ATMOSPHERIC CARBON DIOXIDE: RETRIEVAL FROM GROUND-BASED FOURIER TRANSFORM INFRARED SOLAR ABSORPTION MEASUREMENTS AND MODELLING USING A COUPLED GLOBAL-REGIONAL SCALE APPROACH

Atmospheric carbon dioxide (CO2) was retrieved from ground-based solar absorption measurements using Fourier transform infrared (FTIR) spectrometry in three European stations (Biscarrosse, Bremen and Ny-Alesund). Functioning as an internal standard and a means to determine the dry air mixing ratio, molecular oxygen (O2) was likewise retrieved. The effects of instrumental parameters such as the resolution, the aperture size and high folding limits were assessed to identify correlated errors in both the CO2 and the O2. Changes in the field of view, the maximum optical path difference and the high folding limit seem to change the trace gas column concentrations of O2 and CO2 and just partially cancels out in the CO2/O2 ratio. Correction factors on the O2 column to minimize these instrumental effects were determined and applied to the FTS data. These correction factors seem to be more effective in Ny-Alesund than in Bremen as the degree of these instrumental changes appear to be more subtle in Ny-à �lesund. Additional correction strategies, particularly for the CO2, are still being examined and investigated. Comparisons of FTIR CO2 with integrated CO2 aircraft data were performed in the Biscarrosse station as a means of calibrating the FTIR data. Aside from this, the Stochastic Time Inverted Lagrangian Transport (STILT) model was also used as a 'transfer standard' between FTIR column concentrations and measurements made in situ by a co-located tower. STILT and tower data compared reasonably well. However, comparison of STILT with the FTIR showed a large bias. This bias is attributed to the scaling factor used in calibrating the FTIR data with the integrated CO2 aircraft data. The scaling factor was derived to a large extent from aircraft measurements that sampled within a 50 km distance from the FTS and this introduces spatial heterogeneity in the carbon dioxide volume mixing ratios around the FTIR station. Long-term time series of column averaged carbon dioxide volume mixing ratios for the Bremen and Ny-à �lesund stations were compared to STILT (only the Bremen station) and to CarbonTracker. A 'clear sky' bias was pin pointed as models see increased CO2 during frontal zone conditions - a meteorological condition when FTIRs often cannot measure. The spatial heterogeneity of carbon dioxide around the Bremen station was also assessed by comparing FTIR data with varying resolutions of STILT and it was found that column concentrations are not sensitive to small scale local carbon dioxide emission sources amidst Bremen being situated in an urban setting. The difference in variability between fine to coarse scales are approximately 0.2 ppm.