Soil salinity and its distribution determined by soil sampling and electromagnetic techniques

Abstract

Abstract. Diagnosis of soil salinity and its spatial variability is required to establish control measures in irrigated agriculture. This article shows the usefulness of electromagnetic (EM) and soil sampling techniques to map salinity. We analysed the salinity of a 1‐ha plot of surface‐irrigated olive plantation in Aragon, NE Spain, by measuring the electrical conductivity of the saturation extract (ECe) of soil samples taken at 22 points, and by reading the Geonics EM38 sensor at 141 points in the horizontal (EMH) and vertical (EMV) dipole positions. EMH and EMV values had asymmetrical bimodal distributions, with most readings in the non‐saline range and a sharp transition to relatively high readings. Most salinity profiles were uniform (i.e. EMH=EMV), except in areas with high salinity and concurrent shallow water tables, where the profiles were inverted as shown by EMH > EMV, and by ECe being greater in shallow than in deeper layers. The regressions of ECe on EM readings predicted ECe with R2 > 84% for the 0–100 to 0–150 cm soil depths. We then produced salinity contour maps from the 141 ECe values estimated from the electromagnetic readings and the 22 measured values of ECe. Owing to the high soil sampling density, the maps were similar (i.e. mean surface‐weighted ECe values between 3.9 dS m−1 and 4.2 dS m−1), although the electromagnetically estimated ECe improved the mapping of details. Whereas soil sampling is preferred for analysing the vertical distribution of soil salinity, the electromagnetic sensor is ideal for mapping the lateral variability of soil salinity.