Cooperative Institute for Mesoscale Meteorological Studies

RESEARCH

 

NOAA Strategic Goal 3: Serve Society’s Need for Weather and Water Information

Doppler Weather Radar Research and Development

NSSL Project 7 – Investigation of Advancements in Radar Technology toward the Improvement of Hazardous Weather Detection and Warnings:

Improvement of Spectral Moment and Polarimetric Variable Estimates using Decorrelating Transformations on Oversampled Range Data

Torres (primary – CIMMS at NSSL)

Funding Type: CIMMS Task II

Objectives
Exploit range oversampling followed by a decorrelation transformation for faster data temporal acquisition and denser spatial sampling as needed to satisfy some of the evolutionary requirements for the WSR- 88D.

Accomplishments
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, 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 exploited in a combination of faster data temporal acquisition and denser spatial sampling as needed to satisfy some of the evolutionary requirements for the WSR-88D.

During the past year, we continued our focus on practical issues involving the implementation of oversampling and pseudo-whitening techniques within the WSR-88D operational environment. It was observed that if the amplitude and/or phase mismatch between transmission pulses is disregarded in the formulation of the decorrelation transformation, processing of range oversampled dual-polarization signals with the standard whitening transformation can produce biased polarimetric variable estimates. Our research demonstrated that, by properly accounting for the amplitude and/or phase differences in the two polarization channels, it is always possible to obtain unbiased polarimetric variable estimates. However, the accuracy of these estimators may degrade as the degree of mismatch between the horizontally and the vertically polarized transmitted pulses increases. Optimum estimators can be derived by solving a constrained minimization problem.

This project is ongoing.

Publications
Torres, S. M., 2007: Range oversampling techniques for polarimetric radars with dual transmitters. Preprints, 33rd International Conf. on Radar Meteorology, Cairns, Australia, Amer. Meteor. Soc., P7.5.

Bias and standard deviation of cross-correlation coefficient estimates vs. amplitude and phase polarimetric channel mismatches

Bias and standard deviation of cross-correlation coefficient estimates vs. amplitude and phase polarimetric channel mismatches. Different curves correspond to (a) traditional matched filtering (MFB), (b) oversampling and averaging (OAB), (c) original whitening transformation (WTB), (d) unbiased whitening transformation (UWTB), and (e) optimum, unbiased whitening transformation (OUWTB). OUWTB estimates are unbiased and exhibit minimum variance for the entire range of amplitude and phase channel mismatches.