PNNL

Abstract

Improving the estimation accuracy and reducing the time required for measurement are governing goals for any radar design. Polarimetric weather radar measures the polarization covariance matrix of the signal returns from precipitation volumes in addition to the Doppler parameters. Increasing the equivalent number of independent samples in any estimation process results in decrease in the standard deviation of estimates. Oversampling pulsed Doppler radar returns at a rate larger than the pulse bandwidth, whitening the range samples and subsequent averaging have been pursued as a potential way to decrease the measured standard deviation of signal parameters estimates. It has been shown that the application of oversampling, whitening and subsequent averaging improves the quality of reflectivity and mean velocity estimates in agreement with theory; Oversampled data collected from CSU-CHILL radar are analyzed to evaluate the performance of dual-polarization parameter estimators such as differential reflectivity and differential phase. The reasons that may limit the improvement in estimation quality of polarimetric parameters are investigated. It is demonstrated that the observation of the variability of range signals within subpulses is important for obtaining maximum variance reduction through whitening. Accurate measurement of the amplitude and phase of the transmitted pulse is critically important for effective whitening of the received waveform. The differential phase pattern between the transmit pulses is found to be critical for obtaining unbiased and accurate estimates of polarimetric parameters through whitening. CSU-CHILL radar's transmit pulse sampling capability is used to evaluate the impact of waveforms on oversampling and estimation.