Cyanobacteria detection in small, inland water bodies with CyFi

Harmful algal blooms ( HAB s) pose major health risks to human and aquatic life. Remote sensing-based methods exist to automatically detect large, slow-moving HAB s in the ocean, but fall short for smaller, more dynamic blooms in critical inland water bodies like lakes, reservoirs, and rivers. CyFi is an open-source Python package that enables detection of cyanobacteria in inland water bodies using 10-30m Sentinel-2 imagery and a computationally efficient tree-based machine learning model. CyFi enables water quality and public health managers to conduct high level assessments of water bodies of interest and identify regions in which to target monitoring and responsive actions to protect public health. CyFi was developed in three phases. A machine learning competition leveraged the diverse skills and creativity of data science experts to surface promising approaches for cyanobacteria detection from remote sensed data. Subsequent user interviews and model iteration resulted in a deployment-ready open-source package designed to meet user workflow needs and decision-making priorities. This process illustrates a replicable pathway for developing powerful machine learning tools in domain-specific areas.