Monitoring absolute calibration of a DualPolarization Weather Radar

The German Meteorological Service (DWD) is operating a C-Band polarimetric weather radar network with 17 operational systems. The installation of these systems have been finished early 2015. The new polarimetric moments are primarily used in algorithms for better quantitative precipitation estimation and identification of hydrometeors. In parallel to the introduction of the new radar systems, a new radar data processing chain has been developed which incorporates those algorithms. An important component of the data processing scheme is the radar data quality control. This includes also the monitoring of the radar hardware state such as e.g. antenna pointing but also the accuracy of the radar moments. In this contribution we present the approach to monitor absolute calibration in both polarizations using data from an operational available birdbath scan and in-situ disdrometer data. In order to match radar and disdrometer data a number of assumption have to be made (see e.g. Frech, 2013), because radar data at the beginning of the farfield of the antenna (650 m above the radar) and disdrometer data at the surface are considered. A crucial assumption is that the precipitation process in the vertical column between the farfield and the surface is stationary. The initial analysis shows, that if the data is filtered for stratiform situations, avoiding brightband effects, very good agreement can be achieved between the two sensors. In this contribution we use a co-located vertically looking micro-rain radar (MRR) in order to verify the assumptions. A MRR was operated for the full warm season in 2014 close to the DWD's research radar at Hohenpeißenberg. This setup is ideal as sampling problems related to different time and volume scales are minimized. Here, we analyze the variability of the radar moments in the column between 650 m and the surface in order to verify and optimize the assumption for the disdrometer and radar comparison.

The analysis shows, that the bias between the MRR (rangebin at 650 m) and the disdrometer at the surface is -0.8 dB, and -0.9 dB for the C-Band radar. The bias between the C-Band radar and the MRR at 650 m is 0.5 dB. The results are within the target accuracy of 1 dB for absolute calibration. The results show, that disdrometer measurements close to the surface can be used to monitor the absolute calibration of the C-Band system.

References: Frech, M, 2013, Monitoring the data quality of the new polarimetric weather radar network of the German Meteorological Service, 36rd AMS Conf. on Radar Meteorology, Breckenridge, CO, USA, 16p.