The Value of Using Multiple Hydrometeorological Variables to Predict Temporal Debris Flow Susceptibility in an Alpine Environment

Abstract

Debris flows are typically triggered by rainfall-related weather conditions—including short-duration storms and long-lasting rainfall, in cold climates sometimes in connection with intensive snowmelt. Given the considerable observational uncertainties of rainfall, we tested if other hydrometeorological variables carry enough information content to compensate for these uncertainties and if the combined information of hydrologic catchment state and rainfall can be used to predict the regional temporal susceptibility for debris flow initiation. For this we carried out a probabilistic analysis of variables derived from a conceptual hydrological model for the Montafon region, Austria, where debris flows were recorded on 41 days between 1953 and 2013. Exclusively from hydrological characteristics and, importantly, neglecting precipitation itself, we quantitatively determined different trigger types for historical debris flows. Subsequently, we used four Naive Bayes classifier models, ranging from a simple rainfall-only model to a multiparameter hydrometeorological model differentiating between trigger types, to predict days susceptible for debris flow occurrence in the region. The results suggest that debris flows were triggered by convective rainstorm events on 23 days, on 12 days due to gradual soil moisture buildup in the course of long-lasting rainfall events and on six further days snowmelt played an important role. We find that the differences between the trigger types are statistically significant and that a susceptibility prediction differentiating between trigger types and including hydrological information can outperform simple rainfall-only models. This study thereby contributes to an improved understanding of the hydrometeorological impact on debris flow initiation in a mountain watershed.