During the Holocene terrestrial wetland ecosystems have been a major sink for atmospheric carbon, which is proved by the development of thick peat layers in Northern peatlands. The actual role of pristine peatland in global carbon balance has not been quantified at this time. In particular the sub-arctic peatlands, as extensively present in Western Siberia, are white spaces in knowledge of carbon exchange with the atmosphere. Therefore it is impossible to predict the effects of climate change through changes in summer temperature, permafrost and hydrology on carbon balance of sub-arctic peatlands. By this research project the annual carbon balance of sub-arctic peatlands of Western Siberia will be estimated in key areas by ground flux measurements at the main mire types (wet peatland ecosystem types) and at lakes and rivulets of peatlands. The point fluxes of mires will be validated with measurements of net primary production and recent carbon accumulation. With geographical information technology (GIS), combined with land unit classification by remote sensing techniques using multi spectral satellite images, area fluxes will be calculated, which in turn will be compared with results of interpretation of hyper spectral satellite images on carbon gas concentrations in the lower atmosphere by new developed information technology. By the application of dynamic GIS (cellular automates) for modelling (2D) gas transport in the lower atmosphere the carbon gas concentrations at image pixel size will be calculated from area fluxes interpolated from key areas. This procedure will enhance the possibility to validate the results obtained by remote sensing technology substantially. Validated image interpretation methods will enable to analyse the variation in time and space of the atmospheric gas concentrations of large sub-arctic areas. In the project the effects of climate change as predicted by IPCC scenarios on the carbon balance of sub-arctic peatlands of Western Siberia will be evaluated.