Responses of benthic macroinvertebrates to urbanization in nine metropolitan areas of the conterminous United States

The effects of urbanization on benthic macroinvertebrates were investigated in nine metropolitan areas (Boston, MA; Raleigh, NC; Atlanta, GA; Birmingham, AL; Milwaukee–Green Bay, WI; Denver, CO; Dallas–Fort Worth, TX; Salt Lake City, UT; and Portland, OR) as a part of the U.S. Geological Survey National Water Quality Assessment Program. Several invertebrate metrics showed strong, linear responses to urbanization when forest or shrublands were developed. Responses were difficult to discern in areas where urbanization was occurring on agricultural lands because invertebrate assemblages were already severely degraded. There was no evidence that assemblages showed any initial resistance to urbanization. Ordination scores, EPT taxa richness, and the average tolerance of organisms were the best indicators of changes in assemblage condition at a site. Richness metrics were better indicators than abundance metrics, and qualitative samples were as good as quantitative samples. A common set of landscape variables (population density, housing density, developed landcover, impervious surface, and roads) were strongly correlated with urbanization and invertebrate responses in all non-agricultural areas. The instream environmental variables (hydrology, water chemistry, habitat, and temperature) that were strongly correlated with urbanization and invertebrate responses were influenced by environmental setting (e.g., dominant ecoregion) and varied widely among metropolitan areas. Multilevel hierarchical regression models were developed that predicted invertebrate responses using only two landcover variables—basinscale landcover (percentage of basin area in developed land) and regional-scale landcover (antecedent agricultural land).