Machine learning is helping track human ocean activity

Satellite data and machine learning have been employed to create a global map of maritime infrastructure, providing a more comprehensive understanding of human activity at sea. The research used deep learning to analyse satellite imagery, detecting both stationary and movable objects such as fishing vessels, wind turbines, and oil rigs. The findings revealed that significant human activity in the oceans goes untracked, and estimates based on incomplete data had previously underestimated the extent of fishing operations. The study demonstrates the potential of satellite mapping to aid in environmental conservation, monitor exclusive economic zones, and detect illegal activities like fishing operations. (Nature)