A team of scientists at Los Alamos National Laboratory is leveraging artificial intelligence to forecast toxic algal blooms, known as harmful algal blooms (HABs), which have become increasingly common due to climate change and rising water temperatures. These blooms, which have now been reported in all 50 U.S. states, pose a significant threat to water bodies by closing beaches and lakes, killing aquatic life, and causing billions in economic damage.
“Harmful algal blooms are appearing in areas where, historically, they were never present,” said Babetta Marrone, the project’s lead scientist. The complexity of the ecosystems causing these blooms, combined with dispersed data across various databases, presents a unique opportunity for AI to aid in understanding and predicting these events.
For decades, researchers have known that HABs are often preceded by elevated water temperatures and nutrient influxes, such as phosphorus and nitrogen from agricultural runoff. However, the exact triggers for toxic cyanobacteria, like Microcystis aeruginosa, to dominate and release toxins remain unclear.
The Los Alamos team plans to use AI and machine learning models to mine large genomic datasets of cyanobacterial HABs, uncovering the intricate relationships between cyanobacteria and other microorganisms in the water. This approach aims to pinpoint the functional interactions that lead to toxin production.
A significant challenge in forecasting HABs is the disparate nature of existing data, collected independently by numerous organizations using different methods. Marrone’s team outlines how AI can harmonize and analyze this diverse data, providing deeper insights into the conditions that foster HABs and aiding in their prediction.
“Our goal is to integrate various data sources, including water sampling, weather telemetry, satellite sensing, and biological data, into a comprehensive model,” Marrone explained. This model could predict future algal blooms and assess the impact of climate change on their occurrence and severity.
This research, published in the journal American Chemical Society ES&T Water, represents a significant advancement in using AI to combat the growing threat of toxic algal blooms, ensuring better management and mitigation of these harmful events.