Thursday, 7 August 2003
Polarimetric radar rainfall estimation at X-band: evaluation of attenuation correction
The most commonly used polarimetric radar measurements in rainfall estimation are the reflectivity factor, usually at horizontal polarization (Zh), differential reflectivity (Zdr) and specific differential propagation phase (Kdp). Based on the above three measurements, a number of algorithms have been derived in the literature to estimate rainfall. These algorithms have been derived assuming equilibrium raindrop shapes, described by a specific shape-size relationship. The mean axis ratio versus size relation is crucial for deriving algorithms that use Zdr and Kdp. Chandrasekar et al. (1990) indicated, from Kdp based analysis at S band, that at low rainfall rates, the radar rainfall algorithms have large measurement error. However, Kdp directly scales with frequency, and has a higher dynamic range at X band. This feature makes Kdp at X-band more useful at low rain rates (Matrosov et al. 2002 ). Along with the advantages, X-band measurements come with their own problems, the significant ones being, attenuation, differential attenuation and phase shift on backscatter. In addition, Kdp-based polarimetric radar rainfall algorithms are influenced by deviation from the equilibrium shape of raindrops. Gorgucci et al. (2000) demonstrated that the slope of a linear mean raindrop shape-size relation can be estimated from polarimetric radar measurements which can be used subsequently in rainfall rate estimation parameters. Testud et al. (2000), Bringi and Chandrasekar (2001) presented procedures to correct for attenuation using dual polarization radar observations. Thus any study of polarimetric rainfall algorithms at attenuating frequencies such as X needs to consider the impact of the precipitation path, that exists between resolution volume and the radar. This paper presents an extensive evaluation of the microphysical characteristics of the various precipitation paths on the estimation of the rainfall in the resolution volume. This paper also presents comparison of algorithms to estimate rainfall rate using polarimetric radar measurements at S- and X-band as well as investigate the various effects discussed here. The results show that the variability of the microphysical characteristics of rainfall (such as variability of the DSD parameters) on the propagation path plays an equally important role as the intensity of precipitation along the path in determining the accuracy of the rainfall estimates.
Supplementary URL: http://