Mathematical modeling is an indispensable tool in the field of environmental science, and inevitably, such models must be used as evidentiary tools in environmental-tort litigation.  However, the test for the reliability and relevance (and, ultimately, for the admissibility) of scientific testimony as articulated by the Supreme Court in Daubert v. Merrell Dow Pharmaceuticals, Inc. has proven especially problematic in evaluating the reliability of such mathematical models.

Many factors contribute to this problem: models are often plagued by uncertainty and lack of transparency; in an adversarial context, models are often used to obfuscate rather than clarify empirical evidence; and models can often be overly technical and challenging for judges and juries who may be unfamiliar with even the most basic scientific issues posed by the use of a model.  In light of these problems, this Note argues that Daubert is not the correct framework for addressing the problem of model reliability because it “hits scientific modeling sideways and often obscures the benefits of modeling, while potentially ignoring many of its problems, leading to too many false positives and false negatives in decision making.”

Because the field of modeling constitutes a sui generis problem, the Daubert factors should be abandoned or at least heavily modified in the context of evaluating models.  This Note looks to the use of models by regulatory agencies, particularly the U.S. Environmental Protection Agency (EPA), in formulating a new test.  In line with practices of the EPA, it advocates the creation of a mandatory screening checklist to be used by decision makers.