Evaluating prescribed fire as habitat restoration for an endemic threatened species.

Background/Question/Methods

Fire suppression in west-central Idaho has altered ponderosa pine forest structure by increasing forest density and closing open meadows, which has lead to the subsequent decline of habitat for an endemic species, Urocitellus brunneus (northern Idaho ground squirrel). Urocitellus brunneus was listed as ‘threatened’ under the Endangered Species Act in 2000. Managers surmise that U. brunneusis a fire dependent species, thus burning is the primary restoration tool to recover habitat. Restoration goals are to decrease conifer density immediately post-fire, maintain an open understory by decreasing ladder and understory fuels immediately post-fire and increase understory cover by 10-30% one year post-fire. In order to evaluate the prescribed fire, we measured habitat attributes at three sites: Cap Gun, Summit, and Price Valley.  Each site was divided into two units: a colonized reference unit and an uncolonized burn unit to be treated with fire. Before and after the fall burn, we measured tree density, canopy cover, understory cover, understory visual obstruction, and litter depth. We analyzed variables with a PERMANOVA followed by mixed univariate ANOVAs and Tukey-Kramer pairwise comparisons. Species cover data were run in a non-parametric multidimensional scaling (NMS) analysis.

Results/Conclusions

The prescribed burn did not change conifer density and canopy cover one year post-fire (density: F_1,2 = 0.42, P = 0.58; cover: F_1,2 = 0.31, P = 0.64). The PERMANOVA reported no significant differences in understory functional group cover before and after the burn (pseudo F_1,2 = 0.71, P = 0.62). There is a strong trend, although not statistically significant, that understory fuels, visual obstruction and litter depth were decreased by fire (visual obstruction: F_1,2 = 4.83, P = 0.16; litter depth: F_1,2 = 8.61, P = 0.099). The NMS ordination explained 81% of the variation within the data with one dimension that showed no annual variation in reference unit species composition; burn unit species composition shifted toward reference assemblages after the prescribed fire (Stress < 0.00001, P = 0.099). Restoration goals to reduce conifer density, ladder fuels and increase understory cover were not obtained, but understory fuels were reduced by the burn treatment. Restoration goals might be obtained in subsequent years; thus, we encourage continued monitoring. We also suggest that shifting understory species composition toward reference conditions should be a goal instead of understory cover. Lastly, subsequent fires or thinning treatments might help attain restoration goals.