Disentangling factors controlling earthquake-triggered soft-sediment deformation in lakes

Abstract

In-situ soft sediment deformation structures (SSDS) are commonly used as paleoseismic indicators in marine and lacustrine sedimentary records. Earthquake-related shear can deform sediment in the shallow subsurface through Kelvin-Helmholtz instability. The SSDS related to Kelvin-Helmholtz instability have been used to quantify shaking strength of past earthquakes. However, the relative importance of i) lithology and physical properties, ii) potential basal shear surfaces (e.g. clastic deposits), iii) slope angle, and iv) seismic shaking strength (e.g. peak ground acceleration) for deformation related to Kelvin-Helmholtz instability remains poorly studied. Lake Riñihue (south-central Chile) is chosen as a natural laboratory for disentangling the effect of the aforementioned factors because i) the sediment composition of background sediment varies downcore and ii) volcanogenic clastic deposits are abundant within the sedimentary sequence. A previous study at lake Riñihue identified 25 SSDS intervals induced by historical earthquakes of varying rupture extent in 17 sediment cores taken at slope angles ranging from ~0.2° to ~4.9° (i.e. 16 slope sites and 1 basin site). Our study shows that deformation mostly occurs directly above volcanogenic deposits (i.e. 72 % of SSDS intervals), suggesting that volcanogenic deposits promote earthquake-induced deformation by strain softening, liquefaction or water film formation. Deformation thickness of SSDS increases with higher slope angles (i.e. strong positive correlation). Additionally, deformation thickness commonly corresponds to the stratigraphic depth of the youngest preceding volcanogenic deposit, but for steeper slope angles stratigraphically older volcanogenic deposits can function as basal shear surface. Therefore, we suggest that deformation thickness is primarily regulated by gravitational stress (i.e. slope angle) and secondarily by the stratigraphic depth of volcanogenic deposits. The earthquakes related to strongest shaking caused almost exclusively SSDS with highest deformation degrees (i.e. folds and intraclast breccia) as well as largest spatial extent of SSDS, resulting in highest numbers of related SSDS in the investigated cores. Thinner SSDS have higher deformation degrees at a given shaking strength, as seismically-induced shear energy acts more effectively on thinner deforming sequences. Therefore, we suggest that deformation degree is primarily controlled by shaking strength and secondarily modulated by the thickness of the deforming sequences. We infer that deformation thickness is not a reliable indicator of paleoseismic shaking strength as this relies on many preconditioning factors independent of shaking strength. On the other hand, deformation degree can be a good proxy for shaking strength also in settings with varying lithotypes and intercalated clastic event deposits, provided multiple cores are studied to avoid under- or overestimation of paleoseismic shaking strength.