Bust of shark smugglers in Galápagos waters leads to breakthrough in global transshipment data

  • Global Fishing Watch, a publicly available platform launched by the NGOs Oceana and SkyTruth in partnership with Google, adds a new layer to its map today on “encounters” at sea.
  • The new layer gives unprecedented visibility to the practice of transshipment, which is when vessels meet at sea to transfer fish or even people from one to the other. Transshipment is often used to disguise illegal fishing.
  • Global Fishing Watch now also contains a layer that shows clusters of night lights out at sea where they’re not expected.

On Aug. 13 last year, the Ecuadorian navy intercepted a suspicious-looking vessel near the Galápagos Islands. Climbing aboard, authorities found freezers full of thousands of sharks, some of them endangered hammerheads. It was surely the “biggest seizure of sharks in the history of the Galápagos,” a marine ecologist who was involved in the operation said at the time.

The capture was a lucky accident. The vessel happened to have its automatic tracking system, known as AIS and used by most ships, turned on. Ships fishing illegally typically turn it off.

But it quickly became clear that the Chinese-flagged Fu Yuan Yu Leng 999 had not caught the sharks itself. The vessel wasn’t the right size and lacked the necessary equipment to fish at that scale. Those who had done the illegal fishing were still at large.

It’s likely that the refrigerated cargo vessel, or reefer, as they’re known in the field, acquired the sharks from other boats somewhere in international waters, where there’s little oversight.

Reefers sometimes rendezvous with fishing boats far from shore to take on fresh catch, which allows the fishers to stay out at sea for months without returning to port. This practice is called transshipment.

In the Indian Ocean, off the remote Saya de Malha bank, the refrigerated cargo vessel Leelawadee was seen in 2016 with two unidentified likely fishing vessels tied alongside it. Image courtesy of DigitalGlobe © 2017.

Transshipment is not necessarily illegal. Near a port it is common and often necessary, so that large fishing boats can avoid docking. But undocumented encounters of this kind can be indicative of criminal activity such as the exchange of illegal catch or even trafficked workers. Researchers have called for an outright ban of transshipment on the high seas.

International waters comprise 64 percent of the world’s oceans. In the past, keeping track of all the potentially illegal activity in this vast space was unthinkable. But thanks to advances in data collection and analytics, practices like transshipment have become much easier to spot.

Shortly after the bust in Ecuador, the team behind Global Fishing Watch, an online platform launched by Google and the NGOs Oceana and Skytruth, delved into a dataset they had accumulated over the years, hoping to figure out what the Chinese reefer was up to before it was captured.

What they had was a mass of AIS signals, allowing them to trace its journey with astonishing accuracy.

The Global Fishing Watch team inferred that on July 7, the reefer departed from Fuzhou, a port in eastern China. It traveled across the Pacific Ocean, reaching international waters just outside Ecuador’s jurisdiction by early August. Over the course of two days here, it had encounters with four Chinese longliners.

Each of the four fishing boats spent 12 hours drifting alongside the reefer — plenty of time to transfer catch. Global Fishing Watch even identified the longliners’ names.

The team doesn’t know if the revelations have led to any further arrests. But the activity got them excited about the potential of working with AIS data to map transshipment encounters at sea. Today, Global Fishing Watch is making this type of data public on a new layer on its platform. It’s a tool that gives governments and NGOs an unprecedented view of what’s going on in the ocean.

Global Fishing Watch’s new encounters layer reveals for the first time where and when thousands of vessels are involved in close encounters at sea. To detect pairs of vessels meeting at sea, analysts applied machine learning algorithms to more than 30 billion AIS messages from oceangoing boats to find telltale transshipment behavior, such as two vessels alongside each other long enough to transfer catch, crew or supplies.

When Global Fishing Watch launched in 2016, it already contained a layer called “fishing effort,” which shows all vessels worldwide that are likely involved in a type of fishing activity. This data is also derived from AIS signals. Each signal carries a wealth of information, including a timestamp, the vessel name, its radio call sign, its direction and speed, and information about its width, length and tonnage.

Ships that carry an AIS device emit a signal regularly, up to several times a minute. Not every jurisdiction mandates the use of AIS, but it’s very common, said Paul Woods, who works with Global Fishing Watch and SkyTruth. The satellite companies that capture AIS signals are open to sharing them, which makes them a useful public resource.

Different kinds of fishing vessels tend to exhibit different patterns of movement at sea, so it’s possible to deduce the type of ship, and its activities, from AIS signals alone.

In the early days of the development of Global Fishing Watch, thousands of data points were classified manually with the help of fisheries experts and graduate students, a process that established the rules for what constitutes fishing and what doesn’t. Once these rules existed, they could be applied to the ever-growing AIS dataset.

The latest addition to the platform’s map shows instances of “encounters” — when reefers float alongside fishing boats on the high seas for over two hours, an indication of transshipment. The map reveals where this is most prevalent.

According to Woods, it’s the first global public view of transshipment.

Global Fishing Watch had a dataset of 30 billion AIS signals collected over the course of several years. This year, it added more sources, which means the map is now drawing on some 60 million AIS signals a day, Woods says.

The map is updated close to real time, which means keen observers can identify vessels that might be transferring catch at high seas long before they return to port. The data also makes it possible to predict, for example, when vessels might be going after sharks or other protected species, which could lead to more targeted inspections and law enforcement, Woods explains.

However, working with AIS isn’t failproof. Some vessels, especially small ones, don’t carry the device at all, or tamper with it, or simply shut it off when they’re up to something shady.

But even the smallest boats give off a telltale sign that can be captured by today’s Earth observation techniques: light.

Certain marine species, like squid, are typically caught at night when fishers attract them to a boat’s perimeter with a strong light source.

Satellites with infrared sensors can detect these lights. The Global Fishing Watch platform now also contains a layer that shows clusters of night lights out at sea where they’re not expected — meaning unproblematic sources of light such as oil platforms and cargo vessels on busy shipping lanes can be ruled out.

Global Fishing Watch’s new night-light vessel detection layer uses satellite imagery from the U.S. National Oceanic and Atmospheric Administration to reveal the location and activity of brightly lit vessels operating at night.

The night lights data fills in an important gap in the fishing map, because it reveals boats that don’t use any signal-emitting devices like AIS.

Global Forest Watch was able to identify previously unrecorded squid-fishing activity close to Japan this way, in one of the most opaque fishing areas in the world because of the complicated security issues surrounding North Korea and maritime tensions between Japan, China and South Korea.

“One of the interesting findings from the study of the lights at night is that not only can we detect a large number of these vessels, we can also differentiate between the Chinese vessels and the North Korean vessels based on the intensity of the light they emit. This has allowed us for the first time to map the distribution of the North Korean fleet and see how it’s moving over time,” Woods explains.

With resources like Global Fishing Watch’s map, regulators have a dataset to compare their own recoded data to and come up with stronger policies and enforcement. And because it’s public, NGOs can use the same data to pressure regulators to keep their commitments.

Earth observation technologies are advancing rapidly, driven in part by the growing number of tiny satellites known as Cubesats being shot into orbit by companies like Spire. These satellites are constantly gathering visual data and other signals.

For conservation purposes, this has many useful applications. Data by satellite imaging startup Planet, which has a fleet of microsatellites imaging the whole world every day, is helping with a global coral reef monitoring effort.

Spire has helped boost the number of AIS signals available to Global Fishing Watch, and Woods and his team are now also working with Planet, learning to automatically detect and classify boats and their activity based on images. In the next decade, Woods predicts, all vessels and their behavior will be known.

Follow Nadine Freischlad on Twitter: @texastee

Banner: Peru’s regulated fishing fleet produces more than 6 million tons of fish for the world market annually, while the illegal artisanal fleet produce another estimated 1.2 million tons. Photo by Alex Proimos licensed under the Creative Commons Attribution 2.0 Generic license. 

FEEDBACK: Use this form to send a message to the author of this post. If you want to post a public comment, you can do that at the bottom of the page.

This story first appeared on Mongabay

South Africa Today – Environment

This article is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.

You may republish this article, so long as you credit the authors and Mongabay, and do not change the text. Please include a link back to the original article.