Search

Terraced landscapes represent one of the most widespread human-induced/man-made transformations of hilly-mountainous environments. Slope terracing produces peculiar morphologies along with unusual soil textures and stratigraphic features, which in turn strongly influence slope hydrology. The investigation of the hydrological features of terraced soils is of fundamental importance for understanding the hydrological dynamics occurring in these anthropogenic landscapes, especially during rainfall events. To this purpose, the availability of extensive field monitoring data series and of information on subsoil properties and structure is essential. In this study, multi-sensor hydrological data were acquired over a period longer than 2 years in the experimental site of Monterosso al Mare, in the Cinque Terre National Park (Liguria region, Italy), one of the most famous examples of terraced landscape worldwide. Monitoring data were coupled with accurate engineering-geological investigations to achieve the hydro-mechanical characterization of backfill soils and to investigate their hydrological response at both the seasonal and the single rainstorm scale. The results indicated that the coarse-grained, and anthropically remolded texture of the soils favors the rapid infiltration of rainwater, producing sharp changes in both soil volumetric water content and pore water pressure. Furthermore, the pattern of hydrological parameters showed seasonal trends outlined by alternating phases of slow drying and fast wetting. The study outcomes provide useful insights on the short and long-term evolution of hydrological factors operating in agricultural terraces. These findings represent a useful basis for a better understanding of the time-dependent processes that guide water circulation in terraced systems, which have a key role in controlling the occurrence of erosion and landslide processes. [ABSTRACT FROM AUTHOR]

The assessment of timing and potential locations of rainfallinduced shallow landslides through mathematical models represents a challenge for the assessment of landslide hazard, especially in cases with limited or not available data. In fact, modeling slope hydrological response and stability requires accurate estimates of unsaturated/saturated hydraulic and geotechnical properties of materials involved in landsliding, as well as climate and topography. Such aspect is relevant for the prediction of location and timing of landslide events, which is greatly needed to reduce their catastrophic effects in terms of economic losses and casualties. To such a scope, we present the comparison of results of two physics-based models applied to the assessment of susceptibility to shallow rainfall-induced landslides in Valtellina region (northern Italy). The analyses were carried out considering effects of availability, resolution and type of data concerning spatial distribution, thickness and properties of soils coverings. For such a scope, the Transient Rainfall Infiltration and Grid-Based Regional Slope-Stability (TRIGRS) and the Climatic Rainfall Hydrogeological Modeling Experiment (CHRyME) models were considered. The study emphasizes issues in performing distributed numerical slope stability modeling depending on the availability of spatially distributed soil properties which hamper the quality of physic-based models. Further analyses aimed at the probabilistic assessment of landslide spatial distribution, related to a specific value of rainfall threshold, can be considered as potentially applicable to multi-scale landslide hazard mapping and extendable to other similar mountainous frameworks. [ABSTRACT FROM AUTHOR]

This paper presents a new open-source and physically based model for Spatial Prediction of Rainfall-Induced Shallow Landslides (SPRIn-SL) through the Quantum GIS (QGIS) software. SPRIn-SL consists of a set of shell scripts developed using the Python language that can be directly run from the QGIS processing toolbox through a user-friendly graphical interface. The tool implements the infinite slope method by incorporating the TOPOG and the Green-Ampt models to consider groundwater flow and transient rainfall infiltration, respectively. Furthermore, DEM pre-processing procedures to extract reliable terrain morphometric features, a new statistical method for modelling soil depth and a procedure for predictive accuracy evaluation, were implemented. By using a 1-m resolution DEM, the developed model was tested in a small coastal catchment of Cinque Terre (Liguria, Italy), providing accurate outcomes, and proving to be an easy-to-use tool for landslide susceptibility zoning which can have useful implications on the risk reduction. • QGIS module written in Python for spatially predicting shallow landslides. • DEM pre-processing procedures to extract reliable morphometric predisposing factors. • A new statistical method for modelling soil depth is implemented. • High accuracy in simulating rainfall-induced shallow landslides in a test basin. • Easy-to-use physically based tool for landslide susceptibility zoning. [ABSTRACT FROM AUTHOR]