Mammalian abundance in riparian and adjacent terrestrial habitats prior to and immediately following forest harvesting in the Copper Lake watershed, Newfoundland

The Copper Lake Buffer Zone Study was initiated in 1994 to evaluate the effects of forest harvesting practices on mammalian, avian and piscine assemblages in a small watershed. This portion of the Copper Lake study evaluates mammalian abundance in riparian habitats both prior to and immediately following forest harvesting. Three headwater streams within the Copper Lake watershed received different harvesting treatments; i.e., harvested with no riparian buffer, harvested with a 20 m riparian buffer and not harvested (essentially a buffer of at least 100 m). The riparian habitats and adjacent interior forests on both sides of each stream were evaluated for mammal usage prior to and after harvesting. -- Summer trapping studies of small mammals, i.e., masked shrews (Sorex cinereus), meadow voles (Microtus pennsylvanicus) and deer mice (Peromyscus maniculatus), prior to harvesting indicated that these species were initially low in abundance. Shrew abundances showed a large increase in the summer of 199S in both harvested and unharvested study sites. Post-harvesting data from the summers of 1995 and 1996 indicated no noticeable, immediate effect of harvesting on S. cinereus when the clear-cut sites were compared to uncut sites. Within the 20 m buffers, however, shrew abundances were significantly higher than in the adjacent clear-cuts. M. pennsylvanicus and P. maniculatus showed slight increases in harvested areas. The cyclical nature of such small mammal species may account, in part, for these increases, particularly since abundances were so low in 1994. -- Winter track data for other mammalian species, i.e., pine marten (Martes americana), weasel (Mustela erminea), red fox (Vulpes vulpes), snowshoe hare (Lepus americanus) and red squirrel (Tamiasciurus hudsordcus), indicated that prior to harvesting, winter activity was generally higher in the forest than at the stream edges. After harvesting occurred, I found that more activity occurred in residual patches of forest, including riparian buffers, than in the clear-cuts. Marten, hare and squirrel tracks were significantly more abundant on forested transects. No marten tracks were observed on the clear-cuts without a buffer, while track abundances increased on the uncut sites. The results of this study raise the question of whether a 20 m buffer is adequate to support increased competition for space and prey due to species packing, especially considering the low diversity and densities of small mammals in insular Newfoundland. This can also be compounded with the potential loss of buffer habitat over time due to tree blowdown