Spatiotemporal Prediction of Wildfire Size Extremes with Bayesian Finite Sample Maxima

Wildfires are becoming more frequent in parts of the globe, but predicting whereand when wildfires occur remains difficult. To predict wildfire extremes across the contiguousUnited States, we integrate a 30-yr wildfire record with meteorological and housing data inspatiotemporal Bayesian statistical models with spatially varying nonlinear effects. We com-pared different distributions for the number and sizes of large fires to generate a posterior pre-dictive distribution based on finite sample maxima for extreme events (the largest fires overbounded spatiotemporal domains). A zero-inflated negative binomial model for fire countsand a lognormal model for burned areas provided the best performance. This model attains99% interval coverage for the number of fires and 93% coverage for fire sizes over a six yearwithheld data set. Dryness and air temperature strongly predict extreme wildfire probabilities.Housing density has a hump-shaped relationship with fire occurrence, with more fires occur-ring at intermediate housing densities. Statistically, these drivers affect the chance of anextreme wildfire in two ways: by altering fire size distributions, and by altering fire frequency,which influences sampling from the tails of fire size distributions. We conclude that recentextremes should not be surprising, and that the contiguous United States may be on the vergeof even larger wildfire extremes.