Search for a command to run...
In-situ soil moisture sensors provide continuous information on soil water status and soil–water–plant interactions. Such information can be used for irrigation planning and for evaluating and optimizing irrigation strategies and systems. Wireless sensors also have the advantage that they cause minimal disruption to field operations and can therefore generate spatially explicit data. Sensor performance and practicality depend on soil properties, the moisture range, and implementation conditions. We evaluated wireless dielectric soil moisture sensors under controlled laboratory conditions and on sprinkler-irrigated field plots. Twenty wireless “SoilScout” sensors were used. A dedicated logger recorded the data and transmitted it via the GSM network to a server for processing. In the laboratory, we used fine sand of known bulk density, saturated it, and then allowed it to dry at room temperature. We determined the gravimetric water content (θg) and converted it to volumetric water content (θv) using the bulk density (ρb). In the field, the 20 sensors were installed and operated in the dams of irrigated potatoes from May 2025 to July 2025 and carrots from July 2025 to November 2025. The soil was Sandy Loam. The sensor positions followed a regular grid within the 18 x 18 m sprinkler setup, and a rain gauge was installed at each point to assess distribution uniformity. Sensor data were recorded continuously, capturing both natural conditions (evapotranspiration and rainfall) and irrigation events. After each measurement period, we collected soil samples near the sensor positions to determine and , and computed . These volumetric water contents served as reference values to analyze sensor performance. We estimated the slope and intercept of the corresponding regression lines and assessed precision and accuracy using RMSE, bias, and R2 . To compare with applied irrigation depths measured by the rain gauges, we also analyzed changes in sensor-derived water content during irrigation events (Δθv). Based on these data, we calculated the uniformity of water distribution. Results show a strong correlation between the wireless sensor and the laboratory reference (R2>0.9), indicating reliable tracking of drying in homogeneous media. In the field, agreement with gravimetric sampling converted to θv was less robust. Although absolute values differed in both settings, the dynamics of soil water status were captured very well. Under the canopy, the wireless sensors produced a spatial pattern like the rain gauge data, enabling sensor-based evaluation of distribution uniformity and a rough estimation of application efficiency (and interception losses). The study demonstrates clear advantages of wireless sensors in managed fields, supporting their use for practical irrigation management. However, retrieving the sensors before harvest proved challenging: despite marking and using a metal detector, they were difficult to locate. Further work is needed to quantify the absolute measurement accuracy of the sensors used. Overall, the results support the use of wireless sensors for planning and evaluating irrigation.