Artificial intelligence has made it possible to create the most accurate map to date on the human activity in the sea. This development revealed that most industrial fishing vessels globally lack public monitoring. In addition, it determined that a quarter of the world’s transport and energy vessels are not registered. Until now.
The research was carried out by Global Fishing Watch, a non-profit organization supported by Google. The team analyzed 2,000 TB of images of the satellite constellation Sentinel-1 of the European Space Agency, which cover the period 2017-2021. Sentinel-1 uses advanced radar instruments that allow you to “see in the dark” and through clouds. It also used GPS data from vessels.
The researchers developed three machine learning models to create this map of human activity at sea. Thus, with the help of artificial intelligence, they were able to classify each detected vessel, calculate its size and indicate different types of infrastructure.
The result determined that between the 72% and 76% of the world’s industrial fishing vessels they are hidden. Much of that fishing takes place in South Asia, Southeast Asia and Africa, the study explains. And between 21% and 30% of the activity of transport and energy vessels goes undetected by public surveillance systems.
The map of human activity in the sea reveals an entire industrial machinery
“Vessel activity has historically been poorly documented, limiting our understanding of how the world’s largest public resource, the ocean, is used,” said Fernando Paolo, an engineer at Global Fishing Watch and co-author of the study. “We map undisclosed industrial activity at sea on a scale never seen before.”
More than 1 billion people depend on the ocean as their main source of food, highlights the report published this week in Nature. Additionally, about 80% of all traded goods are shipped across the ocean and nearly 30% of the world’s oil is produced in offshore fields.
The new map of human activity at sea created with artificial intelligence makes it possible to identify everything from suspicious fishing operations to an explosion in marine energy development. Wind turbines outnumbered oil structures at the end of 2020, according to the research. In fact, turbines made up 48% of all ocean infrastructure in 2021, while oil structures accounted for 38%.
All this industrial machinery drives the so-called “blue economy”, which moves between 1.5 and 2.5 trillion dollars a year. It is growing faster than the overall global economy, but it is also causing rapid environmental deterioration.
The exposed environmental risk
One third of fish stocks are exploited beyond the limits of biologically sustainable levels. It is estimated that between 30% to 50% of critical marine habitats have been lost due to human industrialization.
The new map of human activity at sea detected numerous fishing vessels hidden within many marine protected areas. “Vessels that do not have public monitoring systems, often called ‘dark fleets’, pose significant challenges to the protection and management of natural resources,” Global Fishing Watch highlighted in a statement.
They also found a high concentration of ships in the waters of countries that previously showed little or no activity. Publicly available data wrongly suggests that Asia and Europe have similar amounts of fishing within their maritime boundaries. However, new mapping reveals that it is actually much larger in Asia.
“For every 10 fishing boats we found in the water, seven were in Asia, while only one was in Europe”, explained Jennifer Raynor, another of the authors of the study. “By revealing the dark ships, we have created the most complete public picture available of global industrial fishing.”
Global Fishing Watch and other organizations have until now relied on the maritime Automatic Identification System (AIS) to see what was happening at sea. The system tracks vessels carrying a box that sends radio signals, but requirements for carrying AIS vary by country and vessel type. Vessels involved in illicit activities often turn off their AIS radars or manipulate the locations they transmit.