Theory and verification of bias correction for polarimetric phased array radar

Phased array technology in radar meteorology has gained a lot of attention in the past decade due to multiple benefits it provides such as a smart scanning strategy, and more rapid updates. Similarly, the upgrade to dual polarization capability, is a common trend observed in many operational weather radar networks around the world. Implementation of second polarization channel allows significant improvement in the data quality and provides additional information for accurate weather forecasting. The Microwave Remote Sensing Laboratory at the University of Massachusetts has developed a low cost mobile, polarimetric phase-tilt weather radar (see Figure 1). The PTWR is able to provide polarimetric radar products using the alternate transmission and alternate reception (ATAR) mode of operation. However, this is only possible if polarization channels are properly calibrated. There is a major difference between mechanically-scanned reflector antennas and electronically scanning arrays, where the polarimetric properties of antenna may vary with scan angle. Biases in phased array polarimetry are discussed in Zhang et al (2009), Zrnic et al (2011) and Sanchez- Barbetty et al (2012), among others. The PTWR is a 1-D electronic scanning array, which enables electronic scanning in the azimuth plane, while the array is mechanically tilted in elevation. The common source of errors intrinsic to the array antenna itself is a non-optimal cross-polarization isolation or mismatch of the beam patterns in the two polarizations. These are defined at the system design stage and can vary over time due to aging, temperature changes, or other effects. The other error is related to the misprojection of the co- and cross-polar fields onto the local horizontal and vertical directions. In case of the PTWR architecture, synthesized beams remain in the principle plane of the array and hence the “H” and “V” polarizations remain orthogonal across the scan. However, they do rotate as one scans off-boresight at non-zero elevation angle. The effects of this bias are investigated using real data collected using the PTWR. The correction approach is to proceed with rotated polarizations that will result in self-induced canting angle. Fortunately, this bias can be corrected by an appropriate multiplication of the measured scattering matrix with rotation matrix. This paper provides a brief system overview depicting the key features and limitations of the developed system. Then a discussion on origin of polarization rotation and its impact on polarimetric products quality follows. Next, we will define rotation matrix and show how unbiased measurements can be recovered. Due to the fact that PTWR is utilizing ATAR mode of operation, the additional correction due to Doppler effect will be implemented. Finally, we will verify bias correction procedure with the data collected in medium precipitation event. Since PTWR is mounted on a pedestal, it is possible to imitate the operation of “classical weather radar”, by scanning mechanically using only a broadside beam of phased array antenna. This type of scan, which is not affected by rotation of polarizations, will follow immediately after electronic scan. The data collected in this manner will be then compared with bias corrected data collected using electronic scan. Zhang, G., R.J. Doviak, D.S. Zrnic, J.Crain, D.Staiman, and Y. Al-Rashid (2009), Phased array radar polarimetry for weather sensing: a theoretical formulation for bias corrections, IEEE Trans. Geosci. Remote Sensing, 47 (11), 3679-3689. Zrnic, D.S., G.Zhang, and R.J. Doviak (2011), Bias correction and Doppler measurements for polarimetric phased-array radar, IEEE Trans. Geosci. Remote Sensing, 49(2), 843-853 Sanchez-Barbetty, M., R. W. Jackson, and S.J. Frasier (2012), Interleaved sparse arrays for polarization control of electronically steered phased arrays for meteorological applications, IEEE Trans. Geosci. Remote Sensing, 50 (4), 1283-1290