A comparison of vertebral preparation techniques for increasing reader precision and agreement in vertebral band pair identification of Northeast Atlantic skates

Abstract

Direct ageing of elasmobranch species comes with many challenges where the success of preparation methods tends to be species-specific. In this study, we compare age estimations from different preparation methods concerning vertebral location (anterior and posterior), vertebral structure (whole and sectioned), and vertebral staining (stained and unstained) for three skate species: Raja brachyura , Raja clavata , and Raja montagui . Age estimations were derived from modal ages of eight age readers from Dutch, Belgian, and Italian institutions. Only vertebrae from the same individual, where both conditions of the preparation method were available, were used in the analysis. Precision measures based on modal ages were variable but consistent with other elasmobranch ageing studies. For all species, anterior vertebrae showed lower CV values on average compared to posterior vertebrae. APE and CV were lower whole vertebrae compared to sectioned vertebrae regardless of vertebral location or staining, for all species. Age bias plots showed age estimations were significantly higher (p < 0.05) in anterior vertebrae compared to posterior vertebrae regardless of vertebral structure or staining for R. brachyura and R. clavata . Age estimations were similar for whole and sectioned anterior vertebrae for R. clavata and R. montagui , and the effect of staining was variable but minimal across all species. Principal component analyses showed the vertebral preparation method had little effect on the variability of the data. The results from this study show the use of anterior, whole, unstained vertebrae were more precise for age classes of 0–9 years, offering significant reductions in both preparation time and cost by eliminating the need for sectioning or staining. With this study we aim to provide some consistency among future ageing studies concerning Raja species which, in turn, improves data quality and management.