Development of signal analysis algorithm for apparent soil electrical conductivity sensor

For the spatial characterization of the soil in agricultural fields, the use of sensors is an alternative to collecting soil samples for laboratory analysis. Among such sensors, those that measure apparent soil electrical conductivity (ECa) are distinctive. In ECa sensors based on electrical resistivity, perfect soil-electrode contact is required for a reliable ECa measurement. Thus, this study aimed to develop and evaluate an algorithm for the analysis of the signals used in a soil ECa sensor based on electrical resistivity. Two criteria were selected for analysing the signals read by the sensor. The first criterion verifies whether the read signals have a square wave shape. The second criterion evaluates whether the main frequency of a read signal is similar to that of the applied signal, as set by the user. The algorithm was evaluated using signals obtained from soil ECa measurements conducted under laboratory and field conditions. In the laboratory, the absence of soil-electrode contact was simulated by connecting a resistor to the electrodes. In the field, the ECa measurements were performed in clayey and sandy soils under low and high moisture conditions. The simulation of the absence of soil-electrode contact showed that the algorithm was effective for detecting the unsuitability of the read signals for ECa calculation in eight of the ten scenarios. The results of the field measurements in soil with low moisture content demonstrated that the algorithm could detect signals that were inappropriate for ECa calculation. (C) 2021 Published by Elsevier Ltd on behalf of IAgrE.