The effects of forest management on terrestrial salamanders

Jami E MacNeil, Purdue University

Abstract

Maintaining a balance of timber production and conservation in forest management requires an understanding of how timber harvest techniques affect wildlife species. Terrestrial salamanders are useful indicators of mature forest ecosystem health due to their importance to ecosystem processes and their sensitivity to environmental change. The topic of timber harvests and their effect on terrestrial salamanders has received much attention in the literature but disagreements and uncertainty about the severity and duration of these effects persist. I employed artificial cover objects (ACOs) to monitor salamander relative abundance in six forest management treatments both before and after harvests. Prior to harvests, I assessed the efficacy of two cover-object materials in sampling forest herpetofauna. In a side-by-side comparison of 930 pairs of ACOs, salamanders were found more under wood objects than plastic objects (p < 0.001), a result likely due to a greater capacity of wood objects to retain moisture and buffer temperature fluctuations. Pre-harvest data showed little variation in mean salamander counts across treatment types, but counts were greater in spring than fall. Counts varied by slope aspect in the fall but not the spring. Mean encounters of Eastern Red-backed (Plethodon cinereus) and Northern Slimy Salamanders (P. glutinosus) declined from pre- to post-harvest in group selection cuts (p < 0.001); however, mean encounters of red-backed salamanders also declined in uncut control sites (p = 0.025). The effect of sample period was statistically significant for all species tested (p < 0.001), suggesting high temporal variation in salamander abundance. Average daily air temperature was negatively correlated with counts of red-backed and Northern Zigzag Salamanders (P. dorsalis ; rs ≈ -0.43, p < 0.001). In general, harvests which removed the forest canopy had a negative effect on salamander relative abundance during the years immediately following harvest. Salamander declines were not as severe as those seen in most previous research, a fact which may be explained in part by the relatively small size (≤ 4 ha) of the harvests examined here. I found no effect of controls, shelterwoods, and forested sites adjacent to harvests. Much of the variation in salamander counts is likely driven by seasonal patterns in precipitation and temperature. Longer-term monitoring will be necessary to understand the full impacts of forest management on terrestrial salamanders.

Degree

M.S.

Advisors

Williams, Purdue University.

Subject Area

Ecology

COinS