Qiang Li, Tom Gleeson, Sam Zipper , Ben Kerr

Groundwater pumping can cause streamflow depletion by reducing groundwater discharge to streams and/or inducing surface water infiltration. Analytical and numerical models are two standard methods used to predict streamflow depletion. Numerical models require extensive data and efforts to develop robust estimates, while analytical models are easy to implement with low data and experience requirements but are limited by numerous simplifying assumptions. We have pioneered a novel approach that balances the shortcomings of analytical and numerical models: analytical depletion functions (ADFs), which include empirical functions expanding the applicability of analytical models for real-world settings. In this paper, we outline the workflow of ADFs and synthesize results showing that the accuracy of ADFs compared against a variety of numerical models from simplified, archetypal models to sophisticated, calibrated models in both steady-state and transient conditions over diverse hydrogeological landscapes, stream networks, and spatial scales. Like analytical models, ADFs are rapidly and easily implemented and have low data requirements but have significant advantages of better agreement with numerical models and better representation of complex stream geometries. Relative to numerical models, ADFs have limited ability to explore non-pumping related impacts and incorporate subsurface heterogeneity. In conclusion, ADFs can be used as a stand-alone tool or part of decision-support tools as preliminary screening of potential groundwater pumping impacts when issuing new and existing water licenses while ensuring streamflow meets environmental flow needs.