Sliding-type landslide susceptibility paper

####The dataset consists of the shp-file of territorial units classified according to the value of the probability of occurrence of sliding landslides and the corresponding susceptibility class. #### Landslide susceptibility measures the propensity of a territory to produce gravitational failures, that is, it is the spatial probability that new landslides can be generated, regardless of their intensity, even in currently stable areas. A landslide susceptibility zoning divides and classifies the territory based on the predisposition to generate landslides. The maps of landslide susceptibility are therefore a useful tool to deepen the knowledge on the propensity to failure of the territories, for the purpose of their management on a regional and local scale. The Umbria Region’s sliding-type landslide susceptibility map was drawn up under the work programme for the ROP-ERDF funds 2010-2013, Axis II, Activity a1), Action 5 ‘Monitoring plan and systems for the prevention of landslide hydrogeological risk’, under a contract signed by the IRPI-CNR on 22 March 2012 and concluded in March 2014. Ownership of the data is vested in the Region of Umbria within the meaning of Article 9 of the contract. For the zoning of susceptibility, a multivariate statistical model was used, and in particular linear discriminatory analysis, using as a dependent variable the presence/absence of landslide deposits within each reference territorial unit and as independent variables the available geoenvironmental information, in particular: Morphometric variables (altitude/slope statistics) derived from the digital terrain model (DTM) in raster format with a ground resolution of 25x25 m; lithological variables, obtained by reclassifying the formations represented in the regional geological maps; presence/absence of landslides derived from the landslide inventory map; Land use variables obtained from the regional land use map. The model created divides the regional territory into five classes of susceptibility: - very low (less than 0.25) - low (0.25 to 0.45) - average (0.45 to 0.55) - high (0.55 to 0.75) - very high (over 0.75) It has a success rate of 71%, which explains the spatial distribution of landslide deposits in 71.3% of territorial units. The model was validated with information not used for its calibration, in particular using landslides that occurred in major meteorological emergency states on a regional scale. In all cases tested, the model demonstrated a high predictive capacity. Not considering the lowland areas, 47% of the regional hilly and mountainous territory falls into the high and very high susceptibility classes. It is important to underline that the model applied in Umbria refers exclusively to landslides and therefore does not provide information relating to other types of landslides. #####Legislative reference Executive Decision No 9638 of 30 September 2019 ‘Updating and supplementing the reference documents relating to the scenario of danger and risk from landslides in the Umbria Region, referred to in Regional Executive Decision No 384 of 29 April 2013’.