Reliable remote sensing of cloud liquid water content and droplet size could greatly improve aviation safety in winter weather by enhancing the detection of aircraft icing conditions. Previous studies have suggested the potential of a technique using differential reflectivity from dual-frequency radar measurements to estimate these quantities. Under the auspices of the Federal Aviation Administration's Aviation Weather Research Program, the National Center for Atmospheric Research is evaluating this technique with a view toward developing an integrated algorithm that will produce reliable, range-resolved profiles of liquid water content and droplet size.

The Mount Washington Icing Sensor Project (MWISP) and the Alliance Icing Research Study (AIRS) provide ideal datasets for evaluating and enhancing the dual-wavelength technique. Both include coordinated microwave radiometer and X-, Ka-, and W-band polarized Doppler radar measurements. In MWISP, the NOAA Environmental Technology Laboratory's 3-channel radiometer, NOAA/ETL X- and Ka-band radar, and University of Massachusetts Cloud Profiling Radar System (CPRS) Ka- and W-band radar were used; AIRS employed the Radiometrics 2-channel WVR-1100 radiometer, McMaster University's IPIX X-band radar, and the UMASS CPRS Ka- and W-band radar. In both datasets, the radars provide Doppler data, reflectivity, and linear depolarization ratio, while the radiometers provide total-path vapor and liquid water measurements.

The analysis presented in this paper extends earlier work by systematically examining a larger number of cases under varied meteorological conditions, combining X-, Ka- and W-band measurements, and using the radar linear depolarization ratio for data quality control and to enhance the range-resolved liquid water content retrieval. Radiometer data are used principally to evaluate the quality of the radar retrieval, though the information they provide may eventually be an additional input into a combined fuzzy-logic algorithm for producing liquid water and droplet size profiles.