In November 2023, we discovered serious errors in an influential population model that had helped convince the U.S. Fish and Wildlife Service to deny the gopher tortoise federal protection under the Endangered Species Act. Once corrected, the model’s predictions changed from a mild population decline to a rangewide collapse — yet our attempts to get the decision re-evaluated have been met with resistance.
In this PNAS opinion piece, we use the gopher tortoise case to argue for a broader fix: complex demographic models increasingly underpin high-stakes conservation policy, but agencies have no effective mechanism for correcting or re-evaluating decisions once serious flaws in the underlying science come to light. We call for more rigorous, independent technical review of the algorithms behind these models — not just the policy documents that cite them — especially when so much rides on getting the science right.
Citation: Shoemaker, K.T., Loope, K.J. 2025. We need better ways to re-evaluate conservation policies when they’re founded on flawed research. Proceedings of the National Academy of Sciences 122(19):e2426166122. https://doi.org/10.1073/pnas.2426166122