Evaluating Metrics of Lateral Connectivity in a Complex Montane Floodplain.

In undisturbed mountain valleys, interactions between streams and riparian areas generate hydrologically connected, physically complex floodplains with high seasonal water tables and extensive wetlands. These bi-directional lateral exchanges of water and solutes are thought to regulate biogeochemical functioning within floodplain habitats and alter timing and magnitudes of downstream solute fluxes. Within a floodplain, differing habitats exhibit substantial spatiotemporal variation in the magnitude and effect of connectivity. As such, quantifying respective functional roles of floodplain units within the stream reach (i.e. their ability to store, transform and transmit water and solutes) has remained a challenge. Here we evaluate the use of traditional and newly proposed metrics of hydrologic and biogeochemical connectivity and assess each metric's ability to quantify magnitude, timing, duration and downstream effect of connectivity within a complex floodplain in Rocky Mountain National Park. We make use of data from a distributed network of high frequency sensors for water level, temperature, colored dissolved organic matter (CDOM), conductivity and dissolved oxygen and from weekly sampling of conservative (Na, Cl, Ca) and non-conservative solutes (DOC, TDN, NO₃^-, NH₄⁺), and stable water isotope analysis (¹⁸O/²H). Preliminary results indicate that concurrent use of statistical time series approaches such as cross-correlation functions, mutual information and transfer entropy in conjunction with mixing models provide improved insights into the temporal dynamics of connectivity and its effects on downstream fluxes. Developing better approaches for quantifying lateral connectivity will improve our ability to monitor floodplain dynamics and response to disturbance, and design and assess the efficacy of floodplain restoration projects that can improve water quality in river networks.