Is Forest Structure Related to Fire Severity? Yes, No, and Maybe : Methods and Insights in Quantifying the Answer

Document Type

Contribution to Book

Journal/Book Title/Conference

Silviculture in Special Places: Proceedings of the 2003 National Silviculture Workshop

Publication Date


First Page


Last Page



Wildfires in 2000 burned over 500,000 forested ha in the Northern Rocky Mountains. In 2001, National Fire Plan funding became available to evaluate the influence of pre-wildfire forest structure on post wildfire fire severity. Results from this study will provide information on forest structures that are resilient to wild- fire. Three years of data (558 plots) have been collected from forested areas that burned in 2000, 2001, and 2002. Forests used in this study include dry ponderosa pine/Douglas-fir, cold lodgepole pine/subalpine fir, and moist western larch forests. Probability sampling of all areas within a particular fire perimeter was used to locate study sites and a sampling matrix was used to capture variation in weather, topographic setting, and pre-wildfire forest structure of which the fires represented. Fire severity (current state of soils and vegetation after the wildfire) was quantified on adjacent paired plots, with each plot representing a different forest structure. Classification trees and cluster analysis identified relations among forest structure characteristics, physical setting, and fire severity. Probability of a particular forest structure relating to fire severity was computed. This paper describes methodology used in the project, discusses challenges associated with conducting this type of study, and uses preliminary results (probabilities) from the first two years of data collection to show how forest structure relates to both crown and soil surface fire severity.


This item was written and prepared by U.S. Government employees on official time, and is therefore in the public domain.