Polarimetric Radar QPE Based on a New Differential Phase-Based Variational Approach

Quantitative precipitation estimation (QPE) with polarimetric radar observations suffers from different sources of uncertainty. The variational method appears to be a promising way to optimize the radar QPE statistically. In this study, a new variational approach is developed to quantitatively estimate rainfall rate (R) from the differential phase (Φ_DP). A spatial spline filter is utilized in the optimization procedures to eliminate the impact of the random fluctuations in Φ_DP, which can be the major error source in the specific differential phase (K_DP)-based QPE. Additional constraints such as the self-consistency with the horizontal reflectivity factor (Z_H) and the differential reflectivity (Z_DR) can be incorporated into the variational framework easily. The approach is evaluated by an idealized case and a real rainfall case observed by an operational S-band polarimetric radar in South China. The results demonstrate that with a proper spatial filter used, the proposed variational radar QPE agrees well with the rain gauge observations.