Taylor, S. W., Woolford, D. G., Dean, C. B., & Martell, D. L. (2013). Wildfire prediction to inform fire management: Statistical science challenges. Statistical Science, 28(4), 586–615. doi:10.1214/13-sts451
Xi, D. D., Taylor, S. W., Woolford, D. G., & Dean, C. (2019). Statistical models of key components of wildfire risk. Annual Review of Statistics and Its Application, 6, 197–222. doi:10.1146/annurev-statistics-031017-100450
Xi, D. D., Dean, C., & Taylor, S. W. (2020). Modeling the duration and size of extended attack wildfires as dependent outcomes. Environmetrics, e2619. doi:10.1002/env.2619
Fire size and duration are both of interest for prediction. Considers survival models for each (accelerated failure time models). The survival models can be linked to form a joint distribution using copulas. Applied to large fires in British Columbia, the survival models can incorporate covariates for each outcome and also account for dependence between the outcomes (via the copula); they find that “joint modeling outperforms modeling the outcomes separately.” They used covariates derived from vegetation buildup, fire weather, temperature, and local topography (slope, elevation), and these proved useful for prediction.
Also it is excellent that there is a copula named “Frank.”