Regional carbon dioxide fluxes from mixing ratio data

We examine the atmospheric budget Of CO2 at temperate continental sites in the Northern Hemisphere. On a monthly time scale both surface exchange and atmospheric transport are important in determining the rate of change of CO, mixing ratio at these sites. Vertical differences between the atmospheric boundary layer and free troposphere over the continent are generally greater than large-scale zonal gradients such as the difference between the free troposphere over the continent and the marine boundary layer. Therefore, as a first approximation we parametrize atmospheric transport as a vertical exchange term related to the vertical gradient Of CO2 and the mean vertical velocity from the NCEP Reanalysis model. Horizontal advection is assumed to be negligible in our simple analysis. We then calculate the net surface exchange of CO2 from CO mixing ratio measurements at four tower sites. The results provide estimates of the surface exchange that are representative of a regional scale (i.e. similar to 10(6) km(2)). Comparison with direct, local-scale (eddy covariance) measurements of net exchange with the ecosystems around the towers are reasonable after accounting for anthropogenic CO2 emissions within the larger area represented by the mixing ratio data. A network of tower sites and frequent aircraft vertical profiles, separated by several hundred kilometres, where CO2 is accurately measured would provide data to estimate horizontal and vertical advection and hence provide a means to derive net CO, fluxes on a regional scale. At present CO2 mixing ratios are measured with sufficient accuracy relative to global reference gas standards at only a few continental sites. The results also confirm that flux measurements from carefully sited towers capture seasonal variations representative of large regions. and that the midday CO2 mixing ratios sampled in the atmospheric Surface layer similarly capture regional and seasonal variability in the continental CO2 budget.