In lieu of an abstract, here is a brief excerpt of the content:

300 Recent evidence of worldwide amphibian population declines has highlighted the need for a better understanding of both species-specific habitat associations and methodologies for monitoring long-term population trends (Barinaga, 1990; Blaustein and Wake, 1990; Wake, 1991; Lannoo, 1998b). For decades, studies have relied on relative abundance indices to evaluate salamander populations across space and time. However , little effort has been made to evaluate the underlying assumptions of these indices or their relationship to the true population. Heatwole (1962) has shown that eastern redbacked salamanders (Plethodon cinereus) change their microhabitat use in response to precipitation events, differentially using cover objects and leaf litter refugia in response to changing humidity levels. This behavior may suggest that eastern red-backed salamanders are not equally detectable at different moisture levels. Because salamander surface activity is likely to vary with topography, season, humidity, climate, or other landscape variables, the ability to detect animals may also vary across space or time. Comparing two populations over time or space requires that n1/n2  [C1/b1]/[C2/b2]  C1/C2, where n  population size, C  number of individuals counted, and b  the ability of individuals to be detected. To compare two count indices we assume that b1  b2. Finally, we must assume a linear relationship between salamander counts (C) and population size (n), E(C)  bN (Lancia et al., 1994). Great Smoky Mountains National Park (GRSM) 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, as well as evidence that salamanders are finely tuned indicators of environmental quality (Duellman and Trueb, 1986; Corn and Bury, 1989a; Dodd, 2003). Data from ongoing research in GRSM designed to assess spatial and temporal patterns in salamander diversity and abundance are being used to evaluate sampling effectiveness and bias across a variety of habitat types. Here we present evidence that some common salamander sampling techniques may not be appropriate indices for salamander abundance and, therefore , may not be suitable methodologies for use in long-term monitoring programs in the southern Appalachians and perhaps elsewhere. Study Area and Methods Great Smoky Mountains National Park is an internationally recognized refugia of temperate forest biodiversity. Geography and geology, combined with steep, complex topography, promote extreme gradients of temperature and moisture across the park’s environments. In many groups, including salamanders (Jackson, 1989), these gradients produce levels of species diversity that are unmatched elsewhere in North America. Perhaps as many as 10% of the world’s salamander species are found in the region (Petranka, 1998). As part of a larger study designed to assess the distribution, abundance, and habitat associations of salamanders in GRSM (Simons and Johnson, 1999; Hyde, 2000), 111 terrestrial sampling sites were established within the Mount LeConte USGS quadrangle. Sampling sites were stratified by elevation, land use history, and plant community type. Because salamander life histories are variable, we employed four different sampling methods: searches of natural cover objects along transects (Jaeger, 1970, 1994), opportunistic nighttime surveys along transects (Ash and Bruce, 1994), artificial cover boards (Fellers and Drost, 1994; Jung et al., 1997), and leaf litter searches (Pauley, 1995c). Prior to this study, we did not know the relationship between species abundance estimates obtained by these sampling methodologies. The sampling framework at each site was comprised of up to four (approximately parallel) 50 m transects, one transect for each sampling method (Fig. 44-1). Where possible, each site included all three diurnal transects. However, impassable terrain and safety concerns prohibited the inclusion of all transects at some sites. Transects for nighttime surface counts were included at a subset of sites and were opportunistically sampled on warm rainy nights when temperature and humidity conditions favored surface activity by terrestrial salamanders. For a detailed description of our site design or sampling techniques see Hyde (2000). Results All sites were sampled a minimum of three times between 27 May and 5 August 1998, and five times between 5 April and 27 FO RTY-FO U R Monitoring Salamander Populations in Great Smoky Mountains National Park ERIN J. HYDE AND THEODORE R. SIMONS SALAMANDERS IN THE SMOKY MOUNTAINS 301 June 1999. During the 1998 and 1999 field seasons, 98 opportunistic night transect samples were recorded at 39 different sites. In addition to regular sampling, three sites were sampled every 7 to 10 days during the 1998 and 1999 field seasons as part of a separate study and to determine within site variation in salamander abundance...

Share