5. Summary
In this study, we developed and applied an approach to assess the strength of hydrologic connections between a source and target sites within a river-floodplain system using field-based indicators. We defined the source as the river at the upstream boundary of the river-floodplain system with target sites both downstream within the river channel and laterally distributed across the floodplain. Using a field-based dataset, we generated empirical models to describe target site-specific connectivity strength as a function of source stage, and then predicted daily connectivity strength for five years between May and September of 2016 through 2020. Within this approach we also tested the use of aquatic microbiomes as a metric of hydrologic connectivity. By examining similarity in microbial membership, we accurately assessed the presences/absences of surface flows from the source to target sites and found that aquatic microbiomes can provide additional information on residence time dynamics along connected surface flow paths.
Our results demonstrate that connectivity in aquatic water bodies in the river-floodplain system can either be stable or intermittent. Intermittently connected target sites can differ widely in source connection/disconnection thresholds. While some target sites demonstrate binary connectivity behavior quickly changing between high and low connectivity states, others display gradual behavior with substantial durations at intermediate connectivity levels. When aggregated to the river-floodplain system scale, mean system-scale connectivity increased with stage due to the influence of averaging across sites with heterogeneous conditions. However, we demonstrate that mean behavior is a poor descriptor of river-floodplain system behavior because spatial distributions of connectivity across all sites tend toward bimodality at intermediate and lower flows, reflecting conditions when river stage is sufficient to enable connectivity for only a portion of the system. As a result, the spatial heterogeneity of connectivity state peaked at intermediate river stage values. We also demonstrate that connectivity regimes are sensitive to inter-annual variation in streamflow and that while differences in the magnitude of peak flow are important so are changes to the flow duration across the range of differing flow states. Therefore, predictions of how river-floodplain connectivity will change due to climate change and/or other disturbances to hydrologic regimes need to consider the impacts across the full range of potential flows.