Effect of tillage on the hydrology of claypan soils in Kansas

Abstract

The Parsons soil has a sharp increase in clay content from the upper teens in the A horizon to the mid fifties in the Bt horizon. The high clay content continues to the parent material resulting in 1.5 m of dense, slowly permeable subsoil over shale residuum. This project was designed to better understand soil-water management needs of this soil. The main objective was to determine a comprehensive hydrologic balance for the claypan soil. Specific objectives were a) to determine effect of tillage management on select water balance components including water storage and evaporation, b) to quantify relationship between soil water status and crop variables such as emergence and yield, and c) to verify balance findings with predictions from a mechanistic model, specifically HYDRUS 1-D. The study utilized three replicates of an ongoing project in Labette County, Kansas in which till and no-till plots had been maintained in a sorghum [Sorghum bicolor (L.) Moench] – soybean [Glycine max (L.) Merr.] rotation since 1995. Both crops are grown each year in a randomized complete block design. The sorghum plots were equipped with Time Domain Reflectometry (TDR) probes to measure A horizon water content and neutron access tubes for measurement of water throughout the profile. Precipitation, evaporation, and perched water depth were determined at the field scale. Drainage was estimated as negligible after performing hydraulic conductivity measurements on the clayey subsoil. Runoff was determined as the residual in this water balance. Cumulative differences in the hydrologic balances as a result of tillage management were found to be minimal over an entire growing season. However, tillage treatment differences were seen in early season evaporation, surface water content, and the resulting residual runoff values. The chisel-disk treatments had greater evaporation leading to reduced runoff when compared with no-till. There was interaction between tillage treatment and time for surface water content measurements. No effect of tillage treatment was found for whole-profile water content. Crop variables were unaffected by tillage other than the first days emergence, and first days tillering being greater for chisel-disk treatments. No correlation between stored water and crop variables could be found. All aspects of field measurement were well supported by the predictions of the HYDRUS 1-D model.