Absolute Calibration of an X-band Polarimetric Weather Radar: Comparison of Different Methods

Since Summer 2012, MeteoSwiss is equipped with a Gematronik X-band polarimetric weather radar, which is being employed for the detection of wind-shear phenomena and rain rate estimates on Swiss Airports. The main focus of the conducted measurement campaigns is put on the very accurate quantitative estimation of weak to moderate rain rates. Such measurements require a radar system that is calibrated to the highest possible standards in order to guarantee that low rain rates can be detected with satisfying accuracy as well. These high standards in terms of absolute calibration made it necessary to assess a series of different calibration methods and judge their applicability and accuracy. The calibration methods that were used are the following: (1) Corner reflector calibration, (2) comparison of radar derived rain rates with rain gauges, (3) the use of polarimetric self-consistency integrated in a Kalman filter algorithm, (4) comparison with radars from an operational C-band network, (5) usage of the DSD inferred from power spectrum data that is obtained from the vertical pointing of the antenna in stratiform rain and subsequent calculation of the theoretical reflectivity based on the measured DSD, (6) usage of a radar target simulator, i.e., an active transponder system, which allows to simulate an artificial target of a given radar cross section at an arbitrary distance to the radar. Overall, a good consistency was found between the methods, but some of them, e.g., method (1) and (6) are costly to implement and only work for dedicated measurements but not for a continuous monitoring of the radar constant. However, especially at X-band frequencies a continuous monito-ring is necessary since radome attenuation can heavily influence the calibration of the radar. It was found that the methods (3) and (5) can provide very accurate absolute calibration provided that they are applied with care and only during suitable rain conditions. For the self-consistency method these conditions are: large rain cell free of snow or hail with considerable differential phase shift as well as a radar beam that is absolutely free of clutter, partial beam blockage, partial beam filling and side lobe contamination. The vertical pointing method (5) only works in stratiform rain where the radar beam is completely filled. Long term statistics on the behavior the methods during different rain conditions will be presented and discussed.