As anthropogenic changes drive species declines, global biodiversity faces major threats, further exacerbated by climate change. Reptiles are one of the world's most endangered vertebrate classes and among reptiles, turtles are the most threatened group.
This study by researchers from the University of Miami focusses on the endangered wood turtle (Glyptemys insculpta), occurring mainly in the Northeastern United States. Already threatened by habitat loss due to urbanization, this temperature-sensitive species may be particularly vulnerable to climate change, as it relies on cool, fast-flowing streams—a habitat type disappearing with increasing temperatures.
"Our study is part of a collaboration with the Susquehannock Wildlife Society in Maryland, and we wanted to use these habitat suitability outputs to help guide their population survey efforts," explains Caitlin C. Mothes, first author of the paper published in Global Ecology and Conservation. "We expanded the species distribution models to also model climate change impacts, as that hasn't really been studied for this species and they have been under review for listing under the Endangered Species Act since 2012."
Using GBIF-mediated occurrences of G. insculpta, the authors developed a habitat suitability model based on variables deemed important for the turtle—such as distance to nearest stream, canopy cover, elevation and slope, as well as climates variables such as temperature and precipitation. Employing Maxent as the chosen modelling algorithm, the authors used all reptile occurrences as background points to account for sampling bias in the region of interest.
"We tried contacting multiple sources for their occurrence data on this species but never heard back from them, so the open-access data offered by GBIF was vital to the project. As we use target-group background points when training the model to reduce sampling bias, GBIF is pretty much the only database where we could retrieve thousands of occurrence points for all other reptile species in the area."
When projected into future climate scenarios, the authors' model showed the optimal wood turtle habitat decreasing by 62 per cent under low emissions (RCP 4.5) and 86 per cent under high emissions (RCP 8.5) by the year 2070. The main driver behind this decrease is average temperatures, exceeding G. insculpta suitable limits by 2.5 degrees Celsius on average.
With optimal habitat inside protected areas predicted to decrease by as much as 64 per cent, the authors recommend long-term conservation efforts to be directed towards protecting climate refugia present in the higher latitudes of the species' range.
"We hope that our results can help provide additional support for listing this species, and also help guide conservation management strategies."
This paper is the 5,000th peer-reviewed journal article to use GBIF-mediated data to help reach its conclusions. As recommended by GBIF, the authors of the paper cite species occurrence downloads using their assigned DOIs—in this case the G. insculpta data and the overall reptile data. Explore all papers—including more than 750 articles focusing on the effects of climate change—in the GBIF literature index.