In 2004, the Northeast Amphibian Research Monitoring Initiative (NE ARMI) received funding from the National Park Service’s Park Oriented Biological Support, and combined these monies with ARMI funds to initiate a region-wide study on the distribution of vernal pools and estimate the proportion of pools that were occupied by pool-associated amphibians (specifically, wood frogs, Lithobates sylvaticus, and spotted salamanders, Ambystoma maculatum).
At Parks and Refuges where the distribution of vernal pools was known, a random sample of known ponds was chosen for amphibian sampling. At Parks and Refuges where the distribution of vernal pools was unknown, an adaptive cluster sampling (ACS) approach was employed using a random sample from a systematic grid of points overlaid onto each park or refuge. At each sampled point, a 50 x 50m plot was searched for the presence of vernal pools. If a pool was located in the first (primary) plot sampled, 4 adjacent (secondary) plots were added and searched, and this process was continued until no new pools were located in the ‘network’ of plots. While traveling to and between sample points, ‘incidental’ vernal pools were opportunistically encountered and these pools were also sampled.
Pools are searched for egg masses of wood frogs (Lithobates sylvaticus) and spotted salamanders (Ambystoma maculatum) by walking the perimeter of the pool. Counts of the number of egg masses for each species are recorded independently for each observer. Two independent counts allow for better estimation of the detection probability of egg masses of each species. Unless egg masses are detected for both species, the pool is revisited again during the sampling season. To confirm that sites where egg masses were detected in spring surveys result in successful recruitment of juvenile amphibians, sites are revisited late in the breeding season.
From 2004-2011, there was no difference in the average estimated occupancies between the two species; credible intervals overlapped substantially across all years of the study. There were no substantial declines in occupancy at the regional level, though a shallow downward trend is indicated in the data for both species. Individual Parks and Refuges showed differing dynamics across the survey period. The next step in this analysis will be to integrate information about each site (i.e., longitude, spatial position and clustering, precipitation, pool size, depth) to determine the relationship of these covariates on occupancy, site turnover, and regional and local trends.
Recently, surveys at Patuxent Research Refuge have been expanded to include preliminary exploration of techniques for concurrent disease monitoring. The prevalence and distribution of infectious diseases is, for the most part, unknown across the study refuges. Sampling at Patuxent in 2011-2012 coincided with multiple die-off events; these samples are being analyzed in collaboration with Drs. Deb Miller and Matt Gray at the University of Tennessee. Our proposed program of co-monitoring the prevalence of two important amphibian diseases and amphibian occupancy will provide an understanding of the importance of amphibian diseases to amphibian population changes in northeastern National Wildlife Refuges.
sampling a vernal pool
amplexed Lithobates sylvaticus
Ambystoma maculatum egg mass