Optical properties of subvisual cirrus clouds measured by the OSIRIS instrument
Satellite-based measurements with limb viewing geometry give high-quality information about the atmosphere. However, low-altitude measurements are very sensitive to thin clouds, which act as a semi-reflective layer that can screen out the lower atmosphere and the earth's surface. The work described in this thesis demonstrates the ability to retrieve the optical properties of these clouds from observations of scattered sunlight by the OSIRIS instrument. Retrievals from satellite measurements require a radiative transfer model well-suited to the measurement geometry. For this reason the SASKTRAN model is used, which employs full spherical geometry. Since this model was not initially intended to model cloud-particle scattering, several sources for the light-scattering properties of ice crystals -- from both first-principles algorithms and from in-cloud particle measurements -- were incorporated. Also, since these properties violate several simplifying assumptions of the model, modifications to the model were required. With the work described in this thesis, SASKTRAN replicates in-cloud radiance measurements with high accuracy across the measured spectrum. Cloud property retrievals are demonstrated that use SASKTRAN in an iterative retrieval technique. Assuming an effective cloud particle size, cloud particle number densities are retrieved that replicate the spectral measurements with very good accuracy. The effect of cloud properties on ozone and stratospheric aerosol retrievals is investigated. Systematic biases that result from neglecting cloud-particle scattering are discussed. Coincident measurements from another satellite instrument shows very good agreement with OSIRIS cloud properties.
remote sensing, light scattering, clouds, atmospheric science
Doctor of Philosophy (Ph.D.)
Physics and Engineering Physics