Comparative Assessment of Specific Differential Phase Retrieval

Dual polarization observations of precipitation have become ubiquitous and furnish a growing number of hydrological applications and products for the weather and climate research communities. Specific differential phase (KDP) has been shown to improve quantitative precipitation estimation (QPE) and aid in interpretation of cloud microphysical processes. Specific differential phase is defined as the range derivative of the differential propagation phase shift. Estimators of specific differential phase can be tricky because estimation of the derivative of a noisy field can become unstable. Thus, KDP retrieval algorithms often use smoothing or other constraints to obtain a stable solution. In the last decade, several newer techniques have been proposed to promote improved estimates of KDP. These algorithms have been optimized (or tuned) for specific radar frequencies or regimes (regional, process), including several newer algorithms that operate on different underlying assumptions about the precipitation medium to obtain a solution. This paper presents a comparative assessment of KDP estimates retrieved using several algorithms employed by the community. The relative algorithm performance will be evaluated based on data collected from radar deployed by the ARM climate research facility. The ARM climate research facility has recently deployed radars at C-band and X-band frequencies (at various fixed and mobile sites), motivated by demands for targeted improvements to quantitative precipitation estimates and process studies to support climate model forcing and model validation.