A challenging problem in climate modeling is how to deal with interactions and feedbacks across a multiplicity of spatial scales, and how to improve our understanding of the role played by local soil heterogeneities in the climate system. This is of particular interest in northern peatlands, because of the large amount of carbon stored in the soil. Greenhouse gas (GHG) fluxes, such as methane, carbon dioxide and water vapor, vary largely within the environment, as an effect of the small scale processes that characterize the landscape. It is then essential to consider the local heterogeneous behavior of the system components in order to properly estimate water and carbon balances. We propose a novel method to fill the scaling gap from local mechanistic models to large scale mean field approximations. We developed a surface model for peatlands working at the landscape scale, which is able to show the impact of surface microtopography in modeling greenhouse gas fluxes. We tuned our landscape-scale model with data from a peatland site in the Komi Republic of Russia. We simulate surface microtopography and hydrology, and we couple it to a process-based model for methane emissions from the soil (Walter and Heiman, 1996). By partitioning the space in smaller subunits and then analyzing the statistical properties of the tiling, we are able to resolve the small scale processes and investigate their effects at larger scales. We not only investigate the influence of the hummocky surface on GHG emissions, but we are also able to simulate how complex hydrological interactions happening within the system at a subgrid scale affect the landscape-scale land-atmosphere GHG fluxes. We force our model climatology with data from the CMIP5 experiments. Future projections use forcing from the RCP 8.5 scenario, in order to investigate the impact of microrelieves on the future carbon cycle. We also explore potential dynamical feedbacks with the atmospheric water cycle and energy balance by coupling the surface model with an idealized box model of the atmosphere.