Abstract
Using low-cost sensors, the Internet of Things (IoT) on a farm has been popular for monitoring environmental conditions such as temperature, humidity, and CO2 concentration to evaluate mass and energy balances in the field. However, net radiation and soil moisture sensors are still relatively expensive; thus, low-cost sensors have been required to spread IoT technologies on the farm. In recent years, logic ICs and thermoelectric coolers (TECs) have become readily available at very reasonable prices. Here, we introduced low-cost net radiation and soil moisture sensors. First, a TEC was used to measure net radiation after calibration with a linear regression line. The measurement error was approximately < ±50 W m−2 after eliminating data with the air temperature < the dew point temperature. Then, the capacitor and resistor (CR) oscillator circuit was employed to determine soil water content, assuming soil water as a variable capacitor. After insulating sensing rods, the CR oscillator’s output frequency, f , was inversely related to volumetric water contentment, θ . A linear regression line between θ and f yielded a measurement error of < ±0.05 m3 m−3 with minimal temperature dependence of dθ /dT = 0.0006 m3 m−3 ◦C−1 for 10–40 ◦C. The data obtained using those sensors would be accurate enough for practical use with IoT.
https://ror.org/02rqvrp93 
