Berkeley Lab

Simulating groundwater-streamflow connections in the Upper Colorado River Basin

A series of hypothetical numerical experiments with and without lateral groundwater flow is performed to isolate the role of lateral groundwater flow which are often over-simplified in traditional hydrologic model. Results show that that peak flows increase up to 57% when lateral groundwater flow processes are included.

The Upper Colorado River Basin (UCRB) is subject to climate extremes that pose a challenge to simulate and predict. Open questions remain about the correct model physical parameterization for accurate simulation of large, headwaters systems with a mix of terrain. Results show that including lateral groundwater flow component not only improving simulation accuracy but also quantifying the role of subsurface hydrological processes.

Stream flow time series for two representative stations in the Upper Colorado River Basin. Blue lines are observed daily flow, red and green lines are simulated flows from lateral and no lateral simulations respectively.

Summary

The Upper Colorado River Basin (UCRB) is subject to climate extremes that pose a challenge to simulate and predict. Open questions remain about the correct model physical parameterization for accurate simulation of large, headwaters systems with a mix of terrain. This study focusses on the role of lateral groundwater flow on total annual and peak streamflow in the UCRB using a physical hydrology model, ParFlow. Results for the simulated water year of 1983 (the high snowmelt year that almost destroyed the Glen Canyon Dam), suggest an increase in peak flow of up to 57% when lateral groundwater flow processed are included. This is an unexpected result for flood conditions, which are generally assumed to be independent of groundwater. Including lateral groundwater flow produces 25% to 50% more the larger downstream rivers in the domain and 20% to 40% less flow in headwater streams. These changes are mainly driven by hydraulic gradients in the subsurface which are often over-simplified in traditional hydrology models and can only be captured when lateral flow is considered. Lastly, the authors find that the impact of including lateral groundwater flow on the simulated flows exceeds the impact of an order of magnitude change in hydraulic conductivity. While the results focus on the UCRB, the authors feel that these simulations have relevance to other headwater systems worldwide.

Citation

Tran, H., J. Zhang, J.-M. Cohard, L. E. Condon, and R. M. Maxwell (2020), Simulating Groundwater-Streamflow Connections in the Upper Colorado River Basin, Groundwater, 58(3), 392-405, DOI: 10.1111/gwat.13000.