Analysis of the FSI WFM-DOAS retrievals with respect to the groundbased FTIR instrument, located at Egbert, Canada, show goodagreement with an average negative bias of approximately −4.0% witha standard deviation of3.0%. This bias which exhibits anapparent seasonal trend, is of unknown origin, though slightdifferences between the averaging kernels of the instruments and thelimited temporal coverage of the FTIR data may be the cause. Therelative scatter of the retrieved vertical column densities islarger than the spread of the FTIR measurements. Normalizing theCO2 columns using the surface pressure does not affect themagnitude of this bias although it slightly decreases the scatter ofthe FSI data.
Comparisons of the FSI retrievals to the TM3 globalchemistry-transport model, performed over four selected NorthernHemisphere scenes show reasonable agreement. The correlation,between the time series of the SCIAMACHY and model monthly sceneaverages, are0.7 or greater, demonstrating the ability ofSCIAMACHY to detect seasonal changes in the CO2distribution. The amplitude of the seasonal cycle, peak to peak,observed by SCIAMACHY however, is larger by a factor of 2–3 withrespect to the model, which cannot be explained. The yearly meansdetected by SCIAMACHY are within 2% of those of the model with themean difference between the CO2 distributions alsoapproximately 2.0%. Additionally, analysis of the retrievedCO2 distributions reveals structure not evident in the modelfields which correlates well with land classification type.
From these comparisons, it isestimated that the overall bias ofthe CO2 columns retrieved by the FSI algorithm is <4.0%with the precision of monthly 1°×1° griddeddata close to 1.0%.
