INTRODUCTION

Water Data Monitor is a system developed and presented at FACENS across two UPX project cycles (2024). The project addresses the recurring flood risk near the Sorocaba river by implementing a sensor-based data collection system that monitors water level variations in real time, alerts residents upon detection of critical thresholds, and provides a data foundation for future predictive flood modeling.

IMPLEMENTATION

The system employs a resistive water level sensor mounted at the monitoring site, interfaced with an Arduino board for real-time readings. Collected data — including water level, time of day, and seasonal context — is transmitted to a .NET Core API backed by an SQLite database managed through Entity Framework. A web-based dashboard built in HTML, CSS, and JavaScript renders the stored data in an accessible format for public and administrative use.

RELEVANCE

The Sorocaba river, which runs alongside Dom Aguirre Avenue, is prone to overflow during periods of heavy rainfall — causing significant traffic disruption and increasing road accident risk. A reliable, low-cost monitoring and early-warning system for this type of recurring hazard has direct public safety implications, and the data infrastructure developed in this project lays the groundwork for a fully automated predictive alert network.

RESULTS

The core system infrastructure — sensor integration, database storage, and dashboard visualization — was successfully implemented and demonstrated. The AI-based predictive model was not completed within the project timeline. Future iterations should replace the resistive sensor with a buoyancy-based alternative to accommodate wider measurement ranges, and complete the machine learning pipeline to enable automated overflow probability forecasting.