The model assumes a one-to-one relationship between snowflakes above and raindrops below as well as a constant mass flux throughout the melting layer of precipitation. Melting snowflakes are assumed to be spheroidal with aspect ratio dependent on the stage of melting: small ice particles are more oblate than large ones at the top of the melting layer whereas small raindrops are less oblate than large ones at the bottom of the melting layer. The particle orientation is described by an axially symmetric probability density function, the width of which decreases with increasing depth in the melting layer. Backscattering calculations make use of the T-matrix method and of an extension to the technique for averaging over orientations analytically. The outcomes of the theoretical model are height profiles of the polarimetric and Doppler radar observables.
The observational data were obtained during a stratiform precipitation event by means of two Doppler polarimetric radars: the TARA atmospheric radar, which is a 3.3GHz FM-CW system capable of recording the time series of the beat signal, and the KNMI cloud radar, which is a 35GHz pulse system. The two radars were collocated at the Cesar observational site, at Cabauw, the Netherlands and the measurement was made during the Baltex/Bridge Cloud (BBC) campaign. The data exhibit typical features of the radar bright band, that is, peaks of the radar observables that appear in the following order from top to bottom of the melting layer: ZHH, LDR, and ZDR.
The comparisons between simulations and measurements show that the theoretical model is capable of reproducing the aforementioned typical features of the radar reflections from the melting layer of precipitation. The peak of the horizontal reflectivity is more pronounced at 3 than at 35GHz.
Supplementary URL: