Life histories and population dynamics

Sammanfattning: During the last decades there has been a growing interest and attention to random variation in population densities and their causes. It has been shown that even in constant environments (deterministic population growth) the dynamics of populations can display random like behaviour depending on parameter values. Thus, chaotic dynamics, defined as bounded fluctuations in the state variables with sensitive dependence on initial conditions, has been suggested as a source of variability in population densities. However, convincing evidence for chaotic dynamics in natural populations is limited. This may be due to difficulties in detecting chaos in the few time series available or that most populations may actually exhibit chaotic fluctuations. In paper I we present a possible explanation for the lack of chaos. We show that natural selection often lead to stable population dynamics, at least in organisms with plastic growth.Apart from the possibility of internally driven fluctuations, most natural populations are also exposed to an external fluctuating environment. Disregarding more or less deterministic variation like diurnal, lunar and seasonal cycles still leave a large amount of environmental variability that can be considered stochastic. The response of a population to such variation can be expected to depend on the kind of life history that the organism exhibits. Life histories are defined by the pattern in the vital rates (i. e., birth, growth, and survival rates). Responses of these rates to environmental variation determine the structure and dynamics of the population. The incorporation of stochasticity in models of population growth result in a lowered long-run growth rate of the populations. An effect of stochastic variation in vital rates can thus be to drive population numbers into a long-term decline and eventually to extinction. A fundamental issue is, therefore, how environmental variation affects the risk of extinction and evolution in organisms with different types of life histories. We investigate this in paper II and III.In addition to the fact that natural populations fluctuate in density, most long-term studies of population dynamics show an incrasing variance in populations densities with time. Noise that leads to an increasing variance with time is termed red noise. In contrast to white noisered noise causes a temporal con-elation in the vital rates it affects (so called autocon-elation). Which type of environmental noise that is used in population modelling may be important for the resulting dynamics. In paper IV we investigate the effects of different types of noise for approximations of long-run growth rate for organisms with different life histories.