Sensitivity analysis of the MAGFLOW Cellular Automaton model for lava flow simulation

Abstract

MAGFLOW is a physics-based numerical model for lava flow simulations based on the Cellular Automaton approach that has been successfully used to predict the lava flow paths during the recent eruptions on Mt Etna. We carried out an extensive sensitivity analysis of the physical and rheological parameters that control the evolution function of the automaton and which are measured during eruptive events, in an effort to verify the reliability of the model and improve its applicability to scenario forecasting. The results obtained, which include Sobol' sensitivity indices computed using polynomial chaos expansion, confirm the consistency of MAGFLOW with the underlying physical model and identify water content and solidus temperature as critical parameters for the automaton. Additional tests also indicate that flux rates can have a strong influence on the emplacement of lava flows, and that to obtain more accurate simulations it is better to have continuous monitoring of the effusion rates, even if with moderate errors, rather than sparse accurate measurements.