Numerous techniques exist and are under development to infer mesoscale fluxes of CO2 from spatial variations in atmospheric concentrations. At the smallest scales are methods that are based on measuring the build-up of CO2 concentrations above an ecosystem during temperature inversions (usually nocturnal) (Woodwell and Dykeman, 1966; Gallagher et al., 1994). At a larger scale (upward of 50 km2), measurements of concentration are made from tall towers or balloons to approach or reach the top of the boundary layer (Bakwin et al., 1995, 1997). The use of aircraft allows flux to be inferred at still larger scales (Choularton et al., 1995; Hollinger et al., 1995). Simultaneous measurement of CO2 concentrations and other tracers (e.g., radon) and past trajectories of air for continuous measurement sites may also be used to infer fluxes of CO2 and non-CO2 gases.
An example of a technique that infers flux at this intermediate scale is the convective boundary layer (CBL) budgeting approach. The concentrations and fluxes of CO2 in the atmosphere, higher than a few meters above the land surface, are the result of integrated individual contributions from the elements of the landscape at scales on the order of 50-100 km (Stull, 1988). The CBL budget approach exploits the natural integrating properties of the well-mixed atmospheric boundary layer, allowing average surface fluxes to be obtained over relatively large regions for time periods on the order of days rather than hours.
One of the main limits to widespread use of the CBL budgeting technique is the relatively high cost involved, because of the need to employ aircraft to sample above the top of the daytime CBL; these costs generally are on the order of US$1,000 per day. Another limitation is that the same air mass theoretically should be followed as it traverses the landscape. Practical experience has shown that, even with sophisticated weather forecasting tools, this tracking is difficult. Particularly when advection of air masses with different concentrations of the entity of interest occurs, failure to effectively track an air mass can lead to large errors in regional flux calculations using the CBL method. Additional limitations include that fact that the measurements are for relatively short, discontinuous periods and that the resulting flux includes an undefined area of land and all fluxes of CO2 (fossil as well as biotic).
Other reports in this collection