Real-time measurement of the range correlation for range oversampling processing

Abstract

As range-oversampling processing has become more practical for weather radars, implementation issues have become important to ensure the best possible performance. For example, all of the linear transformations that have been utilized for range-oversampling processing directly depend on the normalized range correlation matrix. Hence, accurately measuring the correlation in range time is essential to avoid reflectivity biases and to ensure the expected variance reduction. Although the range correlation should be relatively stable over time, hardware changes and drift due to changing environmental conditions can have measurable effects on the modified pulse. To reliably track changes in the range correlation, an automated real-time method is needed that does not interfere with normal data collection. A method is proposed that uses range-oversampled data from operational radar scans and that works with radar returns from both weather and ground clutter. In this paper, the method is described, tested using simulations, and validated with time series data. © 2013 American Meteorological Society.