Fire history and fire regimes of the central western Cascades of Oregon

Fire history is documented for an 11,000 hectare (27,110 acre) area, including H. J. Andrews ExperimentalForest. Fire scar and tree origin data were collected mainly from stumps at 359 sites. Thirty-five fire eventsare mapped from 1482 to 1952. Mean fire return intervals(MFRI) are derived from data at individual sites, about 5hectares in size (12 acres), rather than from the corrected master fire chronology. Fire frequency is higher than the generally acceptedlevel. Both catastrophic fires and underburning were partof the pre-fire suppression fire regimes. The MFRI of partial or complete stand-replacing fires is 166 years, and the MFRI of all fires is 114 years. Mean fire return intervals vary considerably between different landscape positions. Sites with south aspects or on ridges, at higher elevations, or exposed to east winds,have the shortest MFRI (for all fires, less than 100years). Fire is least frequent at lower elevations, invalley bottoms, on north aspects, and where protected from east winds (MFRI, all fires, 150 years or more). Natural fire rotation is shortest prior to Anglo-settlement (1772-1830), at 78 years, increases to 87 yearsafter settlement (1851-1909), then increases dramatically to 587 years with fire suppression. Aboriginal burning supplemented lightning ignitions, but the extent of aboriginal burning could not be determined. Presumably, aboriginal burning was ecologically important within atleast parts of the study area, e.g., meadows and huckleberry fields. In the absence of logging, the decrease in fire frequency due to fire suppression would greatly influence forest composition and structure, and fuel loading characteristics. No single fire regime adequately describes the spatial and temporal pattern of fire observed here. The study does not exclude fire regimes described previously for the region, rather it includes fire regimes described for otherregions. Further distinction between the frequencies of stand-replacing fires and lower intensity fires is necessary. Future research methods should take into account more extensive underburning than reported here. The methods used in this study may have been too conservative in that they may have discarded a considerable amount of evidence for underburning.