Design and Implementation of an IoT-Based AWD Irrigation Monitoring Prototype for a Rice Field in Minahasa, Indonesia

Abstract

Efficient water management is vital for sustainable rice cultivation, particularly in regions that still rely on conventional irrigation practices. This study presents an Internet of Things (IoT)-based irrigation monitoring prototype implementing the Alternate Wetting and Drying (AWD) method in a rice field plot in Minahasa, Indonesia. The experimental site covers 331.44 m², representing a typical smallholder rice field in the region. The system integrates an ESP32 microcontroller with an A02YYUW ultrasonic sensor for water-level measurement, a DHT20 sensor for temperature and humidity monitoring, and an FC-37 sensor for rainfall detection. Sensor data are processed locally, transmitted via Wi-Fi to the ThingSpeak cloud platform, and displayed on a 0.96-inch OLED module. Powered by a 6 V, 3.8 W solar panel and a rechargeable Li-ion battery, the system operates autonomously with low power consumption. Field testing demonstrated high measurement performance, achieving R² = 0.9994, MAE = 0.18 cm, RMSE = 0.24 cm, and an overall accuracy of 96.78%. The prototype effectively detected AWD thresholds and provided real-time alerts through ThingSpeak, offering a reliable and energy-efficient solution for precision irrigation.