Preliminary verification of a double-moment normalization method for raindrop size distribution retrieval

Retrieving raindrop size distributions (DSDs) from polarimetric radar data can provide a good data support for the study of microphysical rain processes. Limitations in the DSD model lead to limitations of retrieving DSDs. For example, when calculating the μ-Λ relation in the constrained Gamma (C-G) model DSD retrieval method, some data are excluded, and thus the μ-Λ relation cannot represent the characteristics of all precipitation. To solve this problem, a new double-moment normalization method for retrieving DSDs from polarimetric radar data is proposed. On the basis of retaining all valid data, near-linear quantitative relationships are established. DSD moment six and seven are calculated by using Z_H and Z_DR, and then DSD is retrieved based on the double-moment normalization DSD model. Using data of two rainfall events occurred in Guangdong Province of China, the scientific feasibility of the method is verified by the simulation test, ground observations, and retrieval DSD results at different heights. The results reveal that the hypothetical DSD model is in good agreement with the actual observed data, and the method is reasonable and feasible in theory. The new method shows advantages over the traditional C-G DSD model method in both the simulation test and the ground verification, especially for heavy rain. At the same time, the particle size retrievals for the weak echo (＜35 dBz) in the high air derived from dual polarization radar (double-moment normalization method) and cloud radar are consistent. This method shows retrieval reliability based on both theory and ground and high-altitude DSD verification, and provides a good DSD retrieval way to study the DSD characteristics at different heights.