REAL-TIME ASSESSMENT OF POST-FIRE DEBRIS FLOW HAZARDS USING GLOBALLY AVAILABLE DATA

Wildfires are a global phenomenon, and their subsequent effects on landscapes exacerbate mass wasting in the form of post fire debris flows (PFDFs). These events are the direct result of the reduced infiltration capacity of burned soils, causing an increase of available sediment or burned debris that can be mobilized during subsequent storms. Several local- and regionally-based models for PFDF hazard assessment are already in routine operation, such as a statistically-based model operated by the United States Geological Survey (USGS). These models establish regional and local thresholds to constrain the effect of rainfall in recently burned areas, but none have a worldwide application. We present an empirical model that assesses the hazard of post-fire debris flows using freely available global datasets. We show that for global scale assessments, routine remote sensing observations can help detect burned areas, provide quantitative estimates of topographic and burn severity conditions relevant for debris flow initiation, and—importantly—differentiate storms that are more or less likely to trigger debris flows. The use of only globally available data sources provides a statistically robust method for PFDF hazard assessment but remains limited by the spatial resolution of rainfall estimations from NASA’s Global Precipitation Measurement (GPM) products. Furthermore, the dearth of post-fire debris flow inventories outside of the United States (US) presents a challenge for broader geographic application of the modeling framework, but also offers an opportunity for improvement. As the underlying processes that govern post-fire debris flow initiation are not exclusive to the US, our validation dataset would greatly benefit from additional documentation of PFDF occurrences on other continents. In this presentation we demonstrate the feasibility of a global, near-real time model to assess post-fire debris flow hazards worldwide.