Identifying Transshipment From the Data

When two ships meet to transfer goods, it is called transshipment. In the fisheries industry, it is sometimes legal in ports, but usually illegal out at sea where the practice can’t be monitored. [You can read more about it here]. Transshipment can facilitate the mixing of illegal or unreported catch with legal catch, making it easier for illegal operators to “launder” their product. Read more

Witness to a Crime?

Unusual fleet behavior off West Africa spurs our analyst to investigate

Just because something appears suspicious, doesn’t mean laws are being broken. Jumping too quickly to conclusions could embroil a legally operating fishing fleet in unwarranted investigations or accusations. Equally damaging would be to rush in on a crime without solid evidence and lose the chance to shut down a bad actor. Global Fishing Watch is a powerful tool to be wielded with caution.

That’s why we’re taking our time evaluating a series of intriguing vessel tracks, Read more

Where are the Whalers?

Last week, a visitor to our site asked if Global Fishing Watch can be used to track whaling ships.

The short answer is yes, sometimes. At the moment, our machine learning algorithms are being designed to classify three major types of fishing activity—trawling, longlining and purse seining—but some whaling vessels report themselves as “fishing vessels,” and thus appear on our map. Here’s an example of one we recently found.  Below is a screen shot of the map showing the US Exclusive Economic Zone around Hawaii and Midway Island (EEZ delineated in blue).  We’ve circled a string of points that look like a vessel track headed straight toward, or away from, the border of the marine protected area (delineated in red). Read more

Exploring a New World of Data with Team Member Aaron Roan

If you’ve grown accustomed to the idea that everything in our world is traceable—from airplanes to cell phones to personal browsing history and purchasing habits—it may seem like a no-brainer that we can use publicly broadcast Automatic Identification System (AIS) signals from ships at sea to find out where they are and what they’re doing. But in fact, Read more

Teaching Machines to Tell Us About Fishing

None of what we’re doing at Global Fishing Watch would be possible without the advancements in computing power that have occurred in recent years. The volume of data we work with would have been overwhelming in the past. In one random sample, we observed more than 127,000 vessels over a 24-hour period broadcasting the Automatic Identification System (AIS) signals we use to track fishing behavior. Each signal contains multiple messages (vessel location, call sign, speed, etc.), and some of the signals refreshed as often as every five seconds. That means there are billions of data points to feed into our computer systems. It takes tremendous processing power to handle all that data, and it takes an intelligent machine to make sense of it.
Read more

Characterizing Gaps in the Data

Just a few years ago the very idea of collecting billions of radio signals from ocean-going vessels all around the world and creating a global map of their activity in near-real time would have been unthinkable. But today’s cloud computing technology allows us to do amazing things with huge amounts of data. Read more

When Vessels Report False Locations

Occasionally, the AIS messages transmitted from a ship provide a location that makes no sense, say, in the middle of the Antarctic or over a mountain range. In such cases, either the AIS transponder has malfunctioned, the data got scrambled in transmission, or the system has been tampered with in a deliberate attempt to disguise the vessel’s location. Read more

Clarifying Identity: Matching Broadcasts to Vessel Registries

The satellite-based Automatic Identification System (AIS) is great for locating vessels, but it’s not fully reliable for identifying them. AIS broadcasts coded messages that include information about a vessel’s identity such as its name, ship Read more

How Much Fish Can A Fisherman Fish? (and how we’re trying to find out)

To help researchers better understand how much fish is being taken from the ocean, we’re developing ways to use our data for estimating the total potential catch of the global fishing fleet. It’s a big and a complex question to answer, partly because the source of our information, AIS, is limited. It doesn’t tell us most of what we need to know. “If we cannot get the actual amount of catch from a vessel, the next best thing Read more

Going Dark: When Vessels Turn Off AIS Broadcasts

Something we hear often about tracking vessels with Automatic Identification Signals (AIS) is that transponders can be turned off. Fishers that don’t want to be caught doing something illegal or questionable will simply “go dark.” Read more

Spoofing: One Identity Shared by Multiple Vessels

The satellite-based Automatic Identification System (AIS) that we use for tracking vessels is a radio transmission system in which a ship sends a coded message that can be picked up by satellite and land-based receivers. The code includes Read more

What Does Five Billion Look Like?

Every day we download more than 20 million data points, giving us the positions of the about two hundred thousand vessels in the world. To see where these vessels traveled in 2015, I plotted their positions to create the map you see above. This map shows all vessels carrying AIS, and not just fishing vessels, and is created from about five billion data points. Click on the image to enlarge. Read more