Abstract/Summary

Climate change and ecological disturbances are having profound effects on species survival. Predicting how populations will adapt to these changes relies on our understanding of how environment affects species’ life-histories, how this feeds back into abundance, and vice versa. In general, current modelling frameworks over-simplify these relationships, giving spurious predictions. Here, I derive a novel general mathematical modelling framework that links environmentally induced trait variation to population level responses. Applying the framework to classical Nicholson’s blowfly experiments I demonstrate how predictions of population’s responses to environmental change made directly from environment-trait relationships do not always hold. I demonstrate the framework’s accuracy, and how cryptic population dynamics can emerge from mechanisms of environmentally driven trait variation. The framework is then applied to the globally invasive dengue mosquito vector, Aedes albopictus. Developing predictions at a global scale is difficult as environmental variation acts on the vector-pathogen-host triad in complex and non-linear ways. I show how the species population and trait dynamics explain the location, magnitude, and timing of historical dengue outbreaks. By representing the effect of mechanisms of variation on epidemiologically important traits expressed by vectors, I show that the competence of vector populations to transmit disease changes according to their current and historic experience of the environment. Long-lived individuals that developed under favourable environmental conditions can persist within the population long after the environmental conditions that created them have passed and may consequently have a disproportionate effect on pathogen transmission. Importantly, this cannot be accounted for by current modelling approaches that assume all vectors express the same average trait value. This demonstrates that the representation of mechanisms of trait variation is required to produce predictions that account for the underlying complexity inherent in population responses to environmental change, which is required for accurate predictions to inform risk assessment and mitigation strategies.