Ground-based FTIR and MAX-DOAS observations of formaldehyde at Réunion Island and comparisons with SCIAMACHY and the CTM IMAGES
Corinne Vigouroux(1), F. Hendrick(1), T. Stavrakou(1), I. De Smedt(1), B. Dils(1), C. Hermans(1), M. Kruglanski(1), A. Merlaud(1), J.-F. Müller(1), E. Neefs(1), F. Scolas(1), C. Senten(1), M. Van Roozendael(1), S. Fally(2), P.-F. Coheur(2), M. Carleer(2), J.-M. Metzger(3), R. Delmas(3) and M. De Mazière(1)
(1) Belgian Institute for Space Aeronomy (BIRA-IASB), 3, Avenue Circulaire, B-1180 Brussels, Belgium
(2) Université Libre de Bruxelles (ULB), , Brussels, Belgium
(3) Université de La Réunion, , Saint-Denis, France
Formaldehyde (HCHO) has been retrieved from ground-based Fourier transform infrared (FTIR) and UV-Visible MAX-DOAS campaign measurements in 2004-2005 and 2007 at the NDACC site of Réunion Island (21°S, 55°E). The inversion algorithms for both techniques use the Optimal Estimation Method to derive low vertical resolution profiles of the target gases. The FTIR and MAX-DOAS daily mean total columns of formaldehyde have been compared in the common measurement period August to October 2004. All campaigns data have also been compared to correlative SCIAMACHY data. In the comparisons we account for the complete error budget of the involved data sets and for their different vertical resolutions by including their respective averaging kernels.
Despite the difficulty of detecting formaldehyde, the FTIR and MAX-DOAS daily mean total columns agree very well: no significant bias is observed, the standard deviation of the comparisons is only 8% and the correlation between both data sets is 0.73. The FTIR and SCIAMACHY total columns agree also well in 2004, with a standard deviation of 21% and a correlation of 0.66, while in 2007 the standard deviation is larger (54%) and the correlation is only 0.49. Both techniques observe the same seasonal variation with a minimum in local winter. We find that comparing DOAS and SCIAMACHY products is not appropriate because DOAS is sensitive to a change in formaldehyde mostly below the altitude of 1.5 km, where SCIAMACHY has no sensitivity.
The FTIR and MAX-DOAS data are also confronted with modeled HCHO columns from the global CTM IMAGES. We find that the standard model has a significant negative bias: 29% and 23% with FTIR, in 2004 and 2007 respectively, and 15% with MAXDOAS. The seasonal variation has a smaller amplitude in the model than in the observations.