Fluctuations in river and reservoir water levels due to climate extremes disrupt water supply and pose flood risks. Manual monitoring is often delayed and error-prone, especially in remote areas with limited internet. PDAM Bangodua Indramayu lacks a real-time, internet-independent water level monitoring system, hindering rapid response to droughts or floods. This study designed an automated LoRa-based monitoring system using ultrasonic sensors to provide real-time data and early warnings without relying on internet infrastructure. The system integrates a JSN-SR04T ultrasonic sensor, ESP32/ESP8266 microcontrollers, and LoRa SX1278 for long-range communication. Data is transmitted to Firebase and displayed via an Android app built with MIT App Inventor. Testing evaluated sensor accuracy, data transmission success, and alert functionality. The system achieved 100% data transmission success up to 148 meters (line-of-sight) and 80% at 177 meters (with obstacles). It accurately detected water levels (1.03–2.46 m) and triggered alerts via buzzer and app notifications when thresholds were exceeded. This cost-effective, reliable solution enhances disaster preparedness and supports sustainable water management in regions with limited connectivity, offering a model for similar rural applications.