Population consequences of behaviourally-mediated tradeoffs between growth and mortality in age-0 rainbow trout (Oncorhynchus mykiss) cohorts.

摘要

Theory shows that behaviourally-mediated tradeoffs between prey growth and mortality rates is a key mechanism that can predict growth and survival at the population-level, and affect population dynamics. In particular, individuals are predicted adjust behaviour to minimize the ratio of mortality rates to growth rates. Yet, the theory is untested at natural spatial and temporal scales. This thesis tests the existence and population-level consequences of behaviourally-mediated tradeoffs between growth and mortality rates in cohorts of age-0 rainbow trout at risk of cannibalism, in replicated whole-lake experiments.; In the presence of adult rainbow trout, age-0 trout use shallow habitats containing few predators and little food. Greater activity and use of risky (but profitable) habitats by individual age-0 trout in lakes with low food abundance increased predation mortality at the population level. Increased foraging effort by age-0 trout compensated for low food abundance resulting in similar growth between low- and high-food lakes during summer. Greater risk-taking by age-0 trout during summer apparently reduces the size-dependent risk of predation as fish allocated little energy to lipids, thus promoting somatic growth. During late summer, when age-0 trout took similar risks in low- and high-food lakes, they allocated significant energy to lipids, suggesting that size-dependent risk to predation was lower and that minimizing overwinter mortality dominated energy allocation. Indeed, small fish with lesser lipid stores, suffered much greater overwinter mortality than larger fish with more lipids, owing to a critical minimum lipid concentration necessary for survival.; I conclude that behaviourally-mediated tradeoffs between growth and mortality rates exist for individual prey at the whole-system scale, and have significant growth and mortality consequences at the population level. I infer that the size-dependent risk of predation, which is independent of behaviour, promotes risk-taking when fish are small, whereas risk of winter starvation promotes lipid storage when individuals are less vulnerable to predation. These experiments represent the first whole-system test of theory predicting population consequences of a behaviourally-mediated tradeoff between growth and mortality rates. Given the widespread nature of this tradeoff in animals, it is likely a key mechanism that should be incorporated into models of population dynamics.