Landslide hazard assessment using satellite precipitation information

The increasing availability of remotely sensed precipitation and surface products provide a unique opportunity to explore how landslide susceptibility and hazard assessment methodologies may be applicable at larger spatial scales. This research considers a prototype satellite-based global landslide hazard algorithm framework, which evaluates how the landslide susceptibility and multi-satellite precipitation estimates can forecast potential landslide conditions. An analysis of this algorithm using a newly developed global landslide inventory dataset suggests that forecasting errors are geographically variable due the resolution and accuracy of the current susceptibility map and the application of satellite-based rainfall estimates, particularly in complex topographic settings. The current study seeks to improve these identified challenges through considering higher spatial and temporal resolution landslide susceptibility information and testing different uses for current multi-satellite precipitation products worldwide. Satellite precipitation information from the Tropical Rainfall Measuring Mission (TRMM) product is incorporated into rainfall intensity-duration triggering estimates as well as calculations of antecedent rainfall and previous soil moisture conditions. The study tests several different formulations of the global landslide forecasting algorithm to assess its performance accuracy at regional and global scales. Future missions including the Global Precipitation Measurement (GPM) as well as the Soil Moisture Active & Passive (SMAP) provide an important foundation for advancing this system and creating a global landslide hazard forecasting framework with considerable scientific and societal applications.