Rainfall-induced landslide identification using numerical modelling: A southern Chile case

Abstract

Worldwide, Rainfall-Induced Landslide (RIL) affects the population with significant socioeconomic damage and high economic losses in developing countries. Nowadays, RIL hazard has gained attention from the scientific community in developing countries where low instrumental is available for identification and monitoring. However, RIL-prone zones identification is a complex task, mostly in zones with an increase of extreme hydrometeorological events due to climate change. We propose a new coupled methodology to identify RIL-prone zones based on slope instability (SI) during an extreme hydrometeorological event. We consider the Northern Slope of the Biobío River (NSBR, centered at ~36°S, 72°W) in the south of Chile, a densely populated area with historic RIL events. A one-way coupled system was carried out using the Weather Research and Forecasting (WRF) model to a hydrological model over an experimental area of 172 km2. Hydrological model results were used to calculate the factor of safety (FS) evaluating their temporal variations. Our results were compared with in-situ observations and remote sensing of slope-stability to evaluate their performance based on previous studies with good performance. Our results show that areas with low FS values represented shallow RIL and hyper-concentrated flows. Therefore, our approach could identify RIL-prone areas without the need for extensive in-situ information, which will reduce resources. Finally, our methodology allows implementation in countries with scarce or non-existent resources for the instrumental monitoring of hillslopes.