Maternal age, fecundity, egg quality, and recruitment: linking stock structure to recruitment and population recovery using an age-structured Ricker model

Background/Question/Methods

Understanding the process of recruitment is fundamental to fish biology and the management of fish populations.  However, recruitment in natural populations is highly variable and rarely well described by classical stock-recruitment relationships (SRRs). Recent analyses suggest that stock structure, and the age composition of the spawning biomass in particular, may play an important role in the mismatch between SRRs and data. Here we develop a generalization of the Ricker SRR that links the age-structure of the spawning biomass and recruitment by allowing for mortality and fecundity rates that depend on maternal age or maternal mass. We provide a flexible SRR with biologically interpretable parameters that can be estimated from existing fisheries time-series and use a Bayesian framework that enables parameters to be informed by experimental data.

Results/Conclusions

We apply our method to fisheries time-series from 10 Atlantic cod (Gadus morhua) populations and show models that include age-structure effects outperform the classical Ricker SRR in 7 of the 10 populations. Consistent with experimental results on cod and other marine fish, we estimate positive age-dependent production in 6 of the 7 populations (i.e. older fish exhibit higher reproductive rates). We then conducted simulations for cod populations to compare the predicted recovery dynamics of populations following severe over-exploitation with and without age-effects.  Surprisingly, in most simulation recovery dynamics of cod populations were similar between the two models.  Only in cases of severe age-truncation - when essentially all old fish were removed from the population - did populations with age-dependent recruitment exhibit noticeably slower rates of population recovery.  Our results suggest age-dependent reproductive biology is an important consideration for fish biology, but its effect on the management and recovery of over-exploited populations are difficult to generalize and strongly contingent on the biological details of a given population.