Das Abtastverfahren der Interferogramme des flugzeuggetragenen Fourierspektrometers MIPAS-STR

Abstract

The MIPAS-STR (Michelson Interferometer for Passive Atmospheric Sounding -STRatospheric aircraft) is a cryogenic Fourier transform spectrometer for measuring thermal emission of atmospheric species in the mid-infrared spectral range. It is operated on board of the Russian high-altitude aircraft M-55 GEOPHYSICA. Two-dimensional distributions of volume mixing ratios of various trace gases are measured by limb sounding. To optimize the horizontal resolution of the two-dimensional distributions, only one interferogram is measured at each tangent height. The measurement strategy requires low-noise interferograms and measurements that are insensitive to vibrations of the aircraft. The present thesis deals with the effects of vibrations on the MIPAS-STR interferometer and the minimization of perturbations in the spectrum using a novel time-equidistant sampling method for the interferograms. Vibrations of the optical components of the interferometer lead to velocity variations of the optical path difference. In case of different time delays in the IR and reference laser detection chains, sampling errors of the IR interferogram and consequently so-called ghost lines in the spectrum occur. The new sampling method allows for a correction of the electric delay, as a result of which the formation of ghost lines is suppressed. The sampling method was investigated in three steps for its capability of reducing the ghost lines for a given harmonic velocity variation by changing the electric time delay. First, the interpolation filter, i.e. the central part of the sampling system, was tested numerically. In a second step, the interferometer electronics was investigated using signal generators simulating the interferograms. The final test setup yielded - for a velocity variation of 25% - at the optimum time delay ghost amplitudes of 2.0.10 - 4 relative to the amplitude of the original line. In the third step, the complete MIPAS-STR was tested. By using a diode laser as infrared source, the ghost lines were easily identified and a signal-to-noise ratio of about 4000 was achieved. Tests of the system with a velocity variation of 25% resulted in relative ghost amplitudes of 1.0% at the optimum time delay. In contrast to the theory, the amplitudes of the two first-order ghost lines were different in the magnitude spectra. To understand this observation, the theory of ghost lines was further developed and tested with the program GHOST_SIM. It was found, that the remaining ghost amplitudes observed in the test of the complete MIPAS-STR were caused by slightly differing velocity variations during the sampling of the IR- and reference laser interferograms (phase modulation of 2nd kind). The unequal magnitudes of the two first-order ghost lines were explained by a combination of 2nd kind phase modulation and an amplitude modulation induced by the frequency-dependent amplitude response of the signal chain. In particular the investigations of the complete MIPAS-STR will allow to considerably increase the sensitivity of the instrument.