Quantitative Error Analysis of Polarimetric Phased Array Weather Radar Measurements

Phased array radars (PARs) are an emerging technology in the meteorological community. They offer the advantage by providing rapid and timely information that greatly enhances the understanding of severe weather events as they unfold. The 2D electronic scan (E-scan) PARs have great advantages in multifunctionality and flexibility, but achieving high-quality polarimetric weather measurements with such systems is very difficult. As the development of PARs continues, the polarimetric data quality of PARs should be assessed for further meteorological applications. The initial weather measurements, acquired from two polarimetric phased array radars (PPAR) with cylindrical and planar configurations developed and located at the University of Oklahoma (OU; 35.18°N, 97.53°W), were compared with those from the operational KTLX Weather Surveillance Radar-1988 Doppler (WSR-88D) with a dish antenna located in Oklahoma City, Oklahoma (~27 km northeast of OU). The purpose of this comparison is to investigate the error statistics of the polarimetric measurements from each radar during convective rain events. The first event occurred on 29 August 2019, when the cylindrical PPAR (CPPAR) performed a 3.3° elevation plan-position indicator scan at 25 azimuth angles. The second event took place on May 11, 2023, when Horus, a recently functioning S-band fully digital polarimetric phased array with a planar configuration, conducted range-height indicator (RHI) scans based on a single column of 5 panels. The standard deviations based on spatial samples of 11 gates and temporal samples of two consecutive time steps on the assumption of local stationarity and ergodicity are calculated and compared. For further comparisons of the two configurations, volumetric polarimetric radar data provided by KTLX serves as a reference. To ensure temporal and spatial alignment between the radars, reconstructed RHI scans and PPI sectors from KTLX were matched to the corresponding Horus rays and CPPAR domain, respectively. The standard deviations, mean biases, and differences in the measurement results of the two 2D PARs will be presented.