The Microphysical Information of Backscatter Differential Phase in the Melting Layer at X, C, and S-band

Reliable estimates of backscatter differential phase δ can be obtained within the melting layer of stratiform precipitation at S, C, and X band via azimuthal averaging of radial profiles of differential phase Φ_DP at high antenna elevations (i.e., ≥ 7^o). For such elevations, the impact of non-uniform beam filling (NBF) seems to be negligibly small and the “bumps” on the Φ_DP profiles are solely attributed to δ. The forward propagation contribution to the differential phase is reduced leading to increasingly “clean” δ without contamination from specific differential phase K_DP. The backscatter differential phase, which is immune to attenuation, partial beam blockage, and radar miscalibration, would complement the information routinely available from reflectivity Z_H, differential reflectivity Z_DR, and cross-correlation coefficient ρ_hv which are traditionally used for characterizing microphysical properties of the melting layer. The magnitude of δ can be utilized as an important calibration parameter for the improvement of microphysical models of the melting layer. Thus, analyses of the underutilized variable δ, together with Z_H, Z_DR, and ρ_hv within the melting layer measured at different wavelengths and in different climate regimes will be presented to further explore its informative content for microphysics studies as well as quantitative precipitation estimation.

Actual measurements of δ have been performed with a number of polarimetric WSR-88D radars in US, and with the C-band scanning ARM precipitation radars (CSAPR) during the Midlatitude Continental Convective Clouds Experiment (MC3E) at the ARM Southern Great Plains site in central Oklahoma. Similar observations of δ were made in Germany using polarimetrically upgraded C-band radars on the German Weather Service (DWD) network and research X-band radars in Bonn (BoXPol) and Jülich (JüXPol). For comparison a polarimetric model of the melting layer with spectral microphysics is used to simulate vertical profiles of δ within the bright band and its dependence on the density and size distribution of snow aloft for different radar wavelengths.

Strong correlations exist between the radial maxima in azimuth-averaged δ and differential reflectivity Z_DR, the maximum δ and minimum ρ_hv, as well as between the ρ_hv minimum and Z_DR maximum. High values of Z_DR and δ combined with low ρ_hv usually indicate melting of heavily aggregated and less rimed snow. However, such strong correlations between different radar variables are not seen for all events investigated so far. Some events observed with BoXPol show weak or negligible correlation between δ and ρ_hv. The δ-bumps are broader and the minima in ρ_hv are very flat and hard to identify, which possibly hints to a lower amount of large melting snowflakes or heavily rimed snow. The extrema for different moments also occur at different heights, which provides additional information useful for understanding the microphysics of the melting layer.