Climate variability, climate change, and western wildfire with implications for the urban-wildland interface


Keeton, W.S., Mote, P.W., Franklin, J.F. 2007. Climate variability, climate change, and western wildfire with implications for the urban-wildland interface. pp. 225-253. In A. Troy and R. Kennedy (eds)., Living on the Edge: Economic, Institutional and Management Perspectives on Wildfire Hazard in the Urban Interface. Advances in the Economics of Environmental Resources , Vol. 6. Oxford, United Kingdom: Elsevier Sciences JAI Press.


Climatic change during the next century is likely to significantly influence forest ecosystems in the western United States, including indirect effects on forest and shrubland fire regimes. Further exacerbation of fire hazards by the warmer, drier summers projected for much of the western U.S. by climate models would compound already elevated fire risks caused by 20th century fire suppression. This has potentially grave consequences for the urban-wildland interface in drier regions, where residential expansion increasingly places people and property in the midst of fire-prone vegetation.

Understanding linkages between climate variability and change, therefore, are central to our ability to forecast future risks and adapt land management, allocation of fire management resources, and suburban planning accordingly. To establish these linkages we review previous research and draw inferences from our own retrospective work focused on 20th century climate-fire relationships in the U.S. Pacific Northwest (PNW). We investigated relationships between the two dominant modes of climate variability affecting the PNW, which are Pacific Decadal Oscillation (PDO) and El Niño/Southern Oscillation (ENSO), and historic fire activity at multiple spatial scales. We used historic fire data spanning most of the 20th century for USDA Forest Service Region 6, individual states (Idaho, Oregon, and Washington), and 20 national forests representative of the region’s physiographic diversity.

Forest fires showed significant correlations with warm/dry phases of PSO at regional and state scales; relationships were variable at the scale of individual national forests. Warm/dry phases of PDO were especially influential in terms of the occurrence of very large fire events throughout the PNW. No direct statistical relationships were found between ENSO and forest fires at regional scales, although relationships may exist at smaller spatial scales. However, both ENSO and PDO were correlated with summer drought, as estimated by the Palmer Drought Severity Index (PDSI), and PDSI was correlated with fire activity at all scales. Even moderate (&plusmn0.3 C decadal mean) fluctuations in PNW climate over the twentieth century have influenced wildfire activity based on our analysis. Similar trends have been reported for other regions of the western U.S. Thus, forest fire activity has been sensitive to past climate variability, even in the face of altered dynamics due to fire suppression, as in the case of our analysis.

It is likely that fire activity will increase in response to future temperature increases, at the same or greater magnitude as experienced during past climate variability. If extreme drought conditions become more prevalent we can expect a greater frequency of large, high-intensity forest fires. Increased vulnerability to forest fires may worsen the current fire management problem in the urban wildland interface. Adaptation of fire management and restoration planning will be essential to address fire hazards in areas of intermingled exurban development and fire-prone vegetation. We recommend: 1) landscape-level strategic planning of fire restoration and containment projects; 2) better use of climatic forecasts, including PDO and ENSO related predictions; and 3) community-based efforts to limit further residential expansion into fire-prone forested and shrubland areas.