Sampling plethodontid salamanders: Sources of variability

Recent evidence of possible worldwide amphibian population declines has highlighted the need for a better understanding of species-specific habitat associations and methodologies for monitoring long-term population trends. Great Smoky Mountains National Park is committed to incorporating salamander population monitoring into the park's long-term inventory and monitoring program because of the large number of unique species in the park, and evidence that salamanders are finely tuned indicators of environmental quality. We present data on spatial and temporal patterns in salamander diversity and abundance in Great Smoky Mountains National Park and compare the bias and effectiveness of 4 common sampling techniques. We found that large-scale habitat characteristics, including disturbance history, proximity of streams, and elevation are useful to explain patterns of salamander distribution and abundance. With the exception of soil moisture, microhabitat variables were not helpful in understanding variations in salamander relative abundance. Data collected over 2 years suggest that common salamander sampling techniques vary significantly in their effectiveness. and they may often violate assumptions required for comparing salamander population indices over space or time. Salamander counts on our sites were highly variable. Neither sampling variability nor detectability were constant across habitat types or species. These characteristics reduce power for detecting long-term population trends and suggest that some common sampling methods may not provide indices suitable for long-term population monitoring.