NH030-0004
A comprehensive catalog of coseismic landslides in Costa Rica: Toward the development of an empirical landslide hazard model.

Monday, 14 December 2020
Virtual
Paulo Ruiz1, Natalia Rodriguez2, Joan Valverde2, Andrew Marden3, Lea Vecchiarelli4, Dylan Seal5, Martha Anna Nowicki Jessee6 and Michael W Hamburger5, (1)University of Costa Rica, Lanamme, San Jose, Costa Rica, (2)University of Costa Rica, Central America School of Geology, San Jose, Costa Rica, (3)Rutgers University New Brunswick, Civil Engineering, New Brunswick, NJ, United States, (4)Rutgers University New Brunswick, Bioenvionmental Engineering, New Brunswick, NJ, United States, (5)Indiana University Bloomington, Earth & Atmospheric Sciences, Bloomington, IN, United States, (6)Indiana University, Bloomington, IN, United States
Abstract:
Coseismic landslides in Costa Rica are common, destructive (>$100 US million in losses since 2009) and the primary cause of death during recent earthquakes. This work presents a comprehensive and updated catalog of the landslides triggered by earthquakes that have been documented in Costa Rica since 1772. Their genesis, main characteristics and how some of them have progressed over time are presented here. To build this catalog, previous compilations were reviewed and landslide inventories were digitized to be included in a GIS database. For the most recent events (2009 - 2016) field work and remote sensing techniques were used (lidar, satellite images and drone aerial photographs) to compile the landslide distribution data. Since 1772, Costa Rica has experienced at least 23 earthquakes (Mw > 5.5) that have triggered all types of landslides. Five of these events originated from subduction process and at least 12 events by shallow (depth < 23 km) local faults situated near volcanic centers. The comprehensive catalog allows us to make correlations with other factors (topographic slope, geology, soil moisture and ground shaking distribution) in order to find the factors that are most strongly influencing the generation of landslides. The Costa Rican Geographic Institute has topographic data that allowed us to build a high-resolution (20 m contour lines) DEM of the entire country. We also make use of new detailed 1: 50 000 geological maps (mainly in the NW sector of Costa Rica), data on land cover type for the entire country and estimates of local ground shaking based on a dense national seismic network. Combining these observational data with the Nowicki Jessee et al. (2018) methodology from their global empirical model, we seek to develop a local empirical model for near-real-time prediction of coseismic landslides in Costa Rica and to compare results with those from the global model.
<< Previous Abstract | Next Abstract >>