Abstract
AbstractReductions in streamflow due to groundwater pumping (“streamflow depletion”) can negatively impact water users and aquatic ecosystems but are challenging to estimate due to the time and expertise required to develop numerical models often used for water management. Here we develop analytical depletion functions, which are simpler approaches consisting of (i) stream proximity criteria, which determine the stream segments impacted by a well; (ii) a depletion apportionment equation, which distributes depletion among impacted stream segments; and (iii) an analytical model to estimate streamflow depletion in each segment. We evaluate 50 analytical depletion functions via comparison to an archetypal numerical model and find that analytical depletion functions predict streamflow depletion more accurately than analytical models alone. The choice of a depletion apportionment equation has the largest impact on analytical depletion function performance and equations that consider stream network geometry perform best. The best‐performing analytical depletion function combines stream proximity criteria which expand through time to account for the increasing size of the capture zone; a web squared depletion apportionment equation, which considers stream geometry; and the Hunt analytical model, which includes streambed resistance to flow. This analytical depletion function correctly identifies the stream segment most affected by a well >70% of the time with mean absolute error < 15% of predicted depletion and performs best for wells in relatively flat settings within ~3 km of streams. Our results indicate that analytical depletion functions may be useful water management decision support tools in locations where calibrated numerical models are not available.
Highlights
Effective conjunctive management of surface water and groundwater requires information about the impacts of groundwater pumping on streamflow, which is often poorly known
Since we found that the web and web squared depletion apportionment equations, in combination with the adjacent + expanding stream proximity criteria and Hunt analytical model consistently performed the best, we conducted an additional one-at-a-time sensitivity analysis of two parameters: the percent threshold used to define the limit of the adjacent + expanding stream proximity criteria, and the exponent used in the web and web squared depletion apportionment equations
We evaluated the performance of 50 analytical depletion functions to quantify the sensitivity of analytical depletion functions to the choice of depletion apportionment equations, stream proximity criteria, and analytical model under transient conditions; and identify factors describing the landscape and well-stream geometry that influence the performance of analytical depletion functions
Summary
Effective conjunctive management of surface water and groundwater requires information about the impacts of groundwater pumping on streamflow, which is often poorly known. During the development of the Michigan Tool, Reeves et al (2009) found that the best match to numerical model results was the Hunt (1999) analytical model with an inverse distance-based depletion apportionment equation using adjacent catchments as the stream proximity criteria This comparison was only conducted for a single timestep (after 5 years of pumping) and a single stream in Michigan. Zipper et al (2018a) tested 5 depletion apportionment equations across a range of stream network geometries in British Columbia and found that depletion apportionment equations which considers stream network geometry best matched numerical model results across several stream network and aquifer configurations This comparison was under steady-state conditions and did not investigate different stream proximity criteria, analytical models, or performance through time.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
