Montana, Wyoming and New Mexico appear acutely sensitive, especially to temperature changes, and fire seasons there may respond more dramatically to global warming than in states such as California and Nevada.
Researchers have developed statistical relationships between observed climate and an 85-year record of fire extent during the 20th century and used them in conjunction with existing state-of-the-art global climate models.
"Models linking area burned in the Western states with fire-season temperature predict that global warming will bring significant increases in fire extent," said Donald McKenzie of the USDA Forest Service's Pacific Wildland Fire Science Lab. "Such increases could have consequences for threatened and endangered species in ecosystems that experience increased fire because many such species are already restricted to specialized and fragmented habitat."
McKenzie is lead author of "Climatic Change, Wildfire and Conservation," in the August issue of Conservation Biology. His co-authors are Ze'ev Gedalof of the University of Guelph in Ontario, David Peterson of the USDA Forest Service and Philip Mote, UW climate scientist and Washington state climatologist. All are members of the UW's Pacific Northwest Climate Impacts Group.
Their analysis of past climate and fire histories showed that the most important variable is summer temperature, more important than precipitation in most cases.
Using what is considered a low-end climate-change scenario - a 1.6 degree increase in summer temperatures between 2070 and 2100 compared to temperatures from 1970 to 2000 -the area burned will increase by 1.4 to five times in western states except California and Nevada, where the increase isn't so great.
Fire in California and Nevada appears to be relatively insensitive to changes in summer climate and the total area burned in these states may not respond strongly to changed climate. Parts of Northern California where warmer climate could affect the number and extent of fires, for example, are offset by areas in Southern California where fires are almost all caused by human activities and the combination of high temperatures and dry air associated with the Santa Ana winds, not temperature by itself, McKenzie says.
At the other extreme are Montana, Wyoming and New Mexico. Montana is the most sensitive, with the models predicting a 5-fold increase in mean area burned over the observed range in climate, the authors write.
In Washington and Oregon, the effects will be intermediate because precipitation and temperature are both associated with fire extent in these states, rather than temperature alone.
More frequent, more extensive fires in forest ecosystems will likely reduce the number and size of patches of older forests, the authors say. Corridors of wild areas between forests, through which species might migrate if their home territory goes up in flames, also could be affected, possibly eliminated.
"The winners after fires in these cases are the weedy, adaptive, quickly reproducing species," McKenzie says. "The losers are the ones needing more stable environments."