Wild bees in agricultural landscapes: Modelling land use and climate effects across space and time

Abstract: Wild bees are declining in many parts of the world, which poses a risk for crop pollination and ecosystem health. Land use and climate change are two of the main drivers of the decline, and understanding the interactive effects of these two drivers and wild bees is very important to be able to develop proper responses and mitigate future negative effects. In this thesis I have investigated the effects of land use and climate on pollinators and pollination in agricultural landscapes, to increase our understanding of how the variability and interaction between these drivers. To do so, I have used both empirical and ecological modelling approaches, including estimation of effects and prediction on data collected in multiple European regions in multiple years, hypothesis testing on long-term citizen observations of bumblebees, and simulation studies based on a newly developed theoretical model. I demonstrate the limits of transferring statistical models to novel contexts, which is in part due to the quality andtype of data and variables used in these models (paper I). I have also improved methods for studying climate-land use interactions by developing a spatially explicit model of population dynamics that captures how bee populationsrespond to land use and weather at a landscape scale during their active season (paper II, III). The model captures realistic patterns of bee population dynamics. We used the model to investigate the effect of drought-induced reductions in floral resources on population dynamics (paper II), and projected bee population responses to different RCP climate scenarios in multiple types of agricultural landscapes (paper III). Finally, we studied bumblebee queen phenology shifts in Sweden in the last 120 years, and show how some common species keeptrack of global changes while rare and declining species might not (paper IV).Altogether, the thesis contributes to a growing body of knowledge providing evidence of organisms tracking changes in land use and climate, expands the modelling opportunities for exploring interactions between land useand climate change, and provides recommendations for how to improve transferability of results in pollination ecology.