Past logging and future fires: Ever the twain shall meet?

[originally posted on www.foreststeward.com on Oct. 20, 2010]

-> This post graciously provided by the Battles lab at UC Berkeley <-

Article reviewed: Interactive effects of historical logging and fire exclusion on ponderosa pine forest structure in the northern Rockies

By C. Naficy, A. Sala, E. G. Keeling, J. Graham, and T. H. Deluca. Published in Ecological Applications, Vol. 20, No. 7, pp. 1851-1864. Full article available.

The plot line: This study evaluated the contribution of historical logging to the widespread increases in stand density and the abundance of fire-intolerant tree species that are often attributed solely to fire exclusion. The research was conducted in the ponderosa pine/Douglas-fir forests of the northern Rocky Mountains. The study paired 23 historically logged, fire-excluded sites with 23 unlogged, fire-excluded sites. In addition, the study compared stand structure and composition in unlogged, fire-excluded and logged, fire-excluded stands to contemporary unlogged, fire-maintained stands in order to provide a baseline to quantify management-induced changes in forest characteristics.

The authors conclude that historically logged, fire-excluded ponderosa pine forests of the northern Rocky Mountains have much higher average stand density, greater homogeneity of stand structure, more standing dead trees and increased abundance of fire-intolerant trees than similar unlogged, fire-excluded forests. Further, the study found that the interactive effect of fire exclusion and historical logging significantly exceeded the effects of fire exclusion alone. Based on these findings, the authors conclude that historically logged sites are more prone to severe wildfires and insect outbreaks than unlogged, fire-excluded forests and suggest that unique restoration approaches may be required to manage ponderosa pine forests with these distinct management histories.

Relevant quote: “To the extent that modern wildfires are driven by vegetation and fuel characteristics, historically logged stands are likely more prone to severe, stand-replacing wildfires than unlogged, fire-excluded stands.”

Relevance to landowners and stakeholders:

This article challenges the common assumption that fire suppression is the sole cause of the increased forest density that is associated with severe, contemporary wildfires and insect outbreaks in many fire-prone forests of the western United States. The article highlights the lack of data on the long-term effects of various modern silvicultural practices. The authors suggest that where allowance of natural fires is not feasible, the potential negative impacts of alternative fuels treatments requires more careful consideration as part of a longer term approach to fuels management.

Relevance to managers:

For managers attempting to reduce the risk of severe wildfires in ponderosa pine/Douglas fir forests in the northern Rocky Mountains, this study makes several suggestions:

• The current forest structure and composition in historically logged ponderosa pine forests suggests that these forests should be primary targets for fuels reduction efforts.

• While previously logged, fire-excluded forests may benefit from significant mechanical stand manipulations before fire can be safely introduced, unlogged, fire-excluded forests may require much less invasive treatments.

• The authors also conclude that the potential long-term risks associated with mechanical treatments, particularly in unlogged forests, should receive greater attention because the extent to which modern mechanical treatments could have similar long-term counterproductive effects remains largely unknown.

The authors note that the effects of the interaction between historical logging and fire exclusion are likely to vary across broad geographic regions and that long-term responses to timber harvest are likely sensitive to differences in the specific nature and intensity of past silvicultural treatments.

Critique and/or limitations (there’s always something, no matter how good the article is) for the pedants:

While we accepted the general conclusions of this article, we did have some methodological/analytical concerns. For instance, the article claims to be novel in its consideration of the interactive effects of historical logging and fire exclusion. However, this interaction was not actually tested with the statistical approach used in the study. The conclusions regarding the interaction of these disturbances actually examined the additive effects of these disturbances and simply implied the interaction. Many of the potential methodological problems in this study were difficult to avoid given the nature of the data. For instance, the lack of detailed historical data on logging history and the large variation in average fire return interval. We discussed the potential problems of the close proximity of the “paired” stands and the methodological problems with the treatment of the logged and unlogged sites as paired data. The statistical approach used (i.e., paired t-tests) was inappropriate given the selection strategy for the paired stands and was perhaps too simplistic of an approach to test some of the stated objectives, such as the interaction between historical logging and fire exclusion. Finally, one of the final conclusions of this paper was that historical logging had increased forest homogeneity. However, the impact on the homogenization of stand structure was never actually tested.