Spatial structure in lotic macroinvertebrate communities in England and Wales: relationship with physical, chemical and anthropogenic stress variables

Abstract

We describe the relationship between macroinvertebrate community composition, the physicochemical environment and anthropogenic impacts, in running water sites across a range of water qualities in England and Wales. We have also investigated the degree of spatial structure present in both the macroinvertebrate community and the measured environment. Selected explanatory variables could account for 26% of the variation in lotic macroinvertebrate assemblage composition across England and Wales. The explanatory power of the CCA model was based predominantly on a combination of local scale variables (substrate, alkalinity, urban run-off) and regional scale variables (discharge category, northing). The physicochemical gradient associated with changes in stream type from headwaters to estuary dominated assemblage composition. The influence of pollution and habitat modification were of secondary importance. There was a substantial level of spatial structure to both the physicochemical (47% of its explanatory power spatially structured) and anthropogenic stress data (63% of its explanatory power spatially structured), which resulted in a high level of predictable spatial structuring in macroinvertebrate assemblages. Almost 40% of the variation in assemblage composition accounted for by the explanatory model exhibited spatial structure. Positive spatial autocorrelation in macroinvertebrate community composition extended to sites up to 150km apart. As a consequence, community composition could be described from northing and easting with 75% of the explanatory power of the eight physicochemical variables. Our study has confirmed the importance of the longitudinal gradient within catchments, as well as the geographical position of the catchment to macroinvertebrate communities. We have also demonstrated how quantifying the spatial structure in the dataset can improve our understanding of the factors influencing macroinvertebrate community structure.