Predicting Weekly Fishing Concentration Zones through Deep Learning Integration of Heterogeneous Environmental Spatial Datasets
Chaitanya Rele, Aditya Rathod, Kaustubh Natu, Saurabh Kulkarni, Ajay Koli, Swapnali Makdey
TL;DR
The paper addresses the challenge of locating productive weekly fishing grounds (PFZs) in India's North Indian Ocean by introducing SEGN, a deep learning framework that fuses gradient-based oceanographic fronts with multi-year seasonal features. SEGN combines a three-tier architecture with BiLSTM and multi-head attention to predict weekly persistent PFZs, comparing two variants that include or exclude geographic coordinates. The approach demonstrates improved predictive performance over a coordinate-free baseline and highlights practical benefits such as reduced search time and fuel use, supporting sustainable and efficient fisheries management. The work has potential for informing dynamic management, fleet planning, and climate-resilience strategies, aligning with SDG 14, while outlining avenues for incorporating anthropogenic factors and broader oceanographic parameters in future research.
Abstract
The North Indian Ocean, including the Arabian Sea and the Bay of Bengal, represents a vital source of livelihood for coastal communities, yet fishermen often face uncertainty in locating productive fishing grounds. To address this challenge, we present an AI-assisted framework for predicting Potential Fishing Zones (PFZs) using oceanographic parameters such as sea surface temperature and chlorophyll concentration. The approach is designed to enhance the accuracy of PFZ identification and provide region-specific insights for sustainable fishing practices. Preliminary results indicate that the framework can support fishermen by reducing search time, lowering fuel consumption, and promoting efficient resource utilization.