Modellazioni bidimensionali comparate sulla propagazione di debris flow: analisi di performance su alcuni eventi reali

In Alpine regions gravity-driven natural hazards, in particular debris flows, endanger settlements and human life. Mitigation strategies based on hazard maps are necessary tools for land planning. These maps result more precise by using numerical models to forecast the inundated areas after a careful setting of those ’key parameters’ (KP) which directly affect the flow motion and its interaction with the ground surface. Several physically based 2D models are available for practitioners and governmental agencies, but the selection criteria of model type and of the related KP remain flexible and partly subjective. This remark drives to investigate how different models are capable to simulate different types of debris flows (from granular to muddy debris flows, going through intermediate types), in particular when the flow is influenced by the presence of deposition basins. Two commercial 2D physical models (RAMMS and FLO-2D) have been tested for five well-documented debris flows events in mountain catchments were different geology and flow dynamics are observed: 1) a viscous debris flow occurred in 2009 in a catchment with a metamorphic geology (Gadria torrent, Bolzano Province); 2) the 2009 granular debris flow in an granitic geological setting (Rio Dosson, Trento Province); 3-4) two events occurred in 2010 in the rio Val del Lago and ’rio Molinara’ (Trento Province) where porphyritic lithology prevails (intermediate granular debris flow); 5) the Rotolon torrent (Vicenza Province) 2009 debris flow containing sedimentary rocks enclosed in an abundant clay-rich matrix (intermediate viscous case). Event volumes range from 7000 to 50000 cubic meters. The Gadria, Rotolon and Val del Lago events are also influenced by artificial retention basins. Case study simulations allowed delineation of some practical end-user suggestions and good practices in order to support model choice and the KP setting, particularly related to different flow dynamics. The presence of mitigation structures (e.g. check dams and retention basins) demands both the implementation of a precise topography and the introduction of devices to better model sediment trapping and functionality of open/closed check dams. The study results represent: i) a first support for practitioners to directly manage debris-flow simulations; ii) an help for local authorities to give the right value to simulations carried out by practitioners and scientific community; iii) a warning that hazard maps should not be based just on model simulation results.