Articles | Volume 8, issue 3
Atmos. Meas. Tech., 8, 1491–1517, 2015
Atmos. Meas. Tech., 8, 1491–1517, 2015

Research article 23 Mar 2015

Research article | 23 Mar 2015

A comprehensive observational filter for satellite infrared limb sounding of gravity waves

Q. T. Trinh1, S. Kalisch1, P. Preusse1, H.-Y. Chun2, S. D. Eckermann3, M. Ern1, and M. Riese1 Q. T. Trinh et al.
  • 1Institute of Energy and Climate Research, Stratosphere (IEK-7), Forschungszentrum Jülich, Jülich, Germany
  • 2Laboratory for Atmospheric Dynamics, Department of Atmospheric Sciences, Yonsei University, South Korea
  • 3Space Science Division, Naval Research Laboratory, Washington, D.C., USA

Abstract. This paper describes a comprehensive observational filter for satellite infrared limb sounding of gravity waves. The filter considers instrument visibility and observation geometry with a high level of accuracy. It contains four main processes: visibility filter, projection of the wavelength on the tangent-point track, aliasing effect, and calculation of the observed vertical wavelength. The observation geometries of the SABER (Sounding of the Atmosphere using Broadband Emission Radiometry) and HIRDLS (High Resolution Dynamics Limb Sounder) are mimicked. Gravity waves (GWs) simulated by coupling a convective GW source (CGWS) scheme and the gravity wave regional or global ray tracer (GROGRAT) are used as an example for applying the observational filter. Simulated spectra in terms of horizontal and vertical wave numbers (wavelengths) of gravity wave momentum flux (GWMF) are analyzed under the influence of the filter. We find that the most important processes, which have significant influence on the spectrum are the visibility filter (for both SABER and HIRDLS observation geometries) and aliasing for SABER and projection on tangent-point track for HIRDLS. The vertical wavelength distribution is mainly affected by the retrieval as part of the "visibility filter" process. In addition, the short-horizontal-scale spectrum may be projected for some cases into a longer horizontal wavelength interval which originally was not populated. The filter largely reduces GWMF values of very short horizontal wavelength waves. The implications for interpreting observed data are discussed.