Physical Characteristics of Precipitation Types using Three Dimension Radar datasets

The various precipitation systems have different physical characteristics caused by developing mechanisms. Therefore, observation based on studies such as precipitation types from radar composite is important to achieve an improved understanding of these mechanisms. In this study, we attempt to develop the Classification algorithm of precipitation types. Then, we explore the properties of rainfall types in terms of dynamical, microphysical, and thermo-dynamical process.
The algorithm is consisting of two methods, one is the SL3D (storm labeling in three dimensions) and the other is the classification process using feature parameters obtained from VIL (vertical integrated liquid water contents) and mean reflectivity. The 8 categories of cloud regions can be discerned using this algorithm. The physical characteristics of precipitation system (precipitation stratiform, convection, updraft, deep system, and shallow system) are investigated using information associated with the vertical structure of both dual-polarization observation (ZH, ZDR, KDP and ρHV) and dynamical attributes (vertical velocity, vorticity, and divergence). In addition, the generalized microphysical parameters (Dm, log N0’) in the rain regions (altitude below melting layer) are analyzed for precipitation types. Those parameters are obtained using polynomial regressions consisting of ZH and ZDR. The latent heat (LH) acquired from GPM products are examined to identity the thermodynamical properties.
The peak values of vertical profiles for radar observations indicating the melting layer height have appeared at 4.5~5.0 km. The convergence (divergence) patterns are distinctive in lower (higher) atmospheric layers for convection and updraft regions. Both microphysical parameters and LH for convection and updraft regions are relatively higher than those of stratiform.