Automatic identification system (AIS) data has transformed how scientists study global fishing activity - but how accurate are these estimates? A new Editor's Choice article in ICES Journal of Marine Science reveals that in the Northeast Atlantic, Global Fishing Watch's (GFW) AIS-based fishing activity data may significantly overestimate actual fishing effort by pelagic vessels.
AIS and the vessel monitoring system (VMS) are GPS-based tracking systems originally developed for maritime safety and fisheries control. Scientists now commonly use these systems to infer fishing activity, often by analyzing vessel movement patterns, especially speed. Over time, machine learning methods have refined these estimates, and tools like those provided by GFW have become integral to studies on marine protected areas (MPAs), bycatch, and the environmental impact of offshore developments.
However, these systems were not originally designed to indicate when fishing gear is deployed. To test the accuracy of GFW-derived fishing effort estimates, researchers of the latest Editor's Choice article took advantage of a unique dataset: haul-by-haul records collected through self-sampling schemes by two pelagic producer organizations - one based in the Netherlands and the other in the UK.
By comparing the self-reported data with GFW's fishing activity predictions, the study found significant discrepancies. GFW overestimated actual gear deployment duration by 30% to 80%, depending on the metric assessed. These differences affected not only the number and duration of hauls but also the spatial mapping of fishing effort.
Given the growing reliance on AIS-derived products in global fisheries science and policy, the authors urge closer collaboration between data providers like GFW and fishing industry groups with access to high-resolution self-sampled data. Such cooperation could help refine methodologies and improve the accuracy of fishing effort estimates that underpin critical management decisions.
Read the full paper, Bias in Global Fishing Watch AIS data analyses results in overestimate of Northeast Atlantic pelagic fishing impact, in ICES Journal of Marine Science.