Field Performance of Five Soil Moisture Instruments in Heavy Clay Soils

Abstract

The increased use of soil moisture retrieval from satellites has heightened the need for improved accuracy of point measurements that are used to validate remotely sensed soil moisture products. A wide range of devices can be installed for operational monitoring of soil moisture; however, many of these devices have not been tested in situ in soils with a very high reactive clay content. The objective of this study was to evaluate the accuracy in field performance of five soil moisture sensors: the EnviroSCAN probe, the Diviner 2000 (both from Sentek Technologies), the Hydra Probe soil sensor (Stevens Water Monitoring Systems), the ThetaProbe ML2x (Delta-T Devices), and the ECH2O EC-5 (Decagon Devices) in soils that had about 71% clay content. The instruments' default calibrations were tested against observed soil moisture from core samples using the thermogravimetric method. New calibration equations were developed for each device, which were evaluated using an independent data set. The ThetaProbe had the lowest root mean square error (RMSE) of 0.025 m3 m−3 and mean bias error (MBE) of 0.002 m3 m−3 in the precalibration analysis. Although the Hydra Probe showed the highest precalibration errors, the instrument made the greatest improvement in post-calibration analysis, with an RMSE of 0.129 m3 m−3 using the default equation reduced to 0.014 m3 m−3 using in situ calibration, and the 0.110 m3 m−3 MBE was reduced to 0 after applying in situ calibration.