CALYPSO: A Low-Cost Modular System for Water Quality Monitoring and Harmful Algae Bloom Forecasting

Project Calypso addresses critical data gaps in water quality monitoring and Harmful
Algal Bloom (HAB) forecasting by engineering a low-cost, modular, open-source
environmental monitoring ecosystem. The framework utilizes a cost-effective, DIY
approach pairing custom 3D-printed components with a dual-controller hardware
architecture:
• High resolution remote sensing: A Raspberry Pi 5 single-board computer pairs
with a specialized camera module to execute localized multispectral imaging and
real-time machine learning inference.
• Geospatial In Situ Data Logging: A Raspberry Pi Pico microcontroller interfaces
with an in situ water quality sensor suite and a GNSS/GPS module for precise
geospatial telemetry.
• Autonomous Robotic Monitoring Platforms: Custom-built Unmanned Surface
Vehicle (USV) and multirotor Unmanned Aerial Vehicle (UAV) platforms
dynamically deploy these payloads across targeted aquatic environments.
• AI Predictive Data Fusion: Machine learning workflows process both localized
physical sensor telemetry and optical imagery to detect and forecast HABs.
To date, the core hardware and software architectures have been validated through
successful laboratory testing of sensor integration, image acquisition, and initial
predictive machine learning evaluations. The project is currently transitioning into multienvironment field validation across integrated USV, drone, and static buoy
deployments. Ongoing research through September 2026 will expand the sensor array,
optimize edge-connectivity, and train predictive forecasting models on expanded
environmental datasets.