Movement of moisture in the unsaturated zone in a loess-mantled area, southwestern Kansas

Abstract

A study of moisture movement associated with four ponding tests in a loess-mantled area near Garden City, Kans., provides significant information on the potential of using the area for artificial recharge by water spreading. Infiltration during the four ponding tests stabilized at rates ranging from 0.7 to 2.2 feet (0.2 to 0.7 meter) per day. The large differences in infiltration rates reflect changes in the hydraulic conductivity of the soil horizons developed in loess materials. The underlying loess has an appreciably greater hydraulic conductivity than the soil. Removing the soil zone should increase infiltration rates, provided that the underlying loess is not severely compacted during excavation or during subsequent recharge operations. When the wetting front of infiltrated water is in the loess, the moisture-buildup pattern shows the characteristic wetting and transmission zones observed for infiltration in homogeneous materials. When the wetting front penetrates the underlying alluvium, the wetting front becomes indistinct. The loess has the capacity to take large quantities of water into temporary storage. If adequate time is allowed between water applications for the loess to drain, the amount may be as much as 1 cubic foot (0.03 cubic meter) of water for each 6 cubic feet (0.17 cubic meter) of the material. At the ponding site, several fine-grained strata are in the unsaturated alluvium underlying the loess. Because these strata have relatively high hydraulic conductivities, they did not act as effective perching beds during the ponding tests. Strata of this type probably would not cause sufficient mounding to impede infiltration significantly or cause water-logged conditions during water spreading. The ground-water mound that developed after application of 21 feet (6 meters) of water has a maximum thickness of 2 feet (0.6 meter) at the edge of the pond. The boundary of the mound moved laterally a distance of 50 feet (15 meters) from the edge of the pond in 2 days. The mound, which dissipated very slowly because of additional drainage from the unsaturated zone, was discernible 3 months after ponding. Although the mound on the saturated zone spread rapidly as a result of pressure transmission, the recharged water actually spread slowly by lateral displacement. Because accumulated salts are leached from the unsaturated zone, the specific conductance of water arriving at the water table is higher than that of the applied water. The amount of increase is dependent on the extent of leaching from previous applications of water and on dilution by infiltrated water. A general appraisal of the ponding-test data indicates that the study area has an excellent potential for artificial recharge by water spreading. Infiltration rates in the loess-mantled area would be favorable for water spreading, and fine-grained strata in the unsaturated zone would not significantly impede downward percolation. Water that is put into underground storage by utilizing the dewatered part of the reservoir would help to sustain the productivity of the aquifer system. INTRODUCTION The extensive use of ground water for irrigation has caused appreciable declines in the water table in several loess-mantled areas in southwestern Kansas because the rate of withdrawal greatly exceeds the rate of replenishment. The life of the alluvial aquifer underlying the loess could be extended by more efficient irrigation practices and by artificial recharge to supplement natural replenishment. Ponding tests in a loess-mantled area have yielded significant information on the potential of these deposits for artificial recharge by water spreading as well as information basic to the development of more efficient irrigation practices. This report describes infiltration rates under various test conditions, buildup of moisture during ponding and the subsequent redistribution after ponding, mounding on perching beds, and effects of ponding on the water table in terms of mounding and water-quality changes. Owing to the homogeneity of the loess materials, the data provide a classic field demonstration of water movement in the unsaturated zone according to theoretical projections. Green, Dabiri, Weinaug, and Prill (1970) have used a computer to simulate moisture changes with depth, as monitored by neutron moisture logs from this study. Excellent agreement between the calculated and the observed data was obtained. Figure 1 shows the study area, 9 mi (14.5 km) northwest of Garden City in Finney County, south­ western Kansas, and the distribution of loess deposits that mantle about 60 percent of the 11-county area that forms southwestern Kansas. Much of the irriga­ tion in southwestern Kansas takes place in the areas mantled by loess. Because soil surveys show the loess deposits of the 11 counties to be similar in physical and chemical properties, the results of this study have application throughout this area. Details of the ponding site and pertinent facilities of the study area are shown in figure 2. The test pond is MOVEMENT OF MOISTURE IN A LOESS-MANTLED AREA, SOUTHWESTERN KANSAS