A new study calls into question some of the science used by the U.S. Fish and Wildlife Service to justify not listing the gopher tortoise under the Endangered Species Act.
Gopher tortoise populations could decline significantly in the future, and federal agencies use “exaggerated predictions from flawed models” when denying protections to the reptiles, say authors of new study It is said that he did.
“Modeling techniques that worked 20 years ago are now proven to be wrong,” says Kevin Roop, a TWS member and research scientist at Virginia Tech.
USFWS disputes these findings by Loope et al. But if the study is approved by a federal judge, it could ultimately help lead to a reassessment of the decision to list the gopher tortoise (Gopherus Polyphemus).
In October 2022, USFWS released findings on the status of gopher tortoises. They recognized distinct eastern and western population segments separated by the Mobile River in Alabama and the Tombigbee River in Alabama and Mississippi. They classified the western population as endangered under the ESA, but decided not to list the eastern portion.
The latter decision was based on five years of research on the species, part of which was published in 2022 in Global Ecology and Conservation by Brian Folt with the U.S. Geological Survey and colleagues. In this study, researchers from USFWS, USGS and others projected future population numbers of 457 people over 80 years under various threat and management scenarios.
In a study recently published in Global Ecology and Conservation, Loope and his colleagues highlighted what they believe to be a modeling error. Folt and his colleagues acknowledged that there was one error in their reply to Roop and his team, but they believe that the mistake did not significantly change the overall results of the study.
find the problem
Roop and his team wanted to test the conclusions put on the non-decision list. They developed and ran their own model using the same baseline information as the USFWS team, but with different results.
“Our model showed a much larger reduction,” Roop said. The USFWS model showed a more gradual decline.
After taking a closer look at the published model, Loope’s team claims there are two consequential errors that are quite technical. The first concerns the probability that a boy will reach adulthood.
The USFWS model divided adolescents and adults into two age classes and used maturation rates that assumed equal numbers of adolescents in each age class within the adolescent group. However, Loope and his team argue that assuming there are as many nearly mature turtles as hatchlings does not work, as the majority of turtles that die do so during their infancy. . In other words, the model does not take into account the number of premature deaths that can reduce the number of boys old enough to grow into adults each year. This issue in modeling was highlighted in a study published in 2019, Roop said.
In a rebuttal, also published in Global Ecology and Conservation, the USFWS team defended the use of this “common and accepted” model, saying the 2019 study was published at the time it began the review. He said it had just been done. Nevertheless, they acknowledge that their approach may be outdated at this point.
“Since then, age-based models and other approaches have been favored, and we acknowledge that the uniform within-stage age approach is increasingly considered suboptimal,” the USFWS team said. I’m writing.
How many gophers will migrate?
The second problem Loope and his colleagues discovered has to do with the way the USFWS team tried to explain turtle movement between populations. Both groups of scientists agree that it is inaccurate to assume that all groups of turtles are sitting on islands without interacting when modeling populations.
The USFWS team tried to address this by creating a separate immigrant pool for each population in the model, a kind of dummy metapopulation independent of the existing population. “This is a creative solution to a complex situation,” Roop acknowledged.
However, Loope argued that this solution creates a positive feedback cycle and that the model allows small populations to grow exponentially in just a few years.
According to the USFWS model, minority populations and metapopulations grow at the same rate each year. This means that if the population is small, immigration can double the population in some cases in a year. This doubles the metapopulation, bringing more turtles into the immigrant pool and potentially sending even more turtles back to previously small populations the following year.
Although the USFWS team built carrying capacity into its model for the existing population, it did not put a cap on the immigrant metapopulation, Roop said. Although there are only 70,000 gopher tortoises throughout the species’ range, the model shows that an artificial flood of hypothetical tortoises through positive feedback effects could result in some individual metapopulations having as many as 1 million gopher tortoises after 80 years. Monoturtles may be expected to be included, Roop said.
“That’s not possible for turtles. Turtle populations don’t grow that quickly,” Roop says.
The problem, said Kevin Shoemaker, a TWS member, associate professor of quantitative population ecology at the University of Nevada, Reno, and co-author of the study, is that the model’s immigration feature creates turtles out of thin air. He said that.
“Immigration itself cannot create new individuals,” Shoemaker said. “That’s what childbirth is all about.”
Conversely, if we solved this problem by limiting the number of turtles entering this immigrant metapopulation to approximately a 3% increase, many of the populations that the USFWS team predicted would increase would actually remain the same. In the worst case scenario, it will become extinct. .
After solving both the maturation rate and positive feedback cycle issues, Loope said, “You’re talking about de facto extinction.”
He said Loope and his colleagues submitted their findings to the authors and asked if they would like to submit corrections. Publishing such an amendment would require the consent of all co-authors of the paper, but the authors ultimately did not issue the amendment.
“I’m disappointed,” Roop said.
sticky lawsuit
Legally, the situation became complicated. The Center for Biological Diversity (CBD) and Noxe Education, Inc. (the entity associated with the Noxe property housing the relocated turtles near Panama City, Florida) will delist the turtles in September 2023. He sued the department over the decision, calling it “arbitrary.” And it’s illegal. ”
Since the USFWS team did not issue any corrections to the study, Loope’s team published their findings in a letter. CBD and Nokuse are adding Loope’s response to the USFWS study to their lawsuit. But rather than dispute Loop’s science, the USFWS asked the judge not to consider Loop’s findings. “Plaintiffs’ declaration is not an objective technical explanation to the court, but rather provides a post-judgment opinion supporting Plaintiffs’ view on the merits of the case,” USFWS said in court documents. This case is ongoing.
For Loope, excluding peer-reviewed papers from litigation is questionable. “I have a lot of sympathy for the service because it’s really difficult[to model turtle population projections]but we shouldn’t defend bad science,” Roop said. “If they think we’re wrong, they could explain it, but they’re trying to prevent the judge from considering it at all.”
As for why they wanted to disallow the study, rather than trying to disprove the claims made by Loope and his colleagues, Folt himself and a USFWS spokesperson provided a rebuttal to the Loope team’s letter. pointed out. “The agency does not comment on proposed or pending litigation,” a USFWS spokesperson said.
The rebuttal in question is only three paragraphs long. In it, the USFWS defended its approach to dealing with immigrants, stating: The structure we created. Noting that the immigration function developed for this system is hypothetical due to the lack of quantifiable data, we tested the sensitivity of the model output to immigration rates.”
Nevertheless, they seem to acknowledge that setting limits on the number of turtles that may migrate could improve predictions. “Due to its hypothetical nature and high sensitivity, immigration may have room for improvement in future modeling efforts,” Folt and colleagues said.
