You are here:
Advancement of Flow Permanence Prediction in Headwater Systems Using Semi-Distributed Hydrologic Modeling
Citation:
Mahoney, D., J. Christensen, H. Golden, G. Evenson, C. Lane, E. DAmico, C. Barton, T. Williamson, K. Fritz, AND K. Sena. Advancement of Flow Permanence Prediction in Headwater Systems Using Semi-Distributed Hydrologic Modeling. World Environmental & Water Resources Congress 2021, Virtual, May 23 - 27, 2021.
Impact/Purpose:
Presentation to the World Environmental & Water Resources Congress.
Description:
Headwater streams comprise large portions of river networks and provide vital ecosystem functions. Yet these systems are less monitored and more prone to disturbance compared to larger order waterways. Further, while streamflow permanence in headwater systems is poorly understood, it has become a key factor in determining how and whether streams are regulated across the globe. Recent improvements in semi-distributed hydrologic modeling tools have resulted from increased computing power and geospatial watershed data availability. However, most models remain untested in their capacity to simulate streamflow permeance in headwater systems. Our objectives were to: (1) assess the extent to which an existing semi-distributed hydrologic model (Dynamic TOPMODEL) can simulate the magnitude, duration, and frequency of surface flows in headwater stream systems, and (2) compare resulting streamflow permanence measures with previously derived permanence classifications. Our initial study site is a 32 km2 non-mined watershed in the Appalachian Coal Belt region of Kentucky, USA. The region is ripe for study as headwater streams are prevalent in the well-dissected uplands, and they are prone to disturbance from coal mining operations and timber harvest. We applied Dynamic TOPMODEL to assess headwater stream permanence because it simulates saturation-excess overflow at the semi-distributed resolution necessary to capture the full extent of headwater flows. We derived stream channel extents from 2m LiDAR for the watershed and evaluated the model’s performance with an ensemble of spatial and temporal datasets including: (1) streamflow discharge at multiple locations along the stream network; (2) water table height; (3) flow state sensor data; and (4) field reconnaissance of headwater extent. Dynamic TOPMODEL simulated the magnitude, duration, and frequency of streamflow in these headwater systems with relative certainty. Our results suggested that the variability in the morphologic configuration and dynamic hydrologic connections within catchments impacted patterns of stream expansion and contraction within the larger study watershed. Moreover, model-derived maps of perennial, intermittent, and ephemeral stream extent aligned with traditionally derived delineations of headwater stream permanence, emphasizing the efficacy of using dynamic hydrologic modelling tools to characterize and classify stream networks.