Since the last glaciation, boreal wetlands have accumulated substantial amounts of peat, estimated at 180 to 621 Pg of carbon (C). Over the same period of time these wetlands have emitted methane (CH 4 ), the second most important greenhouse gas (GHG) next to carbon dioxide (CO 2 ). Present day estimates range from 32 to 112 Tg CH 4 per year. With these emissions boreal wetlands significantly affect the global cycling of carbon and the CH 4 concentration in the atmosphere. By storing carbon and

more »

... taking up CO 2 from the atmosphere, boreal wetlands have had a cooling effect on climate in the past. Undisturbed boreal wetlands are likely to continue functioning as a net carbon sink. In the future, particularly with regard to future climate change, the carbon balance could be significantly changed since biogeochemical processes of boreal wetlands are sensitive to biotic and abiotic environmental conditions. Earth system models are state of the art tools to investigate carbon cycle dynamics in past and future climates. However, these global models usually neglect biogeochemical processes of boreal wetlands. In order to investigate how boreal wetlands interact with the carbon cycle and with the climate, I developed a model that accounts for the characteristic biogeochemical processes of boreal wetlands (the peatBALANCE model) and implemented it into the land surface model of the Max Planck Institute Earth System Model (MPI-ESM) and also incorporated a CH 4 emission model into it. This dissertation focuses on the peat accumulation and CH 4 emission of boreal wetlands and it describes which processes and parameters are needed to model the carbon cycle dynamics of boreal wetlands in past, present and future times. I analyzed results from numerous model simulations and compared results against data from measurements and other modelling approaches. My research produced a number of key findings: The peatBALANCE model has accumulated 240 Pg of catotelm peat carbon in areas north of 35° during the last 6000 yr of model simulation. The peat accumulation rate for present day to 2100 is 48.88 Tg C yr -1 . The model simulates CH 4 emissions of 49.3 Tg CH 4 yr -1 for 6000 yr BP and 51.5 Tg CH 4 yr -1 for pre-industrial times. At the present day the CH 4 emissions are in the range of 48 to 58 Tg CH 4 yr -1 while showing an annual variability of 10 Tg CH 4 yr -1 . The model predicts large increases of CH 4 emissions up to 78 Tg CH 4 yr -1 in 2100, using the emissions scenario RCP8.5. The main conclusions drawn from this research were that boreal wetlands experience extensive changes and, as boreal wetlands, are complex adaptive systems: they respond to a changing external forcing by adapting C accumulation and CH 4 emission rates. The net effect, however, is more than a combination of impacts, as these drivers interact in complex ways. The results presented in this thesis highlight the importance of boreal wetlands for simulations of the global carbon cycle. My study adds processes to the land surface model of the MPI-ESM and I recommend accounting for these boreal wetland processes when investigating global carbon cycle dynamics with ESMs in past, present and future climates, particularly in the next generation of ESMs.