@article {724, title = {Analyzing extreme disturbance events: Fire in Los Padres National Forest}, journal = {Ecological Applications}, volume = {7}, year = {1997}, month = {1997}, pages = {1252-1262}, abstract = {Extreme disturbance events may strongly influence the structure and functioning of many ecosystems, particularly those in which large, infrequent events are the defining forces within the region. This paper introduces the extremal fire regime (i.e., the time series of the largest fire per year) and the assumptions implicit in its analysis. I describe the statistics of extremes and demonstrate their application to the fire regime of Los Padres National Forest, California, to compare two regions (i.e., Main and Monterey Divisions), to test for a shift in fire regime due to fire suppression, and to examine climatic events as a forcing mechanism for large fires. Despite their similarity and proximity, the Main Division exhibited a much higher frequency of large fires (and shorter return time) compared to the Monterey Division. Comparison of time periods 1911-1950 and 1951-1991 indicated that fire suppression had no effect on the distribution of very large fires in the Main Division, although the frequency of fires smaller than ~4,000 ha declined. Comparing distributions of an index for severity of Santa Ana conditions (i.e., characterized by hot, dry winds) and extreme fire events in the Main Division indicated a convergence of distributions with increasing event size. The distribution of fire events larger than ~4,000 ha appears to be coupled with that of severe Santa Ana conditions, suggesting a strong climatic forcing for extreme fires and a threshold for the transition from small- to large-fire dynamics. Results indicate the usefulness of extremal fire regime analysis for comparisons over space and time and for examining a potential forcing mechanism. This approach can be applied to any disturbance regime in which large events play an important role, providing ecologists and land managers a useful tool for understanding and predicting dynamics of extreme disturbance events.}, author = {Moritz, M. A.} }