Biodiversity of traits and species both show weak responses to hydromorphological alteration in lowland river macroinvertebrates

Abstract

Summary Lotic ecosystems worldwide are being increasingly modified hydromorphologically, for example, by damming, fragmentation, flow regulation and channel modification. Serious threats to riverine biodiversity are suspected, yet available field data are few and rarely address the various taxonomic, functional and phylogenetic components of biodiversity. We examined benthic macroinvertebrate diversity along a gradient of hydromorphological alteration across 65 lowland river sites in the Netherlands, Germany and Poland. We calculated eight metrics describing taxonomic diversity (richness, rarefied richness, Shannon's and Simpson's diversity, Pielou's evenness), functional diversity (mean trait diversity) and phylogenetic diversity (mean taxonomic distinctness and its variance). Threshold Indicator Taxa ANalysis (TITAN) and Non‐metric Multidimensional Scaling (NMS) showed detailed changes in taxonomic and trait composition, respectively, with increasing hydromorphological alteration. TITAN revealed a gradual turnover under increasing impact in which taxon losses were almost balanced by gains. NMS illustrated a high degree of concordance in the traits possessed by the taxa lost and gained at the gradient's end points. Although a biologically meaningful hydromorphological gradient was evident, the biodiversity response was modest (Spearman's ρ < |0.32| for all metrics). Pielou's evenness and mean taxonomic distinctness actually increased at impaired sites. Our results suggest that taxonomic and trait replacement with hydromorphological alteration is not followed by changes in whole‐community diversity. This implies some redundancy in these systems, with ecologically similar species replacing those lost through environmental change. The results highlight how whole‐community‐based diversity, although frequently calculated and interpreted for decades in ecology and conservation biology, is not the all‐purpose concept to detect and assess the various facets of biodiversity loss. Therefore, we suggest that concerted effort is needed to develop and test novel indicators of biodiversity, capable of detecting biodiversity loss, or other changes, in response to the main environmental stressors of lotic ecosystems.