Recognition of lightning echoes with the polarimetric WSR-88D

Two types of lightning effects on polarimetric radar data are considered: 1) echoes from lightning and 2) noise contamination due to broad-band radiation from lightning. The copolar correlation coefficient ρhv drops significantly in lightning echoes, which can cause problems in automatic classification of hydrometeors. Radial extents of lightning echoes can exceed 5 km which is comparable with the radial extent of hail cores. Polarimetric and spectral parameters of lightning echoes are analyzed and classification rules are presented for the polarimetric WSR-88D. The rules are based on measurements of ρhv and the differential phase. It is also shown that the level of Doppler spectrum skirts in the polarimetric channels can be used for lightning recognition. It is found that the lightning recognition rules can be applied for a small number of radar samples, e.g., 3-4, wherein the drop in ρhv is deeper than for 40-60 radar samples. This can be used to recognize and track lightning with rapidly scanning radars such as dual-polarization phased array radars.

The manifestation of the second lightning effect, i.e., increased noise, is seen as streaks of low ρhv in radar echoes. This effect can be mitigated by using autocorrelation estimators of polarimetric variables, i. e. by using estimators immune to noise influence. The performance of the estimators is demonstrated on data collected with dual-polarization WSR-88D.