Rainfall-induced landslide early warning system based on corrected mesoscale numerical models: an application for the southern Andes

Abstract. Rainfall-induced landslides (RILs) are an issue in the southern Andes nowadays. RILs cause loss of life and damage to critical infrastructure. landslide early warning systems (RILEWSs) can reduce mitigate economic social damages related RIL events. The do not have operational-scale RILEWS yet. In this contribution, we present a pre-operational based on Weather Research Forecast (WRF) model geomorphological features coupled logistic models Andes. been forced using precipitation simulations. We correct derived from WRF 12 weather stations through bias correction approach. were trained 57 well-characterized validated by ROC analysis. show that has strong limitations representing spatial variability precipitation. Therefore, accurate needs study zone. used simulation slope, demonstrating high predicting capacity (area under curve, AUC, 0.80). conclude our proposal could be suitable at operational level determined conditions. A reliable database networks allow real-time mesoscale implemented zone needed. RILEWSs become support decision-makers during extreme-precipitation events climate change south