Development of a solar-powered soil moisture monitoring node for smart irrigation systems

Abstract

This project presents the development of a solar-powered, LoRa-based soil moisture monitoring sensor node aimed at improving irrigation scheduling in smart agricultural systems. The system addresses challenges such as unreliable grid power, limited internet connectivity, and short-range communication in traditional IoT-based irrigation technologies. The sensor node integrates an Irrometer Model 200SS Watermark tensiometric sensor, Arduino Uno microcontroller, LoRa SX1276 transceiver, and a solar-powered energy system. Deployed at the MUARIK irrigation demonstration site, it successfully monitored real-time soil water tension and transmitted data to a LoRa gateway without internet dependency. The watermark sensor was calibrated and validated against the gravimetric method, producing two linear equations (R² > 0.96), while validation using a two-tailed t-test gave a p-value of 0.673, confirming correlation between the sensor and gravimetric results. Real-time soil water tension was categorized by a Python-based decision algorithm into wet, moist, and dry. When soil reached the dry threshold (≥60 kPa, 17.521 g/g), a downlink message opened a Strega smart valve for irrigation, which continued until tension dropped to ≤9 kPa (32.581 g/g), automatically closing the valve and preventing over-irrigation. The process was autonomous with a 1-minute update cycle. Field tests showed best performance when the gateway had clear line of sight, with higher RSSI and SNR at the farm shelter compared to CAEC. Overall, the system reduced manual labour, improved water use efficiency, and proved viable for off-grid deployment. Recommendations include placing the gateway in clear line of sight, deploying more nodes for zoned irrigation, and integrating a mobile dashboard for real-time monitoring.