Long-term partial cutting impacts on Desmognathus salamander abundance in West Virginia headwater streams

To understand long-term impacts of partial cutting practices on stream-dwelling salamanders in the central Appalachians, we examined pooled abundance of Desmognathus fuscus and D. monticola salamanders (hereafter Desmognathus) in headwater streams located within long-term silvicultural research compartments on the Fernow Experimental Forest, Tucker County, West Virginia. We sampled Desmognathus salamanders in 12 headwater streams within silvicultural research compartments that have been subjected to partial cutting for approximately 50 years. We used an information-theoretic approach to test five a priori models explaining partial cutting effects at the compartment-level on Desmognathus abundance and eight a priori models explaining stream reach-scale habitat effects on Desmognathus abundance. Our modeling efforts resulted in the selection of two competing models explaining partial cutting effects on Desmognathus abundance at the compartment-level. The VOLUME model, which incorporated cumulative timber volume harvested within compartments, received the greatest support and indicated that Desmognathus abundance was impacted negatively by increased timber volume removal. The second model, LASTDISTURB, incorporating the single variable of time since last harvest activity, indicated that Desmognathus abundance increased with time since last harvest at the compartment-level. For stream reach-scale habitat variables, the EMBEDDED model incorporating the percent of embedded substrate within streams, received the strongest support for explaining Desmognathus abundance. Our results suggest that long-term partial cutting suppresses Desmognathus abundance, possibly by increasing stream sedimentation and thereby reducing available cover for juvenile and adult salamanders. However, these practices do not appear to have threatened long-term persistence of Desmognathus in central Appalachian headwater streams.