Amphibian Research and Monitoring on National Wildlife Refuges in the Southeastern United States

C. Kenneth Dodd, Jr., Margaret S. Gunzburger,
Jennifer S. Staiger, William J. Barichivich

Florida Integrated Science Center, 7920 NW 71st Street, Gainesville, Florida 32653, USA

Presented at the 2005 Joint Meeting of Ichthyologists and Herpetologists, Tampa, FL, 6-11 July 2005.

The Southeastern ARMI (SEARMI) Program

For the purposes of the USGS Amphibian Research and Monitoring Initiative (ARMI), the United States is divided into seven blocks of States that are the focus of regional herpetological investigations. The Southeastern Region encompasses the states of Florida, Georgia, Alabama, Tennessee, South Carolina, and North Carolina, as well as Puerto Rico and the U.S. Virgin Islands, an area approximately 1300 x 1660 kilometers, excluding the Caribbean territories.  In this poster, we present some of the results obtained by SEARMI during field sampling in 2004

SEARMI monitors amphibians based on a 3-tiered approach involving sites with intensive research (apex sites), sites that form the basic areas for the core of monitoring activities (mid-level sites), and sites where inventories are conducted (base sites). With few exceptions, federal lands in the southeast have not been surveyed for amphibians or their habitats, thus requiring inventories before proceeding to more intensive study. Most current research focuses on mid-level and base sites. Information from SEARMI's inventory and monitoring program (data collected from the mid-level and apex sampling sites) are being used to assess the status of amphibians on DOI lands using Percent of Area Occupied analyses. By making probabilistic arguments, PAO uses an estimation of site occupancy rate to measure species detection probabilities. SEARMI biologists are collecting extensive data on species and their habitats that will allow for an assessment of distribution patterns and trends, and the initiation of research on declines or problem areas should they be identified.

Objectives

The general objectives of the SEARMI program are: 1) to conduct an inventory of the amphibian species on selected lands administered by the Department of Interior, 2) collect environmental data to identify factors influencing distribution, activity, and detection probability of amphibians, 3) establish long-term monitoring programs for amphibians on DOI lands, and 4) establish baseline data on the health of amphibians through disease monitoring. An additional objective at SMNWR is to resample sites surveyed in the late 1970's by researchers with the U.S. Fish and Wildlife Service.

Methods

We use a variety of methods to sample amphibians.  Terrestrial sampling techniques included field searches (i.e., visual encounter surveys) for animals in the open and under cover objects, drift fences with associated funnel and/or pitfall traps, PVC pipe refugia, and road cruises. Aquatic sampling consists of using dip nets, crayfish traps, and aural surveys (using custom built automated frog-call data loggers). We have used drift fences at two refuges (SMNWR, LSNWR) to compare current species richness and relative abundance with historic data collected at the same sites using the same techniques, and to assess the logistic use of drift fences in large-scale monitoring programs. Each refuge is visited 3-4 times per year.

Current Study Sites: 4 Atlantic Coastal Plain National Wildlife Refuges

Harris Neck National Wildlife Refuge (HNNWR) is located ca. 46 km south of Savannah and 31 km north of Darien, in McIntosh Co., Georgia. The refuge comprises 1,255 hectares of mostly coastal deciduous and oak woodlands, grasslands, former cropland, and some pine. The refuge is surrounded by salt marshes and tidal creeks, and has a long history of human habitation.

Lower Suwannee National Wildlife Refuge (LSNWR) is located along Florida's Big Bend region on the Gulf of Mexico, approximately 80 km WSW of Gainesville. LSNWR encompasses approximately 21,425 hectares of upland and wetland habitats. Established in 1979 to preserve unique coastal, flood plain, and upland ecosystems at the lower reach of the Suwannee River, the refuge stretches 42 km north to south in Levy and Dixie Counties, including lands along both banks of the Suwannee River from Yellow Jacket landing southwest to the Gulf of Mexico.

St Marks National Wildlife Refuge (SMNWR) is located in Florida's panhandle approximately 25 km south of Tallahassee. SMNWR encompasses 27,500 hectares of diverse upland and wetland habitats, including pine flatwoods, sandhills, coastal palm hammocks, and fresh and saltwater waterfowl impoundments.  Established in 1931 to provide wintering habitat for migratory birds, SMNWR extends along the Gulf coast in Taylor, Jefferson, and Wakulla Counties.

Savannah National Wildlife Refuge (SVNWR) comprises 11,320 hectares in Georgia and South Carolina immediately upstream along the Savannah River from the city of Savannah. The refuge has an extensive history of human occupation and use, from Amerindian through the plantation era, when the bottomlands and freshwater tidal marshes were extensively diked and modified for rice production.  Designated in 1927, SVNWR is primarily managed for waterfowl, and water levels within the former rice fields (1,364 hectares) are carefully controlled.  The northern part of SVNWR consists mostly of extensive islands of bottomland hardwoods (cypress, gum, maple) that may or may not be periodically flooded. These islands contain creeks and an extensive number of woodland pools and channels which hold water for varying amounts of time. River bluffs and upland terraces on the refuge are few, but some uplands and slopes are present on the north side of the river along Dodge Tram Road and on the south side of the river east of O'Leary.

Amphibian Species Richness at Four Coastal Plain National Wildlife Refuges, through May 2005

Collecting water quality data is vital to SEARMI. For example, we collected water-quality data at 15 of the 17 ponds sampled in 2004 at SMNWR. We measured all parameters at most of the ponds more than once.  All of these wetlands were relatively isolated ponds in the Panacea Unit.  Values for pH were acidic, with a range of 3.62 - 6.67 and a mean = 4.47.  Conductivity was low, only exceeding 57 µS/cm at one of the 15 sites. Dissolved oxygen (DO, mg/l) and % DO varied considerably, ranging from 3.27 mg/l and 36.6% to 11.33 mg/l and 139.9%. Water temperatures ranged from 14.63 ºC to 34.23 ºC, with the lower temperatures in March 2004 and the higher ones in the summer and early fall samples. These variables may be important covariates in explaining patterns of species richness, abundance, and distribution.

Selected Results for 2004

•  SMNWR: We have detected a total of 31 species of amphibians at SMNWR through Oct. 2004 (20 frogs and 11 salamanders). Our species detection rate in sampling has decreased, indicating we have probably recorded most species for this refuge using conventional sampling techniques. We captured 1,560 individuals of 20 species (15 frogs, 5 salamanders) at the drift fences. We detected 14 frog species and 8 salamander species among the 17 wetlands sampled during 2004. Since the start of SEARMI surveys at SMNWR, we have detected 16 frog species and 8 species of salamanders among 58 sampled wetlands.

•  LSNWR: We detected a total of 23 species of amphibians (19 frogs, 4 salamanders) through Sep. 2004. Our detection rate of species previously unrecorded in our sampling has decreased, indicating we have probably recorded most species for this refuge we are likely to capture using our sampling techniques. We had 485 new amphibian captures (13 species of frogs) at the drift fences through Sep. 2004. The cumulative number of species captured at each site ranged from 6 to 15. The number of captures per site varied considerably, from 18 to 103. Treefrogs were almost always captured in PVC pipes. We detected 9 frog species and 3 species of salamanders at three wetlands sampled in 2004. We collected water-quality field parameter data at 3 ponds in Apr. 2004. pH values were acidic in all ponds; conductivity and DO values were relatively low. Pond water temperatures varied from 17.7 to 22.75 ºC.

•  HNNWR: We made three intensive sampling visits in Apr., Jun., and Oct., detecting 12 species of amphibians (11 frogs, 1 salamander) in 2004. Hyla gratiosa was added in 2005. Our detection rate of species previously unrecorded during sampling is high, suggesting that we should detect additional species in 2005; this prediction has been verified. We detected 11 frog species and 1 species of salamander at 19 wetland sites (14 distinct ponds, ditches, or other sites) sampled in 2004. pH values were acidic in most ponds. Dissolved oxygen values were very low in all wetlands, but conductivity was relatively high everywhere. Percent DO values usually were very low, and pond water temperatures varied between 19.3 and 27.6 ºC during sampling.

•  SVNWR: We made three intensive sampling periods in Apr., Jun.-Jul., and Oct. We detected 21 species of amphibians (15 frogs; 6 salamanders) in 2004. Eurycea quadridigitata was added in 2005. Our detection rate of species previously unrecorded during sampling was high and constant, indicating that we are likely to detect additional species from SVNWR. We detected 15 frog species and 6 salamander species at 24 wetland sites, including 11 sites in old rice fields. The most amphibian species-rich wetland is Kingfisher Pond (8); most of the rice fields are depauperate in species, although biomass is undoubtedly quite high.  These are the only areas where Siren lacertina and Amphiuma means are found. Of particular interest was the finding of Ambystoma opacum on Bear Island in the Savannah River at several sites, D. auriculatus on both sides of the Savannah River, Rana heckscheri at Kingfisher Pond, and possibly Bufo fowleri on Bear Island and at Kingfisher Pond. We collected water-quality field parameter data at 7 locations in Jul. and 6 locations in Oct. 2004.  Values for pH were generally only slightly acidic in most wetlands; dissolved oxygen values were very low in all sampling sites; conductivity was relatively high everywhere; % DO values were low everywhere. Pond water temperatures varied between 19.3 ºC and 27.4 ºC. Three areas and on adjacent river bluffs were searched using time constrained methods, and yielded the only records of Plethodon variolatus.

Amphibian Species Richness Accumulation Plots, through May 2005

Our goal is to develop monitoring at SMNWR into a long-term, apex-level research site for SEARMI.  At LSNWR we are continuing to monitor amphibians at various wetlands, and we are expanding our wetland sampling to more localities on both refuges. At HNNWR and SVNWR, we are continuing to sample wetlands and terrestrial habitats to complete our inventory of amphibians at these sites. We are also conducting a pilot study assessing the potential for disease transmission through fish stocking at these refuges. At the end of the 2005 field season, a comprehensive plan will be prepared to guide future monitoring of amphibian status and trends on DOI Atlantic Coastal Plain refuges.