To support efficient fisheries monitoring, control and surveillance, Starboard maintains the IUU listed, Previously IUU listed, and Closely linked to IUU listed risk indicators. We assign these indicators to vessels that have engaged in illegal, unreported, or unregulated (IUU) fishing, as well as vessels that are closely linked to those through encounters.
🔴 IUU listed
Vessels that are currently on RFMO IUU vessels lists
🟠 Previously IUU listed
Vessels that were historically on RFMO IUU vessels lists
🟡 Closely linked to IUU listed vessels
Vessels with strong direct or indirect links to IUU listed or previously listed vessels
Clicking on the risk indicator assigned to a vessel will show further details about why a particular vessel has this risk assigned.
What are IUU listed vessels?
IUU listed vessels are those that appear or have historically appeared in IUU vessels lists published by Regional Fisheries Management Organisations (RFMOs) and similar organisations. These lists are collated in the Tryg Matt Tracking (TMT) Combined IUU Vessel List, which we use to define whether vessels are IUU listed or previously IUU listed.
We match vessels from the TMT Combined IUU Vessel List to the Starboard database in the same way we match vessels from fishing registries using a combination of IMO numbers, callsign, and vessel names. Unmatched vessels can occur if the vessel is not in the Starboard database or if AIS reports of vessels contain incorrect or insufficient information. This could result in vessels which are in the IUU Vessel List but do not have the IUU listed or Previously IUU listed risk indicator in Starboard.
We import and update the TMT IUU Vessel List regularly. Nevertheless, there might be a delay between the time when a vessel is listed or delisted by an individual RFMO and when that change is reflected in the TMT IUU Vessel List.
What are closely linked to IUU listed vessels?
Closely linked to IUU listed vessels are those that, based on their on-water relationships, have a particularly strong relationship with IUU listed vessels. To find those closely linked, we use a network model that incorporates millions of vessel encounters to reveal, sometimes hidden, associations to IUU listed vessels. Roughly speaking, closely linked vessels are those with the shortest network distance to IUU listed vessels.
With the exception of buoys, we assign the risk indicator to all types of closely linked vessels. This includes other fishing vessels, tankers, and carriers that may regularly service IUU listed vessels, as well as vessels of unknown type.
Closely linked vessels may be noteworthy because of their disproportionate connection to IUU listed vessels. This does not necessarily mean that they are involved in IUU fishing themselves, but rather signals that further investigation may be helpful. Although closely linked vessels may have a higher risk of infringement in certain cases, the network model does not evaluate risk. Instead, the model highlights on-water relationships that otherwise may not have been considered or require substantial manual analysis.
Using a network model to find closely linked vessels
Our network model is a mathematical representation of vessels and their relationships. This model allows us to quantify the strength of association between vessels even in complex scenarios with thousands of encounters or when 2 vessels might be indirectly associated, for example because they encounter the same reefers, bunkering vessels, or even fishing buoys.
We determine closely linked vessels based on the overall network distance to IUU listed vessels. Intuitively, the network distance between vessels depends on two factors:
The recency of encounters–All else being equal, 2 vessels that encountered recently will be closer to each other than 2 vessels that encountered a long time ago. This is because recent encounters are more relevant than older encounters. Mathematically, we assume that the relevance of encounters decays exponentially; a recent encounter is about twice as relevant than one that occurred 6 months ago and about 4 times as much as one from a year ago. We use encounters from the last 2 years in the network model.
The frequency of encounters–All else being equal, 2 vessels that encountered repeatedly will be closer to each other than 2 vessels that encountered only once.
Because associations with currently listed IUU vessels are more relevant than associations with previously listed IUU vessels, we adjust the network distances such that the contribution of previously listed vessels is inversely proportional to the time passed since they were delisted.