VPR correction of bright band effects in radar QPEs using polarimetric radar observations

Abstract

Vertical profile of reflectivity (VPR) correction of bright band (BB) effects has been a challenge for single‐polarization radar quantitative precipitation estimations (QPEs) for mesoscale convective systems and for cool season stratiform precipitation when the freezing level is low. BB is often found in the radar observations of stratiform precipitation, and the inflated reflectivity intensities in the BB often cause positive biases in radar QPEs. A VPR correction is desirable to mitigate the BB contamination and reduce the bias. However, a well‐defined BB bottom, while critical for an effective correction of the bias, is often not found in the VPRs. Fortunately, polarimetric radar variables, especially the copolar correlation coefficient (ρHV), can provide a much better depiction of vertical BB structure than does reflectivity. In the current study, an apparent vertical profile of ρHV (AVPρHV) correction scheme is developed. For each tilt of the radar volume scan data, the precipitation echoes are segregated into convective and stratiform regions. An apparent VPR (AVPR) and AVPρHV are computed for the stratiform region in the given tilt. Then the bright band top, peak, and bottom are identified from the AVPR and AVPρHV, and a linear VPR correction model is fit to the AVPR in the BB layer. VPR corrections are applied to the reflectivity field in the given tilt based on the linear correction model, and radar QPEs are derived from the corrected reflectivity field. The new AVPR and AVPρHV combined scheme was tested on two mesoscale convective system events and one cold season event in the United States and was shown to be more effective in reducing the radar QPE bias associated with the BB than did the AVPR correction alone.