Accuracy of ZDR data and implications to operational radar measurements

Radar-meteorological algorithms require an accuracy of differential reflectivity (ZDR) data of about 0.1 dB. In principle, a radar system can be calibrated to provide ZDR with that accuracy. However, there are several sources of bias depending e.g. on azimuth angle or time of observation. This type of bias cannot easily be accounted for by a conventional calibration.

The following sources of systematic ZDR bias are analysed:

• Partial beam blockage from sharp objects (buildings) cause no ZDR bias, but blockage due to trees can cause bias of several dB, depending mainly on the distance to the trees. The bias may vary significantly with season, depending on the type of trees.
• For radars without radome, a slight azimuthal ZDR variation with a magnitude of a few tenth of a dB was found even in unblocked areas. This bias is supposed to be due to the wetting of the antenna dish from rain, which is depending on wind direction. No azimuthal variation was observed in case of snowfall. Rain on radome may cause similar bias depending on the wind direction.
• The rotary joint causes negligible azimuthal ZDR variation. It was found to be of the order of a few hundredth of a dB only.