The interest in carbon dynamics and the interactions between ecosystems and the atmosphere has increased during the last decade due to the postulated threat of anthropgenically induced global and climate change. Northern peatlands, with their large stores of organic carbon and long-term net accumulation of atmospheric carbon dioxide are key ecosystems in these interactions. Furthermore, peatlands transform organic carbon to methane, which also is an important greenhouse gas.

The findings reported in this thesis and in the accompanying papers are based on both laboratory and field investigations of carbon transformation dynamics on the process scale and at the resolution of individual peatland plant communities. The data from one of the studies also is extrapolated in an attempt to identify environmental controls on regional scales in order to predict the response of northern pcatlands to climate warming.

The laboratory experiments focus on how climate variations, inducing fluctuations in groundwater level and also soil freeze-thaw cycles, influences organic matter mineralisation to carbon dioxide and methane. The field studies investigate year-to-year variations and interdecadal differences in carbon gas exchange at a subarctic peatland, and also how the physiological activities of vascular plants control methane emission rates.

The main conclusions presented include:

Soil freeze-thaw events may be very important for the annual carbon balance in northern peatlands, because they have the potential to increase mineralisation rates and alter biogeochemical degradation pathways.

Vascular plants exert a strong influence on methane flux dynamics during the growing season, both by mediating methane transport and through substrate-based interactions with the soil microbial community. However, there are important species-related factors that govern the nature and extent of this influence.

Caution has to be taken when extrapolating field data to estimate regional carbon exchange because the relevance of the specific environmental parameters that control this exchange varies depending on resolution. On broad spatial and temporal scales the best predictor of peatland methane emissions is mean soil temperature, but also microbial substrate availability (expressed as the organic acid concentration in peat water) is of importance. This temperature sensitivity represents a strong potential feedback mechanism on climate change.