WHOLE-BAND ANALYSIS AND INSTRUMENTAL EFFECTS IN FTIR SPECTRA

$^{1}$G. A. Vanasse and H. Sakai, in: Progress in Optics, Vol. 6, Chapter VII, North Holland, Amsterdam (1967). $^{2}$ E. Niple, W.G. Mankin, A. Goldman, D.G. Murcray and F.J. Murcray, Geophys. Res. Lett. 7, 489-492 (1980)Author Institution:Vanasse and $Sakai^{1}$ and others have described instrumental effects which affect the appearance of spectra obtained by Fourier Transform Spectroscopy. In particular, an error in locating the Zero path Difference (ZPD) in an interferogram, or an asymmetry of the interferogram, may cause phase errors and apparent line shape distortions; and the finite solid angle of the beam traversing the interferometer causes a loss of resolution in addition to that caused by the finite maximum path difference. We have found it necessary to model these effects in our analysis of high resolution spectra of $CO_{2}$ and $N_{2}O$ by spectral curve fitting. By including these effects the mean residual difference between the observed and calculated spectra have been reduced to approximately 1.5 times the estimated noise level for some spectra. The phase errors retrieved in these analyses would correspond, if due solely to an error in determining ZPD, to a ZPD error of about one-twentieth of a wavelength of the Helium-Neon laser used to measure path difference. The retrieved resolution degradation due to aperture size is in agreement with that calculated from the actual aperture size of the instrument. The ability of the whole-band spectral curve fitting technique to account for these and similar effects is particularly useful in analyzing field $measurements^{2}$, but our results indicate that it is also important in determining band parameters from laboratory data