A Simple, First-Order Approach for Assessing Aquifer Response to Anthropogenic and Climatic Stresses: New Insights into the Future of the High Plains Aquifer in Kansas

The main driver of water-level changes in many regional aquifers such as the High Plains aquifer (HPA) in the United States is the amount of water pumped for irrigation, which is often primarily a function of meteorological conditions. Correlations between climatic indices and annual water-level changes and between reported water use and annual changes may explain most of the observed water-level declines in these regional systems. In that case, these correlations can be used for rapid, first-order assessment of an aquifer's response to future climatic and development stresses. The potential of this general approach is demonstrated for the relatively data rich portion of the HPA in Kansas. The Kansas HPA is overlain by five groundwater management districts (GMDs) that spatially coincide well with four climatic divisions. Each winter, water levels are measured in about 1400 wells distributed approximately evenly over the GMDs. The spatial average of the annual water-level changes for a GMD, the reported water use within the GMD, and a climatic index (Standardized Precipitation Index [SPI]) for the spatially coinciding climatic division are the three time series considered here. The strong linear correlations (coefficients of determination from 0.71-0.78) between the SPI and annual water-level changes indicate that, under average (historic norm) climatic conditions, water levels will decline 0.2-0.6 m/yr across the Kansas HPA. As a result of such declines, there is growing interest in reducing pumping to extend the 'usable lifetime' of the aquifer. The key issue is how much reduction is needed to significantly moderate the declines. Correlations between reported water use and annual water-level changes, in conjunction with correlations between the SPI and annual changes, can be used to show that pumping reductions of 20-30% would likely stabilize water levels, at least in the short term, over much of the Kansas HPA. Although this stabilization may be a product of enhanced recharge produced by past inefficient irrigation practices, and thus only of limited duration, it should help extend the usable lifetime of the resource and serve as a bridge to an economy based on a different mix of agricultural practices. This correlation-based approach is not envisioned as a replacement for process-based modeling. However, the rapid, first-order assessments that can be readily obtained should be of considerable value for those responsible for the management of heavily stressed aquifers.