Modeling snowmelt influence on shallow landslides in Tartano valley, Italian Alps

Shallow landslides (SLs) are rapid soil mass movements, typically occurring in the mountain areas, involving the most superficial soil layers up to 5 to 10 m in depth. Damages, and casualties due to shallow landslides are recorded globally, and in literature a variety of models to study landslides have been implemented hitherto. Often times, shallow landslides occur in the wake of snowfall events, when sudden temperature increase triggers fast snow thaw, and soil moisture increases thereby. Several models studied the influence of intensity, and duration of rainfall upon shallow landslides, but the effect of snow melt in spring/summer was little considered so far. Thus, we developed a simple but robust, and parameter-wise parsimonious model, that mimics the triggering mechanism of SLs, accounting for the combined effect of precipitation duration and intensity, and snowmelt at thaw. The model is here applied to the case study of the high altitude Tartano basin, paradigmatic of SLs in the Alps of Lombardia. Our results showed that about 26 % of the Tartano basin slopes display unstable conditions. Using a traditional (i.e. rainfall-based) approach, the occurrence of shallow landslides was predicted in ca. 19 % of the basin, mainly during storms in October and November. In contrast, when snowmelt was included, the model was able to mimic potential SLs even during April and May, when snow melt rate is the highest, and may increase SLs triggering potential, to ca. 26 % of the treated area. With better spatial and temporal description of slope failure as achieved here, validated against observed failures, a public authority may be prepared to implement emergency plans, to prevent injuries, causalities, and damages to infrastructures even during springtime, when shallow landslides may occur in response to fast snowmelt, even during dry, clear sky days, and with scarce/null precipitation.