Statistical Biases and Standard Deviations of Polarimetric Variables.

摘要

Simultaneous transmission and reception of electromagnetic waves with horizontal and vertical polarizations has been chosen as a main mode for polarimetric prototype for the WSR-88D networking weather radar (Doviak et al. 2000 ). We will refer to this mode as SHV , i.e., Simultaneous transmission and reception of Horizontally and Vertically polarized waves. In the mode, six basic radar moments are measured in each radar resolution volume: reflectivity (Z), Doppler velocity (v), spectrum width (ov), differential reflectivity (ZDR), the differential phase (odp), and the modulus of the copolar correlation coefficient (nco). The first three values are the base radar moments in terms of products of the WSR-88D, the latter three are base polarimetric moments. An additional polarimetric variable, the specific differential phase (Kdp), is used in rainfall measurements and hail recognition and is calculated from odp. Radar signals reflected from weather objects are random. The estimates of the radar moments are calculated on a pulse train consisting of a given number of samples M. The statistical errors of the estimates are characterized with bias and standard deviations. Main properties of the base radar estimates are well known (Doviak and Zrnic 1993). The standard deviations of the polarimetric variables in SHV mode for strong signal-to-noise ratios are known also (see, e.g., Bringi and Chandrasekar, 2001). Here we consider noise influence on the radar moments and calculate biases of polarimetric variables. We also consider some new algorithms to calculate polarimetric moments that have no noise dependent bias.