Landscape demography: Population dynamics across spatial scales
Demography, the enumeration of survival and reproduction and the consequent patterns of changes in the dynamics of populations, is central to both fundamental and applied ecological and evolutionary processes. Demographic processes are at the heart of the decline of endangered populations, population responses to climate change, natural selection, and the expansion of invasions. While most demographic studies are carried out at local scales, there are compelling reasons to understand demographic dynamics of collections of populations at landscapes and larger scales. Attempts to extrapolate local study results to larger scales may yield biased, inaccurate, and incomplete answers. We propose a new theoretical framework – landscape demography – to address such questions as: what is the distribution of population growth rates at different spatial scales, and how does this determine the collective change of these populations at landscape scales? How does this depend on interactions between local populations (e.g., dispersal, competition), and on feedbacks between local populations and environmental factors that drive their fluctuations (e.g., predators, pathogens, fire, or heterogeneous soils)? When must we account for differences in the spatial or temporal scales on which these populations and their environmental drivers fluctuate?
We have developed a theoretical framework adapted from economic portfolio theory for considering the collective behavior of many local populations at a range of spatial scales. Using the means, standard deviations, and covariances of local population growth rates, our approach allows us to quantify not only the collective growth rate across all populations, but also related quantities like landscape-scale extinction risk. Perhaps of greatest interest, our method permits us to ask questions about how the collective growth rates across populations depend on the distribution of the relative sizes and growth rates of the different local populations. We apply this method to nine time series of sets of wild populations of organisms, and estimate not only the projected collective growth rates but also the limits of possible collective growth rates. Our approach yields new insights, provides testable theory, and offers quantitative predictions for a range of important basic and applied questions.