Introduction The accurate temporal and spatial quantification of sources and sinks of radiatively or chemically active atmospheric trace gases, constitutes a considerable scientific challenge. However, this quantification is needed for two purposes. Firstly, it provides critical data for the evaluation of process-based prognostic models, which are used to predict the evolution of the atmospheric composition as functions of anthropogenic impacts and environmental changes. Secondly, in the context of international negotiations to curb the emissions of greenhouse gases, an accurate quantification is indispensable to verify reduction targets claimed by individual nations or groups of nations. One approach to this problem involves the extrapolation of local flux measurements using geographically referenced databases of properties of the surface (e.g. vegetation cover, topography, soil properties etc.) in conjunction with climatic variables (e.g. temperature, precipitation, insolat