Reflectivity field of storms in central Oklahoma processed using traditional sampling (top) and oversampling and whitening (bottom). As expected, “smoother” fields obtained with oversampling and whitening are an indication of reflectivity estimates with lower statistical errors than those obtained with traditional processing.
Range oversampling followed by a decorrelation transformation is a novel method for increasing the number of independent samples from which to estimate the Doppler spectrum, its moments, as well as several polarimetric variables on pulsed weather radars. Since errors of estimates increase with increased antenna rotation speed (for the same spatial resolution), the decreased errors associated with decorrelation permit the antenna to rotate faster while maintaining the current errors of estimates. It follows that storms can be surveyed much faster than is possible with current processing methods. Alternatively, for a given volume scanning time, errors of estimates can be greatly reduced. These are important considerations in WSR-88D operations. This technique can be advantageously exploited in a combination of faster data temporal acquisition and denser spatial sampling as needed to satisfy some of the evolutionary requirements for the NEXRAD network.
Recently, we've focused our research on the practical issues involving the implementation of oversampling and adaptive pseudowhitening techniques within the WSR-88D operational environment. The NWRT PAR is a natural platform for range oversampling research because, by default, the system oversamples in range. Initially, a simple pseudowhitening strategy was implemented and tested on the NWRT using a fixed transformation matrix. The success of this evaluation led to the design and implementation of an adaptive pseudowhitening algorithm. This adaptive algorithm chooses the proper pseudowhitening transformation to get the optimum trade-off between variance reduction and noise enhancement for different conditions. This work represents a significant step towards establishing range oversampling techniques as operationally viable on weather surveillance radars.
For a tutorial on range oversampling click here. If you want all the nitty-gritty details, please read two of my papers that appeared on the Journal of Oceanic and Atmospheric Technology in November and December of 2003.
More recent work is documented in these two AMS Radar Conference papers from 2005 and 2007, these Journal of Oceanic and Atmospheric Technology papers from 2009, 2011, and 2012, and this AMS Annual Meeting paper from 2010.
The Spring 2014 software release on the NWRT PAR is now operational. This new release enables multifunction capabilities and adds new signal processing and adaptive scanning functionalities.
During PARISE 2013, NSSL scientists worked with NWS forecasters to understand the strengths and limitations of weather-radar scan strategies in their warning-decision process.
We recently received National Science Foundation (NSF) funding for our research project titled "Understanding Polarimetric Radar Tornadic Debris Signatures Using Modeling, Simulations, and Field Measurements."