NASA scientists have developed a new artificial intelligence tool capable of tracking harmful algal blooms by fusing data from five different satellite datasets. Published in the journal Earth and Space Science, the study demonstrates the system's ability to detect blooms in western Florida and Southern California, addressing a long-standing challenge in ocean monitoring. Coastal algal blooms pose significant risks to public health and local economies, costing the United States tens of millions of dollars annually. In the Gulf of Mexico, the species Karenia brevis causes toxic blooms that kill marine life and sicken humans through airborne toxins, while on the West Coast, Pseudo-nitzschia blooms have recently poisoned marine mammals. Current monitoring relies on manual water sampling, a process that takes days to complete and often lacks the foresight to predict where blooms will start. While satellites provide broad coverage, distinguishing blooms from natural coastal variations like sediment or vegetation has historically been difficult. The new AI system, developed by researchers from NASA's Jet Propulsion Laboratory and Spatial Informatics Group, overcomes these hurdles by integrating data from five space missions, including the Plankton, Aerosol, Cloud, Ocean Ecosystem (PACE) satellite and the TROPOMI sensor. The tool utilizes a self-supervised machine learning approach, allowing it to learn patterns from diverse data streams without requiring pre-labeled training data. The system was trained on satellite imagery from 2018 and 2019, using field measurements to add real-world context. This approach enables the AI to distinguish between deep water and coastlines, recognize specific algal species across different data sources, and filter out environmental noise such as runoff. Initial evaluations show the tool performs accurately in complex coastal environments, correctly identifying and mapping blooms, including specific species like K. brevis. Michelle Gierach, a co-author and scientist at NASA JPL, noted that the tool helps agencies determine precisely where and when to collect water samples as blooms begin. Lead program scientist Nadya Vinogradova Shiffer highlighted that applying self-supervised AI to massive satellite data streams is rapidly becoming a powerful method for generating actionable ocean intelligence. The research team, which also included Kelly Luis and Nick LaHaye, is now working to expand the system with data from additional coastlines and other water bodies like lakes. The ultimate goal is to make the technology accessible to decision-makers in industries ranging from aquaculture to tourism. By bridging advanced technology with user needs, the project aims to enhance coastal safety and economic stability through better predictive capabilities.