Date of Award

12-2024

Degree Type

Report

Degree Name

Master of Science (MS)

Department

Civil and Environmental Engineering

Committee Chair(s)

Pin Shuai (committee chair)

Committee

Pin Shuai

Committee

Bethany Neilson

Committee

Wei Zhang

Abstract

Baseflow, commonly referred to as the groundwater contribution to streamflow, constitutes approximately 50% of streamflow in mountainous regions of the Western United States. Accurately quantifying the amount of baseflow is critical for water management and decision-making, as it significantly impacts stream water quality, low flow availability, and ecosystem health. Traditionally, baseflow has been estimated using conceptual and automated baseflow separation methods, which are known to be both arbitrary and ambiguous, posing a challenge to validate them. In this study, we developed an integrated hydrologic model that seamlessly integrated the exchange between surface and subsurface flows to physically quantify the baseflow component in a snow dominated catchment—Coal Creek Watershed (CO, USA). The simulated baseflow and streamflow from the numerical model were then used as a controlled experiment to evaluate the performance of four commonly used baseflow separation methods, including the Pettyjohn and Henning (PH) graphical, the United Kingdom Institute of Hydrology (UKIH) graphical, the Eckhardt digital filter, and conductance mass balance (CMB) methods. Simulated baseflow has an average baseflow index (BFI) of around 53% with a higher BFI in dry years versus that in wet years. In comparison to the numerical baseflow, both UKIH graphical and Eckhardt digital filter methods performed relatively well with high modified Kling-Gupta Efficiency (mKGE) (0.72 and 0.68, respectively) and Nash-Sutcliffe Efficiency (NSE) (0.58 and 0.7, respectively) values. However, UKIH graphical method performed poorer than the Eckhardt digital filter method in average and dry years when stream hydrographs resemble unimodal peaks, which are typical in snow-dominated catchments. Additionally, the Eckhardt digital filter showed better performance in matching the temporal dynamics of baseflow with a smoother hydrograph. Both the PH graphical and CMB methods did not perform satisfactorily in estimating baseflow with both mKGE and NSE values less than 0.3. Among them, the PH graphical method has consistently overestimated baseflow with an average BFI of 85%, whereas the CMB method has consistently underestimated baseflow with an average BFI of 24%. Overall, the Eckhardt digital filter method is promising for automated baseflow separation in snow-dominated catchments, though the relative importance of baseflow contribution to streamflow may be underestimated in dry years. Our findings suggest that integrated hydrologic models, when calibrated, provide a quantitative way to evaluate and improve existing baseflow separation methods. Caution should be exercised when applying automated baseflow separation methods in snow-dominated catchments, and future investigations are warranted to thoroughly evaluate these methods in catchments with diverse hydroclimate conditions.

Download

Share

COinS

Copyright for this work is retained by the student. If you have any questions regarding the inclusion of this work in the Digital Commons, please email us at .