Abstract

More than one-half of the world's population lives in urban areas, so quantifying the effects of urbanization on ecological communities is important for understanding whether anthropogenic stressors homogenize communities across environmental and climatic gradients. We examined the relationship of impervious surface coverage (a marker of urbanization) and the structure of stream macroinvertebrate communities across the state of Maryland and within each of Maryland's three ecoregions: Coastal Plain, Piedmont, and Appalachian, which differ in stream geomorphology and community composition. We considered three levels of trait organization: individual traits, unique combinations of traits, and community metrics (functional richness, functional evenness, and functional divergence) and three levels of impervious surface coverage (low [<2.5%], medium [2.5% to 10%], and high [>10%]). The prevalence of an individual trait differed very little between low impervious surface and high impervious surface sites. The arrangement of trait combinations in community trait space for each ecoregion differed when impervious surface coverage was low, but the arrangement became more similar among ecoregions as impervious surface coverage increased. Furthermore, trait combinations that occurred only at low or medium impervious surface coverage were clustered in a subset of the community trait space, indicating that impervious surface affected the presence of only a subset of trait combinations. Functional richness declined with increasing impervious surface, providing evidence for environmental filtering. Community metrics that include abundance were also sensitive to increasing impervious surface coverage: functional divergence decreased while functional evenness increased. These changes demonstrate that increasing impervious surface coverage homogenizes the trait diversity of macroinvertebrate communities in streams, despite differences in initial community composition and stream geomorphology among ecoregions. Community metrics were also more sensitive to changes in the abundance rather than the gain or loss of trait combinations, showing the potential for trait-based approaches to serve as early warning indicators of environmental stress for monitoring and biological assessment programs.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call