Evaluating a sensor setup with respect to near-surface soil water monitoring and determination of in-situ water retention functions

Abstract

Monitoring water status near the soil surface is a prerequisite for studying hydrological processes at the soil-atmosphere boundary and an option for calibrating remotely sensed water content data, for instance. As the water status of the uppermost soil layer is highly variable in space and time, adequate sensors are required to enable accurate measurements. Therefore, a sensor setup was tested and evaluated in the laboratory and in the field for such a purpose. The arrangement included Hydra Probe and MPS-2 sensors to measure water content and matric potential, respectively.Performance of the MPS-2 was validated in the laboratory by comparing sensor readings with the water potential of a soil, drained to equilibrium for certain pressure steps inside a pressure plate apparatus. Afterwards, six Hydra Probes and twelve MPS-2 sensors were installed in bare soil at a small field plot of about 9m2. The measurements represented soil water status to a depth of 6cm from surface. Core samples were repeatedly excavated around the measurement spots. Their water content was determined and the samples were further utilized to analyze water retention characteristics.The tested setup properly reflected changes of near-surface soil water status due to rainfall and evaporation. However, some shortcomings weakened the potential of the chosen arrangement. Site-specific calibration of the Hydra Probes – implemented by relating sensor readings to the water content values of the core samples – confirmed the applicability of the recommended standard calibration parameters for the respective soil texture. The derived user calibration enabled a measurement accuracy of 0.02cm3·cm−3. Further improvement was restrained by the spatial variability of soil moisture. In this context, spots that were permanently drier or wetter than the others were discovered by means of a temporal stability approach. Performance of MPS-2 sensors was more critical with respect to the objectives. Sensor-to-sensor variation was small at the applied pressure steps of −20, −50, and −100kPa, but the respective averaged readings were −18, −37, and −57kPa. At matric potentials of −200 and −300kPa, the MPS-2 revealed substantial sensor-to-sensor variation. The large deviation of the sensor readings in the field confirmed that the calibration of the MPS-2 should be improved. However, in spite of this inaccuracy, the wide measuring range of the MPS-2 offers suitability to a wide range of potential applications. As an example, water retention functions were calculated from the in-situ data and compared to retention data from the core samples.