SOUTHEAST  AMPHIBIAN RESEARCH AND MONITORING INITIATIVE

Activity Summary for Okefenokee National Wildlife Refuge

December 2001

Project Description

Okefenokee National Wildlife Refuge is one of five sites in the southeast where USGS amphibian research and monitoring programs are underway. The Refuge encompasses nearly 80% of the Okefenokee Swamp, one of the largest freshwater wetlands in the southeastern United States.  Conservation issues of concern at the Refuge include: assessing the impact of surrounding land uses, heavy metals and decreasing pH on the aquatic ecosystem. There has been little recent work on amphibians on the Refuge.  Therefore, our approach to the amphibian inventory and monitoring initiative at Refuge has included: 1) surveys of historic localities of rare or unusual species; 2) "extensive" sampling to determine species presence and distribution on the Refuge; 3) "intensive" sampling at permanent monitoring sites within the Refuge; and 4) development of a protocol for monitoring aquatic salamanders. In addition, researchers with the Maine Cooperative Fish and Wildlife Research Unit are developing spatial models that may determine species-habitat associations based on our extensive survey results.

Current Results

Field surveys began in August 2000.  We have established 16 permanent monitoring sites throughout the Refuge [forested wetlands (4), wet prairies (4), lakes (2), ponds (2), shrub wetlands (2), and pine flatwoods (2)].  These sites are sampled quarterly and we have completed three quarterly sampling sessions (March-April; June-July; September-October).  Quarterly sampling includes three visits to each site over a five day period and methods include both active (litter search) and passive (pvc refugia, funnel traps, automated acoustic recorders/frog loggers) techniques for detecting amphibians. These methods are used to generate species lists at each sampling period for analysis with proportion of area occupied (PAO) models.  Multiple visits to each site per sampling period should enable us to generate estimates of detectability with confidence intervals for individual species.  We also are conducting mark-recapture studies of hylid frogs at all 16 sites (using pvc refugia), in an attempt to monitor populations of these species. We plan to complete a full year of quarterly sampling in early January 2002.

We conducted 94 extensive surveys across the Refuge over the past year.  Extensive surveys involved the use of a variety of sampling techniques to detect species presence during a single visit to a site (1-day extensive surveys).  We recently expanded our effort on these surveys to include a three day sampling effort ("3-day extensive surveys"), where we set and check traps at some sites over a 3-day period, so that we can estimate detectability.  A portion of the 1-day extensive and 3-day extensive sites will be revisited on an annual basis in an attempt to monitor changes in species composition over time.  We originally chose extensive sampling sites somewhat arbitrarily, i.e., by traveling along boat canals and stopping to sample at regular time intervals.  However, we now have a pool of randomly chosen sites to chose from, based on a computer generated series of grid cells.  Access to sampling points in the Refuge continues to be a problem; therefore, we are only considering sampling points within 0.5 km of boat trails (but a minimum of 50m from the trail edge).

Data for intensive and extensive surveys have been entered in a Microsoft Access database. The database is structured so that we can evaluate the success of various techniques at capturing amphibians; structure the data for PAO and mark-recapture analysis.

An aquatic salamander monitoring project was initiated in August 2001 and will be continued through August 2002. The project includes an evaluation of capture techniques (crayfish traps and minnow traps) and a mark-recapture study at four sites in the Refuge.

Plans for Next Year

The quarterly monitoring effort will be scaled back beginning in Spring 2002.  Several of the sites have yielded extremely low numbers of amphibians, despite our best efforts. Furthermore, frequent visits to these sites have begun to alter the habitat (e.g. peat and soil subsidence). Therefore, only 10 of the original 16 intensive sites will be sampled quarterly, although passive traps will be left at the sites so that they can be re-sampled on a less frequent basis if needed. An increased emphasis will be placed on visiting randomly chosen "3-day extensive" sites; a portion of these sites will be re-visited each year in an attempt to detect changes in species composition over time.  We also plan to increase the use of passive means of detecting amphibian presence by using timed frog loggers (devices which record frog calls at specified intervals).  We will deploy frog loggers at selected sites over extended periods for use in PAO models.  The aquatic salamander monitoring project will be completed in August 2002. Over the first three months of the study, trapping success exceeded 30% for two-toed amphiuma (Amphiuma means).  Final results of this project may be used to design a larger scale population monitoring study for amphiuma at the Refuge.

Collaborators

U.S. Fish and Wildlife Service:  Sara Aicher, Okefenokee National Wildlife Refuge, Route 2, Box 3330, Folkston, GA 31537

Vegetation Mapping/GIS Support:  Cynthia S. Loftin, Maine Cooperative Fish & Wildlife Research Unit, 5755 Nutting Hall, Room 230, Orono, ME 04469

Jon McClosky, Maine Cooperative Fish & Wildlife Research Unit, University of Maine, 5755 Nutting Hall, Room 230, Orono, ME  04469

Dean Easton, U.S. Geological Survey, 412 NE 16^th Avenue, Room 250, Gainesville, FL 32601—GIS support.