Weather Observation Using the Atmospheric Imaging Radar and Adaptive Beamforming

With the need for faster update times, the next generation of weather radars is likely to exploit phased-array technology. Rapid evolution of severe thunderstorms, and associated hazards such as tornadoes and damaging wind, requires higher temporal resolution, which can increase warning lead-time and improve forecasters' confidence. Furthermore, high temporal resolution can lead to better scientific understanding of rapidly evolving weather phenomena, such as tornadogenesis or rapid changes in tornado intensity. Research has been conducted using the National Weather Radar Testbed phased-array radar (NWRT PAR) to demonstrate that phased-array weather radars can achieve faster update times while maintaining data quality. An advantage of phased-array radars that has not been fully investigated is the application of digital beamforming for weather observations. With adaptive beamforming, the radar can adapt the beam pattern pulse-to-pulse and gate-to-gate to reduce the impact of ground clutter and other interference on the estimation of meteorological variables. Imaging techniques that were previously used in the profiling community can be used with weather radars to achieve even faster volume update times. Many adaptive beamforming methods have been developed for applications involving signals that can be modeled as point sources. However, the received signal by weather radars is from distributed targets, namely the hydrometeors that fill spaces much larger than the radar resolution volume. Thus, direct application of adaptive beamforming methods intended for point sources to weather radars could lead to significant biases in estimated signal parameters. This work will discuss the challenges of using adaptive beamforming for weather radars and present some necessary bias correction methods so that meteorological variables can be accurately estimated. Simulated data and severe storm data collected by the University of Oklahoma's (OU) Atmospheric Imaging Radar (AIR) will be used to demonstrate the effectiveness of the algorithms. Recent field experiments have provided data with 6 – 10 s update times from tornadic and non-tornadic supercells, and squall lines. These data will be used to illustrate the improvement in the estimation of meteorological variables, and highlight some potential avenues for future meteorological research with such unique, rapid-scan volumetric radar measurements.