Flooding has always been one of the major problems faced in the Philippines. Though flood warning systems exist, they are often costly, sophisticated, and limited. This study proposes SIRENAS: A Smart Inundation Response and Early Notification Arduino-based System, a scalable, cost-efficient prototype that sends early flood warnings for local risk reduction. SIRENAS integrated an Arduino microcontroller, Red Rainwater Level Sensor, and a SIM800L GSM Module to detect rising water levels and send alerts through text messages. Three parameters were tested: Notification Delivery Time in varying distances — 0m, 300m, and 500m; Sensor Accuracy in detecting low, medium, and high-water levels; and Sensor Sensitivity under weak and heavy rain. Results showed that distance affected notification delivery time, where significant differences were seen between 0m and 300m, 0m and 500m, while 300m and 500m did not significantly differ. The sensor accuracy test demonstrated 100% accuracy as the water levels were correctly indicated in all trials. The sensor sensitivity test exhibited no false alarms under weak rain but had an average of 3.7% false alarms under the heavy rain intensity. These findings suggest that SIRENAS is a viable and low-cost option for localized risk reduction; with further scaling and refinements for sensor sensitivity, it may be integrated into localized flood preparedness initiatives in flood-prone areas.