NASA Deploys AI to Detect Harmful Algal Blooms from Space
NASA scientists have developed an artificial intelligence tool that fuses data from multiple satellites to detect harmful algal blooms (HABs). The tool was reported in a study published in AGU Earth and Space Science.
"The aim of this work is to start to bridge technologies to better serve end users and their needs, from aquaculture to tourism." — Kelly Luis, NASA JPL
How the Tool Works
- Self-Supervised Learning: The AI uses a self-supervised machine learning system to learn patterns from various satellite data sources without pre-labeling.
- Training & Validation: It was trained on satellite data from 2018 and 2019 and validated with field and lab measurements.
- Species Identification: The system can identify and map specific toxic algae species, such as Karenia brevis in the Gulf of America and Pseudo-nitzschia off the U.S. West Coast.
- Data Sources: Inputs include NASA's PACE satellite (hyperspectral sensor) and the TROPOMI instrument, among others.
Why This Matters
- Current Limits: On-site testing currently requires manual water sampling and lab analysis, taking days.
- Strategic Guidance: The AI tool aims to help health agencies prioritize where to test as blooms develop, allowing for faster, more targeted responses.
- Future Expansion: Researchers are expanding tests to more coastlines and lakes, with a goal of making the tool accessible to decision-makers in coming years.
Expert Perspectives
"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." — Michelle Gierach, NASA JPL
"Applying self-supervised AI to massive streams of satellite data is rapidly becoming a powerful tool for generating actionable ocean intelligence." — Nadya Vinogradova Shiffer, NASA Headquarters