MEPS 627:125-139 (2019)  -  DOI: https://doi.org/10.3354/meps13061

ABSTRACT: For sustainable fisheries management, fish individuals in a mixing area need to be separated according to their stock affiliation. The assignment of individuals to one of the stocks requires reliable stock discrimination methods with high assignment accuracy. In the Baltic Sea, 2 genetically differentiated Atlantic cod Gadus morhua stocks, the western (WBC) and eastern Baltic cod (EBC), coexist in the Arkona Basin, inducing uncertainties in the stock assessments. Here, we evaluated a suite of non-molecular stock discrimination techniques (otolith shape analysis, stable isotope analysis on otolith nuclei, otolith readability and diameter of translucent zones [TZs]) on the same set of genetically validated Baltic cod baseline samples from the mixing area (Arkona Basin) and adjacent areas (Belt Sea, Øresund and Bornholm Basin). Otolith shape and stable oxygen isotope analyses showed the highest classification accuracies; between 80 and 84% of cod individuals were correctly assigned to their respective stock of origin. Stable carbon isotope analysis, otolith readability and the diameter of the first 2 TZs yielded classification accuracies of only 52 to 61%. Given the high assignment accuracy and the availability of archived otoliths, otolith shape and stable oxygen isotope analyses on otolith nuclei are powerful separation methods that allow for high-throughput quantification of present and past mixing proportions of Baltic cod stocks. This study provides the most comprehensive approach of genetically validated stock discrimination techniques currently available for Baltic cod, and evaluates the applicability and reliability of otolith-based methods for future research studies and for fisheries management purposes.

KEY WORDS: Stock separation · Baltic Sea · Single nucleotide polymorphisms · SNPs · Otolith shape analysis · Fourier analysis · Stable isotope analysis · Translucent zones · Readability