Monitoring polarimetric weather radar calibration

Calibration of weather radars is essential for accurate measurement of precipitation and for operational applications should be maintained continuously and be as automated as possible. In case of polarimetric radars, relations containing both reflectivity Z and differential reflectivity ZDR can lessen the effects of drops size distribution variation on quantitative precipitation measurements provided both variables are free of instrument bias. Establishing absolute bias (calibration) requires significant efforts and is hard to automate the usual approach is to perform it at reasonable time intervals but monitor and correct drifts automatically and continuously. This paper proposed use of ground clutter to monitor calibration from scan to scan.

It is demonstrated that urban ground clutter has sufficiently stable reflectivity factor and differential reflectivity. The median reflectivity and the median differential reflectivity from urban ground clutter are good indicators of stability of polarimetric radar and thus can be used to solve the precision part of radar calibration problem.

In the presence of up to moderate rain or snow a reflectivity threshold suffices for separating significant clutter from precipitation observed with a 3 cm wavelength radar. A test on a month of data suggests the scheme is worthy considering for operational applications because it can neutralize variations as small as 0.5 dB in system gains and 0.05 dB in differential gains.

Preliminary test on 10 cm wavelength polarimetric radar indicates that a similar procedure is feasible. The stability is being tested and will be reported.