Parameter estimation of fire propagation models using level set methods

The availability of wildland fire propagation models with parameters estimated in an accurate way starting from measurements fronts is crucial to predict the evolution and allocate resources for firefighting. Thus, we propose approach estimate a model combining empirical rate spread level set methods describe front over time space. estimation affecting performed by using shapes measured at different instants as well wind velocity direction, landscape elevation, vegetation distribution. Parameter done solving optimization problem, where objective function be minimized symmetric difference between predicted instants. Numerical results obtained application proposed method are reported two simulated scenarios case study based on data originated 2002 Troy Southern California. showcase effectiveness both qualitative quantitative viewpoints.