Monitoring near-surface soil water loss with time domain reflectometry and weighing lysimeters

Abstract

Three goals of this research were: 1) to develop a field-scale research facility that could be used for conducting a variety of soil water experiments in both deep (greater than 2 meters) and near-surface soils where the soil water balance could be accurately determined; 2) to develop a transient experimental technique for calibrating time domain reflectometry (TDR) probes; and 3) to study the use of vertically-installed TDR probes for measuring near-surface soil water movement in a field setting, and to compare these measurements with those made by the weighing lysimeter. The weighing lysimeter facility consists of two lysimeter tanks, 4.0 m deep and 2.5 m in diameter, which rest atop a scale with a resolution of ±200 g, equivalent to ±0.04 mm of water on the surface. Data collection is completely automated with a data logger and personal computer. Both lysimeters are instrumented with TDR probes, tensiometers, and pore water solution samplers; thermocouples are installed in one lysimeter for measuring temperature. The TDR probes were calibrated using a transient method known as upward infiltration. The method is rapid, allows the soil to remain unchanged during the experiment, and provides many data points. The upward infiltration method was tested using two different length probes in soils of three textures. Results show that the upward infiltration method is stable, repeatable, and provides accurate dielectric constants and calibration curves. Four, vertically-installed TDR probes of different lengths (200, 400, 600, and 800 mm) were placed in the lysimeter at ground surface to measure water added and water lost during a one-month period in the presence of daily irrigated turfgrass. The purpose of this study was to compare changes in soil water storage as measured by the TDR system, against measurements made using the weighing lysimeter. The TDR probes detected diurnal changes in water content due to irrigation and evapotranspiration, even when these amounts changed slightly from day to day. The TDR probes underestimated the measurements of both water added and water loss, as confirmed using measurements from the weighing lysimeter. The presence of a 47-mm thick biomass above the TDR waveguides retained water that otherwise would have percolated the soil surface into the measurement domain of the probes. Addition and loss of water in the biomass were recorded by the lysimeter, but not by the TDR probes, thus explaining the underestimation. Modeling of near-surface water movement with the HYDRUS model showed very similar water movement behavior as measured by the TDR probes. This confirms our hypothesis that TDR would a useful tool for measuring diurnal changes in water content for irrigation scheduling.