Effects of Widespread Mass Wasting Triggered by Hurricane María in the Lago Caonillas Watershed, Puerto Rico, in Applying the Multiscale Model-to-Model Cloud Comparison (M3C2) method to evaluate landslide source area metrics in Puerto Rico

Hurricane María triggered more than 70,000 landslides across Puerto Rico in 2017, and 8,617 of those sites were identified in the 125 km2 Lago Caonillas watershed, one of the watersheds of the Río Grande de Arecibo basin, which had the highest event landslide density across the island. The focus of this study is to investigate the landscape connectivity between hillslope failure sites and the 2.4 km2 reservoir sediment sink. The geology, soils, topography, and historical land use practices of the basin make it a hotspot for shallow mass wasting activity. LiDAR-derived digital elevation model differencing is being used in the target watershed using pre- and post-event 1-m datasets from 2016 and 2018, and an event headscarp point and polygon inventory is being used to estimate landslide area, volume, and sediment delivery to the fluvial network using a semi-automated GIS work ow. The United States Geological Survey (USGS) conducted bathymetric surveys in the Lago Caonillas reservoir in 1990, 1995-2000, and 2012. The reservoir was constructed in 1948 and had an initial storage capacity of 55.66 million cubic meters (Mm3). Results from the 2012 survey of Lago Caonillas showed that the lake was 29% sedimented since its creation and its storage was reduced to 39.55 Mm3. Given an established high sediment trapping efficiency (93%) for the reservoir, post-Hurricane María bathymetric surveys of Lago Caonillas are important to obtain refined insights into sediment transportation, sediment rates, and reservoir life expectancy. A post-Hurricane survey was conducted in 2020 and results indicate a storage capacity of 34.69 Mm3. The 2020 survey data will be compared to the sedimentation rates determined from the previous surveys carried out by the USGS between 1990 and 2012. Results of both the LiDAR analysis and 2020 bathymetric survey will be presented and considered in the context of Hurricane María. Given the vulnerability to sedimentation of the target lake and similar reservoirs in Puerto Rico, this study will serve to project potential future impacts of extreme precipitation events and widespread landslide triggering events on the fragile freshwater infrastructure in the island.