Calibration and Evaluation of a Frequency Domain Reflectometry Sensor for Real-Time Soil Moisture Monitoring.

RoTimi Ojo, E., Bullock, P.R., L'Heureux, J., Powers, J., McNairn, H., and Pacheco, A. (2015). "Calibration and Evaluation of a Frequency Domain Reflectometry Sensor for Real-Time Soil Moisture Monitoring.", Vadose Zone Journal, 14(3), pp. 12 pages. doi : 10.2136/vzj2014.08.0114  Access to full text


The accuracy of various calibration procedures of a soil moisture sensor used for real-time soil moisture monitoring was tested. Calibration equations developed using the soil’s cation exchange capacity may provide a suitable alternative to in situ calibration. Soil spatial heterogeneity poses a challenge to accurate soil moisture determination. Remote sensing, in particular, using sensors that acquire data at microwave frequencies, is being used to overcome this challenge. In situ soil moisture monitoring can be used to validate remotely sensed surface soil moisture estimates and as inputs for agronomic and hydrologic models. Nine in situ soil moisture stations were established in Manitoba (Canada) and instrumented with Stevens Hydra Probes. The sensors were installed in triplicate with vertical orientation at the surface and with horizontal orientation at the 5-, 20-, 50-, and 100-cm depths. To ensure accuracy of the measured soil moisture, both laboratory and field calibrations were conducted. These calibrated soil moisture values were compared with the probe default values and those generated using published calibrations. Overall, the results showed that the field calibration was superior (coefficient of determination r2 of 0.95) to the laboratory calibration (r2 of 0.89). In addition, coarse-textured sites generally performed better than the fine-textured, high cation exchange capacity (CEC) sites. At the Kelburn site with high clay and CEC, the use of field calibration reduced the root mean square error from 0.188 to 0.026 m3 m-3. However, at the low clay and CEC Treherne site, gains in accuracy were minimal, about 0.005 m3 m-3. The laboratory calibration consistently underestimated soil moisture at all the evaluation sites, whereas both Topp and Logsdon calibrations overestimated soil moisture.

Date modified: