PREDICTING EFFECTS OF CLIMATE CHANGE ON SVALBARD REINDEER POPULATION DYNAMICS: A MECHANISTIC APPROACH

Svalbard represents a hotspot for climate change, with the strongest temperature increase in Europe. These changes may have large impact on wildlife. The main goal in REINCLIM was to understand how weather and climate cause changes in the environment and in turn affect Svalbard reindeer and their population fluctuations. In order to achieve this, the project integrated national and international collaborators across disciplines, unique high-quality and long-term datasets, and state-of-the-art biostatistical modelling tools. In a joint effort from otherwise competing research groups REINCLIM has taken important steps towards an understanding of how climate change affects tundra ecosystems. The inter-disciplinary project group and continuous time-series data made it possible to estimate how climate alters important environmental components driving the dynamics of both the reindeer and other species. Extreme weather events in winter lead to dramatic environmental changes with a multitude of effects across nature and society. Warm spells and heavy rain-on-snow causes ground-ice to encapsulate the vegetation, and many reindeer die from starvation or lose their fetus. An experimental study recently showed that such icing also affects vegetation in itself negatively. Our climate projections suggests that these types of weather events, which have already increased in frequency, will be increasingly common throughout the century due to continued winter warming. On the other hand, summer warming leads to higher food abundance for the reindeer and other herbivores. The amount of green plant biomass follows the annual fluctuations in summer temperature; a two degrees increase predicts 50% increase in biomass. The basis for a mechanistic understanding of climate change effects on the reindeer and their fluctuations in numbers lies in the development of the novel age-structured modelling tools that have given important insights into demographic processes. While the summer warming and increase in plant biomass have positive effects on the reindeer's condition, reproduction, survival and population growth, this is at least partly counteracted by winter warming and icing. Poor food accessibility due to rain and icy pastures reduce the body condition and can lead to starvation or female abortion, and hence low reproduction. In extreme winters this results in a population crash. The relative strength of the effects of summer and winter warming, and hence the net effect of climate change, varies locally, leading to contrasting population trends. Such spatial heterogeneity may buffer climate change effects and maintain the subspecies' viability. At the individual level, icing episodes can also be buffered through adaptive changes in behavior. Some reindeer feed on kelp and seaweed while others climb steep mountains, seek less icy valleys or cross sea-ice in the search for less icy pastures. In that respect, the observed and predicted reduction in sea-ice may increase dispersal barriers and reduce the buffering capacity towards climate change. The inter-disciplinary approach and exchange of data and expertise have resulted in high-impact publications and associated media attention, guiding the way forward in both ecosystem management and research. A paper in Science showed that the drivers of the reindeer population dynamics act across species and trophic levels. In particular, the negative effects of icy pastures on reindeer population growth are shared across the herbivore guild, and the herbivore fluctuations are further reflected in the Arctic fox, which partly depends on reindeer carcasses as food source in late winter. As a result, climate fluctuations generate cross-species population synchrony in all the overwintering vertebrate species on the tundra. This is the first time it has been demonstrated that climate effects can operate across an entire community. Our climate projections indicate that winter rain and icing will become more frequent, and these results provide a significant contribution to our understanding of climate change effects on ecosystems. The REINCLIM project has applied a multitude of dissemination forms in order to reach the different end-users. This includes peer-reviewed journals, chronicles, reports, conferences and talks, student teaching and supervision, as well as national and international mass media. The results are integrated parts of the curriculum at both the BSc, MSc and Phd levels at UNIS, and one PhD and eight MSc projects have been linked to the project. The documented historical changes in the environment and the future projections may provide guidelines and background tools for future research on climate change and their ecosystem effects. In particular, our inter-disciplinary studies have contributed to a current focus within research and management on the changes in winter climate and snow and ice conditions, with potentially great implications for nature and society.

Reindeer and caribou are key environmental indicators that represent great socioeconomic and cultural value across the Arctic. In the light of global warming, the circumpolar decline in population sizes has thus raised major concerns. However, identifying how specific climate drivers may have contributed to this decline and will influence future population trajectories has been challenging. This is partly because proper estimation of population-dynamical parameters requires demographic models that include a mechanistic understanding of climate effects. In addition, our empirical linkage between climate and variation in key environmental variables, such as food availability during critical seasons, has been weak. Based on these challenges, this cross-disci plinary project will apply a mechanistic approach towards a predictive understanding of future population trajectories in wild reindeer on Svalbard, where a relatively simple trophic system offers a unique opportunity to examine the relative importance of climate drivers and net future implications of climate change. We will do this by taking advantage of a novel theoretical framework for age-structured population modelling and by integrating time-series data on climate, vegetation, snow-ice, and reindeer . Specifically, we will (1) estimate key population parameters and the age-specific components of climate effects on reindeer population dynamics using individual-based data; (2) identify the linkages between climate and reindeer food availability; (3) qu antify the relative importance of climate change during different seasons from long-term population time-series in three contrasting areas; and (4) predict population trajectories and time to local extinction under different climate scenarios, and how thi s varies in space. The project will provide tools for deriving a sustainable management of Svalbard reindeer and improve our predictive ability and mechanistic understanding of climate effects in arctic ungulates.