Hybrid IoT-Based Early Warning System for Peatlands Groundwater Monitoring

Peatlands are critical ecosystems but remain highly vulnerable to groundwater depletion and wildfires, posing ecological and socio-economic risks. This study introduces a hybrid IoT -based monitoring system that integrates a human-as-sensor approach to enhance groundwater monitoring. The system combines ultrasonic sensors, humidity sensors, and community reports, with LoRa W Anenabling stable long-distance data transmission in remote areas. To enhance data reliability, the Local Outlier Factor (LOF) algorithm is applied to detect anomalies, automatically flagging inconsistencies in community-reported measurements and ensuring more precise groundwater monitoring. Field tests demonstrated an accuracy rate of 99.64%, with an average measurement deviation of 0.51 cm, confirming the system's reliability. The integration of LOF -based anomaly detection strengthens real-time monitoring, supports data validation, and improves decision- making in peatland management. This study highlights the novel integration of Io%, anomaly detection, and community participation, providing a scalable and adaptive solution for mitigating environmental risks and fire hazards.