EP031-0002
Characterization of Extreme Rainfall Events and Landslide Hazard in the Himalaya from High-Temporal Resolution Precipitation Records

Thursday, 10 December 2020
Poster
Madeline Hille, University of Michigan Ann Arbor, Ann Arbor, MI, United States, Marin Kristen Clark, Univ Michigan, Ann Arbor, MI, United States, Dimitrios Zekkos, University of California, Department of Civil and Environmental Engineering, Berkeley, CA, United States, A. Joshua West, University of Southern California, Department of Earth Sciences, Los Angeles, CA, United States, Deepak Chamlagain, Tribhuvan University, Department of Geology, Kathmandu, Nepal, Dalia Kirschbaum, NASA Goddard Space Flight Center, Hydrological Sciences Laboratory, Greenbelt, MD, United States and Robert Emberson, NASA Goddard Space Flight Center, Greenbelt, MD, United States
Abstract:
Variation in extreme rainfall plays an important yet poorly understood role in the weathering and erosion of mountainous landscapes. In the high-relief and steep terrain of the Himalaya, threshold hillslopes are vulnerable to the intensity and duration of storm events and frequently experience soil and shallow bedrock landslides during the summer monsoon. Characterizing monsoon-triggered landslide hazards requires an understanding of the intensity and duration of precipitation at sub-daily resolution and fine spatial scales. However, local ground-based precipitation records offer daily resolution and sparse spatial coverage insufficient for interpreting and modeling regional-scale landsliding. Here we use NASA’s Global Precipitation Measurement (GPM) IMERG 30-minute, 0.1x0.1 degree product scaled to match local rain gauge records to identify extreme rainfall events (ERE’s) over a study area in central Nepal, which was affected by widespread landsliding both during and following the 2015 Mw7.8 Gorkha earthquake. We present a systematic analysis of merged precipitation datasets with the goal of identifying and characterizing ERE’s and their spatial variability. To produce a precipitation model that captures both short-duration and high-intensity events, we scale the 30-minute GPM time series at selected gauge locations so that daily accumulated rainfall is equal to the daily record of the local rain gauge, and then isolate storm events in the scaled time series with a minimum inter-event time between storm arrivals. We fit the site-specific distribution of average storm intensity with a lognormal function and define events that fall over the 90th percentile as ERE’s. Preliminary results show variability in the intensity and duration of ERE’s with geographic and topographic location (i.e. sites on ridges, hillslopes, or in valley bottoms) and that average ERE intensity increases with distance from the Himalayan range front in a pattern consistent with the orographic effect on precipitation in mountainous regions. The analysis we present is a novel approach to merging high-temporal resolution GPM IMERG data with local rain gauge records and provides an opportunity to understand spatial and temporal patterns in extreme rainfall as they relate to landsliding in complex, data-poor settings.
<< Previous Abstract | Next Abstract >>