Scenarios of probability and potential power of vegetation fires in the department of Landes, France

摘要

Although the southwestern region of France has not experienced a major wildfire in several decades, the wildfire potential of this area remains high. This threat not only represents a matter of human safety, especially given the substantial influx of tourists during the summer season, but also endangers the important forest resources of this region. Widely used in North America, modelling wildfire probability at the landscape level is now a proven method for wildfire risk mapping. The aim of this study is to adapt Burn-P3, a Canadian simulation model for burn probability mapping, to the geographic context of the Landes department, France, in order to calculate wildfire probability and potential fire intensity (considered together as the two main components of wildfire hazard) for several weather scenarios. This exercise involved a conversion of the vegetation of the Landes region to the forest fuel types derived from the Canadian method. Results reveal a highly heterogeneous wildfire probability and head fire intensity across the study landscape, with the two modelled components exhibiting divergent spatial configurations. They also show a rapid and nonlinear increase in wildfire risk according to the severity of weather conditions. These findings represent the first detailed mapping of wildfire risk in southwestern France using a probabilistic simulation model. This product can be readily integrated in risk analysis and guide targeted mitigation actions to improve human safety and optimize the protection of forest resources.