Hooking the future: Scientists deploy AI to track Bengal’s hilsa, but fishermen fear it will catch more than necessary

In a move that could reshape traditional fishing practices, Indian scientists have introduced an AI-powered prediction system to help fishermen spot hilsa shoals in the Bay of Bengal, the Times of India reported. This is the first effort to promise bigger catches, but it is already raising concerns about sustainability.

Developed by India’s National Marine Information Services Center under the Ministry of Earth Sciences, the Hilsa Fisheries Advisory Service uses machine learning model XGBoost to predict where rare fish are most likely to be found. The system utilizes historical data collected between 2012 and 2016 and has been validated with a more recent dataset from 2021 to 2023. According to INCOIS, the service is already operational.

At the core of the model, it decodes the complex environmental signals that influence hilsa movement, tracking variables such as water temperature, salinity, current speed, and flow direction. “Hilsa does not follow simple patterns. Traditional methods struggle to handle such complexity, but machine learning can recognize such changing relationships much better,” said Sandip Giri, a scientist leading the project.

The output will be converted into a digital map highlighting hilsa-rich zones, providing fishermen with a data-backed guide to improve fishing efficiency. The advisory is designed with safeguards such as compliance with seasonal fishing bans and a five-kilometre coastal buffer zone to prevent overfishing near the coastline.

To fine-tune the model, the researchers deployed GPS-enabled boats in the Hooghly estuary and used government-approved gillnets to record detailed data on catches, fishing periods, and locations. The results showed “encouraging predictive accuracy,” Giri said. The underlying research was published in a journal Fisheries oceanography.

Efforts are also underway to disseminate this technology to the grassroots. Vidyasagar University is conducting training and awareness programs to familiarize small-scale fishermen with the advisory system and associated mobile applications. “The model takes into account biogeochemical parameters such as water salinity, wind direction and speed, and temperature. We are currently testing its accuracy,” said Sourav Maity, a scientist at the university’s Coastal Observatory and Outreach Center and a member of the project team.

But beyond laboratories and pilot projects, awareness remains limited and skepticism is growing. “We don’t know anything about these things,” said Shyam Sundar Das, general secretary of the Digha Fishermen and Fisheries Trade Association, reflecting the information gap at the ground level.

Others are more cautious about the implications. “We are not aware of such a move. However, if it were to occur, we question how sustainable it would be to bypass conventional knowledge and use AI to locate fish. Targeting specific concentrations of hilsa could lead to overfishing, disrupt ecological balance, and ultimately cause severe stock depletion,” warned Debashis Shyamal, general secretary of the Dakshinbhanga Matsyajivi Forum.

However, researchers say support efforts are still ongoing and will be stepped up in the future. “We have already conducted awareness camps for fishermen at Digha and Kakdwip and now, as Hilsa fishing is prohibited from April to June, we plan to organize more such camps, including at Frasergunj and Namkhana,” Maity said.

The Hilsa project sits at a delicate crossroads as India marries artificial intelligence with traditional fishing. While technology promises efficiency, sustainability may ultimately determine a project’s success.