Tracking the signal in year-class strength of Northeast Arctic cod through multiple survey estimates of egg, larval and juvenile abundance.

Egg and larval abundances for Northeast Arctic cod estimated annually from 1959-1993 were combined with estimates of spawning stock biomass (SSB) and stage abundance estimates for the larvae, pelagic juveniles and demersal juveniles to determine when a correlation with recruitment could first be detected and to track how the correlation changed with increasing age/stage. The largest increase in explanatory power occurred going from SSB to the egg stage suggesting that the transition from potential to realised egg abundance is a critical stage and/or that SSB is a poor index of the potential total egg production. Egg stage abundance was significantly correlated with recruitment;, however, the portion of the recruitment signal explained by this index differed from the portion explained by subsequent stages suggesting that the signal in year-class strength undergoes modification between the egg and larval stages. Contrary to the expectation that the signal in year-class strength should become stronger as the penultimate recruitment stage is approached, the significance of the correlations with recruitment decreased for the larval and juvenile stages relative to the egg stage suggesting that the measurement error for these latter stages is high. Temperatures during the post-settlement period explained a significant portion of the residual variation in stage abundance/recruitment relationships for the larval, pelagic juvenile and demersal juvenile stages. A multivariate recruitment model was therefore developed and tested using independent data for the periods preceding and following the model calibration period. (C) 2003 Elsevier B.V. All rights reserved.