You are here:
Advancing prediction of headwater flow permanence and stream expansion and contraction using a process-based hydrologic model
Citation:
Mahoney, D., J. Christensen, H. Golden, C. Lane, G. Evenson, E. White, K. Fritz, E. DAmico, C. Barton, T. Williamson, AND K. Sena. Advancing prediction of headwater flow permanence and stream expansion and contraction using a process-based hydrologic model. Kentucky Water Resources Research Annual Symposium, Lexington, KY, September 13, 2021.
Impact/Purpose:
Headwater streams transport materials and energy to downstream waters but how often they flow is often difficult to characterize. We used a process-based model to estimate the occurrence of stream flows in a headwater catchment in Kentucky using gauge and logger data to calibrate and verify the model results. The model helps us determine what processes are important to determining the probability of stream flow and can be applied in other catchments.
Description:
Streamflow permanence in headwater systems supports both ecosystem functioning and the state of water quality within stream networks by facilitating hydrologic connectivity between upstream and downstream watershed compartments. Additionally, the quantification of streamflow characteristics in headwater systems has become particularly relevant in recent years for jurisdictional determination of waters protected under the Waters of the United States rule. Despite its apparent importance with respect to both ecosystem functioning and environmental policy, characterization of the frequency, magnitude, and duration of flow in headwater streams remains limited – largely due to the scarcity of monitoring efforts in these types of environments. Recent advancements in process-based, semi-distributed hydrologic models show promise for characterizing streamflow in headwater systems at semi-distributed spatial resolutions where data collection may be difficult; however, the effective simulation of streamflow permanence using process-based hydrologic models remains largely untested at the watershed scale. The objectives of this study were to: (1) develop and test an approach for simulating the frequency, magnitude, and duration of headwater streamflow with a process-based, semi-distributed hydrologic model and (2) apply model outputs to map the spatiotemporal dynamics of headwater expansion and contraction throughout the stream network. A spatially resolved hydrologic model (Dynamic TOPMODEL) was applied to a 1-km2 headwater network in University of Kentucky’s Robinson Forest, located in the Appalachian region of Kentucky. The model was forced using local climate data and structured with high-resolution LiDAR and geospatial data. Model performance was evaluated with several datasets including discharge data at the watershed outlet, flow-state sensor data, and observed headwater extent collected from field reconnaissance. The model framework simulated flow across the watershed uplands and within reaches at a high spatiotemporal resolution relevant for characterization of important stream dynamics. For example, the model predicted the probability of streamflow permanence at the reach scale as well as network scale dynamics of streamflow expansion and contraction. This study underscores the potential for watershed-scale, process-based hydrologic models to characterize headwater streamflow dynamics in systems throughout the eastern United States.
