To know regional-to-continental scale CO₂ fluxes between atmosphere and terrestrial biosphere using an inverse model, the CO₂ measurements on plural towers situated in a thousand square kilometer area of West Siberia have been carried out since 2002. The CO₂ concentrations at 80m of the tower during daytime afternoon well represents those of PBL with its difference in ±3 ppm, and 90% of them in ±2 ppm, in clear sky day, when no strong inversion is occurred in winter. The tower observation expands to five sites to date, and additional four sites will be established in a year.

Continuous atmospheric measurements from tall towers have the capability to bridge an observational gap between hemispheric and local scales. We present first results from measurements made at such a tower in Germany. We show anti-correlated O₂ and CO₂ high frequency temporal variations which are caused by regional land biotic and fossil fuel emissions. We also show correlated changes in CO₂ concentration with air mass back trajectories, for example showing elevated CO₂ from air masses derived from eastern Europe, and lower, “background” concentrations from air masses derived from the North Atlantic.

First atmospheric δO₂/N₂, CO₂ and δ¹³C flask measurements from vertical aircraft sampling in the lower troposphere above Griffin Forest (GRI), Perthshire, UK, (56°37’N, 3°47’W) and from ground based flask sampling at the high altitude site Jungfraujoch (JFJ), Switzerland (3580m above sea level (a.s.l.), 46°33’N, 7°59’E), and the mountain site Puy de Dôme (PUY), France (1480m a.s.l., 45°46’N, 2°58’E) are presented.

Measurements of concurrent changes in both the atmospheric O₂ and CO₂ mixing ratios have been proven to be useful independent information for the partitioning of anthropogenic CO₂ into its different sinks [e.g. Keeling et al., 1996]. This information is used along with the “classical” partitioning models that make use of CO₂ concentration and (radioactive as well as stable) isotopic composition information [e.g. Keeling et al., 1995]. Global carbon budget reconstruction needs long time series observations of global means. Downscaling to a more regional assessment introduces a closer relation to possible annual and regional variations in prescribed oxidative ratios of biospheric and combustion processes. With the goal of improving the knowledge on the temporal and local variability of the O₂/ CO₂ signal, we present the results of the analysis on an extended data set from the remote station of Lutjewad (The Netherlands) and compare them with the findings of different other sampling stations in Europe, starting from 2001 till present.

Our understanding of biogeochemical and physical processes in the Southern Ocean, which are critically important to future anthropogenic CO₂ uptake and global climate, is limited by the sparse spatial and temporal coverage of existing oceanographic and atmospheric measurements. We will present high-precision horizontal atmospheric O₂ and CO₂ concentration gradients over the Southern Ocean from three independent observing networks. These measurements reveal that, relative to southern mid-latitudes and Antarctica, CO₂ concentrations over the Southern Ocean are high during winter and low during summer (Fig. 1). This suggests a seasonal variation between net CO₂ summertime uptake and wintertime release that is in disagreement with the T99 [Takahashi et al., 2002] dissolved pCO₂ climatology, which predicts year‑round CO₂ uptake, and with the OCMIP‑2 biological ocean general circulation models [BOGCMs, Doney et al., 2004], which either predict year-round CO₂ uptake or opposite seasonality with wintertime uptake and summertime release.

We used recent ocean bomb radiocarbon inventory estimates for the time of GEOSECS (mid-1970s) and WOCE (mid-1990s) from Peacock [2004] and Key et al. [2004], corrected for missing ocean areas [Naegler 2005], to develop a new parameterisation of the piston velocity – wind speed relationship of CO₂ air-sea gas exchange. For monthly mean climatological winds on a 1°x1° grid, this results in a gas exchange parameter a_q,660 of 0.32±0.04 (in cm hr^-1 m^-2 s²) and a net oceanic CO₂ uptake of 1.53±0.18 PgC/yr for the mid-1990s, when using the Takahashi et al. [2002] pCO₂ data.

Air-water CO₂ fluxes were up-scaled to take into account the latitudinal and ecosystem diversity of the coastal ocean, based on an exhaustive literature survey. Marginal seas at high and temperate latitudes act as sinks of CO₂ from the atmosphere, in contrast to subtropical and tropical marginal seas that act as sources of CO₂ to the atmosphere. Overall, marginal seas act as a strong sink of CO₂ of about -0.45 Pg C yr^-1. This sink could be almost fully compensated by the emission of CO₂ from the ensemble of near-shore coastal ecosystems of about 0.40 Pg C yr^-1.

The current WMO CO₂ Mole Fraction Scale consists of a set of fifteen CO₂ –in-air primary standard calibration gases ranging in CO₂ mole fraction from 250 to 520 micromol/mol. Since the WMO CO₂ Expert Group transferred responsibility for maintaining the WMO Scale from the Scripps Institute of Oceanography (SIO) to the Climate Monitoring and Diagnostics Laboratory (CMDL) in 1995, the fifteen WMO primary standards have been calibrated at regular interval, between one and two years, by the CMDL manometric system. From mid-1996 to 2001, the assigned CO₂ values of the WMO Primaries have been jointly based on the SIO and CMDL manometric measurements, and completely on the CMDL manometric measurements alone from 2001 to present. The uncertainty of the 15 primary standards is estimated to be 0.07 micromol/mol in the one-sigma absolute scale. Manometric calibration results indicated that there is no evidence of overall drift of the Primaries from 1996 to 2004. In order to lengthen the useful life of the Primary standards, CMDL has always transferred the WMO Scale to the Secondaries via NDIR analyzers. The uncertainties arising from the analyzer random error and the propagation error due to the uncertainty of the reference gas concentration are discussed. Precision of NDIR transfer calibrations is about 0.01 micromol/mol from 1979 to present. Propagation of the uncertainty is calculated theoretically. In the case of interpolation, the propagation error is estimated to be between 0.05 and 0.07 micromol/mol when the Primaries are used as the reference gases via NDIR transfer calibrations.

Mode Waters provides a privileged pathway for the transport of heat, salt and anthropogenic CO₂ into the ocean interior. The carbon cycle decadal variability in response to environmental changes is investigated using historical and recent data collected during the INDIGO (1985-1987) and OISO (1998-2003) oceanographic campaigns conducted in the South West Indian Ocean, an important zone for Mode Waters formation. The observed change in dissolved inorganic carbon over the 15-year period was 8 µmol/kg in Subantarctic Mode Water (500-800m), which is less than the anthropogenic carbon increase alone (13 µmol/kg). This difference may be explained by natural or climate-induced changes in ocean processes. Predictions from a global ocean-carbon model (OPA-PISCES) are used as a means to help interpret changes in the controlling processes: ocean dynamics, biological activity and air-sea interactions.

We present preliminary results of a modeling experiment that compares observed vertical profiles of CO₂ with those generated by an atmospheric transport model (ATM). The ATM is driven by CO₂ flux fields generated from the inversion of monthly averaged CO₂ surface data (GLOBALVIEW). We note large differences between the best fit to the observations produced in the inversion and the same quantity simulated by the forward model. This difference arises from the nonlinearity of the advection scheme used in the transport model. When comparing with vertical profiles, we note that much of the difference between simulated and observed concentration has the same structure as the impact of this nonlinearity. Inversion schemes must therefore take nonlinearity into account. Despite these differences, the profiles are able to distinguish among inversions that fit subsets of the surface data, suggesting they are a useful validation dataset.