Comparing the Ability of Multiple Soil Sensors to Predict Soil Properties in a Scottish Potato Production System

A soil survey with two soil sensors – an electromagnetic induction (EMI) sensor and a gamma radiometer – was conducted on a farm in south-east Scotland. The collected sensor data were used to direct soil sampling on the farm. The soil samples were then regressed against the sensor output to identify how well the sensor output predicted individual soil properties. The gamma radiometer produced better prediction fits in the topsoil than did the EMI sensor; however, the EMI predicted clay content better in the subsoil. Combining the sensor outputs produced improved fits for the topsoil data but not the subsoil. Neither sensor, individually or combined, produced good fits of soil pH. For potato production systems, topsoil properties are the dominant production determinants; thus a gamma radiometer using current configurations would be the preferred sensor in a single-sensor system assuming that all costs were equal. The economics of single vs. multiple sensor surveys is still unclear.