Microphysical Characteristics of Raindrop Size Distribution and Implications for Radar Rainfall Estimation Over the Northeastern Tibetan Plateau

Abstract

AbstractRaindrop size distribution (DSD) observations with OTT PARSIVEL2 disdrometers at Dari and Delingha in the northeast of the Tibetan Plateau (TP) during summer 2019 and 2020 were utilized to investigate the DSD characteristics and DSD‐based polarimetric radar quantitative precipitation estimation (QPE) relations on the X, Ku, and Ka bands. The DSD characteristics were examined for different rain rates (R) and for stratiform and convective rainfall types. The raindrop spectra became wider and flatter as the R increased. Gamma parameters N0, μ, and λ (representing intercept, shape, and slope parameters, respectively) were smaller for convective compared to stratiform rainfall, corresponding to the broader and flatter convective raindrop spectra. The μ–λ and Dm–log10Nw relationships were also examined in our study. There was a significant μ–λ correlation, with a quadratic function relationship. The summer convective rainfall at both Dari and Delingha showed continental‐like clusters based on the Dm–log10Nw relationship. The DSD‐based two‐parameter polarimetric radar QPE relations (R(Zh, ZDR), R(KDP, ZDR)), and one‐parameter relations (R(Zh), R(KDP)) at the X, Ku, and Ka bands were proposed, and the R computed from DSD was implemented to evaluate these relations. The performance of two‐parameter models was stronger than that of one‐parameter models. Moreover, R(Zh, ZDR) and R(KDP, ZDR) performed better for the X and Ku bands, while the R(KDP, ZDR) model performed best for the Ka band. The ground‐based Ka/Ku band measurement Z at Dari was used to evaluate the Z–R and R−Dm QPE relations. The R−Dm relations had a better performance compared to the Z–R relations. The Ku radar‐derived R series are more consistent with the DSD measured counterparts. The research of the dual‐polarization radar parameter models is vital to develop radar QPE algorithms on the TP.