Rocky Mountain Research Station

Title: Modeling containment of large wildfires using generalized linear mixed-model analysis

Author: Finney, Mark; Grenfell, Isaac C.; McHugh, Charles W.

Date: 2009

Source: Forest Science. 55(3): 249-255.

Description: Billions of dollars are spent annually in the United States to contain large wildland fires, but the factors contributing to suppression success remain poorly understood. We used a regression model (generalized linear mixed-model) to model containment probability of individual fires, assuming that containment was a repeated-measures problem (fixed effect) and individual fires were random effects. Changes in daily fire size from 306 fires occurring in years 2001-2005 were processed to identify intervals of high spread from those of low spread. The model was tested against independent data from 140 fires in 2006. The analysis suggested that containment was positively related to the number of consecutive days during which the fire grew little and the number of previous intervals. Containment probability was negatively related to the length of intervals during which the fire exhibited high spread and the presence of timber fuel types, but fire size was not a significant predictor. Characterization of containment probability may be a useful component of cost-benefit analysis of large fire management and planning systems.

Keywords: wildfire containment, large wildfires, fire suppression, generalized linear mixed models

View and Print this Publication (329.65 KB)

Publication Notes: 

• We recommend that you also print this page and attach it to the printout of the article, to retain the full citation information.
• This article was written and prepared by U.S. Government employees on official time, and is therefore in the public domain.

 Get the latest version of the Adobe Acrobat reader or Acrobat Reader for Windows with Search and Accessibility

Citation

Finney, Mark; Grenfell, Isaac C.; McHugh, Charles W.  2009.  Modeling containment of large wildfires using generalized linear mixed-model analysis.   Forest Science. 55(3): 249-255..