Drying determines the temporal dynamics of stream invertebrate structural and functional beta diversity

Assembly processes shaping ecological communities can vary over time following variations of environmental conditions at different scales. Such temporal dynamism is exacerbated by climate change and increasing extreme events, and recent evidence suggests that, in turn, community composition and functions can vary substantially. However, empirical relationships between the spatio-temporal dynamics of communities and that extreme events altering ecosystems are poorly investigated. We quantified the temporal dynamics of stream invertebrate communities over two years across 11 river basins prone to drying, covering a large geographical area of France. We tested predictions on the influence of the spatial arrangement and temporal dynamics of drying events across river networks. Combining a high temporal resolution of community description from taxonomic and functional perspectives, we quantified beta diversity over time and space and partitioned them into additive components: replacement of taxa and richness difference. Frequency and duration of drying events were precisely quantified and basins were classified based on the location of the drying events. We found a strong influence of the spatial drying pattern on the dissimilarities of community composition between sites. The high temporal variability of community structure was directly related to the frequency and duration of drying events. This temporal dynamism of communities was also strongly affected by the spatial drying pattern, indicating that fragmentation had a stronger effect on recolonisation processes for upstream-drying basins. Finally, biological traits were unevenly distributed in space and time, suggesting a lack of functional redundancy that could have strong implications for ecosystem functions and services. The high temporal dynamics of communities highlighted in this study challenge the current definition of reference conditions in intermittent rivers, and the community sensitivity to frequency and duration of drying suggest that climate change might lead community dynamics to be increasingly driven by stochastic environmental variability.