Quantifying age-reading error for use in fisheries stock assessments, with application to species in Australia’s southern and eastern scalefish and shark fishery

Abstract

Age-reading error occurs when estimates of age based on reading hard structures differ from the true age of the animal concerned. This error needs to be accounted for when conducting stock assessments. Common methods for quantifying age-reading error include the average percent error, the coefficient of variation, age bias plots, and age difference tables, but these techniques cannot be used to construct age-reading error matrices. A method for constructing age-reading error matrices that accounts for both ageing bias and ageing imprecision is outlined. Simulation evaluation of this method suggests that it is able to estimate both ageing bias (assuming that one reader is unbiased) and ageing imprecision for relatively large sample sizes and for the ages that constitute the bulk of the ages in the sample. However, the performance of the method is poor when sample sizes are small, age-reading error is correlated among readers, when both readers are biased, and for ages that are poorly represented in the sample. The method is applied for illustrative purposes to data on multiple-aged fish in Australia’s southern and eastern scalefish and shark fishery.