Genus-level, trait-based multimetric diatom indices for assessing the ecological condition of rivers and streams across the conterminous United States

Taxonomic inconsistency in species-level identifications has constrained use of diatoms as biological indicators in aquatic assessments. We addressed this problem by developing diatom multimetric indices (MMIs) of ecological condition using genus-level taxonomy and trait-based autecological information. The MMIs were designed to assess river and stream chemical, physical and biological condition across the conterminous United States. Trait -based approaches have the advantage of using both species-level and genus-level data, which require less effort and expense to acquire than traditional species-based approaches and eliminate the persistent taxonomic biases introduced over vast geographic extents. For large-extent assessment programs that require multiple taxonomic laboratories to process samples, such as the United States Environmental Protection Agency's (U.S. EPA's) Na-tional Rivers and Streams Assessment (NRSA), the trait approach can eliminate discrepancies in species-level identification or nomenclature that hinder diatom data interpretation. We developed trait-based MMIs using NRSA data for each of the three large ecoregions across the U.S. -the East, Plains, and West. All three MMIs performed well in discriminating least-disturbed from most-disturbed sites. The MMI for the East had the greatest discrimination ability, followed by MMIs for the Plains and West, respectively. The performance of the MMIs was comparable to that observed in existing NRSA fish and macroinvertebrate MMIs. Our research shows that trait -based diatom indices constructed on genus-level taxonomy can be effective for large-scale assessments, and may also allow programs such as NRSA to assess trends in freshwater condition retrospectively, by revisiting older diatom datasets. Moreover, our genus-based approach facilitates including of diatoms into other assessment programs that have limited monitoring resources.