Large scale macroecological and metacommunity research is developed to identify patterns and processes of ecosystems and biotic community changes. Too often though, studies of change in the occurrence of species and communities fail to identify the drivers of change, as they often lack the biogeographical context, or the fine-grained data needed to identify the relevant parameters and processes. More specifically they fail to consider or express the hierarchical relationships between the drivers and stressors among the different spatial scales.  Here, we coupled both a large scale biogeographical and metacommunity structure analysis with a fine-grained community analysis, over an extended dataset of macroinvertebrates samples in space and time. Hereto we gathered or broke up as many catchment units as feasible across France from our data by specifying a catchment scale of interest for looking at metacommunity patterns. Through statistical analysis we partitioned the total variance between the different scales, with emphasis on differences between catchments, to identify at which ones do communities structure themselves. Thus the identified metacommunity structures can be linked to community trait characteristics to identify drivers of (bio)geographic character and anthropogenic stress gradients.