Evolutionary applications to cellular automata models for volcano risk mitigation

Abstract

A GPGPU accelerated evolutionary computation-based decision support system for defining and optimizing volcanic hazard mitigation interventions is proposed. Specifically, the new Cellular Automata numerical model SCIARA-fv3 for simulating lava flows atMt Etna (Italy) and Parallel Genetic Algorithms (PGA) have been applied for optimizing protective measures construction by morphological evolution. A case study is considered, where PGA are applied for the optimization of the position, orientation and extension of earth barriers built to protect a touristic facility located near the summit of Mt. Etna (Italy) volcano which was interested by the 2001 lava eruption. The methodology has produced extremely positive results and, in our opinion, can be applied within a broader risk assessment framework, having immediate and far reaching implications both in land use and civil defense planning.