Understanding the Role of Model Structure and Watershed Properties on Streamflow Response

Abstract

Hydrologic models may be used for both prediction and understanding of hydrologic responses. In this dissertation, I first evaluated the influence of representation of storage-discharge relation on streamflow prediction accuracy. Results suggest that derived storage-discharge relations vary significantly depending on the choice of recession analysis method, and this difference is large enough to appreciably affect the streamflow response obtained using it. Next, a process-explicit model was used to evaluate the relative role of different processes and states on hydrologic response. Specifically, I explored the causes of variability in flood response to large hurricane-season storms, and identified the dominant controls on this variability to be soil saturation near the ground surface and evapotranspirative losses. I also evaluated the role of the process controls on both intra- and inter-seasonal variations in daily peak time in a snow-dominated watershed, and identified subsurface flow as the most dominant control on daily peak time delays, with contribution from melt water translation through the snowpack comes a close second. Finally, I developed and implemented a physics-based 2D hydrologic model that accounts for interception, evapotranspiration, and subsurface flow processes to investigate how geomorphological modification of the landscape might have affected hydrologic responses of the piedmont hillslope. Results indicate that with increasing gully incision, groundwater flow out of the hillslope increases, while groundwater table, root zone moisture, transpiration and surface runoff reduces. The impact of gully incision can also propagate far and wide in the hillslope, with its extent mainly determined by hydraulic conductivity and hillslope steepness. Studies in this dissertation could help prioritize measurements during observation campaigns and could also aid in risk management under climate change and other disturbance conditions.