Application of statistical inversion to ground-based microwave remote sensing of temperature and water vapor profiles

Surface-based observations of downwelling microwave thermal emission are related to temperature and humidity profiles via a standard integral equation of radiative transfer. Both in clear and in cloudy atmospheres, statistical inversion techniques are used to retrieve profiles from a data vector of brightness observations and surface meteorological constraints. For the clear case, accuracy predictions and profile retrievals are illustrated for: (1) single frequency angular scanned data; (2) multi-frequency angular scanned data; and (3) multi-frequency zenith data. For the last case predicted and achieved accuracies were compared in a recently conducted radiometric experiment. Retrievals of cloud contaminated radiometric data are elaborated.