Assessing field-scale soil water distribution with electromagnetic induction method

Abstract

Summary We evaluated the sensitivity of electromagnetic induction (EMI) method to variation in soil water in an irrigated cotton field by measuring apparent electrical conductivity (EC a ) with a Geonics EM38. EC a was measured at a fixed position in each plot (within the furrow adjacent to stands of cotton planted on ridges) at ground level as well as at 0.1 and 0.4 m height above the ground. Measurements were made in both vertical and horizontal modes (VM and HM) of EM38 to allow soil water sensing within 1.5 and 0.75 m depths at each location. Surrogate values of soil water were obtained with a locally calibrated neutron probe adjacent to EM38 measurements. All measurements were carried out within three replicate plots of four irrigation treatments at various times during cotton growth. As EC is sensitive to variation in temperature, soil and air temperature was also measured at the time of EM38 measurements. Temporal patterns of variation in EC a and soil water were broadly similar for shallow and deep soil layers. Values of EC a over the season increased nonlinearly with increased values of accumulated soil water within specific depths with high degree of confidence ( P ⩽ 0.001) and high coefficient of determination ( R 2 ) for fitted models. Although both soil and air temperature varied over the season (21.2–39.1 °C for soil and 26.1–33.2 °C for air), dependency of EC a on temperature was weak. The relationship between EC a and soil water was greatly affected by surface configuration in crop fields (i.e. whether the crop was planted on a flat bed or a raised bed) and season. Thus, there is a need to calibrate the EMI equipment to suit local condition in order to measure soil water distribution at field scale.