Recovery Plan for the Northern Spotted Owl - DRAFT

ums and all rebums are thought to have been human-ignited), the reburnhypothesis is likely to be true in certain areas. However, it is not clear whetherrebums were a common event prior to European settlement in the moistportion of the Douglas-fir region.After crown closure, potential surface fire behavior declines, and then graduallyincreases in the old-growth seral stage (Agee and Huff 1987). Rebums inroughly 100-year-old stands during the late 1400s (Henderson and Peter 1981)may suggest that crown fire behavior independent of surface fuels in thesethick-canopied stands may be an additional significant type of fire. Currentknowledge is insufficient to tell.For many years, the pattern of stand replacement fire summarized above was aparadigm of fire for the west side Douglas-fir region. Recent work, particularlyin the Oregon Cascades in drier western hemlock plant associations, suggestsa higher fire frequency, and different ecological role, for fire in mesic to dryDouglas-fir forest, reinforcing the output of the climate-based fire cycle model(Agee 1991 a). A site in the western Oregon Cascades (Stewart 1986) near theH. J. Andrews Experimental Forest regenerated after a stand replacement firein ca. 1530, but had experienced three partial mortality fires since then, in ca.1660, 1860, and 1890. Some of these were in the settlement period and probablyreflect human-caused fires of that period, but the partial mortality associatedwith them is significant. Over a broader area several miles to the southeastencompassing similar forest types, Morrison and Swanson (1990) suggesta natural fire rotation of 95 to 145 years over the last 5 centuries, well belowthat of the moist Douglas-fir forests of Washington. The patchiness of at leastsome of the fires is illustrated by the fire severity maps in Morrison andSwanson (1990). A similar fire regime was noted by Means (1982) on dry Douglas-firsites in the western Oregon Cascades and by Agee and Dunwiddie(1984) for dry Douglas-fir forests in Washington's San Juan Islands. Anotherfire history analysis was completed by Teensma (1987) near the area studiedby Morrison and Swanson. Using conservative methods that did not recognizeunderburns with no resulting regeneration or substantial fire-scarring of trees,Teensma estimated a natural fire rotation of 100 years over the last 5 centuries.If fires of moderate severity are removed from the analysis, a stand-replacementmean fire return interval is 130 to 150 years, suggesting that intensefires are a significant part of the natural fire regime in this area, but thatfires of lower severity also occur. Other stands of 500 years age or older existwithout much evidence of recurrent fire.These studies indicate that a variable fire regime with shorter fire return intervalsthan moist Washington Douglas-fir forests occur in the central OregonCascades, and in other mesic to dry Douglas-fir forests. It is, in a sense, atransitional area to the Klamath subregion in terms of fire history.Huff (1984) has summarized the species response to disturbance regimes formoist Douglas-fir forests. If fire is absent for 700 to 1,000 years on wet sites,Douglas-fir will drop out of the stand, and western hemlock, Pacific silver fir,or western redcedar will be the primary seed source for post-fire regeneration.On sites with fire return intervals in the 300- to 600-year range, well withinthe longevity of individual Douglas-fir, mixed dominance of Douglas-fir andwestern hemlock or Pacific silver fir will result from a typically severe standreplacement fire (Figure F.4). By age 200+years, the characteristics of oldgrowthare almost always present. The Douglas-fir component, having developedafter the previous centuries-old fire, provides the large live tree criterion.Both Douglas-fir and the more dense western hemlock begin to supply thelarge log component as they begin to die from suppression, disease, orwindthrow.445