Maturation shifts in a temperate marine fish population cannot be explained by simulated changes in temperature-dependent growth and maturity

To disentangle genetic and environmental influences on phenotypic traits that influence maturation of fish, it would be useful to predict the expected change due to environment alone to compare with observations. This requires a realistically scaled, species-specific life history model of environmentally determined variation in individual growth and maturation. In this study, inter-annual variability in the proportion of mature haddock in the west North Sea was predicted using a stochastic, individual-based simulation model incorporating a temperature-dependent maturation threshold. This species and region are particularly relevant to the debate about the relative importance of genetic and climate change because North Sea haddock have experienced both high fishing mortality and substantial warming in recent decades. Using observed temperatures in combination with temperature-dependent models for growth and maturation, the simulation model predicted year-to-year variation in length and maturity at age expected for cohorts produced from 1979 to 2006. The simulated proportions mature at age 2 were then compared to the observed proportions in an annual bottom trawl survey. Although the model explained much of the high-frequency variation in maturation, the simulated time trend under-represented the rate of increase in the observed trend in proportions mature. This inability of the temperature-dependent life history model to predict the magnitude of change appears consistent with a long-term decline in the maturation threshold. This result provides indirect support for a genetic change in a key life history trait.