Is speed through water a better proxy for fishing activities than speed over ground?

Understanding fishing vessel dynamics at a fine spatial scale is of great interest for defining appropriate spatial management plans. Different models have been developed to detect fishing activity from Vessel Monitoring System (VMS) data. While mathematical and statistical methods differ, all rely on the idea that vessel speed over ground provides information on fishing vessel activity. However, trawling with constant speed relative to the water mass may as well prove a winning strategy for both technical (to ensure sufficient trawl opening) and biological (to limit escapement of fish) reasons. Therefore, considering speed through water instead of speed over ground might provide insights into fishing activities. We developed a method combining surface currents derived from ocean circulation models with VMS data to estimate vessel speed through water. We then used vessel speed through water as input to two previously published segmentation methods to infer fishing activity during a fishing trip. We illustrate the approach by analysing trajectories of trawlers operating in the Eastern English Channel. All vessels were equipped with VMS and part of the Obsmer national discard sampling programme. Overall, results showed that surface currents influenced fishing behaviour, and trawling preferentially occured parallel to surface currents. Speed over ground associated with trawling behaviour was much more variable than speed through water, suggesting that trawling occured at constant engine regimes. However, for both segmentation methods using speed through water instead of speed over ground did not improve our capacity to identify trawling sequences. In both cases the amount of time spent trawling during a trip was overestimated