Joint simulation of regional areas burned in Canadian forest fires: A Markov Chain Monte Carlo approach

Abstract

Areas burned annually in 29 Canadian forest fire regions show a patchy and irregular correlation structure that significantly influences the distribution of annual totals for Canada and for groups of regions. A binary Monte Carlo Markov Chain (MCMC) is constructed for the purpose of joint simulation of regional areas burned in forest fires. For each year the MCMC prediction is a binary vector with regions classified to a large fire year (LF) or a small fire year (SF). The regional area burned is then obtained from empirical quantile functions; separately for LF and SF years. The MCMC results were unbiased with respect to: the annual number of LF regions, national totals, and variances of area burned. Approximately 65% of the observed regional covariance was captured in the results.

Parent Publication

• Forest Inventory and Analysis (FIA) Symposium 2008; October 21-23, 2008; Park City, UT

Citation

Magnussen, Steen. 2009. Joint simulation of regional areas burned in Canadian forest fires: A Markov Chain Monte Carlo approach. In: McWilliams, Will; Moisen, Gretchen; Czaplewski, Ray, comps. Forest Inventory and Analysis (FIA) Symposium 2008; October 21-23, 2008; Park City, UT. Proc. RMRS-P-56CD. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 10 p.