Summertime Siberian CO2 simulations with the regional transport model MATCH : a feasibility study of carbon uptake calculations from EUROSIB data

Biogenic surface fluxes Of CO2 over Europe and Siberia are implemented in the regional tracer transport model MATCH. A systematic comparison between simulated and observed CO2 fluxes and mixing ratios is performed for two observational sites in Russia taking into account both surface observations and vertical profiles of meteorological parameters and CO2 in the lowest 3 km from the summer months in 1998. We find that the model is able to represent meteorological parameters as temperature, humidity and planetary boundary layer height consistent with measurements. Further, it is found that the simulated surface CO2 fluxes capture a large part of the observed variability on a diurnal time scale. On a synoptic time scale the agreement between observations and simulation is poorer which leads to a disagreement between time series of observed and simulated CO2 mixing ratios. However, the model is able to realistically simulate the vertical gradient in CO2 in the lowest few kilometres. The vertical variability is studied by means of trajectory analysis together with results from the MATCH model. This analysis clearly illustrates some problems in deducing CO2 fluxes from CO2 mixing ratios measured in single vertical profiles. Studies of the regional variability Of CO2 in the model domain show that there exists no ideal vertical level for detecting the terrestrial signal Of CO2 in the free troposphere. The strongest terrestrial signal is found in the boundary layer above the lowest few hundred metres. Nevertheless, this terrestrial signal is small, and during the simulated period it is not possible to detect relative variations in the surface fluxes smaller than 20%. We conclude that a regional flux cannot be determined from single ground stations or a few vertical profiles, mainly due to synoptic scale variability in transport and in CO2 surface fluxes