Estimating age composition for multiple years when there are gaps in the ageing data: the case of western Atlantic bluefin tuna

Abstract

Abstract Age–length key (ALK) methods generally perform well when length samples and age samples are representative of the underlying population. It is unclear how well these methods perform when lengths are representative but age samples are sparse (i.e. age samples are small or missing in many years, and some length groups do not have any age observations). With western Atlantic bluefin tuna, the available age data are sparse and have been, for the most part, collected opportunistically. We evaluated two methods capable of accommodating sparse age data: a novel hybrid ALK (combining forward ALKs and cohort slicing) and the combined forward-inverse ALK. Our goal was to determine if the methods performed better than cohort slicing, which has traditionally been used to obtain catch-at-age for Atlantic bluefin tuna, given the data limitations outlined above. Simulation results indicated that the combined forward-inverse ALK performed much better than the other methods. When applied to western Atlantic bluefin tuna data, the combined forward-inverse ALK approach was able to track cohorts and identified an inconsistency in the ageing of some samples.