The global carbon cycle is one of the large biogeochemical cycles spanning all living and non-living compartments of the Earth system. Against the background of accelerating global change, the scientific community is highly interested in analyzing and understanding the dynamics of the global carbon cycle and its complex feedback mechanism with the terrestrial biosphere. The international network FLUXNET was established to serve this aim with measurement towers around the globe.

The overarching objective of this thesis is to exploit the powerful combination of carbon flux measurements and satellite remote sensing in order to develop a simple but robust model for the gross primary production (GPP) of vegetation stands. Measurement data from FLUXNET sites as well as remote sensing data from the NASA sensor MODIS are exploited in a data-based model development approach. The well-established concept of light use efficiency is chosen as modeling framework. As a result, a novel gross primary production model is established to quantify the carbon uptake of forests and grasslands across a broad range of climate zones. Furthermore, an extrapolation scheme is derived, with which the model parameters calibrated at FLUXNET sites can be regionalized to pave the way for spatially continuous model applications.