| A new method to detect and predict refractivity ducts based on passive radiometric measurements is presented in this thesis.;A refractivity duct occurs when a layer of air above the ocean surface traps electromagnetic energy by the effects of water vapor on the vertical gradient of the refractive index of air. The presence of this duct can have serious effects on short wavelength electromagnetic propagation, including ship-to-satellite communication links, and over-the-horizon propagation.;In order to detect and predict the refractivity ducts, we require an accurate estimate of the humidity profile. Retrieving humidity profiles from integrated water vapor profiles is a difficult problem, and is much worse under ducting conditions due to the sharp variations in the water vapor gradients. In this thesis, a new method is developed to retrieve water vapor profiles from radiometer measurements, including, those which demonstrate sharp gradients.;This method was explored using a statistical approach applied to synthetic radiometer data calculated from balloon measurements. Zenith and off-vertical radiometer integrated water vapor measurements were simulated and used in a self-training statistical retrieval algorithm. The retrieval of the profiles in general shows improvement over previous work. Results indicate that a complete set of radiometer measurements as well as an average profile is adequate to predict the profile shape. Surface measurements of dewpoint temperature and surface air temperature are redundant in this case. Particular emphasis in this retrieval was paid not just to the profile retrieval, but also to the retrieval of the water vapor profiles, which display refractivity duct characteristics. The characteristics of radiometers capable of making such measurements are also discussed. |
