How disappearing ships could hold the clue to stopping illegal fishing
On an early January afternoon, a South Korean fishing boat, the Oyang 77, departed from Montevideo, Uruguay, for the plentiful waters off the coast of Argentina. A month later, it would be apprehended by the coast guard for illegal fishing.
As it intruded into Argentina's territorial waters, the boat left a telltale digital signature, one that could be used to catch outlaw fishermen across the planet.
The Oyang 77 headed out to sea before turning southwest to track the Argentine coast. On Jan. 10, 2019, it turned off its location transponder and disappeared.
Over the next 17 days, the vessel disabled its transponder eight more times. Its precise whereabouts during the gaps in its location data are unknown.
Finally, on Feb. 7, the Oyang 77 reappeared and made a beeline for Comodoro Rivadavia, Argentina. It was escorted by the coast guard, under arrest for illegally fishing in Argentine waters.
Commercial fishermen regularly disable their location transponders, sometimes for innocent reasons, but often to hide illegal activity.
Researchers and government officials say the gaps in transponder data can be used to track illegal fishing - and to fight back.
Today's high seas are a bit like the Old West: a wilderness too vast to police, offering riches to those with a tolerance for danger and, sometimes, a dubious commitment to the law.
Why it matters:
- It's bad. In part due to illegal fishing, more than one third of the world's fishery stocks have fallen below biologically sustainable levels, threatening the more than 3 billion people who depend on seafood for animal protein and further imperiling maritime ecosystems already stressed by elevated levels of carbon dioxide.
- It's rampant. Seafood is one of the last animal proteins that people hunt in large quantities. Although farmed seafood has grown to more than half of global consumption, more than one hundred million tons are caught in the wild every year. By one estimate, illegal and unreported fishing accounts for one-fifth of that total.
- It's on your store shelves. As much as 11 percent of U.S. seafood imports came from illegal, unreported and unregulated fishing.
A stark reality confronts anyone who seeks to rein in bad behavior on the ocean: it is very, very big. To pinpoint where in the haystack to search for needles, a team of data scientists and machine learning engineers called Global Fishing Watch collects data from fishing boats' Automatic Identification System (AIS) transponders, whose signals are picked up by satellites and land-based receivers.
Recently, the team at Global Fishing Watch had a novel idea: instead of looking for where fishing boats broadcast their positions, what if they looked for where they hid them? "The AIS data tells us a lot, but the absence of it does as well," Tyler Clavelle, a data scientist at Global Fishing Watch, told me.
Together with scientists from the University of California at Santa Cruz and NOAA Fisheries, Global Fishing Watch analyzed more than 28 billion AIS signals from 2017 to 2019. The researchers identified more than 55,000 gaps in the data and discovered that disabled transponders hide about 6 percent of the globe's commercial fishing activity.
Fishing boats often hide their signals on the edge of Exclusive Economic Zone (EEZ) boundaries, where countries have the right to exploit the resources within 200 nautical miles off their shoreline. That's just what the Oyang 77 did in early 2019 when it vanished and reappeared near the boundary of Argentina's EEZ.
The South Korean-flagged trawler belonged to the fleet operated by the Sajo Oyang corporation, notorious for its record of high seas transgressions, as documented by The Guardian. In recent years, the Oyang 77 had gotten in trouble in New Zealand for illegally dumping dead fish overboard, underreporting catch and failing to pay workers, according to a report from Oceana, a nonprofit focused on ocean conservancy.
In February 2019, the Argentine Coast Guard discovered the trawler with its nets extended inside the EEZ. They found more than 310,000 pounds of seafood on board. Leaving nothing to chance, they deployed a helicopter and an airplane to assist the Coast Guard in escorting the Oyang 77 to shore, releasing it after confiscating its fishing equipment and extracting a fine of 25 million Argentine pesos, or about $550,000.
Sajo Oyang did not respond to multiple requests to comment.
Global Fishing Watch's data did not help catch the Oyang 77, but patrol boats from the U.S. Coast Guard and Canada's Department of Fisheries and Oceans already make use of it to decide which boats to pursue. Countries inspecting fishing boats in port can use Global Fishing Watch's analytical tools to narrow their search.
"You can look up a vessel and see a history of its activity and quickly filter out vessels that appear to be operating above board, or identify vessels that have big gaps in their data or are operating in ways that are suspicious," Clavelle said.
Using ships' AIS data to enforce illegal fishing laws is not as easy as it sounds. AIS is not universally mandated aboard commercial fishing boats. Many fleets, including U.S. fishing boats, are tracked with a separate technology called the Vessel Monitoring System, which is visible to authorities but hidden from other vessels.
AIS was created in the 1990s as a way to keep oil tankers from crashing. In the 2000s, private companies began launching satellites that could capture AIS signals from space, and a new industry emerged to supply government agencies with ships' location data. Yet even law-abiding fisherman sometimes wish to hide their location, either to conceal good fishing spots from competitors or to avoid capture in waters where pirates lurk.
"We thought we might have a pure illegal fishing story," said Heather Welch, a NOAA affiliate and marine biologist at UC Santa Cruz, who led the research with Global Fishing Watch. "And it became very clear that that's not fair to the fishermen, that that's not the story we're seeing here."
The researchers used a method of machine learning to separate the innocent AIS disabling from the nefarious. For instance, behavior that could have looked like statistical noise appeared to the researchers' computer program as "loitering," when boats with disabled transponders were motionless long enough to offload their catch to giant floating refrigerators called reefers.
This is not always illegal, but it can be a way for boats with illegal catch to get rid of the evidence. "It's a way to launder illegally caught seafood into the supply chain," Welch told me. That makes it harder to know whether the fish at the grocery store was caught legally or not. The analysis could help coast guards pinpoint where and when illegal transshipment is likely to take place.
Some in the fishing industry say that the rise of farmed seafood will reduce the opportunities for illegal fishing. The share of seafood from aquaculture grew from 6 percent in 1960 to 58 percent today, according to figures from the U.N. Food and Agriculture Organization.
That share is likely to keep rising, said Gavin Gibbons, vice president of communications at the National Fisheries Institute, an industry lobby. "There's only going to be more farmed going forward. Period. End of story," Gibbons told me. "Farming will have to increase in order to feed a growing planet."
Yet aquaculture has only kept pace with the growth in seafood consumption and has not replaced wild caught seafood. Since the 1990s, wild seafood catch has stayed steady at about 100 million tons per year.
In short, fish farming has not lowered the pressure on wild marine life. Even if aquaculture continues to grow, there will always be demand for wild caught fish, which many people prefer to the farmed variety.
If aquaculture is not the solution to overfishing on the high seas, perhaps technology is. Global Fishing Watch and allies like Oceana, which co-founded the project in 2015, have pushed to require AIS on more commercial fishing boats. Global Fishing Watch's next goal is to learn to detect fishing boats directly from satellite imagery, which would reveal far more activity than AIS signals alone.
"What gets monitored gets managed," Clavelle told me. "So if you can't see what's happening on the ocean, how do you expect to manage it properly?"
Related Content
How Michael Phelps learned to make the right calls
A year after Uvalde, officers who botched response face few consequences
Objection to sexual, LGBTQ content propels spike in book challenges
