Longitudinal variation of macroinvertebrate communities in a Mediterranean river subjected to multiple anthropogenic stressors

River impoundments and waste water discharge are a serious threat to the integrity and biodiversity of river ecosystems, especially in central Italy. Benthic macroinvertebrates were sampled in autumn and summer along the Aniene River to assess the cumulative biological effect of the numerous dams and sewage treatment plants that affect its middle and lower course. We hypothesized that (i) increasing habitat impairment would promote the formation of nestedness in species assemblage, where species poor locations support only a sub-set of organisms from richer sites; (ii) specific life-history traits would confer sensitivity to habitat degradation. Patterns of macroinvertebrate richness and diversity along the river tracked the distribution of dams and sewage treatment plants. Partial Mantel test showed that dissimilarity in assemblages increased with the number of dams and treatment plants between reaches after controlling for longitudinal distance. Assemblages were significantly nested, and nestedness appeared related to both water quality gradients (phosphorous, turbidity) that reflected anthropogenic inputs, and to natural gradient in altitude. Reaches with nested assemblages (supporting a sub-set of the species pool) were characterized by greater representations of taxa with shorter life cycles, while, in contrast, species rich sites supported taxa with longer life cycles and lower dispersal ability. These results suggest that the cumulative effect of dams and sewage treatment plants promoted the formation of nested subsets in species distribution. Moreover, it appeared that certain functional traits that conferred sensitivity also dictated the progressive non-random loss of taxa in face of multiple anthropogenic stressors. These findings have conservation implications in the regions, but need to be considered preliminary since anthropogenic and natural factors co-varied systematically along the study river precluding the identification of single factor effects.