P5.5
Estimating Polar Precipitation with CloudSat, AIRS and Microwave Radiometers
Ziad S. Haddad, Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA; and K. Park, F. J. Turk, and N. Y. Wang
While measuring and monitoring precipitation in polar regions is difficult, recent studies have shown that microwave radiances measured by operational high-frequency sounders, such as the Advanced Microwave Sounding Unit (AMSU) and the Microwave Humidity Sounder (MHS), are sensitive to falling snow, though the ambiguous radiometric signature of the frozen surface shows through the drier polar atmosphere, making it difficult to extract the signature from falling. We capitalized on the typically small delay between CloudSat and the NOAA-18 platform to compile a database of non-precipitating surface emissivities and used a principal component analysis to identify a subset of channel combinations that are essentially insensitive to the radiometric signature of the surface, and whose deviation from the mean can be related to the precipitation. The CloudSat-detected precipitation cases were then processed through several radar-only vertical retrievals (assuming different snow habits and hydrometeor size distributions, as well as super-cooled cloud liquid at the top of the cloud), and the retrieved candidate profiles of water content were then combined with estimates of water vapor (from AIRS as well as from ECMWF analysis) to forward-calculate the associated brightness temperatures, and compare the results with the surface-insensitive principal components of the measured brightness temperatures. The comparisons produce weights which quantify the consistency of the candidate profiles with the nearly-simultaneous radar and radiometer measurements. Not only does this Bayesian approach produce estimates of snowfall for each CloudSat measurement, its results are also being compiled into a database that can be used to make snowfall estimates from the MHS (and from the SSMIS) brightness temperatures alone. In turn, this will help evaluate the realism of numerical models and their microphysical assumptions, particularly as the latter appear to have significant difficulties representing Arctic clouds accurately.
Poster Session 5, Algorithms Exploiting the Synergy of Multiple Satellite Sensors, Satellite/Model Fusion, and Blended Products - Posters
Wednesday, 29 September 2010, 3:00 PM-5:00 PM, ABC Pre-Function
Previous paper Next paper