Design and experiment of soil water sensor monitoring soil moisture at three depths

Abstract: Difficulties exist to measure the soil water content at different depths in plant root zone, such as sensor installing, disturbance to in-situ soil environment, and poor coordination between sensors at different depths. We designed a sensor based on the impedance principle to measure soil water content at three different soil depths. The three-depth sensor can simultaneously measure the soil water contents and minimize in-situ soil disturbance during sensor installing. Experiments using clay, sandy and loam soil samples were conducted to evaluate the effectiveness of the sensor, and the results demonstrated that the sensor can be applied to a variety of soil texture types with sensitivity coefficients greater than 1 mV/(cm3·cm3). A strong linear relationship between the sensor outputs and the soil volumetric water contents was supported by the coefficients of determination (R2) of 0.983, 0.965 and 0.975 for the clay, the sandy and the loam soil, respectively. We also measured soil water contents of five loam soil samples with different soil water contents by oven-dried method and compared them with the sensor outputs. The results were highly consistent, with R2 of 0.973 and root-mean-square error (RMSE) of 0.013. A good linear relationship between sensor outputs and temperature variations which was used as the sensor temperature compensation and this compensation was proved in the temperature calibration experiment. The sensor sensitivity radius was 3-4.1 cm. Therefore, the distance between adjacent pairs of sensor electrodes should not be less than 5 cm. We verified the sensor consistency of the sensor electrodes at three different depths using loam and sandy soil samples. The high consistency for the sensor electrodes at three different depths can be clearly seen in the verification experiments for the loam and sandy soil samples. The maximum absolute relative error for the loam soil sample was less than 2%, and for the sand soil sample was no more than 5%. We further tested the three-depth soil water content sensor by soil water infiltration experiments with loam and sandy soil samples in PVC cylinders (height is 50 cm, diameter is 10 cm). The three-depth soil water content sensor detected the remarkably different rates of water infiltration in the sandy and the loam soil with high fidelity and specificity. In the loam soil sample, it took about 15 minutes, 3 hours and 8 hours for the infiltration water to reach the depth of 5 cm, 25 cm and 45 cm, respectively. As the water infiltration continued, the infiltration rate gradually slowed down. On the other hand, due to good permeability and poor water holding capacity, the whole infiltration process in the sandy soil sample experienced less than half an hour. We finally tested the sensor using a field experiment with rainfall. The dynamic relationship between the sensor outputs at different depths and the amount of precipitation again showed that the three-depth soil water content sensor had excellent reliability and stability. Depending on the different crop root systems and actual needs, the three-depth soil water content sensor can be easily modified into multiple-depth sensors with variable number of sensor electrodes or variable distance between sensor electrodes. Thus, the sensor designed here was highly modular and has a wide potential market.