Constraining the global carbon budget from global to regional scales - The measurement challenge

The Global Carbon Cycle can be modelled by a Bayesian synthesis inversion technique, where measured atmospheric CO2 concentrations and isotopic compositions are analysed by use of an atmospheric transport model and estimates of regional sources and sinks of atmospheric carbon. The uncertainty associated to carbon flux estimates even on a regional scale can be improved considerably using the inversion technique. In this approach, besides the necessary control on the precision of atmospheric transport models and on the constraints for surface fluxes, an important component is the calibration of atmospheric CO2 concentration and isotope measurements. The recent improved situation in respect to data comparability is discussed using results of conducted interlaboratory comparison exercises and larger scale calibration programs are proposed for the future to further improve the comparability of analytical data. (author)