Estimativa do conteúdo de humidade da vegetação através da modelação numérica

Abstract

In recent years, Portugal suffered several devastating wildfires, such as in 2003, 2005, and 2017. Wildfires are related to fuel load, climate, and ignition factors. Currently, land use and occupation management is a way to reduce wildfires. The work’s objective is to improve the fuel moisture content (FMC) representation across mainland Portugal, through numerical modelling and machine learning. Initially, numerical simulations were performed using the Applications of Research to Operations at MEsoscale (AROME), a limited-area non-hydrostatic operational atmospheric model, creating forcing files to initialize the SURFEX surface model. SURFEX output variables were used as predictors to estimate the FMC through a machine learning-based classifier. These results are useful for understanding the FMC spatiotemporal variability in Portugal and important to identify high-fuel load areas which is crucial for integrated fire management. This work was funded by the Foundation for Science and Technology, I.P., under the PyroC.pt project (Ref. PCIF/MPG/0175/2019) and PhD Grant (2022.11960.BD).