Inventory and Assessment of the Reptile and Amphibian Community of Bull Neck Swamp, Washington County, North Carolina.

Abstract

Recent declines in reptile and amphibian populations across the globe have encouraged an increased desire to discover, document, and monitor these taxa. Arguably, the greatest cause is land-use change. Management interests for Bull Neck Swamp (BNS) encouraged research to inventory the reptile and amphibian community and to document possible impacts of land-use practices, such as silviculture and site preparation. Four habitat preserves were delineated based on plant community, leaving 1, 554ha (3, 841ac) available for management. Comparisons between habitat assemblages were used to determine if preserves were occupied by more vulnerable species and land-use effects on these species. However, variations in behavioral or environmental variables, and detection probabilities between capture techniques could provide misleading data for assemblage comparisons of community parameters. Therefore, 11 different capture techniques were employed to obtain better samples of habitat assemblages. To determine the accuracy of sampling techniques at inventorying species, techniques were categorized into primary (i.e., drift fence arrays with pitfall and funnel traps, visual encounter surveys, and coverboard arrays), secondary (i.e., road searches, polyvinyl chloride (PVC) piping grids, auditory surveys, and line transects), and tertiary (opportunistic encounters, aquatic funnel traps, crayfish traps, and basking traps) methodologies. All techniques had variable distributions and were evenly represented in all five areas when possible. All captured individuals were marked; snakes were double-marked with visible implant fluorescent elastomer to augment a concomitant laboratory experiment. Initial capture data were used to derive estimates of species richness (S) and modified Chao — Jaccard similarity indices (JSI). During May to August, 2005 and 2006, 1, 581 total captures represented 33 species, giving an estimated species richness of 34. Primary techniques sampled an estimated species richness of 14 and two unique species, species detected by only one sampling technique. Estimated species richness for secondary and tertiary techniques was 29 and 25, with three and seven unique species, respectively. If primary techniques alone were used, 59% of the reptile and amphibian community, including 10 unique species, would have been missed. Observed and estimated species richness for habitats ranged from 7 to 32 and 13 to 44, respectively. Chao — Jaccard similarity indices ranged from 0.59 to 1.0, with nine comparisons over 0.75, which indicated high similarity between habitat assemblages. These results suggested that land-use practices should be carefully planned and implemented to reduce effects to the reptile and amphibian community of BNS. Empirical results supported the use of elastomers for snakes. It is recommended that future inventory studies for all taxa employ as many capture techniques as logistically and spatially possible to derive accurate species richness. Also, assemblage comparisons should rely on species composition when determining conservation plans.