C-band Dual-polarization Radar Data Processing And Its Application To Precipitation Type Classification And Rainfall Estimation

With the rapid development of dual-polarization radar, its advantage is more and more obvious which is mainly reflected in rainfall estimation, hail identification,phase identification of precipitation particles and the recognition of drop-size distribution, etc. But in actual detecting process, the data quality is vulnerable to radar system, signal attenuation and ground clutter contamination, etc. In this paper,we use some observation data of NUIST C-band dual polarization radar to carry out some research about C-band dual polarization radar data processing, precipitation type classification and quantitative precipitation estimate. The main contents and conclusions are as following:(1) Wavelet analysis is employed to preprocess ?pp for generating KDP based on three methods , then we use least square method to calculate KDP, and KDP is thereby used for attenuation correction. Validation of attenuation correction is executed by comparing pixel-to-pixel the reflectivity factor of a S-band radar with that of C-band radar before and after attenuation correction. The corrected C-band radar reflectivity factor is getting close to that of S-band radar; moreover, the corrected relationships between KDP vs. ZH and ZDR vs. ZH are also close to the theoretical curve.(2) Because ground clutter will affect the accuracy of rainfall estimation, we should identify it from the radar data. Using reflectivity factor ZH and differential reflectivity factor ZDR can identify ground clutter when C-band dual polarization radar scanning in low elevation.(3) Use the fuzzy logic algorithm to identify the type of precipitation particles in stratiform cloud, 4 kinds of echo parameters are involved to obtain 7 kinds of precipitation particle types. The results shows that: This method can accurately identify the type of precipitation particles in stratiform cloud. The results coincide with the theory of cloud microphysical structure.(4) For rainfall estimation, different schemes lead to diverse results. Rainfall amount estimated from S-band radar is more reliable than that from C-band radar if corresponding Z-R relationships are used. But estimation from C-band radar using R(ZH,KDP) relation with attenuation correction becomes better than that of S-band radar using R(Z) equation. Gage-adjustment is necessary in case of rain-rate over 10 mm/hr. After adjustment, the low-biased radar estimation is calibrated, and the gap between radar estimation and guage measurement goes down.