Ageing bias and precision for deep-water snappers: evaluating nascent otolith preparation methods using novel multivariate comparisons among readers and growth parameter estimates

Tropical deep-water snappers (Etelinae) support valuable fisheries across the Indo-Pacific, with stock assessments reliant on age-based information in the absence of reliable catch and effort statistics. These long-lived species have been considered notoriously difficult to age. However, nascent developments in ageing protocols, particularly thinner transverse sections of otoliths (similar to 180-200 mu m), are providing improvements in growth zone clarity, interpretation and repeatability of annuli counts. At a recent international workshop, thin sectioned otoliths from three deep-water snappers were read under reflected light by eight fisheries scientists from across the Indo-Pacific, with various levels of fish-ageing experience. Precision and bias were assessed using traditional ageing precision metrics (index of average percent error, IAPE; and coefficient of variation, CV), and a novel approach using multivariate analyses (metric multidimensional scaling, mMDS) based on Euclidean dissimilarity among readers' counts and subsequent von Bertalanffy (vB) growth parameter estimates. Annuli counts between the primary reader and all other readers were within two for 80% of estimates, with uniform variation across a wide age range for Etelis carbunculus (intraclass correlation coefficient (ICC) = 0.924, n = 20, 3-25 annuli) and Etelis sp. (ICC = 0.933, n = 15, 2-27 annuli). In contrast, annuli counts for Pristipomoides filamentosus (n = 14, 4-49) were less precise (i.e. ICC = 0.835, 66% of counts within two of primary reader) with a bias toward greater variation in younger, pre-maturational life stages (<= 5 annuli). Traditionally accepted ageing precision (IAPE <= 5.5%, CV <= 7.6%) was achieved for each species, but was commensurate with reader experience. The multivariate mMDS ordination was more informative in identifying both distance (i.e. dissimilarity) and direction (i.e. form) of variations in annuli counts and vB growth parameter estimates among readers. The acceptable level of ageing precision and bias achieved among most readers indicated that deep-water snappers can be aged precisely when appropriate otolith preparation methods are used. This study contributes towards ageing protocols that can be used