Tracking an endangered right whale and its prey at once — binding separately gathered ocean data together with AI
The National Science Foundation awarded roughly $1.09 million to Rutgers University New Brunswick to develop new AI methods for understanding marine food webs. Focusing on the North Atlantic right whale and the zooplankton it preys on, the work joins oceanographic and species data that have been gathered separately.
Grant overview (primary data)
- Award amount$1,086,074
- RecipientRutgers University New Brunswick (New Jersey)
- ProgramAI and Geosciences
- Period2026-09-01 〜 2029-08-31
- FunderU.S. National Science Foundation (NSF) / NSF
Key points
- New AI methods for understanding marine food webs, focused on the North Atlantic right whale and the zooplankton it preys on ($1,086,074; 2026-09-01 to 2029-08-31).
- The institution is Rutgers University New Brunswick, under the AI and Geosciences program.
- Because currents, weather and nutrients shape the distribution of both, oceanographic and species data of different provenance are joined with deep learning.
- The framework is not confined to these two species and is stated to apply to other predator-prey interactions.
- Among the NSF awards this site holds as of 2026-09-02, AI and Geosciences accounts for 6; the median obligated amount across the records is $1,955,351.
1Watching the predator alone does not answer it
Sometimes tracking an animal on its own will not tell you where it is. The North Atlantic right whale appears where the zooplankton it eats gather, and the distribution of that zooplankton is set by ocean conditions such as currents, weather and nutrients.
Predicting where the whale will be therefore requires handling the observation record of the whale, the physical and chemical state of the ocean, and the distribution of the prey, all at once.
2Data that resists being joined
Ocean observations and species observations differ in who gathers them, at what interval and over what extent. Fitting what a satellite measures broadly and thinly into the same frame as what a ship measures narrowly and densely is not straightforward. That is why the project reaches for deep learning: joining data of different provenance is itself posed as the technical problem.
The method is not confined to these two species and is stated to apply to other predator-prey interactions.
3The geosciences and AI program
AI and Geosciences accounts for 6 of the records this site holds as of 2026-09-02. The same program includes an award on super-resolving satellite data for wildfire detection and prediction, and a common direction is visible: reworking planetary-scale observation with AI. The means of observation are already abundant; what has become the problem is binding them into predictions that mean something.
4Between conservation and the economy
The description notes that the United States depends on healthy oceans for food production, energy, transportation and commerce, and that the work lays a foundation for ocean management balancing conservation with economic use. The North Atlantic right whale is known as an endangered species, with vessel routes and fishing gear posing recurring hazards.
Predicting where it will be with greater precision allows regulation to be applied more finely. It is a domain where understanding an ecosystem translates directly into the precision of the rules. Amounts are the obligated amount as of the check date and may change.
Why it matters
The means of observation are already abundant, and the problem has moved to binding them into predictions. Better prediction of habitat allows vessel routes and fishing rules to be designed more finely, connecting conservation and economic activity directly.
FAQ
Why handle predator and prey together?
Why deep learning?
Sources (primary)
Source: NSF Award Search (U.S. National Science Foundation, public domain). Amounts are the obligated amount. For privacy, we do not handle principal investigator names.
- NSF Award (original, official)
- NSF Award ID: 2615949