Knowledge of past fire regimes associated with mountain big sagebrush-dominated landscapes is inadequate for accurate assessment of current departures from historic conditions. Scientifically-based understanding of the role of fire in big sagebrush ecosystems is particularly critical when managing landscape mosaics that include habitat for sagebrush obligate species such as Greater sage-grouse and pygmy rabbit (Crawford et al. 2004, Welch 2005). Inferences of fire frequency based upon proxy fire-scarred trees have been widely utilized for management purposes but are untested, and may have limited application, because of the limited number and restricted distribution of supporting studies. Finally, these studies fail to address the uncertainties associated with fire behavior across the fuels threshold at the forest/shrubland ecotone. Alternative estimates of big sagebrush fire frequency have been based upon the time required for big sagebrush to recover after fire, as measured by percent cover and plant density (Harniss and Murray 1973, Humphrey 1984, Wambolt et al. 1999, 2001). Similarly, maximum fire-free intervals for big sagebrush communities prone to invasion by conifers are estimated based upon the time required for tree invasion and dominance (Miller et al. 1999, Miller and Tausch 2001). Published studies of mountain big sagebrush post-fire recovery are also limited primarily to higher latitudes, and as a rule are not linked to tree invasion studies. This project will provide proxy tree-based estimates of fire frequency for mountain big sagebrush from multiple sites located in the southern half of its distribution. It will generate estimates for shrub recovery and tree invasion rates for the same region. It will provide a test of the appropriateness of using scar-based proxy trees to estimate fire frequency for mountain big sagebrush. Landscape-scale modeling of fire and plant community response will assess the adequacy of tree scar-based estimates of fire frequency for mountain big sagebrush and provide a framework for extrapolating our results to un-sampled landscapes.
|
