Light PhD Seminar: Fire-spotting modelling in wildfire simulations. Motivation, state of the art and test over a real case

Location: Maryam Mirzakhani Seminar Room at BCAM

Title
Fire-spotting modelling in wildfire simulations. Motivation, state of the art and test over a real case.


Abstract
Wildfire propagation is a non-linear and multiscale phenomenon in which there are involved multiple physical and chemical processes. One critical mechanism in the spread of wildfires is the so-called fire-spotting: a random phenomenon which occurs when embers are transported over large distances by wind, causing the start of new spotting ignitions which jeopardize firefighting actions. Due to its nature, fire-spotting is usually modeled as a probabilistic process. Three principal processes are involved during the fire-spotting: firebrands´ generation, transport and landing, and spot ignition. In this light seminar, we will present the physical parametrization of fire-spotting RandomFront, developed by the Statistical Physics research group in BCAM. RandomFront has been implemented in the operational wildfire spread simulator PROPAGATOR, based on a cellular automata approach. For validation purposes, the laboratory-scale framework is not a suitable setting because of scaling laws. Therefore, we have reproduced the evolution of a real wildfire occurred in Italy, in which the fire-spotting effects played a critical role in it spread, to test how RandomFront is able to reproduce wildfires at its natural scale. In addition, we have implemented established fire-spotting empirical parametrizations for cellular automatabased wildfire models to compare the performance between them. The results show that RandomFront parametrization, on the one hand reproduces the main spotting effects given by the available literature parametrizations, while, on the other hand, generates a variety of fire-spotting situations as well as long range fluctuations of the burning probability. The physical parametrization allows for complex patterns of fire spreading in this operational simulator context. No advanced knowledge about wildfire modelling or probability distributions are needed to follow this seminar.