An Improved Attenuation Correction Methodology for Polarimetric Weather Radar

Attenuation correction of measured reflectivity (Z_H) and differential reflectivity (Z_DR) is critical in short wavelength radar applications, especially during extreme precipitation events such as heavy rain. This paper develops an improved attenuation correction method based on the self-consistency of polarimetric radar observables. In particular, a non-negative constraint on differential propagation phase (Φ_DP) is applied for traditional ZPHI method. The copolar correlation coefficient (ρ_HV) is used to further partition the Φ_DP profiles to range segments characterized by different hydrometeor phases such as pure liquid rain or mixed-phase precipitation. In addition, a cost function on the difference between processed Φ_DP and reconstructed Φ_DP is applied to ensure convergence in the computation. Data collected by a C-band polarimetric radar (Hangzhou-CPOL) at Da’ming mountain in Hangzhou of China during two extreme precipitation events are used to demonstrate the improved attenuation correction method. Measurements from disdrometers and a nearby radar operating at non-attenuated frequency (i.e., S-band) at Huangshan mountain (HSM) are used to quantitatively evaluate the attenuation correction performance of Hangzhou-CPOL. This paper also investigates the impact of the improved attenuation correction on quantitative precipitation estimation. Results show that the polarimetric radar measurements from Hangzhou-CPOL are effectively enhanced and are more consistent with S-band observations as well as simulated radar moments based on disdrometer data. Hourly rainfall products derived from R(Z_H) and R(Z_H, Z_DR) are significantly improved due to the enhanced attenuation correction, and their performance is comparable to R(K_DP)-based estimates.