North Atlantic Mesoscale Eddy Detection and Marine Species Distribution

Abstract

An eddy detection workflow was developed and applied to reference series of delayed time maps of sea level anomaly (Ref DT-MSLA) published by Aviso Altimetry, France in the North Atlantic region between 30-55° N and 30-80° W. The eddy detection parameters, maximum/minimum Okubo-Weiss parameter to assign to the eddy core/ring, minimum eddy core area, minimum eddy core area to perimeter ratio, and minimum eddy duration, were set to -0.2/0.2 standard deviation, 5 cells (6869 km2), 0.4, and 10 images (70 days), respectively. Using these parameters, 635 anticyclonic eddies and 930 cyclonic eddies were detected between October 14, 1992 and December 31, 2005. The eddy structure of the 62103 pelagic longline fishery catch records in The Logbook System maintained by Southeast Fisheries Science Center was sampled. One-way ANOVA and t-test were conducted to compare the mean catch-per-unit-effort (CPUE) of bluefin tuna (Thunnus thynnus), yellowfin tuna (Thunnus albacores), bigeye tuna (Thunnus obesus), and swordfish (Xiphius gladius) at different eddy structures. For bluefin tuna and bigeye tuna, the mean CPUE is higher in the eddy area than in the non-eddy area. For yellowfin tuna and swordfish, the mean CPUE is higher in the non-eddy area than in the eddy area. For all three species of tuna, the mean CPUE is higher in the anticyclonic eddies than in the cyclonic eddies. For swordfish, the mean CPUE is higher in the cyclonic eddies than in the anticyclonic eddies. The results suggest different eddy structures make different habitats for large marine predators and eddy activities contribute to marine species distribution.