Evaluation of the applicability of sediment transport models to dam filling prediction in different Italian geological contexts

Abstract

Artificial reservoirs are crucial infrastructures, since they allow for the management of the water resource they store and for their role in the hydroelectric production. Water is modulated and stored inside artificial basins built upstream of the dams for hydroelectric, industrial, drinking, irrigation, and flood mitigation purposes. However, as time passes, the deposition of solid material, transported by rivers flowing into the reservoir, reduces its storage capacity. In this work, a novel physicallybased erosion and sediment transport model developed by Politecnico di Milano (SMART-SED) is used to estimate dam filling in different Italian geological contexts. The present model differs from others in the literature since it can automatically detect drainage zones and it works at a basin scale, requiring few input parameters that can be easily downloaded from global or government databases. Moreover, it is based on robust and unconditionally stable numerical techniques, and it guarantees the mass conservation. The proposed model was applied to four watersheds, two in the Southern Alps and two in the Southern Apennines, with catchment areas ranging from 5 to 48 km2. After the model calibration on the local geological context, a validation of the obtained results was accomplished considering different time intervals and the available sediment filling data. It was observed that the SMART-SED model, initially developed for the evaluation of fluvial hazard and of sediment transport in mountainous areas, was also suitable for the estimation of dam filling. Results indeed show a precise approximation of the order of magnitude of the cumulative sediment volume produced inside the catchments and transported to the water basins. In the framework of a complex problem affecting the production of “clean” energy in a changing world, SMART-SED could become a useful tool also for dam management.