The three most fatal Italian energy-related disasters ever occurred as a result of landslides or floods triggered by intense precipitation are the Vajont, Gleno, and Molare disasters. Numerical simulation of historical disasters enables improved computational codes for hydrogeological risk analyses associated with the electrical system and its vulnerabilities to extreme natural events, which are also dependent on climate change.
SPHERA (RSE SpA) is a 3D computational fluid dynamics (CFD) research code, based on the Smoothed Particle Hydrodynamics (SPH) method, dedicated to flash floods and rapid landslides at the sub-basin scale for the safety of hydropower plants, power substations, and power line supports.
In 2021 SPHERA was improved and applied to the ex-post fluid dynamics analysis of the Vajont Dam overflow flood.The resistance strength on the topographic surface depends on the CORINE Land Cover class through the characteristic length of the roughness. The overflow rate from the Vajont reservoir has been reconstructed by a simple numerical model fed by the information available in situ.
The 3D CFD-SPH simulation showed the 3D nonlinear characteristics of the Vajont Dam overflow flood. The SPH simulation does not require calibration: each input data is in accordance with the relevant literature. The code correctly simulates the flooded areas in 89.2% of the examined surface. Validation shows that “false positives” are 11.5% of positive cases, whereas “false negatives” are 9.7% of negative cases. Comparison of simulated maximum water head in the outlet section of Vajont canyon YSPH=77.0m and the associated measurement Ymeas=70m (Reberschack, 2003, [1], Bosa & al.; Petti, 2013, [2]) shows an overestimation of 10%. In this section, the arrival time estimate falls within the range defined by the associated measure.

Two appendices required by the current research project are given: a feasibility analysis on possible future 3D CFD-SPH research studies of the Gleno and Molare disasters, characterized by collapse-triggered flooding; and a comparison of SPH results of the Kagerplassen earthen embankment collapse, a hydrogeological risk study already done in System Research (2019) for flood protection works and damage to power line supports, with published measurements in 2021.
The SPHERA results are also available in “.asc” format for GIS tools, for the FloodRisk2 damage model (University of Basilicata, RSE SpA) and the Relief v.2 power grid model .0 (RSE SpA). RSE’s SPHERA numerical chain tools are available for free on github.com .