Abstract Dual-polarization (dual-pol) radar variables provide information about the quantity, type, size, and water content of hydrometeors. Assimilating these dual-pol radar variables into numerical weather prediction models can enhance forecast accuracy. Observation operators are essential for radar data assimilation. This study focuses on applying a realistic dual-pol radar observation operator to more accurately calculate dual-pol radar variables. Previously reported dual-pol radar observation operators tended to overestimate radar variables near 0°C in convective precipitation and simulate unrealistic dual-pol radar variables in subfreezing regions. To address this, the improved operator (KNU dual-pol radar observation operator; K-DROP) limits the distribution of mixed-phase hydrometeors, which have both solid and liquid properties, in areas with strong updrafts and downdrafts, improving the overestimation of radar variables near the melting layer. Additionally, by applying the observed snow axis ratio during winter to K-DROP, the issue of differential reflectivity ( Z DR ) being calculated as a constant value in subfreezing regions has been improved. By incorporating the observed maximum radius of hydrometeors into K-DROP, the overestimation of reflectivity ( Z H ) in subfreezing regions, the overestimation of Z DR in warm regions, and the underestimation of specific differential phase ( K DP ) in subfreezing regions and overestimation in warm regions, are improved. Compared to previous operators, the enhanced version reported in the present work produces more realistic dual-pol radar variables.
Lee et al. (Sat,) studied this question.