Year of Award


Document Type


Degree Type

Doctor of Philosophy (PhD)

Degree Name


Department or School/College

College of Forestry and Conservation

Committee Chair

LLoyd P. Queen

Commitee Members

Carl A. Seielstad, Ronald Wakimoto, Anna Klene, Colin Hardy


Fire Regime, Fire-on-Fire Interaction, MTBS, Re-burn, Severity


University of Montana


Current knowledge about wildfire occurrence is not complete. Fire researchers and managers hold the assumption that previous wildfires affect subsequent wildfires; however, research regarding the interactions of large wildfires at their common boundaries is missing from the literature. This research focuses on understanding the influence of previous large wildfires on subsequent large wildfires in three wilderness areas: The Greater Bob Marshall, the Selway-Bitterroot, and the Frank Church. Data from the Monitoring Trends in Burn Severity (MTBS) project, which mapped large wildfires in the western United States occurring since 1984, are used for the research. The combination of using wilderness areas and remotely sensed images allows an objective and consistent analysis of fire-on-fire interaction that is extensive in both time and space. Standardized methods for analyzing fire interactions do not currently exist, therefore methods were developed, tested, and refined to describe, quantify, and compare once-burned and re-burned locations within a subset of ten fires in terms of size, location, timing between fires, and severity. These methods were then used to address the question of whether re-burns occur within each of the three wilderness areas. Edge and re-burn characteristics were also derived and quantified. Results were statistically and empirically compared to randomized fire intersections and to published fire history research for each area. Although a low proportion of each study area burns or re-burns, when a new fire encounters a previous fire it re-burns onto the previously burned area approximately 80% of the time. Current large wildfires are behaving in a typical fashion, although on some landscapes the amount of re-burn is not different from what would be expected due to chance. Lastly, the complexity of the post-fire landscape was assessed using texture metrics. Pre-fire and post-fire landscapes were shown to be different, with post-fire landscapes exhibiting greater diversity than pre-fire landscapes. When re-burned areas were compared to those locations that had only burned once, however, landscapes generally became less complex. Although wildfires and wildfire effects in each wilderness area differ, the overall results of this study confirm that previous fires do affect subsequent wildfires.



© Copyright 2012 Casey Teske