A blue fishing boat in water near land.
Credit: Peter Grima, CC BY-SA 2.0

Illegal, unreported, and unregulated (IUU) fishing costs coastal and island states between $26 billion and $50 billion annually. Ships involved in these operations often use clever tactics to hide their identity, making it difficult for regulators to monitor or address the problem.

Using machine learning techniques, researchers found that nearly 20% of high seas fishing is carried out by vessels that are not publicly authorized or regulated. This new approach to monitoring fleets was published in the journal Science Advances.

Ships Hide Their Identity

Large ships are required by the International Maritime Organization to transmit their location for safety purposes using the automatic identification system (AIS). Global Fishing Watch, an international nonprofit dedicated to advancing ocean governance, designs open access tools using these data to track fishing vessels far from shore.

To hide their identity and avoid oversight, some ships involved in IUU fishing change the flag of their home country by registering under another country’s official records. This “reflagging” is legal, but it can allow vessels to comply with the laws of a country with more relaxed regulations than their own. Reflagging can hinder authorities’ ability to hold ships accountable for their activities.

“The act of reflagging is fairly concentrated in a small number of flag states.”

To address this problem, researchers designed a machine learning model to cross-reference around a hundred billion GPS positions with information from more than 40 public vessel registries. This approach allowed them to identify and track 33,000 fishing vessels and 2,000 support vessels (such as refrigeration and refueling ships). The resulting data, the authors said, “can track vessels from shipyard to ship graveyard, reveal potentially unauthorized fishing activity, and reconstruct vessel history to map patterns of reflagging.”

“The act of reflagging is fairly concentrated in a small number of flag states,” said Jennifer Van Osdel, a data scientist with Global Fishing Watch and a coauthor of the study. Of the 116 states that can register ships under their flags, only 20% are responsible for 80% of all reflagging, according to the analysis. The most common is Panama, followed by Russia, Kiribati, Norway, and Liberia.

“Those flag states also correlate with higher rates of foreign ownership, so you see correlations between potentially risky behaviors or behaviors that just make it more difficult to investigate potential IUU [fishing],” explained Van Osdel.

Tania Arosemena Bodero, a fisheries lawyer from the nongovernmental organization MarViva in Panama, explained that countries like hers “with good legislation but no claws and no strength to implement controls end up being associated with IUU fishing.”

Far-Reaching Effects

Arosemena added that the consequences of IUU fishing can be severe and wide-ranging. For example, she highlighted the economic effects of yellow cards and red cards issued by the European Union, legal measures designed to ensure that countries comply with international regulations. Panama, she said, has already been dealt a yellow (warning) card, and a red card would ban Panamanian fisheries products from being sold in the EU.

IUU fishing also poses a threat to ocean ecosystems, as crews may not abide by catch limits, gear restrictions, or area closures. In particular, overfishing and disregard for birds, mammals, and corals inadvertently caught in nets (bycatch) harm the overall health and sustainability of food webs and ecosystems.

“The key for tackling illegal fishing in Southeast Asia is a concrete, shared budget between countries.”

To combat these economic and environmental consequences, Van Osdel and her fellow researchers stressed the need for greater transparency from public registries.

Asmiati Malik, a political economist from Bakrie University in Indonesia, agreed. She pointed out that many vessels in Southeast Asia simply turn off their AIS to stop broadcasting their locations. She believes that “the key for tackling illegal fishing in Southeast Asia is a concrete, shared budget between countries.”

Arosemena added that what’s needed is “a single registry that includes all vessels involved in fishing activities at global, regional, and national levels.”

—Roberto González (@PerRoberto_G), Science Writer

Citation: González, R. (2023), Machine learning helps researchers track illegal fishing, Eos, 104, https://doi.org/10.1029/2023EO230072. Published on 1 March 2023.
Text © 2023. The authors. CC BY-NC-ND 3.0
Except where otherwise noted, images are subject to copyright. Any reuse without express permission from the copyright owner is prohibited.