Despite great advances in carbon cycle research during the past decade the climatic impact of terrestrial ecosystems is still highly uncertain. Although contemporary studies suggest that the terrestrial biosphere has acted as a net sink to atmospheric carbon during the past two decades, the future role of terrestrial carbon pools is most difficult to foresee. When land use change and forestry activities were included into the Kyoto Protocol in 1997, the requirements for scientific precision increased significantly. At the same time the political expectations of carbon sequestration as climate mitigation strategy added uncertainties of a social kind to the study of land-atmosphere carbon exchange that have been difficult to address by conventional scientific methods. In this paper I explore how the failure to take into account the effects of direct human activity in scientific projections of future terrestrial carbon storage has resulted in a simplified appreciation of the risks embedded in a global carbon sequestration scheme. I argue that the social limits to scientific analysis must be addressed in order to accommodate these risks in future climate governance and to enable continued scientific authority in the international climate regime.

In theory, the interaction between the worlds of environmental science and policy may seem straightforward. From a realm outside politics and power, scientists provide relevant knowledge about nature upon which informed policy decisions could be based. However, in reality this linear model tends to be replaced by a much more complex relationship where the distinction between facts and values, knowledge and interests is less clear cut. In this paper, I explore links between science, policy and power through an interview study conducted with Swedish carbon cycle scientists and government negotiators to the UN Framework Convention on Climate Change. Drawing on a co-production model of science–policy interplay this paper addresses the implications of a mutually constitutive relationship between carbon cycle science and climate policy.

International Relations have increasingly projected an image of the world where territoriality has lost its organising force. The global movements of people, information, capital and pollution are seen as signs of increasing deterritorialisation. Climate change is one of these issues ‘beyond borders’ that due to its global framing has been established within the international. This article is an investigation into the political geography of the carbon cycle. We approach the tension between the representations of climate space as global and deterritorial on the one hand, and political practices that reterritorialise the climate on the other. We trace the political transformation of the global carbon cycle into ‘national sinks’ and argue that the two tendencies of deterritorialisation and reterritorialisation of climate space mirror the spatial assumptions of IR; the national inside and global outside.