Dual-polarization radar retrieval during Typhoon Lekima (2019): Seeking the best-fitting shape–slope relationship depending on the differential–horizontal reflectivity relationship

Abstract

In recent years, Doppler radars have been or are being upgraded worldwide with dual-polarization capability, and polarimetric radar data are used extensively in retrieving the raindrop size distribution (RSD) to understand the microphysical processes of precipitation. However, the conventional constrained–gamma (C–G) retrieval method for dual-polarization radar is sensitive to local shape–slope (μ–Λ) relationships. This explains the major discrepancies between retrieval results based on different μ–Λ relationships derived from various rain regions of Typhoon Lekima (2019). Three μ–Λ relationships and corresponding differential–horizontal reflectivity (ZDR–ZH) relationships are recategorized according to precipitation types that are dominated by small, midsize, and large raindrops. This linkage is then used to modify the C–G retrieval method with a best-fitting μ–Λ relationship for each of the three ZDR–ZH relationships. The modified method is verified in terms of overcoming deficiencies in the conventional method, such as the insufficiency of retrieved samples when using the μ–Λ relationship from eyewall rain and the underestimation of retrieved raindrop diameters when using the μ–Λ relationship of outer rainbands. The modified method is also more able to elucidate microphysical processes such as collision–coalescence process.