Nonprofit journalism dedicated to creating a Human Age we actually want to live in.

The vast ocean is crisscrossed by 1000s of ships that have been virtually invisible. Until now.


The vast ocean is crisscrossed by 1000s of ships that have been virtually invisible. Until now.

By combining satellite technology with state-of-the-art machine learning, researchers mapped undisclosed industrial activity at sea on a scale never done before.
January 10, 2024

Let the best of Anthropocene come to you.

An ocean conservation group just released a system for tracking the whereabouts of many of the world’s ocean-going ships, an approach that promises to reveal the whereabouts of a navy’s worth of previously unseen fishing vessels.

The database, developed by scientists at several U.S. universities and two nonprofit conservation groups, SkyTruth and Global Fishing Watch, could help identify illegal fishing hotspots and give a more accurate picture of fishing vessels’ locations. That’s vital information at a time when many parts of the ocean are threatened by overfishing.

“Historically, vessel activity has been poorly documented, limiting our understanding of how the world’s largest public resource—the ocean—is being used,” said Fernando Paolo, an engineer at Global Fishing Watch. “By combining space technology with state-of-the-art machine learning, we mapped undisclosed industrial activity at sea on a scale never done before.”

Much of the information we have about the movement of ships at sea comes from transponders known as AIS (automatic identification system), that are required on most industrial-scale ships ranging from oil tankers to fish processors. These systems report the movements and identities of vessels plying the ocean. But the map created by these signals is incomplete. Transponders can be disabled to hide the location of illegal fishing or manipulated to hide a ship’s real location. In some areas, radio or satellite receivers don’t have good coverage, creating neglected blind spots.

To get a more complete view, Paolo and his collaborators tapped into the growing constellation of satellites encircling the Earth. A quartet of European Space Agency satellites supplied them with more than 750,000 radar-generated images and nearly 2.5 million optical pictures of the ocean between 2016 and 2021. A bevy of privately-owned satellites provided 53 million “pings” from ship AIS units.


Recommended Reading:
Eyes on the High Seas


To make sense of all this data—more than 2 million gigabytes worth—the researchers used a type of computer program called a convolutional neural network, which can “learn” to identify objects in images. Using satellite images matched to boats identified by their AIS signals, the scientists trained the computer to learn what characteristics matched a particular kind of boat, such as a fishing vessel. The computer then scanned the millions of images and picked out the different boats that turned up, regardless of whether they had AIS signals.

The analysis revealed that as many as three quarters of the industrial fishing vessels at sea weren’t publicly tracked over the five years, with particularly high concentrations of these shadow fleets in Indonesia, south and southeast Asia, and the northern and western coasts of Africa, the scientists reported earlier this month in the journal Nature.

“Publicly available data wrongly suggests that Asia and Europe have similar amounts of fishing within their borders, but our mapping reveals that Asia dominates—for every 10 fishing vessels we found on the water, seven were in Asia while only one was in Europe,” said Jennifer Raynor, a natural resource economist at the University of Wisconsin, Madison, who was part of the research team. “By revealing dark vessels, we have created the most comprehensive public picture of global industrial fishing available.”

The results can highlight previously unknown fishing pressure. For instance, some of the same scientists had previously used satellite data to identify Chinese ships illegally fishing for squid in waters off North Korea. The new analysis revealed this fishing was far more extensive than previously known. Between 2017 and 2019 that area had the highest concentration of fishing vessels in the world, the peak coinciding with an annual moratorium on fishing in China’s own waters.

The scientists also found that even famous protected areas, including the waters surrounding the Galapagos Islands and Australia’s Great Barrier Reef, aren’t immune to untracked fishing boats.

In a column published in The Conversation, Raynor ticked off a list of ways the new data might be used: informing fishery managers in developing countries that lack the resources to track fishing; targeting illegal fishing spots; and catching illegal trade, such as violations of the United Nations sanctions North Korean seafood.

“Previously, this type of satellite monitoring was only available to those who could pay for it. Now it is freely available to all nations,” said David Kroodsma of Global Fishing Watch. “This study marks the beginning of a new era in ocean management and transparency.”

Paolo et. al. “Satellite mapping reveals extensive industrial activity at sea.” Nature. Jan. 3, 2024.

Image: ©Sea Shepherd

Our work is available free of charge and advertising. We rely on readers like you to keep going. Donate Today

What to Read Next

Anthropocene Magazine Logo

Get the latest sustainability science delivered to your inbox every week


You have successfully signed up

Share This

Share This Article