# NASA researchers have developed an AI system that fuses data from multiple satellites to detect harmful algal blooms in U.S. coastal waters.

*genai · news · 2026-05-27 · Orbital Today*

## Key points

- NASA's SIT-FUSE AI system fuses data from five satellites to detect harmful coastal algal blooms.
- SIT-FUSE uses self-supervised machine learning, reducing dependence on pre-labelled field datasets.
- The system identified species-specific blooms, such as Karenia brevis, in complex U.S. coastal regions.
- Research is underway to expand SIT-FUSE for inland waters and broader decision-maker accessibility.
- Improved remote sensing like SIT-FUSE could reduce U.S. economic losses by up to $234 million annually.

NASA researchers have developed an AI system that fuses data from multiple satellites to detect harmful algal blooms in U.S. coastal waters. SIT-FUSE System The study was published on 18th May in AGU Earth and Space Science. It introduces a system called SIT-FUSE (Segmentation, Instance Tracking, and data Fusion Using multi-SEnsor imagery). The tool identified toxic algal blooms in western Florida and Southern California. It combined observations from five satellite missions, including the NASA PACE satellite and the TROPOMI. Related: Space Force Awards SpaceX $2.29b Contract For Space Data Network Backbone Toxic Algal Blooms Threaten Coasts Harmful algal blooms pose serious risks to public health and coastal economies. They cost the United States tens of millions of dollars each year. In Florida, blooms of Karenia brevis have been linked to wildlife die-offs, beach closures, and human respiratory illness. On the U.S. West Coast, Pseudo-nitzschia blooms have been responsible for poisoning marine mammals, including dolphins and California sea lions. Some algal toxins can even become airborne, triggering respiratory symptoms in humans. Improving How Blooms Are Detected Traditional monitoring methods rely on water sampling and laboratory analysis. They can take more than 24 hours and require prior knowledge of where to test. This delay makes it difficult to track fast-developing blooms. “At the very least, a tool like this can help us know where and when to collect water samples as an algal bloom is starting,” said Michelle Gierach, a scientist at NASA Jet Propulsion Laboratory and co-author of the study. SIT-FUSE addresses this gap using self-supervised machine learning. Instead of relying on pre-labelled datasets, the system learns patterns directly from large volumes of satellite data. It then aligns those patterns with field observations. This allows it to integrate information with different spatial, spectral, and temporal resolutions while reducing dependence on ground-truth labelling. Performance and Future Applications The model was trained on satellite observations from 2018 and 2019, supported by field and laboratory measurements. When tested on later data from the same regions, the system successfully detected and mapped harmful algal blooms. This included species-specific events such as Karenia brevis, even in complex coastal environments. “Applying self-supervised AI to massive streams of satellite data is rapidly becoming a powerful tool for generating actionable ocean intelligence,” said Nadya Vinogradova Shiffer, lead program scientist at NASA Headquarters. By combining data from multiple satellites, the system enables more frequent and detailed tracking of bloom development and severity over time. The research team is now expanding the system with additional coastal datasets and extending its application to inland waters such as lakes. The goal is to make the tool more widely usable for decision-makers in the coming years, from aquaculture operators to tourism managers. “This work aims to start to bridge technologies to better serve end users and their needs, from aquaculture to tourism,” said study author Kelly Luis. Recent estimates suggest that improved remote sensing tools could reduce economic losses from harmful algal blooms by $158–$234 million annually.

**Companies:** NASA
**Countries:** United States

[Read the full story on Orbital Today](https://orbitaltoday.com/2026/05/27/nasa-builds-ai-system-to-map-harmful-algal-blooms-in-near-real-time/)

---

Canonical: https://newsio.io/n/37dc66f3-c023-41ba-a66e-22017823154c/nasa-researchers-have-developed-an-ai-system-that-fuses-data-from-multiple-satel
Summarized by Newsio from Orbital Today. https://newsio.io/how-it-works