael Hilbe, Achim Kopf, Agos�ny Lontsi, Valen�n Nigg, Anastasiia Shynkarenko, Sylvia Stegmann, NH5.1 Michael Strupler, David F. Vetsch & Stefan Wiemer EGU2020 -4711 *[email protected] Context © Authors. All rights reserved 480000 500000 520000 540000 560000 580000 600000 620000 640000 660000 680000 700000 720000 740000 760000 780000 800000 0 0

In public perception, tsunamis are still related to marine settings and to as direct triggers. However, 0 0 6 2 Lake Constance

tsunamis, which are generated by a sudden displacement of the water column, can occur in every aquatic system. 0 0 - 1720 0 Alps 0

4 2 triggered delta The most prominent cases in are: - 1601 earthquake 0

0 collapse? 0 0

- 563 AD wave in Lake Geneva caused by a rockfall-induced delta-collapse. Modelled wave heights 2 - 1687 delta collapse 2

0 Lake Neuchatel

0 Lake

reached up to 8 m. 0 0 0

2 - 1898 earthquake - 1584 AD tsunami in Lake Geneva generated by an earthquake-triggered Rhone delta failure. - 1806 rockfall 0 triggered delta 0 0 0 8 - 1601 AD tsunami wave in Lucerne due to earthquake-triggered sublacustrine slope failures and a rockfall at 1 collapse? 0 0 0

Bürgenstock. Wave heights reached 4 m in Lucerne. 0 6 Lake Brienz 1

0 - 1996 Aare delta collapse 0 0 0 4 The goals of this project are to understand governing mechanisms of genesis and propagation of tsunamis in lakes. 1 0 0 0

0 Lake Geneva 2

Through a multidisciplinary approach involving limnogeologists, seismologists, geotechnical specialists, hydraulic 1 Rhone delta collapse 0 0 0

engineers and hazard specialists, we are developing key concepts and factors relevant for causes and consequences 0 - 563 rockfall triggered 0 1 - 1584 earthquake triggered

of these tsunamis (Fig. 2). To reach these goals, ve Work Packages (WP) exist: WP response, WP delta, WP wave, WP 0 0 0 0 paleo and WP hazard. 8 0 25 50 km 0 0 0 0 6 Figure 1: Historical Swiss tsunamis and their causes that have been documented in wri�en archives. Lakes marked with a red line will be used as case study in this project in one or more work packages.

th Chat time 4 of Project description Chat time 6th of May at May at 16:15-18:00 NH5.6 14:00-15:45 D2060 NH5.1 WP response EGU2020 WP wave D1899 EGU2020 -6979 -20733 a b Goals: Goals: - Evaluate the sediment-mechanical characteristics - Determine the governing wave generation and propagation processes in a narrow and con ned - Estimate the volume of potentially mobile sediments sediment lake basin. - Estimate the stability of sediments under seismic mechanical - Select a numerical model that allows for a suciently detailed and ecient prediction of runup shaking heights and inundation depths. characteristics - Provide a set of experimental data for numerical model validation. Methods: & - Ocean bottom seismometer record the seismic signal c Methods: (mainly ambient vibrations and earthquake monitoring) sediment - Laboratory experiments (Fig. 4) to allow a better insight into the underlying physical processes - Cone Penetration Test (CPT) measurements are used to volumes and act as a benchmark for simulation models. de ne the geotechnical characteristics of the sediment. - Numerical modelling to compute tsunami scenarios for entire lakes at prototype scale (Fig. 5) - A new instrument combines standard OBS and piezo meter probes. 2670000 2675000 2680000 2685000 2690000 d 0 0 0 0 1 2 1 Figure 3: a) OBS - b) CPTu - c) Iden�fica�on of OBS array geo- metry with mul�beam bathyme- 0 0

0 try- 5 0 2 1 Figure 4: Wave ge- nera�on by a suba- SED queous in WP

0 hazard

0 High : 433 m VAW the VAW laboratory. 0 CPT 0 Univ. Bern 0 WP Figure 5: Numerical simula�on of wave propaga�on 2 Low : 219 m OBS (single station) wave 1 WPdelta OBS (array geometry) Sediment cores 0 0 0

5 inundation extend & ow characteristics 9 MARUM/SED 1 1 0 1.25 2.5 5 Kilometers WPresponse Univ. Bern WPpaleo -> calibration of the model WP paleo sediment-mechanical Figure 2: Sublacustrine mass movements (at deltas or hemipelagic slopes) do generate Goals characteristics of deltas tsunamis. This sketch shows the lake-system approach adopted in this project and the focus - Identi cation of lake tsunami deposits of the different research sub-projects. The ins�tutes responsible for the individual work packages are also men�oned. 250 m Methods: WP delta - Sedimentological analysis on on- & o-shore sediment cores (Fig. 7) Goals: 14 a CT scan Grain size C dating Lithology Density CaCO3 TOC Core Core - Understand the causes of delta collapses cal. AD Core (g / cm-3) (μm) (wt.%) (wt.%) photo photo 1000 1500 photo 1 2 0.1 10 1000 20 60 0 2 4 CT scan CT scan - Distinguish between delta collapse turbidites and 0

other turbidites (oods) core location 20 carbonate mud 5 m water depth 10 m

14 Methods: modern C uvial sediments Upper Marine Molasse 40 1364 AD articial landll Lower Freshwater Molasse - Repeated multibeam bathymetry surveys of deltas Figure 7: Right: Sediment core loca�on 1365 AD - Sediment cores (in dierent lakes such as Lake 60 at Lake Lucerne (Bay of Lucerne). Le�: Lucerne and Lake Brienz; Fig. 6) Transect along recovered sediment Figure 6: Mul�beam bathymetry map of the 80 1416 AD cores show a fining upwards sand se-

Aare delta that collapsed in 1996. sand ning upwards quence. Based on radiocarbon da�ng of

depth below lake oor (cm) depth below 1412 AD 100 coniferous needles, the fining upward 1217 AD sandois interpreted to be reworked and deposited during the 1601 Lake Lucerne detail in greater Modelling of high risk areas 120 glaciolacustrine sediments tsunami.

WP hazard Figure 8: Example of the flow Goals: depth alongshore Lake Zurich, Frequency - Evaluate the tsunami potential of Swiss lakes calculated for tsunamis that are of tsunamis - Produce a rst generation of intensity maps for relevant return periods (case studies for expected to be caused by landsli- des triggered by earthquakes with some lakes such as Lake Lucerne, Lake Brienz, Lake Zurich, Lake Constance, Fig. 8) an occurrence probability of 0.5% - Evaluate the technologies related to warning in 50 years (Strupler et al. 2018, Swiss Journal of Geosciences) Methods: - Simple classi cation of the tsunami potential on Swiss lakes, using geospatial analyses - Creation of conceptual workow for the rapid estimation of the tsunami hazard (Strupler et al. 2019) - Probabilistic modelling of tsunami eects Frequency of delta collapses? - delta stability model of delta collapses? Frequency

References Strupler, M., Anselme�, F. S., Hilbe, M., Kremer, K., and Wiemer, S. (2019). A workflow for the rapid assessment of the landslide-tsunami hazard in peri-alpine lakes. Geol. Soc. London, Spec. Publ.

Acknowledgments WP response, WP delta, WP wave, WP paleo are financed by the Sinergia Program of the Swiss Na�onal Science Founda�on (nr. 171017). WP hazard is financed by the Federal office of environ- ment (FOEN) of Switzerland (17.0092.PJ). We thank the „German Pool for Amphibian Seismology“ (DEPAS) for borrowing us eight OBS from the Alfred Wegener Ins�tut (AWI) in Bremerhaven. We also thank EAWAG and colleagues for their precious help during field work.