The Center for Atmospheric Sciences | » Ground-based retrievals using the ASSIST

  • Research Projects

  • Ground-based retrievals using the ASSIST

     

    I also work for a Department of Energy contractor where I have participated in over a dozen field campaigns, collecting and processing data  from the Atmospheric Sounder Spectrometer for Infrared Spectral Technology (ASSIST) for various atmospheric studies. The ASSIST is a Fourier transform infrared spectrometer with an internal calibration unit dedicated to support applications that require very high radiometric accuracy. The interferometer contains two detectors, Mercury cadmium telluride (MCT) and Indium Antimonide (InSb), which give the ASSIST a spectral range of 2 – 20 µm (500 – 5,000 cm-1).  The unapodized spectral resolution of the ASSIST is 0.5 cm-1 and the temporal resolution is approximately 1 interferogram per second.  The collection of ASSIST data has been instrumental in my research for the validation and improvement of a retrieval algorithm developed by my advisor, Dr. Bill Smith.

    Ultraspectral infrared systems contain vast amounts of information on surface and atmospheric properties within their thousands of spectral measurements. Each piece provides a piece of information (i.e. temperature, moisture, trace gases) as to the final picture of the column of atmosphere or surface. These pieces of spectral information can be used together to extract atmospheric temperature, water vapor, as well as cloud and radiative properties. The retrieval process converts the thousands of discrete independent spectral measurements into an accurate depiction of the atmosphere through a physically based optimal estimation program. This physical-statistical profile retrieval algorithm combines (1) a statistical regression using a pre-calculated statistical dataset containing the properties of atmospheric state parameters including temperature and moisture profiles, surface characteristics and cloud parameters, and (2) a physical estimation using the ‘best guess’ as an initial condition, rather than the mean of the statistical regression sample, and constraint within theoretical calculations of the inversion of the radiative transfer equation in order to find a fit to the atmospheric and cloud conditions in various clear and cloud trained datasets [1,2,3].

     

    References

    [1] Smith, William L. Sr.; Weisz, Elisabeth; Kireev, Stanislav V.; Zhou, Daniel K.; Li, Zhenglong and Borbas, Eva E. Dual-regression retrieval algorithm for real-time processing of satellite ultraspectral radiances. Journal of Applied Meteorology and Climatology, Volume 51, Issue 8, 2012, 1455–1476.

    [2] Zhou, D. K., W. L. Smith, J. Li, H. B. Howell, G. W. Cantwell, A. M. Larar, R. O. Knuteson, D. C. Tobin, H. E. Revercomb, and S. A. Mango (2002). “Thermodynamic product retrieval methodology and validation for NAST-I”. Applied Optics, Volume 41, Issue 33, 2002, 6957-6967.

    [3] Smith, W. L., L. Rochette, R. B. Pierce, “Determination of Greenhouse Gas Profiles From Ground based Infrared Radiance Measurements”, SPIE 2012 Optics and Photonics, 12-16 August, 2012, Dan Diego CA.