pipette
ENEnglish

FireGen: Quantifying Wildfire Risk by Simulation

Allyson Hineman, William Kleiber, Stephan Sain, Alexis Hoffman

PreprintUso en el mundo real

En palabras de los autores

Wildfire regimes in Northern California exhibit strong spatio-temporal variability, heavy-tailed fire size distributions, and sensitivity to climatic conditions. We develop a hierarchical statistical framework to model wildfire occurrence, geometry, and burned area in Northern California from 1984-2023. Fire centroids are modeled as a spatio-temporal point process with covariate-driven intensity, extended by a multiplicative gamma-based stochastic shock to better capture the distribution of monthly fire counts. Conditional on fire location, burned area polygons are represented using a parametric ellipse model that separates scale, shape, and orientation, with ellipse parameters estimated from standardized fire perimeters. Total burned area is modeled using a heteroskedastic lognormal regression incorporating climatic and spatial covariates, including vapor pressure deficit anomalies. The framework enables Monte Carlo simulation of wildfire processes and provides a probabilistic tool for assessing wildfire risk and cumulative burned area under observed climate variability.

Resultado principalEl resumen no menciona limitaciones.

Apareció: miércoles, 7 de octubre. arXiv. Preprint, todavía sin revisión por pares.

Comentario de los autores: 40 pages, plus 38 pages of supplementary material