Use of Ground Water Models for Resource Assessment of Watersheds Affected by Irrigation Pumping: Impact of Model Input and Parameter Uncertainty

Ground water is an important resource since it provides baseflow to streams and rivers as well as a supply for irrigation and municipal demands.   During growing seasons, irrigation pumping is a significant stress on the coupled surface-subsurface system.  Numerical ground water flow models are often used to estimate pumping impacts on baseflow and to inform water allocation policies.  Model parameters are estimated based on nonlinear regression techniques that assume model inputs such as pumping rates are known.  Although this is a reasonable assumption for municipal or industrial pumping, irrigation pumping rates are subject to considerable uncertainty and bias since there is normally no reporting requirement for farm operators.

We use hypothetical scenarios and real-world data to demonstrate that failure to account for input uncertainty during the calibration process can lead to biased estimates of the hydrogeological parameters and potentially erroneous forecasts of the impact of irrigation pumping.  As a case study we consider the Republican River Compact Model.  We demonstrate how parcel-level tax assessors’ data and a spatial economic optimization model for irrigation pumping can help to reduce the uncertainty about irrigation pumping rate input.  Integrating economic analysis into hydrogeologic modeling will lead to more reliable and less biased forecasts and thus to more technically defensible policy decisions.