D?AMBROSIO D. (1), DI GREGORIO S. (1), IOVINE G. (2), LUPIANO V. (2), MERENDA L. (2), RONGO R. (3), SPATARO W. (2), 2002, Simulating the Curti-Sarno debris flow through cellular automata_ the model SCIDDICA (release S2), Physics and chemistry of the earth (2002) 27 (2002): 1577–1585. doi_10.1016/S1474-7065(02)00179-1,
URL: http://www.cnr.it/prodotto/i/41488

Cellular automata (CA) are based on a regular division of the space in cells. Each cell embeds an identical finite automaton, whose input is given by the states of neighbouring cells. The transition function of the CA is made of a set of rules, simultaneously applied, step by step, to each cell of the cellular space. Rules are derived by subdividing, in computational terms, the physical phenomenon into a set of independent, elementary processes. By properly combining each elementary result, the behaviour of the phenomenon can be simulated. Debris flows are dense mixtures of sediment and water, which surge down the slopes and along the drainage system, characterised by severe destructive potential. They can be described in terms of local interactions among their elementary portions, and can thus be efficiently modelled through CA. Debris-flows rheologic equations cannot be easily solved without making substantial simplifications. By applying CA, a phenomenological description - able to overcome resource computational limits - can be obtained. On May 1998, hundreds of soil slip-debris flows were triggered by exceptional rains in Campania (Italy), mostly on the slopes of Pizzo d’Alvano. Aiming at modelling purposes, the Curti debris flow was selected as a case study, among the whole population of landslides triggered by the event. The general frame of SCIDDICA_S2 is inherited from previous releases, recently applied for the simulation of the 1992 Tessina (Italy) earth flow and of the 1984 Mt. Ontake (Japan) debris avalanche. Since its S1 release, the model satisfactorily simulated the Curti-Sarno debris flow. Latest improvements to the transition function led to the S2 release, and to better simulations (presented here). SCIDDICA exhibits a notable flexibility in modelling and simulating flow-like landslides. It could be usefully applied in hazard mapping (also through a statistical approach), and in evaluating the effects of either human works or ‘‘accidents ’’along the path of the flow.

Data from https://intranet.cnr.it/people/