Forecasting through simulations the shape of lava invasions in a real topography represents a challenging problem, especially considering that the phenomenon usually evolves for a long time (e.g. from a few to hundreds of days) and on very large areas. In the latest years, Cellular Automata (CA) have been well recognized as a valid computational approach in lava flow modelling. In this paper we present some significant developments of SCIARA, a family of deterministic CA models of lava flows which are optimized for a specific scenario through the use of a parallel genetic algorithm. Following a calibration-validation approach, the model outcomes are compared with three real events of lava effusion.
Modelling macroscopic phenomena with cellular automata and parallel genetic algorithms: An application to lava flows
D'AMBROSIO D;DI GREGORIO S;RONGO R;SPATARO W;
2007-01-01
Abstract
Forecasting through simulations the shape of lava invasions in a real topography represents a challenging problem, especially considering that the phenomenon usually evolves for a long time (e.g. from a few to hundreds of days) and on very large areas. In the latest years, Cellular Automata (CA) have been well recognized as a valid computational approach in lava flow modelling. In this paper we present some significant developments of SCIARA, a family of deterministic CA models of lava flows which are optimized for a specific scenario through the use of a parallel genetic algorithm. Following a calibration-validation approach, the model outcomes are compared with three real events of lava effusion.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
