Surface Flows Modeling: Cellular Automata Simulation of Lava, Debris and Pyroclastic Flows.

Cellular Automata (CA) are a computational paradigm, a valid alternative to standard methods with differential equations for modelling and simulating complex systems, whose behaviour may be specified in terms of local interactions in a context of discrete time and space. Some surface flows may be approximated to such a type of complex systems. The Empedocles Research Group developed an empirical methodology for modelling this kind of macroscopic phenomena. The CA space for surface flows is divided in hexagonal cells, whose specification (state) describes the physical characteristics (substates) relevant to the evolution of the system and relative to the space portion corresponding to the cell. The cell neighbouring, specifying the interaction range, is given by its adjacent cells. The evolution of the phenomenon is obtained by updating the values of the substates simultaneously at discrete time steps in all the cellular space according to the CA transition function, which is split in sequential "elementary" processes. This CA methodological approach for modelling large scale surface flows was applied to lava flows (the model SCIARA), pyroclastic flows (the PYR model) and debris flows (the SCIDDICA model). Satisfying simulations of real events are exhibited: the NE flank lava flows of the 2002 Etnean eruption, the pyroclastic flows invading the Sacobia area during the 1991 eruption of Mount Pinatubo in the island of Luzon (The Philippines Islands), the Chiappe di Sarno (Italy) catastrophic debris flows on 1998.