Managing flood flow connectivity to landscapes to build buffering capacity to disturbances: An ecohydrologic modeling framework for drylands

Increased flooding, droughts, and sediment transport are watershed-scale problems negatively impacting agriculture and ecosystems in drylands worldwide. Vegetation loss in upland watersheds is leading to scouring floods, which in turn decreases infiltration, soil moisture levels, and downstream groundwater recharge. Management to confront these intractable problems has been hindered by a lack of accessible decision support tools for both land and water managers that synthesize the watershed processes that buffer against dryland disturbances. Flood flow connectivities across the landscape create buffer zones through replenishing soil moisture and reducing flood energy, which in turn support multiple functions. This study developed a decision support tool, the Flood Flow Connectivity to the Landscape (FlowCon) framework that quantifies the most efficient management efforts to increase the key watershed buffering functions of increasing infiltration and reducing flow energy. FlowCon links three spatially explicit, process-based, and predictive models to answer two critical management questions: what key processes acting in what optimal areas are drivers of infiltration dynamics and what roles do peak flows of differing scales of energy play. The spatial models delineated the buffer zone to characterize the heterogeneous and optimal infiltration dynamics across the landscape. The hydrologic process model, using a curve number technique, identified the key ecohydrologic processes that affect infiltration and characterized peak flows and flow regime variability. The predictive flood routing model quantified the potential management benefits. We calibrated the models with measured runoff and the corresponding rainfall events for a six-year period, which included thirty-six flow events. The synthesized ecohydrologic indicators provided critical calibrations, improving the relationship between the hydrologic modeling results and observed data by 12% for the linear regression R² and 69% for the root mean square error (RMSE). Implementation of prioritized management is estimated to reduce peak flow by half, with interventions focused on 24% of floodplains that infiltrate three times the flow volume per area than the floodplain average. FlowCon provides an efficient assessment framework that integrates watershed process understanding in an accessible decision support tool to achieve tangible improvements in dryland watershed management.