Abstract:
During 1988-1989, stream habitat variables were compared between beaver-dam sites and unoccupied-stream sites to identify variables that may have been important for beaver (Castor canadensis) in selecting dam sites in the streams of the Drift Creek basin, Lincoln County, Oregon. Increasing valley floor width and grass/sedge cover and decreasing stream width, stream gradient, red
alder (Alnus rubra) cover, and shrub cover had positive effects on selection of dam sites. A discriminant function model correctly classified 83% of beaver-dam sites and 88% of unoccupied-stream sites with
chance-corrected classification rate of 69% (Kappa statistic). Applicability of the U.S. Fish and Wildlife Service Habitat Suitability Index (HSI) model for beaver was
tested in the basin. HSI-model scores were highly influenced by the water life requisite and differed between darn sites and unoccupied sites. Scores for the food life requisite did not differ between dam sites and unoccupied sites; food may not be a factor influencing selection of dam sites. The water fluctuation variable was subjective and was not sensitive enough to
differentiate darn sites from unoccupied sites. Three geomorphic characteristics (stream width, gradient, and valley floor width) were used in developing a new Habitat
Suitability Index (HSI) model for the basin. The HSI model scores produced by the new model were different between darn sites and unoccupied sites. Capture frequencies of small mammal and amphibian species were compared between beaver-occupied reaches and
unoccupied reaches in 5 streams in the Oregon Coast range during the fall of 1989. Species richness, species diversity, and equitability of small mammal and amphibian
communities did not differ between occupied and unoccupied reaches. The capture frequency of Microtus spp. was consistently higher at occupied reaches than at unoccupied
reaches. None of small mammal and amphibian species had consistently higher or lower capture frequencies at occupied reaches than at unoccupied reaches in each of 5 streams. Inconsistent abundance of small mammal and
amphibian species may reflect diverse habitat conditions among occupied reaches.
