Southeastern Naturalist A. Parlin, S.A. Dinkelacker, A. McCall, M.S. Gosselin, C. Mettey, and R. Tibbert 2019 Vol. 18, No. 3 430 2019 SOUTHEASTERN NATURALIST 18(3):430–440 Temporal and Spatial Assessment of the Maritime Forest Herpetofauna Diversity on a Barrier Island in North Carolina Adam F. Parlin1, Stephen A. Dinkelacker1,*, Aaron McCall2, Michael S. Gosselin1, Colin Mettey1, and Ryan Tibbert1 Abstract - Monitoring biodiversity over time allows for temporal comparisons of community composition and potential shifts in community resilience. We surveyed the herpetofauna assemblage at a maritime forest in Dare County, NC, during late spring and summer of 2012, which was 25 years since the last survey. Our goal was to resurvey the preserve and compare the alpha (α), beta (β), and gamma (γ) diversity values between 1987 and 2012 to quantify changes in assemblage composition using similarity indices. We found that assemblage structure became less similar temporally and spatially at the different habitats sampled but remained similar between survey years. Our study shows the importance of resurveying preserve diversity to determine shifts in assemblage and loss of biodiversity. Introduction Monitoring community assemblage is important in quantifying changes in biodiversity that are reflective of habitat quality in different environments. Environments with high biodiversity typically have communities with higher resilience and resistance that can rebound faster after a catastrophic event (Downing et al. 2012). However, current climate conditions pose a risk to many terrestrial ectotherms, as some species are insufficiently tolerant of high temperatures (Sunday et al. 2014). Ecosystems also require habitats with enough environmental heterogeneity to support microclimate selection for herpetofauna (Sears et al. 2016). Formal surveys provide temporal comparisons of communities over time, serving as indicators for shifts in community and species composition in an area, and as a baseline for future comparisons (Noss 1990). Monitoring and assessing herpetofauna diversity is imperative for understanding the potential impacts of environmental stochasticity and severe storm events on the resilience and resistance of the community. Along the coast, identifying high-risk ecosystems is crucial because estuarine and coastal seas are under constant threat of degradation due to human impact and severe storm events (Lotze et al. 2006). Barrier islands are coastal dune formations that absorb most of the energy from waves and wind coming off the ocean, and consequently are prone to the disturbance caused by severe storms (Snyder and Boss 2002). Precipitation is the primary source of fresh water on these islands, although, due to the sandy 1Department of Biology, Framingham State University, 100 State Street, Framingham, MA 01701. 2The Nature Conservancy, Nags Head Woods Ecological Preserve, 701 West Ocean Acres Drive, Kill Devil Hills, NC 27948. *Corresponding author - sdinkelacker@framingham.edu. Manuscript Editor: Kristen Cecala Southeastern Naturalist 431 A. Parlin, S.A. Dinkelacker, A. McCall, M.S. Gosselin, C. Mettey, and R. Tibbert 2019 Vol. 18, No. 3 composition of the area, water can easily percolate through the sand (Bellis 1995). Low freshwater retention and high rates of salt spray make barrier islands challenging for flora and fauna with respect to survival, reproduction, and dispersal (Snyder and Boss 2002). Special adaptations of plant species in response to the harsh conditions of barrier islands create vastly different habitats in contrast to those found on the mainland (Bellis 1995), leading to habitat types known as maritime forests. These forests make it possible for ecological communities to establish on barrier islands by stabilizing the sand, reducing evaporation of fresh water, providing organic nutrients for the soil, creating habitat for wildlife, and protecting against salt spray (Bellis 1995, Bourdeau and Oosting 1959, Boyce 1954). Communities on barrier islands need to be resilient in order to tolerate the harsh environmental disturbances and ambient conditions. Barrier islands attract a high level of human presence, and increased human encroachment in an area has been shown to negatively impact native species while increasing the risk of invasion by non-native species (Ficetola and Padoa-Schioppa 2009). One of the priorities of the North Carolina Wildlife Action Plan (NCWAP) is to survey the species of the maritime forests on the Outer Banks (OBX), a chain of barrier islands along the East Coast, to better manage the communities living within this unique habitat. Updated species lists will allow for long-term monitoring programs to track population changes over time (NCWRC 2005). Prior to our study, Braswell (1988) conducted the last survey of the reptile and amphibian assemblage of the maritime forest in Nags Head Woods Ecological Preserve (NHWEP) on the OBX of North Carolina. That survey added an additional 16 species to the official list, several unique to NHWEP; species were categorized as relict populations, on the periphery of their distribution, or associated with mainland-like habitats similar to the barrier islands. Braswell (1988) emphasized the importance of the freshwater ponds in maintaining diversity and cautioned that destabilization of the maritime forest’s buffering ability could adversely affect the plant and animal communities on the island. Since then, the size of the preserve has increased, as has human presence on the OBX surrounding the preserve (US Census Bureau 2012). In addition to increased human presence, the OBX has experienced numerous tropical storms and hurricanes that may have altered the herpetological assemblage of NHWEP since the 1987 survey. It is unclear what the potential effect of increased human presence and the tropical storms have had on the island since the previous survey. We were interested in resurveying the preserve to compare community assemblage using similar capture methods and techniques as the previous study. Although the North Carolina Museum of Natural Sciences (NCSM) database has documented vouchers dating from between the studies, there has been no formal survey since Braswell (1988). Our objectives were to (1) compare the alpha (α), beta (β), and gamma (γ) diversity values between the 2 surveys (1987 and 2012) and habitats sampled; and (2) use dissimilarity indices to quantify any changes in assemblage and overlap. We hypothesized that (1) overall herpetofauna assemblage has remained similar in species richness since the previous survey 25 y ago, and (2) assemblage composition at drift-fence locations between 1987 and 2012 would Southeastern Naturalist A. Parlin, S.A. Dinkelacker, A. McCall, M.S. Gosselin, C. Mettey, and R. Tibbert 2019 Vol. 18, No. 3 432 be similar 25 y later. In addition to focusing on sites monitored by Braswell (1988), we surveyed newer areas in the preserve to develop a more comprehensive understanding of the herpetofaunal assemblage in the area. Field-Site Description We performed the herpetological survey at NHWEP, located in the towns of Kill Devil Hills and Nags Head, Dare County, NC. NHWEP is a 567-ha (1400-ac) preserve of maritime forest managed by The Nature Conservancy. Braswell (1988) conducted the most recent survey previous to ours. There are several major habitat types in NHWEP, including maritime deciduous forest, maritime shrub forest, maritime swamp forest, and interdunal ponds (for detailed descriptions, see Bellis 1995). Soil in these habitats is very deep, excessively drained, rapidly permeable, and formed in thick, sandy sediments adjoining beaches and waterways along the coast. Cover varies from low to moderately high tree canopy dominated by combinations of Quercus virginiana Mill. (Southern Live Oak), Q. hemisphaerica Bartram ex Willd. (Laurel Oak), and Pinus taeda L. (Loblolly Pine). Methods We conducted sampling between 16 May 2012 and 10 August 2012 to seasonally coincide with Braswell (1988), who survey from 11 April 1987 to 15 September 1987. We used sampling techniques comparable with Braswell (1988), including drift fences, fyke nets, hand captures, day and night visual searches, and minnow traps. We did not listen for frog calls during our survey, although we made note of species by visual encounters and used PVC pipes. We placed PVC pipes (7.5 cm diameter, 90 cm long) 30 cm in the ground at a ~ 45° angle (Boughton and Staiger 1996, Zacharow et al. 2003) and checked them daily. We also recorded incidental captures and hand captures made on hiking surveys. We installed 2 drift fences within NHWEP at locations used by Braswell (1988) for drift fences in similar habitat types. We set the first drift fence during our study from 28 May 2012 until 10 August 2012 and a second drift fence from 8 June 2012 until 10 August 2012. Drift fences in the previous survey were up from 29 May 1987 to 15 September 1987, with an inactive period from 19 June 1987 to 6 July 1987. The total time that the drift fences were active was similar between the 2 surveys. The habitat types compared included a drift fence placed on the boundary of a hardwood forest and an old field, and a hardwood forest with mixed pine (Fig. 1). We built the drift fences using 90-m linear stretches of construction silt fencing. We buried the fence bottoms 20-cm deep to prevent animals from escaping underneath (Heyer et al. 1994). We placed box traps at the ends and middle (at 45 m) of each fence underneath the canopy to prevent direct sunlight from overheating the animals. The box traps at the ends were 60-cm long, 30 cm wide, 30 cm high, and had a single throat. The traps in the middle were 90 cm long, 30 cm wide, 30 cm high, and were double-throated. We placed 3-m sections of silt fence at each corner of the middle box traps and on each side of the fence at the ends. These winged sections helped funnel animals into the traps. We constructed pitfall traps Southeastern Naturalist 433 A. Parlin, S.A. Dinkelacker, A. McCall, M.S. Gosselin, C. Mettey, and R. Tibbert 2019 Vol. 18, No. 3 using 19-L buckets and installed them between each box trap. Both box and pitfall traps were located on each side of the fence, remained in place for the entire survey, and were checked twice a day to ensure animals did not overheat. Following the recommendation from Braswell (1988), we placed fyke nets in interdunal ponds and in the sound along the preserve to capture turtle species. The bodies of water were deep enough to set up the nets so that the throats of the nets were submerged but a small pocket of air at the top of the net allowed any captured turtles to breathe (Harless and Morlock 1989). We used a total of 8 nets simultaneously during the survey. We kept the nets in a body of water for 2 weeks before moving them to another pond, for a a total of 27 separate trapping locations during the study. Figure 1. Study site at Nags Head Woods Ecological Preserve, Kill Devil Hills, NC. The solid black lines represent the outer portions of the barrier island and the dotted lines within represent the preserve boundaries. Major habitat types are classified as wetland, maritime forest, maritime shrub, dune, and maritime swamp. The black triangles outlined with white mark the locations of drift fences from the 2012 survey and are based on the previously sampled locations from Braswell (1988) in the west and south locations. The southern drift fence was in the maritime forest and maritime shrub habitats while the western drift fence was in the wetland, maritime swamp, and maritime forest intersection. Southeastern Naturalist A. Parlin, S.A. Dinkelacker, A. McCall, M.S. Gosselin, C. Mettey, and R. Tibbert 2019 Vol. 18, No. 3 434 Our analysis of assemblage structure employed calculations to quantify alpha (α), gamma (γ), beta-additive (βa), and beta-multiplicative (βm) to compare captures at drift fences in 1987 and 2012, as well as to compare the species found during both years (Crist et al. 2003). Beta-additive indicates species found in 1 year or location and not the other, while beta-multiplicative is the effective number of communities, and indicates similarity between the time or location. We calculated assemblage overlap between 1987 and 2012 with Sørensen and Jaccard dissimilarity indices and the drift fence captures during each year using the ‘betapart’ in R package (Basegla et al. 2018) for comparison (for detailed description of Sørensen and Jaccard dissimilarity indices and robustness see Schroeder and Jenkins 2018). Values determined by the equations span from 0 to 1, where 1 indicates no overlap, in terms of species, and 0 indicates complete overlap where species occur in same proportions. The equations only need presence–absence data and thus allow temporal comparison of whole species lists between 1987 to 2012 and comparison between drift fences in each year separately. We were unable to accurately quantify abundance between the years and were unable to consider the detection variation between species; however, our protocol was similar to Braswell’s (1988) and assume that our sampling period was a closed period. We conducted our analyses in program R (R Core Team 2018). Results We documented a total of 41 reptile and amphibian species in NHWEP during our study from 16 May 2012 to 10 August 2012, which was the same amount found in 1987 from 11 April 1987 to 15 September 1987, though not all of the species found during each survey were the same (Table 1). When we compared the results of the 2 studies, the gamma value indicated that there are 47 total documented species in NHWEP with 6 of the species not captured during each year (βa = 6), and we detected little change in the community structure between the 2 years (βm = 1.14; Table 2). In 1987, α was the same (21) for both drift fences, and 6 species were Table 1. Reptile and amphibian species documented in Nags Head Woods Ecological Preserve, Kill Devil Hills, NC, during surveys in spring and summer of 1987 and late spring and summer of 2012. The table lists species documented during the 1987 or 2012 surveys; presence (+) and absence (-) are noted for each species found. [Table continued on following page.] Species 1987 2012 Turtle species Chelydra serpentina (L.) (Common Snapping Turtle) + + Chrysemys picta (Eastern Painted Turtle) - + Clemmys guttata (Spotted Turtle) - + Deirochelys reticularia (Eastern Chicken Turtle) + + Kinosternon baurii (Striped Mud Turtle) - + Kinosternon subrubrum (Eastern Mud Turtle) + + Pseudemys rubriventris (LeConte) (Northern Red-bellied Cooter) + + Terrapene carolina (Eastern Box Turtle) + + Trachemys scripta (Thunberg in Schoepff) (Yellow-bellied Slider) + + Southeastern Naturalist 435 A. Parlin, S.A. Dinkelacker, A. McCall, M.S. Gosselin, C. Mettey, and R. Tibbert 2019 Vol. 18, No. 3 captured at one or the other (βa = 6.0; Table 2). There was an overall decrease in α values in 2012 at both locations (17 species were captured at drift fence 1 and 9 species were captured at drift fence 2), with 8 different species captured at one drift fence but not the other (βa = 8.0; Table 2). We found a decrease in gamma diversity (γ) from 1987 to 2012, but an increase in the beta additive (βa), indicating more species found at one drift fence than the other, and an increase in beta multiplicative Table 1, continued. Species 1987 2012 Snake species Agkistrodon piscivorus (Cottonmouth) + + Carphophis amoenus (Eastern Worm Snake) + + Cemophora coccinea (Blumenbach) (Scarlet Snake) + - Coluber constrictor (Northern Black Racer) + + Crotalus horridus (Timber Rattlesnake) + - Diadophis punctatus (Southern Ringneck Snake) + + Elaphe obsoleta (Black Rat Snake) + + Farancia abacura (Eastern Mud Snake) - + Farancia erytrogramma (Rainbow Snake) + - Heterodon platirhinos (Eastern Hog-nosed Snake) + + Lampropeltis getula (Eastern Kingsnake) + - Liodytes pygaea (Black Swamp Snake) + - Thamnophis sauritus (Eastern Ribbon Snake) + + Nerodia erythrogaster (Red-bellied Water Snake) + + Nerodia sipedon (Northern Water Snake) - + Opheodrys aestivus (Rough Green Snake) + - Rhadinaea flavilata (Pine Woods Snake) + + Storeria dekayi (Brown Snake) + + Lizard species Anolis carolinensis (Carolina Anole) + + Cnemidophorus sexlineatus (Six-lined Racerunner) + + Eumeces inexpectatus Southern Five-Lined Skink) + + Ophisaurus ventralis (Eastern Glass Lizard) + + Scincella lateralis (Ground Skink) + + Amphibian species Acris gryllus (Southern Cricket Frog) + + Ambystoma opacum (Marbled Salamander) + + Amphiuma means (Two-toed Amphiuma) - + Anaxyrus fowleri (Fowler’s Toad) + + Anaxyrus terrestris (Southern Toad) + + Gastrophryne carolinensis (Eastern Narrow-mouthed Toad) + + Hyla chrysoscelis (Cope’s Gray Treefrog) + + Hyla cinerea (Green Treefrog) + + Hyla squirella (Squirrel Treefrog) + + Lithobates catesbeianus (American Bullfrog) + + Lithobates clamitans (Green Frog) + + Lithobates sphenocephalus (Southern Leopard Frog) + + Notophthalmus viridescens (Eastern Newt) + + Plethodon cinereus (Redback Salamander) + + Scaphiopus holbrookii (Eastern Spadefoot) + + Southeastern Naturalist A. Parlin, S.A. Dinkelacker, A. McCall, M.S. Gosselin, C. Mettey, and R. Tibbert 2019 Vol. 18, No. 3 436 (βm) for the habitat types surveyed, indicating that the community assemblage at each drift fence location was less similar after the 25 y. Species captured in each year totaled 41 different species. In total, there have been 47 species found in the NHWEP (as indicated by γ), but in each year 6 species were found that were not found in the other (βa) The overall herpetofauna assemblage is still similar between the 2 years based on the low βm (close to 1). The herpetofauna had similar community overlap temporally from 1987 to 2012 (Sørensen = 0.14, Jaccard = 0.26); however, the assemblages had less overlap spatially, indicating changes in herpetofauna within habitats sampled at each drift fence location (Table 3). Discussion Formal surveys that determine the presence and absence of species capture moments in time that allow us to assess the quality of a habitat through the measure of biodiversity. As a result, multiple surveys over time are needed to best assess changes in community assemblages at any given location. Areas that are protected change over time through addition or loss of land; thus, they need frequent checks to quantify changes in species assemblage. It is important to acknowledge that despite highly intensive search efforts, true absence may be difficult to verify. Therefore, standard temporal and spatial assessments of species assemblages with surveys and vouchers improve our ability to assess habitat quality and inform management decisions. Table 2. Community diversity (α, γ, βa, βm) values for drift fence 1 (DF1) and drift fence 2 (DF2) in 1987 and 2012, as well as whole preserve comparison of the community diversity indices for species captured in 1987 and 2012 at Nags Head Woods Ecological Preserve (NHWEP). Community indices 1987 2012 NHWEP 1987 α - - 41 2012 α - - 41 DF1 α 21 17 - DF2 α 21 9 - γ 27 21 47 βa 6.0 8.0 6.0 βm 1.3 1.6 1.1 Table 3. Dissimilarity indices indicating assemblage overlap between 1987 and 2012, and within years comparing drift fences (n1987 = 2, n2012 = 3). Drift fence 3 was omitted from the comparison between 1987 and 2012 because it was a new location that was not surveyed in 1987. For the 1987 to 2012 comparison, we combined the species captured at drift fence (DF) 1 and 2, while within each year we compared the drift fence locations. Sørensen and Jaccard dissimilarity indices values range from 0 to 1. In terms of species, values of 1 indicate no overlap and 0 indicates overlap, that is, species occur in similar proportions. 1987 2012 Indices 1987 and 2012 DF 1 and 2 DF 1 and 2 Sørensen 0.14 0.28 0.53 Jaccard 0.26 0.44 0.50 Southeastern Naturalist 437 A. Parlin, S.A. Dinkelacker, A. McCall, M.S. Gosselin, C. Mettey, and R. Tibbert 2019 Vol. 18, No. 3 Based on our analyses, there has been little change in the herpetofaunal diversity and community structure at NHWEP over the 25 y since the last official survey was performed in the area, thus supporting our first hypothesis that there would be similar species richness in the herpetofaunal assemblage temporally in the preserve. Although some species found in 1987 were not captured in 2012, we documented several new species occurrences in the area (Table 1). Braswell (1988) predicted that species could become extirpated from the island because of the increasing human presence on the OBX. For instance, Crotalus horridus (Timber Rattlesnake) would likely be extirpated from the island due to limited availability of undisturbed habitat needed by the species. Timber Rattlesnakes require large amounts of undisturbed habitat and feed specifically on small mammals (Clark 2002), while many of the other snake species caught feed on a wide array of amphibian species. Braswell (1988) provided a list of 12 species expected for the preserve, of which we found 4 in the 2012 survey—Amphiuma means (Two-toed Amphuma), Clemmys guttata (Spotted Turtle), Farancia abacura (Mud Snake), and Nerodia sipedon (Northern Water Snake). During our study, we did not find the remaining 8 expected species Braswell (1988) mentioned, and currently, there are no vouchers reported for those species in the preserve—Elaphe guttata L. (Rat Snake), Malaclemys terrapin (Schoepff) (Diamondback Terrapin), Pseudarcis crucifer (Wied-Neuwied) (Spring Peeper), Hyla femoralis Bosc (Pine woods Tree Frog), Nerodia taxispilota (Holbrook) (Brown Water Snake), Anaxyrus quercicus Holbrook (Oak Toad), Thamnophis sirtalis L. (Common Garter Snake), and Siren lacertina L. (Greater Siren). It is plausible that we did not detect some species, despite the high search intensity. Whether these species were truly absent or just not detected during our survey is an important consideration that emphasizes the need for frequent surveys to determine true absence and true presence of a species. As a result, caution should be taken with regards to the resilience of the herpetological community, as we did not find several species during our survey that may not have been truly absent. Additional surveys could determine whether species have been truly extirpated from the area or merely have decreased in abundance. Our second hypothesis, that there would be similar assemblage composition between drift fence locations 25 years later, was not well supported. Our analyses indicated an increasing dissimilarity between drift fences 1 and 2 within each year from 1987 to 2012. Drift fence locations in 1987 had similar assemblage composition while the drift fence locations in 2012 were less similar, despite being in similar habitat types. We found a shift in the effective number of communities between the 2 drift fence locations from 1 to 2 communities. Barrier islands are subject to major storms and environmental stochasticity, including droughts and hurricanes (Clinch et al. 2012). Prior to the 1987 study, a major drought was reported, which was the worst in over 40 y. Prior to the 2012 study, Hurricane Irene had inundated the preserve with seawater at both drift fence locations for several hours (A. McCall, The Nature Conservancy, NHWEP, pers. comm.). The preserve has also changed over the past 25 y, including addition of new land, removal of a trash pile, changes in vegetative cover, and new trails. It is reasonable to assume Southeastern Naturalist A. Parlin, S.A. Dinkelacker, A. McCall, M.S. Gosselin, C. Mettey, and R. Tibbert 2019 Vol. 18, No. 3 438 that the habitat structure of NHWEP has changed over the past 25 y, resulting in an increasing dissimilarity between previously monitored drift fence locations. The question of whether the herpetofaunal assemblage at NHWEP is becoming 2 separate effective communities requires further monitoring of both the weather and human presence. Despite these events, the overall herpetological assemblage has remained resilient, with little change in the species richness. Given that NHWEP is a protected preserve, it is more likely that severe storms and ambient temperatures rahter than human activities are impacting the herpetological diversity. Many of the species found on the barrier islands are also found in Alligator River National Wildlife Refuge (ARNWR). A survey of the reptile and amphibian species in ARNWR, on the mainland of North Carolina, resulted in 28 reptile and 11 amphibian species (Meyers and Pike 2006). Although we documented fewer reptiles during our survey in 2012, we documented 4 more amphibian species in NHWEP compared to ARNWR. It is plausible that many species could have migrated to the barrier island, either through displacement by storms or other methods, or are relicts of former populations that have persisted. Species found on the mainland and the nearby barrier island appear to be associated with similar habitat structure found in the maritime forest. In NHWEP, much of the marsh on the western side of the preserve still has not been surveyed due to difficulty accessing the area. Sampling effort in these areas, as well as for species seen but not documented, should be intensified. Further studies comparing nearby mainland species may help to elucidate dispersal of certain species. Six snake species were not found during the survey in 2012 that had been documented in the 1987 survey at NHWEP (Cemophora coccinea (Scarlet Snake), NCSM 9224; Crotalus horridus (Timber Rattlesnake) and Farancia erytrogramma (Rainbow Snake) NCSM 28647; Lampropeltis getula (Eastern Kingsnake), NCSM 28646; Opheodrys aestivus (Rough Green Snake), NCSM 28963; Liodytes pygaea (Black Swamp Snake), NCSM 28680). Many of the snake species caught in 2012 feed on the wide array of amphibian species abundant throughout the preserve. During the survey, we caught a total of 2 Blarina brevicauda (Say) (Northern Shorttailed Shrew) in the pitfall traps of the drift fences, compared to the thousands of amphibians captured. Drift fences and pitfall traps are commonly used methods for surveying small mammals (Brannon 2005, DeSa et al. 2012, Kapfer and Muñoz 2012), and additional studies on the small-mammal community and habitat composition surveys within the preserve may provide evidence for why Timber Rattlesnakes and possibly other species of snake are absent. We speculate that the small-mammal population would have increased due to loss of predation pressure from Timber Rattlesnakes. It is also plausible that many of the snake species are highly cryptic (e.g., Black Swamp Snake) and would require extensive trapping efforts to locate. Our study shows the resilience in the herpetofaunal assemblage at NHWEP over the 25 y since the last survey. There has been additional fieldwork done at NHWEP between 1987 and 2012, though mostly on a casual, opportunistic, and less intensive basis. These encounters are still important for monitoring whether certain easily detectable species are present. However, there is the possibility of Southeastern Naturalist 439 A. Parlin, S.A. Dinkelacker, A. McCall, M.S. Gosselin, C. Mettey, and R. Tibbert 2019 Vol. 18, No. 3 biodiversity loss or lower detection of several species due to factors such as severe storms and human encroachment. Long-term monitoring can address these issues, as prior studies serve as a baseline for future comparisons, whether every decade or after a major storm event. Repetition of these survey studies requires large-scale effort, but is feasible and important for monitoring ecosystem integrity through assessing species composition. Use of a standardized approach during each survey provides a comparative approach for future studies to observe changes in assemblage composition over time and inform management decisions. Acknowledgments We thank W. Stroud and A. Beck for their field assistance, M. Mahon for help with R code for community analyses, and B. Boutin for his guidance during the study. We also thank J. Beane from the NCSM for providing database information on species captured at NHWEP. Funding was provided by The Nature Conservancy and Framingham State University. Research was conducted under North Carolina Wildlife Resources Commission Permit #12-SC00491 and in accordance to Framingham State University Institutional Animal Care and Use Committee protocol #2011-008. Literature Cited Baselga, A., D. Orme, S. Villeger, J. De Bortoli, and F. Leprieur. 2018. betapart: Partitioning beta diversity into turnover and nestedness components. R package version1.5.1. Available online at https://CRAN.R-project.org/package=betapart. Accessed 9 January 2019. Bellis, V.J. 1995. Ecology of maritime forests of the Southern Atlantic coast: A community profile. Biological Report 30. US Department of Interior, National Biological Service, Washington, DC. 95 pp. Boughton, R.G., and J. Staiger. 1996. Use of PVC pipe refugia as a sampling technique for hylid treefrogs. American Midland Naturalist 144:168–177. Bourdeau, P.F., and H.J. Oosting. 1959. The maritime Live Oak forest in North Carolina. Ecology 40:148–152. Boyce, S.G. 1954. The salt spray community. Ecological Monographs 24:29–67. Brannon, P.M. 2005. Distribution and microhabitat of the Woodland Jumping Mouse, Napaeozapus insignis, and the White-footed Mouse, Peromyscus leucopus, in the Southern Appalachians. Southeastern Naturalist 4:479–486. Braswell, A.L. 1988. A survey of the amphibians and reptiles of Nags Head Woods Ecological Preserve. Association of Southeastern Biologists Bulletin 35:199–217. Clark, R.W. 2002. Diet of the Timber Rattlesnake, Crotalus horridus. Journal of Herpetology 36:494–499. Clinch, A.S., E.R. Russ, R.C. Oliver, H. Mitasova, and M.F. Overton. 2012. Hurricane Irene and the Pea Island Breach: Pre-storm site characterization and storm surge estimation using geospatial technologies. Shore and Beach 80:38–46. Crist, T.O., J.A. Veech, J.C. Gering, K.S. Summerville, and W.J. Boecklen. 2003. Partitioning species diversity across landscapes and regions: A hierarchical analysis of α, β, and γ diversity. The American Naturalist 162:734–743. DeSa, M.A., C.L. Zweig, H.F. Percival, W.M. Kitchens, and J.W. Kasbohm. 2012. Comparison of small-mammal sampling techniques in tidal salt marshes of the central Gulf Coast of Florida. Southeastern Naturalist 11:G17–G21. Southeastern Naturalist A. Parlin, S.A. Dinkelacker, A. McCall, M.S. Gosselin, C. Mettey, and R. Tibbert 2019 Vol. 18, No. 3 440 Downing, A.S., E.H. van Nes, W.M. Mooji, and M. Scheffer. 2012. The resilience and resistance of an ecosystem to a collapse of diversity. PLoS ONE 7:e46135. DOI:10.1371/ journal.poone.0046135. Ekner, A., Z. Sajkowska, K. Dudek, and P. Tryjanowski. 2011. Medical cautery units as a permanent and non-invasive method of marking lizards. Acta Herpetologica 6:229–236. Ficetola, G.F., and E. Padoaf-Schioppa. 2009. Human activities alter biogeographical patterns of reptiles on Mediterranean islands. Global Ecology and Biogeography 18:214–222. Harless, –ommunities at sites in the Piedmont of North Carolina. Southeastern Naturalist 11:65–88. Lotze, H.K., H.S. Lenihan, B.J. Bourque, R.H. Bradbury, R.G. Cooke, M.C. Kay, S.M. Kidwell, M.X. Kirby, C.H. Peterson, and J.B.C. Jackson. 2006. Depletion, degradation, and recovery potential of estuaries and coastal seas. Science 312:1806–1809. Meyers, J.M., and D.A. Pike. 2006. Herpetofaunal diversity of Alligator River National Wildlife Refuge, North Carolina. Southeastern Naturalist 5:235–252. North Carolina Wildlife Resources Commission (NCWRC). 2005. North Carolina wildlife action plan. Raleigh, NC. Noss, R.F. 1990. Indicators for monitoring biodiversity: A hierarchical approach. Conservation Biology 4:355–364. R Core Team. 2018. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Schroeder, P.J., and D.G. Jenkins. 2018. How robust are popular beta-diversity indices to sampling error? Ecosphere 9(2):e02100.10.1002/ecs2.2100. Sears, M.W., M.J. Angiletta, Jr., M.S. Schuler, J. Borchert, K.F. Dilliplane, M. Stegman, T.W. Rusch, and W.A. Mitchell. 2016. Configuration of the thermal landscape determines thermoregulatory performance of ectotherms. Proceedings of the National Academy of Sciences of the United States of America 113:10595–10600. Snyder, R.A., and C.L. Boss. 2002. Recovery and stability in barrier-island plant communities. Journal of Coastal Research 18:530–536. Sunday, J.M., A.E. Bates, M.R. Kearney, R.K. Colwell, N.K. Dulvy, J.T. Longino, and R.B. Huey. 2014. Thermal-safety margins and the necessity of thermoregulatory behavior across latitude and elevation. Proceedings of the National Academy of Sciences of the United States of America 111:5610–5615. US Census Bureau. 2012. Kill Devil Hills (town), North Carolina. Available online at https://www.census.gov/quickfacts/killdevilhillstownnorthcarolina. Accessed 25 June 2013. Zacharow, M., W.J. Barichivich, and C.K. Dodd. 2003. Using ground-placed PVC pipes to monitor hylid treefrogs: Capture biases. Southeastern Naturalis t 2: 575–591.