KDP Estimation Based on Spline Smoothing with the Self-Consistency Principle

We present a specific differential phase (KDP) estimation method developed in the PREVENIR project between Argentina and Japan, supported by Science and Technology Research Partnership for Sustainable Development (SATREPS) in collaboration between Japan Science and Technology Agency (JST, JPMJSA2109) and Japan International Cooperation Agency (JICA). KDP is an important polarimetric parameter because it is strongly correlated with the rainfall intensity. However, since KDP is derived from an observed noisy differential propagation phase (PHIDP) as its derivative information, it is inherently difficult to estimate small KDP values while preserving the resolution of radar data. To accurately estimate KDP while preserving the original resolution as much as possible, the proposed method exploits the self-consistency principle with the radar reflectivity factor. Specifically, below the melting layer, the initial KDP estimate is first obtained with the ZPHI method as preprocessing. Then, we construct a clean PHIDP as a monotonically increasing smooth spline function that matches the initial KDP in the light rain area and fits the observed PHIDP well in the heavy rain area. The final KDP estimate is derived instantaneously from the derivative information of the clean spline PHIDP. We applied the proposed method to C-band radar data acquired in Buenos Aires, Argentina, and confirmed that the proposed method gives a high-resolution KDP estimate consistent with both observed PHIDP and reflectivity factor.