SPATIAL VARIATION IN SPECIFIC YIELD IN THE HIGH PLAINS AQUIFER IN KANSAS DERIVED FROM A NEW WATER-BALANCE APPROACH

We have estimated specific yield (S_y) for the zone within which water levels vary for the High Plains aquifer (HPA) of Kansas using a recently developed water-balance method (Butler et al., GRL, 2016). The method is based on the relationship between average annual water-level change and annual water use for an area. We have determined values of S_y for different areal scales, from regions (groundwater management districts [GMDs], thousands of km²), to counties, and to circles around individual monitoring (index) wells (several km²). Average S_y for the three GMDs in the Ogallala region of the HPA in western Kansas (semi-arid climate) range from 0.06 to 0.10 based on 2005-2017 data. Average S_y for each of the two GMDs in the Quaternary region of the HPA in south-central Kansas (sub-humid climate) is 0.03 for the same period. Values of S_y in counties range below and above the average for the GMDs in which they occur depending on the lithology in the zone of water-level change; values for index wells in areas of substantial pumping are generally higher than for the GMD and county in which they are located. For example, the range in S_y for eight counties with substantial water use in GMD4 in the Ogallala region is 0.03–0.15 compared to 0.08 for the entire GMD, and is 0.15 for an index well in GMD4 compared to 0.11 for the county in which it is located. The range in S_y for seven counties with substantial water use in GMD5 in the Quaternary region is 0.022–0.072 compared to 0.029 for the entire GMD, and is 0.069 for an index well in GMD5 compared to 0.041 for the county in which it is located. The higher S_y estimates for the index wells may reflect more permeable zones in those localities, which led to a higher volume of pumping. All the S_y estimates for GMDs are smaller than used in existing numerical models of the Kansas HPA. A main reason is that the models typically use values that represent the lithology in well logs of high-capacity wells, which are biased towards more permeable sediments. In actuality, the average HPA lithology is finer grained and more heterogeneous. In addition, potential entrapment of air in the zone in which water levels vary may decrease the S_y. Selection of S_y values for regional groundwater flow models is generally non-unique. Use of S_y from the new water-balance method provides data-based values that are a good guide for use in models.