Spatiotemporal modelling for the annual egg production method of stock assessment using generalized additive models

Generalized additive models (GAMs) are used to model the spatiotemporal distribution of egg density as a function of locational and environmental variables. The main aim of using GAMs is to improve precision of egg abundance estimates needed for the annual egg production method. The application of GAMs requires a survey design with good coverage in space and time. If the only results available are from less optimal survey designs, they can be improved by using historical data for spawning boundaries. The method is applied to plankton egg survey data of Atlantic mackerel (Scomber scombrus) in 1995. The GAM-based method improves the precision of estimates substantially and is also useful in explaining complex space-time trends using environmental variables.