MASSACHUSETTS — MIT Sea Grant and the Woodwell Climate Research Center published a study in February 2026 demonstrating a deep learning system to automate fish monitoring in Massachusetts rivers. The open-access paper in Remote Sensing in Ecology and Conservation detailed how computer vision technology can supplement traditional volunteer-based counting methods for river herring populations.

Researchers from the Woodwell Climate Research Center, MIT Sea Grant, MIT CSAIL, MIT Lincoln Laboratory, and Intuit developed the monitoring method using underwater video and computer vision. The paper, titled "From snapshots to continuous estimates: Augmenting citizen science with computer vision for fish monitoring," outlined how object detection, tracking, and species classification technologies can automate fish counting.

The team built a pipeline from in-field underwater cameras to video labeling and model training to achieve automated computer vision-powered fish counting. They collected videos from the Coonamessett River in Falmouth, the Ipswich River in Ipswich, and the Santuit River in Mashpee. For training dataset preparation, the team selected video clips with variations in lighting, water clarity, fish species and density, time of day, and season.

The researchers used an open-source web platform to manually label video frames with bounding boxes to track fish movement. The team labeled 1,435 video clips and annotated 59,850 frames. They compared computer vision counts with human video reviews, stream-side visual counts, and passive integrated transponder tagging data.

Models trained on diverse multi-site and multi-year data produced season-long, high-resolution counts consistent with traditional estimates. The system provided insights into migration behavior, timing, and movement patterns linked to environmental factors. Using video from the 2024 Coonamessett River migration, the system counted 42,510 river herring.

The system revealed that upstream migration peaked at dawn and downstream migration was largely nocturnal. Fish utilized darker, quieter periods to avoid predators. "MIT Sea Grant has been funding work on this topic for some time now, and this excellent work by Zhongqi Chen and colleagues will advance fisheries monitoring capabilities and improve fish population assessments for fisheries managers and conservation groups. It will also provide education and training for students, the public, and citizen science groups in support of the ecologically and culturally important river herring populations along our coasts," said Robert Vincent, researcher at MIT Sea Grant.

River herring migrate from Massachusetts coastal waters to freshwater rivers and streams to spawn each spring, with the annual run beginning in March. Volunteer visual counts are limited to brief daytime sampling windows and miss nighttime fish movement and short migration pulses when hundreds of fish can pass by within a few minutes. Traditional monitoring will continue until fisheries management agencies fully implement automated counting systems to maintain consistency in long-term datasets. This work was funded by MIT Sea Grant with additional support from the Northeast Climate Adaptation Science Center, an MIT Abdul Latif Jameel Water and Food Systems seed grant, the AI and Biodiversity Change Global Center supported by the National Science Foundation and the Natural Sciences and Engineering Research Council of Canada, and the MIT Undergraduate Research Opportunities Program.