A fresh, far more in-depth look at the marine sector is provided by satellite imaging and machine learning, particularly with regard to the quantity and operations of transport and fishing vessels at sea. Policymakers should take note of this fact as it turns out that there are far more of them than publicly available data would imply.
The oceans are everyone’s business since they are a common global resource, yet naturally, no two nations or regions have the same traditions, legal frameworks, or even driving forces.
Although it is not yet widely used, the automated identification system (AIS), which uses shipboard transponders to precisely track activity, is becoming more and more popular worldwide. Because of this, crucial information such as the number of fishing vessels, their operators, and the quantity of fish they capture is frequently erratic and composed of a patchwork of local, private, and officially sanctioned figures.
This not only makes determining policy difficult and imprecise, but it also gives the business a feeling of lawlessness, as innumerable ships secretly visit restricted or protected regions or drastically overfish in order to rapidly deplete supplies.
An alternative viewpoint on this puzzle is provided by satellite imagery: There’s nowhere you can run from a starry eye. However, the industry’s scope and the amount of photography that records it are both enormous. The good news is that machine learning can now handle the millions of vessel recognition and tracking tasks required to precisely follow the tens of thousands of ships at sea at any given time.
In an analysis of two petabytes of orbital imagery from 2017–2021, Fernando Paolo, David Kroodsma, and the Global Fishing Watch team (along with collaborators from several universities) identified millions of seafaring vessels and cross-referenced them with reported and known coordinates for vessels tracked via AIS. The results were published in a paper published in Nature.
According to the survey, about a third of all transport and energy vessels as well as about three-fourths of all industrial fishing vessels are not publicly tracked. The illicit fishing business is enormous, possibly even surpassing the size of the legally recognized sector. (The imaging also recorded increases in the locations of wind turbines and other renewable energy sources, which can be equally challenging to monitor.)
Now, “not publicly tracked” does not imply complete absence of traceability.
Paolo told TechCrunch, “There are a few reasons why these vessels are missing from public tracking systems.” For example, vessels that intentionally disable their transponders or evade detection are just as “untracked” as smaller ones or those operating in regions with minimal or no satellite coverage or AIS infrastructure.
It is noteworthy that certain nations possess other, exclusive methods for monitoring ships in their territorial seas. However, the vessels that these private systems can track are the only ones they can monitor, and this data is not shared with other countries,” he went on.
It is becoming more and more important that information like this is known outside of a country’s borders and internal agencies as the population increases and the oceans warm.
Since fish are a dynamic and valuable resource that migrate around, it is essential to openly track fishing vessels in order to monitor fish supplies. Without every vessel disclosing its location and activities to the public, it is impossible to comprehend and record the entire ecological footprint of those vessels, according to Paolo.
The visualizations show that tracking is strongest in Iceland and the Nordic region and lowest in Southeast Asia, with virtually no tracking off the coasts of Bangladesh, India, and Myanmar.
The fact that their activities are not shared, as is required by law in the Nordic countries, does not imply that they are all unlawful, as was previously mentioned. To what extent do these locations fish? The Asian fishing sector is systematically underrepresented, according to a research that the world only learns about through hearsay.
Based on AIS data, around 36% of fishing activity occurred in European waters, while 44% occurred in Asian waters. However, the satellite data entirely refutes this, demonstrating that a startling 71% of fishing vessels are in Asian waters and only 10% are in European waters. Indeed, it seems that China alone is responsible for about 30% of global fishing!
This is merely intended to highlight the errors in our understanding of the size of the worldwide fishing industry, not to place blame or criticism on particular nations or areas. Furthermore, without solid data, both research and policy will out to be essentially faulty.
However, the satellite research also revealed that fishing boats are frequently present in places that are protected, such as the Galapagos Islands, despite the fact that this is illegal under international law. You can bet a little more care was given to those black vessels.
The next stage is to evaluate these new maps in collaboration with regional authorities. We have probably discovered some fishing in certain instances inside marine protected regions or limited areas, which calls for more research and safeguarding, according to Paolo.
Although the data collection and analysis is still in its early stages, he expects that better data can assist inform policy.
He declared, “This is just the initial iteration of our open data platform.” As fresh radar data is gathered from the Sentinel-1 satellite, we are analyzing it to detect activities all across the world. Our website, globalfishingwatch.org, provides access to this data, which is up to date as of three days ago.
