The role of weather in regulating European bird and butterfly populations

Understanding the causes of temporal variation in abundance is a fundamental question within population ecology, and one with a number of implications for how we think about and manage our natural systems. In particular, there is much interest in the role of climatic variation, or weather, in determining population processes, due both to the observation that these variables can be substantially involved in regulating demographic and population processes, as well as the current context of climate change, which is forecast to become a major determinant of population change over the course of 21 st century. In this thesis, I address the role of weather in determining variation in abundance of European birds and butterflies, using large-scale monitoring datasets from 11 countries across Europe. In chapter II I assess the sensitivity of 12 univoltine butterfly species to climatic variation across their geographic range, finding that weather appears to be relatively more important towards latitudinal range margins. In chapter III I address the extent to which weather variables contribute to observed temporal variation in abundance in 153 species of European bird. I find that, while there is a statistical signal of weather in the population growth rates of the majority of species, explanatory capacities are typically low, with the effect that models that contain large sets of weather variables perform fairly equivalently to those that contain none at all. There are a number of potential causes of this result, which I discuss, as well as make recommendations for further work to discriminate between these. In chapter IV I test whether including information
about short-term thermal variation in population growth rate models enhances our ability to explain variation in abundance for 32 species of butterfly in the UK. I find that, for the majority of species, models of population growth are more able to explain variation in abundance when they include information about fine-scale thermal variation. These results suggest that it may be important to consider thermal variation acting at short timescales to understand the temporal dynamics of populations. Looking across the thesis as a whole, my findings indicate that using monitoring scheme datasets to relate inter-annual variation in abundance to weather is not straight forward; future work with these datasets needs to do more to address the measurement process, as well as working to identify the limitations of models that do not contain population detail such as demographic structure.