‘Ribbit Radio’ Shows Frog Population Estimates Are Likely Flawed
For Immediate Release
Scientists track amphibian populations because these animals are sensitive to changes in their environment and can serve as “canaries in the coal mine” to give researchers early warnings about pollution or other ecological problems. But new research from North Carolina State University shows that data from the largest amphibian monitoring program in the country may have flaws that, if uncorrected, could result in overestimates of frog populations.
The U.S. Geological Survey launched the North American Amphibian Monitoring Program (NAAMP) in the late 1990s to collect information on the diversity and abundance of frog species. The program is important because interest in amphibian populations has grown over the past 15 years, as scientists have learned that amphibians are environmental indicators that respond very quickly to pollution and environmental degradation.
“Understanding changes in amphibian populations can help us identify potential environmental problems,” says Dr. Ted Simons, a professor of biology at NC State, co-author of the new study, and assistant unit leader at the U.S. Geological Survey (USGS) Cooperative Fish and Wildlife Research Unit.
The NAAMP is a citizen-based program that relies on volunteers to travel to sample locations and listen for calling frogs at designated times of the year. These volunteers then report their findings, including which species are heard as well as a rough estimate of the population size, to the USGS.
Simons and his co-authors wanted to test the accuracy of these surveys by using the “Bird Radio” system Simons developed previously to test the accuracy of bird census methods. The system, renamed “Ribbit Radio,” consists of a series of remotely controlled playback devices that can be used to mimic populations of calling frogs. The researchers set up “Ribbit Radio” in a field and used it to test how well observers identify frog species.
Simons says the researchers immediately noted a lot of “false positives” in the data – meaning that some observers were saying they heard species that were not played by the “Ribbit Radio” system. “A lot of it is human nature,” Simons says. “Observers want to include as many species on their lists as possible, but sometimes our ears play tricks on us.” This could be an important issue for the NAAMP, because the survey uses the presence or absence of species to monitor changes in populations. Including false positives in the analysis, if not accounted for, can produce a serious bias in these population estimates. USGS has published on how to handle false positives and false negatives in assessing amphibian occurrence data and has ongoing research in this area.
Researchers hope that better training can help address this potential problem and improve the accuracy of the NAAMP in the future, Simons notes. “Training observers to help reduce false positive errors should improve the quality of the monitoring data,” he says.
“Dr. Simons’ work adds to our understanding of observer error in frog call surveys and more research on this topic is needed,” says Linda Weir, USGS NAAMP Coordinator and one of the participants in the study. Dr. Simons and NAAMP will continue working together this fall on a follow-up study using “Ribbit Radio.”
The current study, “Unmodeled observation error induces bias when inferring patterns and dynamics of species occurrence via aural detections,” is published in the August issue of Ecology. It is co-authored by Simons, NC State’s Kenneth Pollock, Larissa Bailey of Colorado State University and Brett McClintock of the University of St. Andrews.
NC State’s Department of Biology is part of the university’s College of Agriculture and Life Sciences.
Note to editors: The study abstract follows.
“Unmodeled observation error induces bias when inferring patterns and dynamics of species occurrence via aural detections”
Authors: Brett McClintock, University of St. Andrews; Larissa Bailey, Colorado State University; Ted Simons, Kenneth Pollock, North Carolina State University
Published: August 2010, Ecology
Abstract: The recent surge in the development and application of species occurrence models has been associated with an acknowledgment among ecologists that species are detected imperfectly due to observation error. Standard models now allow unbiased estimation of occupancy probability when false negative detections occur, but this is conditional on no false positive detections and sufficient incorporation of explanatory variables for the false negative detection process. These assumptions are likely reasonable in many circumstances, but there is mounting evidence that false positive errors and detection probability heterogeneity may be much more prevalent in studies relying on auditory cues for species detection (e.g., songbird or calling amphibian surveys). We used field survey data from a simulated calling anuran system of known occupancy state to investigate the biases induced by these errors in dynamic models of species occurrence. Despite the participation of expert observers in simplified field conditions, both false positive errors and site detection probability heterogeneity were extensive for most species in the survey. We found that even low levels of false positive errors, constituting as little as 1% of all detections, can cause severe overestimation of site occupancy, colonization, and local extinction probabilities. Further, unmodeled detection probability heterogeneity induced substantial underestimation of occupancy and overestimation of colonization and local extinction probabilities. Completely spurious relationships between species occurrence and explanatory variables were also found. Such misleading inferences would likely have deleterious implications for conservation and management programs. We contend that all forms of observation error, including false positive errors and heterogeneous detection probabilities, must be incorporated into the estimation framework to facilitate reliable inferences about occupancy and its associated vital rate parameters.