Study of Ku- and Ka-band Spaceborne Radar for Hydrometeor Phase Identification

The Ku- and Ka-band dual-wavelength precipitation radar (DPR), which will be on board the Global Precipitation Measurement (GPM) core satellite, is expected to provide 3-dimensional storm structure and precipitation rate profiles. Knowledge of hydrometeor phase states, such as snow, mixed-phase and rain along the radar path, is critical to accurately estimate radar signal attenuation resulting from hydrometeor absorption and scattering. Because the DPR measures only the co-polar radar reflectivity factor, most of the existing algorithms/techniques used for ground-based polarimetric radar for hydrometeor phase identification are not applicable to the DPR. A recent model study has shown potential for identifying phase states of hydrometeors by using the differential frequency ratio (DFR), defined as the difference of radar reflectivity factors between Ku- and Ka-bands, and the radar reflectivity factor at Ku-band, Z(Ku). This is based on the fact that there is a clear separation between snow and rain in the Z(Ku)-DFR plane assuming that the snow follows the Gunn-Marshall size distribution and rain follows the Marshall-Palmer size distribution. In an effort to verify the simulated results, the data collected by the Airborne Second Generation Precipitation Radar (APR-2) in the Wakasa Bay AMSR-E campaign are employed. Using the signatures of Linear Depolarization Ratio (LDR) at Ku-band, the APR-2 data can be easily divided into the regions of snow, mixed phase and rain for stratiform storms. These results are then superimposed onto the theoretical curves computed from the model in the Z(Ku)-DFR plane. For over 90% of the observations from a cold-season stratiform precipitation event, snow and rain can be distinguished if the Ku-band radar reflectivity exceeds 18 dBZ (the minimum detectable level of GPM DPR at Ku-band). This is also the case for snow and mixed-phase hydrometeors.

Although the DPR has a potential to differentiate precipitation phase states for stratiform rain, its applicability to convective rain has not yet been tested. Determination of the regions of snow, rain and mixed-phase along the radar path for convective rain is imperative for improving accuracy of rain estimates. To extend our validation of Ku- and Ka-band dual-wavelength radar application to convective storms, we will check the phase identification techniques against the APR-2 products on phase states obtained from the combined measurements of the APR-2 and other meteorological data. One of the important aspects in this study is to determine whether the dual-wavelength radar phase algorithms that are developed from the model simulations and trained on radar measurements of stratiform rain, are applicable to the convective cases. The development of an effective dual-wavelength method to distinguish the rain, snow and mixed-phase hydrometeors will be an important step toward an accurate, efficient DPR profiling algorithm.