ERRORS ASSOCIATED WITH "BEST-VALUE" REFERENCE CRITERIA IN MONTANA'S RAPID BIOASSESSMENT PROTOCOLS.

B.D. Marshall and B.L. Kerans. Department of Ecology, Montana State University, Bozeman, MT 59717

Definition of reference condition is arguably the most important element for development of defensible biological assessment programs. Montana's rapid bioassessment protocols use three levels of reference criteria. The regional reference criteria are based on an "ecoregion" classification scheme. Internal references are near to (and similar to) study sites but are minimally influenced by disturbances; such as upstream sites, or sites from nearby watersheds. The third set of reference criteria has been called the "best value" reference. These criteria are developed by combining data from all samples and selecting the highest-scoring (best) metric values to define reference condition – regardless of where they were collected. Thus, impaired streams may actually represent the reference for some metrics. We believe that this violates the underlying assumption that metrics represent and summarize the condition of a least-disturbed macroinvertebrate assemblage because combined metrics may not represent real (or attainable) communities. We compared the results of assessments using "best-value" references to the results attained from using regional and internal references. Although we examined surveys from only one ecoregion, our preliminary results indicate that this practice usually obscures anthropogenic impairment, rather than increasing the sensitivity of bioassessments.