Retrieval of Line Parameters from High Resolution Fourier Transform Laboratory Spectra in Support of Atmospheric Spectroscopy

Abstract

Fourier transform infrared (FTIR) spectroscopy has become a powerful tool for the detection and measurement of gas emissions (or absorption) in terrestrial and planetary atmospheric remote sensing. Besides the improving quality of current instruments (signal to noise ratio, spectral resolution, sensors’ response…) and the advances in computational facilities, one needs precise knowledge of molecular spectroscopic line parameters to deduce accurate atmospheric state (e.g., temperature and molecular concentrations). Consequently, scientific studies of the major environment questions of global warming and Ozone depletion require global precise datasets of atmospheric constituents. These databases are supplied with molecular line parameters from laboratory spectroscopy. In fact, measurements of positions, intensities, pressure broadening and pressure shifting coefficients…or other parameters of lines are in general long, very difficult, fastidious and even impossible for weak, blended, large,… or superposed lines. That is why it is imperative to have theoretical models which permit calculating these parameters. But models are reliable only if they are built up using correct data concerning line parameters obtained by using adequate line shape as Lorentz, Voigt, Rautian, Galatry, Dicke profiles…according to experimental conditions (e.g. temperature, pressure, buffer gases,…), and taking into account instrumental parameters for modeling the Instrumental Line Shape (ILS). Positions of lines produced by experiments are directly used to determine quantum mechanical Hamiltonians’constants. At the opposite, intensities, pressure broadening and pressure shifting aren’t directly used as produced by experiments. They can’t be used as raw data. This crucial work located between the experimentation and the theoretical modeling of spectra is a vital intermediate step in the treatment of the data. It needs the use of adequate, efficient, reliable computation codes adaptable to each particular case. Here we describe this intermediate work. As example, we retrieve line parameters of few lines of a pure rotational spectrum of Ozone molecule (O3).