A Comparison of Dynamic-State-Dependent Models of the Trade-Off Between Growth, Damage, and Reproduction
Abstract
Fast growth can be costly, so trade-offs between growth and fitness are to be predicted when organisms adjust their growth to compensate for earlier environmental conditions. We developed four generic models of increasing complexity with different processes to predict the indeterminate growth of vertebrate ectotherms, which is sensitive to ambient temperature even when food is not limiting. We contrast the predictions of the models with observed experimental data on growth trajectories, feeding activity, and reproductive investment of three-spined sticklebacks and inferred patterns of accumulation of biomolecular damage arising from activity and growth. All models predicted observed patterns of compensatory growth (both accelerating and decelerating) in response to earlier temperature perturbations, but the more complex models provided the best fit to experimental data. Growth trajectories influenced future reproductive investment regardless of final body size at breeding. Our findings suggest that while models with fewer parameters can predict basic patterns of growth in stable conditions, they cannot capture the costly long-term effects of deviations from steady growth trajectories. In contrast, models in which foraging activity is assumed to carry costs are capable of predicting the complex patterns of feeding, growth, and reproductive investment seen in animals, with the cost of a heightened mortality risk (e.g., through predation) being more important than the cost of increased physiological damage.