Zdr calibration instability: an experimental analysis of the responsible radar components

Calibrating Zdr to a tenth of a dB is a challenging endeavor. Probably the most accepted technique for Zdr calibration is the use of vertical pointing (VP) data in light precipitation, however, there are other techniques such as the crosspolar power (CP) method, which relies on the principle of radar reciprocity. The VP and CP techniques have been show to be equivalent via experimental data from NCAR's (National Center for Atmospheric Research) S-Pol. The CP technique requires the measurement of the two crosspolar powers and the measurement of the horizontal and vertical solar powers. NCAR has developed a technique where the CP Zdr bias can be estimated from a single solar scan in an automated fashion. S-Pol participated in a 2014-2015 winter experiment. Using the VP technique in the winter is difficult since the storms can be shallow and the ice crystals can have near zero velocities (prone to ground clutter contamination from side lobes) and thus the CP technique is more practical. CP technique revealed that the Zdr bias is a strong function of the ambient temperature at the S-Pol site with a dependence of about 0.01dB/Co. Measurements show that antenna is the source of the Zdr bias variation. The cause of this temperature dependence along with a general analysis of Zdr variability for S-Pol is presented in this paper.