Modelling climate-driven range changes of European butterflies and their host plants

The loss of biodiversity is among the most important and widely discussed effects of anthropogenic climate warming. To escape extinction, species may respond to a changing climate in two ways: they can adapt to the changing conditions in situ or they can shift their ranges to track the climate suitable to them. In this project we focus on species range shifts, which are under the joint control of various processes including the magnitude of climatic change, the growth rate of existing populations, the mobility of individuals and (trophic) interactions with other species within the biological community. These processes have hardly ever been considered simultaneously so far. Thus, our ability to predict individual species ranges and their combined consequences for regional or global biodiversity is still limited. In this project, we focus on one of the most important biotic interactions, the one between host- specific animals and their food plants. More specifically, we propose a new approach by adapting the dynamic plant range model CATS for application to the interacting species groups of European butterflies and their larval host plants. CATS proceeds by annual time steps and starts, each year, from simulating changes to the growth of local populations in response to changing climatic conditions and subsequently simulates the geographical redistribution of their offspring. The proposed adaptation will additionally link these demographic and dispersal processes of butterfly populations to the simultaneously changing geographical distribution of their host plants. The adapted model will be applied to forecast range shifts of c. 30 butterfly species in Europe under different scenarios of 21st century climate change. The results of these simulations will be used to (1) answer theoretical questions on the possible effects of biotic interactions on species migration under climate change; and (2) evaluate whether and to which extent the existing European protected area network Natura 2000 allows the coupled butterfly-host plant systems to track the changing climate.

Anthropogenic climate change is likely to change environmental suitability in different directions across species ranges and beyond. As a consequence, species may go locally or regionally extinct but also colonize new areas beyond their former range boundaries. Resulting range shifts are under the joint control of various processes including the magnitude of climatic change, the availability of suitable habitats, the growth of populations, the mobility of individuals and interactions with other species. Understanding how these processes interact in determining range shifts requires, in the first place, a well-founded knowledge of what drives their current distributions. In this project we evaluated the importance of species and landscape traits for the current ranges of European butterflies. More specifically, we identified various components of the ecological niche (i.e., the climatic, the habitat and the diet niche breadth) as well as large-scale pressures (i.e., airborne nitrogen deposition and climate-warming induced changes in temperature and precipitation) as factors determining a species range and overall range size. Furthermore, we applied the dynamic range model CATS to predict range dynamics of 22 butterfly species in response to changing climatic conditions and the availability of larval host plants in Europe until the end of the 21st century. Range losses (of up to ~90%) mainly in southern and south-eastern regions were only partly offset by gains of areas (of up to ~40%) predominantly at higher latitudes and altitudes. However, most European butterflies were able to keep stable populations until the end of the century for unexpectedly large parts of their range. The projected northward shifts reached up to 45 km per decade but were <10 km for the majority of species. Despite these shifts the proportion of populations covered by the European protected area network Natura 2000 did not significantly change over time. Although every 3rd species showed considerably larger ranges when hosts were assumed to occur everywhere, unrestricted host availability did neither affect the overall changes in range size nor the magnitude of range shifts of European butterflies. Range dynamics of European butterflies are dominated by range losses. Consequently, the effect of larval host availability is low and restricted to species whose hosts are more sensitive to climate-induced changes than the butterfly causing losses of populations even in areas climatically still suitable for the butterfly. Based on a better understanding of the determinants behind the current distribution, these findings contribute to an improved knowledge of the fate of butterflies in Europe in a changing climate until the end of the 21st century.