Accurate identification of juvenile scombrids is essential for understanding recruitment and spawning dynamics but remains challenging due to morphological similarity and rapid ontogenetic change. We applied an integrative framework combining geometric morphometrics (GM), probabilistic classification, and mitochondrial DNA barcoding to investigate species identification, genetic structure, and spawning activity of bullet tuna ( Auxis rochei ), little tunny ( Euthynnus alletteratus ), and Atlantic bluefin tuna ( Thunnus thynnus ) in the Adriatic Sea. A total of 738 juvenile scombrids collected from fisheries-dependent samples were analysed using landmark-based GM and Gaussian mixture model (GMM) clustering, achieving a 99.7% correct classification rate with full concordance to COI-based molecular identification. Ichthyoplankton surveys conducted during two consecutive summer seasons (2019–2020) detected eggs, larvae, and juveniles of A. rochei and E. alletteratus along the eastern Adriatic coast. The occurrence of early life stages, together with the short pelagic larval duration of both species, supports local spawning within the Adriatic Sea. Overall, these findings indicate that the Adriatic Sea contributes to regional recruitment processes for small tunas. The validated GM–GMM framework provides a scalable and transparent tool for juvenile species identification, with direct applications for fisheries-dependent monitoring, bycatch assessment, and evidence-based management in the Adriatic Sea. • Geometric morphometrics with GMM achieved 99.7% accuracy in juvenile identification • Probabilistic shape classification fully validated by COI barcoding • Evidence of Auxis rochei and Euthynnus alletteratus spawning in the Adriatic Sea • Eggs, larvae and juveniles indicate local recruitment processes • Framework enables scalable fisheries-dependent monitoring
Žuvić et al. (Wed,) studied this question.