On Tuesday, April 30th, the National Research Council will release a highly anticipated study on pesticides and endangered species. The study is a major milestone toward resolving the clash over how to evaluate the effects of pesticides on endangered species. To date, the U.S. EPA has disagreed with the U.S. Fish & Wildlife Service and the National Marine Fisheries Service about which scientific methods and assumptions to use when completing these evaluations.
The study will address six key areas of scientific disagreement or lack of consensus among the agencies. But it will not address any related policy questions, which are an equally important source of disagreement. To think that the study will resolve the disagreements between the Services and EPA is to overlook the distinction between a science question and a policy question.
Here is one example of this distinction. To determine the effects of a pesticide on an endangered species, the EPA relies on laboratory tests on surrogate species. For example, EPA’s guidance document identifies bobwhite quail and ducks as a surrogate for endangered reptiles. Likewise, minnow and sunfish are surrogates for endangered amphibians. But of course, California tiger salamanders are not exactly like sunfish, and bog turtles are not exactly like ducks. When you extrapolate data on ducks to turtles, you must make a lot of assumptions about how ducks are like (and unlike) turtles. You will never have perfect information about the relationship between the two animals and their response to pesticides. As a result, you will have scientific uncertainty.
Scientific uncertainty is one of the topics the NRC study will cover. Specifically, it will evaluate the use of “uncertainty factors” (also known as safety factors) to account for incomplete data when extrapolating from surrogate species to endangered species. Think of uncertainty factors as a safety margin. If you assume that an endangered species is 100 times more sensitive to a chemical than is its corresponding surrogate species, then you might use an uncertainty factor of 100. So if your surrogate species does not show any significant adverse effects up to 10 mg/kg of exposure from a chemical, then you would divide that value by 100 to get 0.01 mg/kg for the endangered species.
But why use a safety factor of 100? Why not 200, or 38, or 4, or zero? For many endangered species, we do not know the extent to which they are more sensitive than surrogate species to any particular pesticide. So we need to make assumptions about how protective we are of endangered species. A safety factor of 100,000 would be very protective, while a safety factor of zero would offer no protection. This question of how protective we should be is a policy question, not a science question. Science can recommend methods for addressing uncertainty (safety factors are not the only method, and some would argue they are not the best). But science cannot tell us whether we “should” use a safety factor of 0, 10, 100, or 10,000. The Services and EPA must make those decisions based on their interpretations of the ESA and FIFRA. In fact, the NRC made a similar observation in its 1995 study titled “Science and the Endangered Species Act.” There, it noted that “[e]ven though estimates of risk are grounded in scientific information, those implementing the [ESA] often make value judgments when making decisions about listing, jeopardy, etc.” Thus, the NRC explained, “science by itself is not sufficient input to policy decisions, apart from the objectives and values it serves.”
Unfortunately, agencies are often very reluctant to acknowledge that they are making these value and policy judgments every day. But unless they do so, we will never fix the current debacle over pesticides and endangered species. We will never agree on how much risk of harming an endangered species is acceptable under the ESA and FIFRA. And we will continue quarrelling. This is why, when I presented at the CropLife America and RISE Spring Conference last week, I emphasized the need for both the Services and EPA to create a framework that clearly describes how they will address risk and uncertainty when evaluating the effects of pesticides on endangered species. Let’s hope the NRC study motivates the agencies to do so.
* * * *
Additional information on a way forward for pesticides and endangered species.