Data-driven ocean watch

CEO of Global Fishing Watch, Tony Long, on why open-access data, transparency and strategic partnerships are vital elements in the clampdown on illegal activities at sea.

Illegal fishing can lead to overfishing of fish stocks which impacts on marine biodiversity. The dumping of toxic substances pollutes waters and harms habitats. Both are vastly underreported activities that require monitoring and reporting. And thankfully, modern technology is tracking what’s happening out at sea. But up until now, the organisations and governments that produced the data tended to keep it to themselves.  

In response to the dearth of open-access data, Global Fishing Watch (GFW) was established. Today it tracks, processes and visualises billions of coordinates from tens of thousands of vessels, making use of GPS coordinates, tracking datasets and vessel registries. Artificial intelligence and machine learning also play their part, as do optical and radar satellites. Once the datasets are cleaned, the findings are made publicly available as open-access data, providing a unique resource for researchers monitoring fishing at sea, and for those wishing to expose illegal dumping and other dark activities.  

CEO Tony Long explains, “The basis of Global Fishing Watch's platform is open data. We’re trying to shine a light on human activity at sea through the provision of open and free to access data.” The GFW approach is collaborative, in partnership with organisations, technology companies and NGOs.  

“It is challenging to work with governments and drive change when there are many competing priorities,” says Long, “but we are steadily seeing governments realise the value of open data and adopt greater transparency. So far, 10 countries have committed to sharing their data.” It’s a promising start, and GFW is hoping that many more countries will follow suit. 

Detecting dark vessels

Earlier this year, five developers were named winners of Defense Innovation Unit’s (DIU) Dark Vessel Detection Challenge, which challenged 1,900 participants worldwide to develop machine learning models that identify dark vessels, by analysing synthetic aperture radar data. Jared Dunnmon, Technical Director at DIU says that “instead of asking human beings to look through specific satellite images we think may be important – which may take several hours per image – we can use modern computer vision algorithms to look through every single satellite image we record in a matter of minutes.” These models can help pinpoint illegal fishing activity at speeds not possible for human analysts to perform. 

GFW partnered up with DIU to host this three-month contest. “We've got some very good algorithms and technology,” says Long, “but there's a lot of other people out there that are also good at this work. We wanted to crowdsource the best ideas and mechanisms." 

Outsourcing helps scale-up the global mission for open data, but the true power lies in how the information is utilised.

Aerial drone view of fishing trawlers in the ocean in Asia (Shutterstock)

Exposing invisible illegal activity

Open data is helping to expose all kinds of unruly activities including that of industrial fishing vessels which have chosen not to broadcast their locations. GFW has published studies into such dark fishing fleets, reporting on hundreds of vessels that had originated from China and were violating UN sanctions by illegally fishing in North Korea, and similarly thousands of North Korean vessels that had been fishing illegally in Russian waters.  

Partnering with the Earth Observation Group at Colorado School of Mines, GFW has worked with the team of researchers that draw on data from the NASA Suomi NPP satellite with the help of a highly sensitive infrared imaging suite camera. Being able to identify vessels equipped with bright lights that may be fishing at night has been instrumental in many discoveries.  

Transhipment is also on the GFW radar, and its analysis of the automatic identification system (AIS) that transmits a ship’s location, speed, direction and identity information means it’s possible to track where a vessel is or was last located and help identify vessels that are potentially transhipping at sea. GFW partners with policy-driven organisations that can take this big data and convert it into actionable information, revealing abuses to the system.  

Partnering with news teams is also effective and GFW has collaborated with SkyTruth, which is using open-access data to visualise events such as oil slicks. Honing skills developed from its analysis of the Deepwater Horizon oil spill in the Gulf of Mexico in 2010, Skytruth provides a toolkit for researchers and journalists looking to estimate the impact of any new oil spill. More recently, the firm is deploying AI techniques to detect pollution from illegal bilge dumping, and helping investigative journalists report on the matter. 

Democratisation of data

By sharing data openly, we have a better understanding of what’s happening in the world. International collaborations led by data-driven organisations such as GFW are pushing for increased transparency to clamp down on illegal fishing, and other activities that endanger the oceans, to ultimately enable a fair and sustainable use of the planet. 

Clarissa Wright is a freelance journalist.