Using Bragg Scatter to Estimate Systematic Differential Reflectivity Biases on Operational WSR-88Ds

The National Weather Service recently completed a dual polarization upgrade to its fleet of 160 operational WSR-88Ds. The upgrade provides the opportunity to improve estimates of precipitation using the new Quantitative Precipitation Estimate (QPE) algorithm. The QPE algorithm uses differential reflectivity (ZDR) to estimate rain rates in light to heavy rain. However, radar calibration errors can lead to systematic ZDR biases and QPE biases. For example, at constant reflectivity, a 0.25 dB bias in ZDR will yield about a 20% error in rain rate estimates. Currently, the Radar Operations Center (ROC) uses plan position indicator (PPI) scans of light rain (weather method) to monitor systematic ZDR biases. While the weather method estimates of ZDR biases can identify radar calibration trends, they do not have sufficient accuracy mainly due to contamination by big drops. This study examines the feasibility of using Bragg scatter to estimate systematic ZDR biases on operational WSR-88Ds. Because Bragg scattering (turbulence at the top of the convective boundary layer) is not polarized, its values of ZDR should be near zero. Therefore, any biases away from zero represent a systematic ZDR bias. Data filters using the radar base data moments, dual polarization parameters, and statistical filters were used to isolate Bragg scatter from clutter, biota, and precipitation. Six radars were examined in detail for May and June 2013 from 1400-2200 UTC each day. Systematic ZDR bias estimates from Bragg scatter were compared to reliable estimates from the weather method. Bragg scatter derived systematic ZDR biases were comparable to those estimated by the weather method; most cases were within 0.20 dB. More cases of Bragg scattering were found in May than in June (23 and 15, respectively). This study demonstrates that Bragg scattering offers an alternative method for monitoring systematic ZDR biases on the WSR-88D fleet.