PS 83-18
Development of rapid salamander monitoring and habitat assessment protocols for the Delaware River basin

Friday, August 14, 2015
Exhibit Hall, Baltimore Convention Center
David H. Keller, Fisheries Section, The Academy of Natural Sciences of Drexel University, Philadelphia, PA
Richard J. Horwitz, BEES, Drexel University, Philadelphia, PA
Stefanie A. Kroll, BEES, Drexel University, Philadelphia, PA

Salamanders are recognized as being abundant, important members of stream communities and sensitive to habitat loss and stream impairment. However, most states in the Delaware River basin do not have monitoring programs/protocols for assessing stream salamanders and their habitats. Existing state protocols use minimal salamander-specific habitat assessment. As a result, determinations from salamander monitoring may confound water quality impairment with habitat quality. Our objectives were to 1) determine the optimum reach length for determining richness, 2) identify habitat variables that account for variability in salamander richness and relative abundance, and 3) propose rapid assessment protocols. From 2013-2014, we surveyed 48 sites across the Delaware River Basin using visual encounter surveys and fine-scale habitat assessment. In preliminary analyses, optimum reach length was determined graphically and individual Principle Components Analyses (PCA) of aquatic, terrestrial, and geographic factors were used to identify variables and factors for inclusion in a Redundancy Analysis (RDA).


Richness was not even across the basin and the optimum reach length for determining richness differed amongst regions. Richness was highest in least degraded streams. When pooling all samples, a 40 m reach collected the most species after which a 60 m reach collected more but the discovery rate was greatly reduced.  A uniform 40 m reach length may be adequate to index richness throughout basin, however a 60-80 m reach may be needed to improve precision. 48.6% of the variation in salamander relative abundance was explained by 13 environmental variables/factors. Fine-scale habitat variables explaining this variation included: terrestrial measures (within 1 m of the wetted stream edge) of bedrock, cobble, and embeddedness, and instream measures of boulder, coarse gravel, and roots. Aquatic, terrestrial, and geographic variables influenced salamander assemblages. In addition to geographic factors, in-stream and out-of-stream substrate needs to be taken into account when assessing salamander richness and relative abundance.