Mobile high-resolution X-band polarimetric Doppler weather radar measurements (XPOL): Evaluation and application.

摘要

This Ph.D dissertation focuses on applications of a mobile high resolution X-band polarimetric Doppler weather radar observations in quantitative rainfall and microphysical estimation. X-band tends to be an attractive radar frequency for hydrologists and hydrometeorologists who are more interesting in high-resolution measurements over small watersheds. However, the drawback with X-band radar is severe attenuation of the electromagnetic signal in significant rainfall, which affects the radar observations and introduces errors in rainfall estimation.; The major advantage of the polarimetric weather radar is that it has the ability to transmit and receive both horizontal and vertical polarization. This capability introduces two radar measurements apart from the horizontal reflectivity (ZH). These are the differential reflectivity (ZDR), which is the ratio of horizontal (H) to vertical (V) polarization and the differential phase shift (phiDP), which is the difference in phase between the H and V polarization signals.; This additional information helps to increase the correlation (r 2 > 0.95) between attenuation-corrected (National Observatory of Athens X-band polarimetric) XPOL versus the non-attenuated ZH and ZDR X-band parameters derived from (NCAR S-band polarimetric radar) S-Pol. Error statistics show that the selected algorithm with the least systematic error than the other methods and axial ratio models, converge to below 10% (50%) at path integrated attenuation (differential PIA) values greater than 10 dB (2.5 dB). Overall, the combined uncertainty in the estimation of specific and differential attenuation parameters represent about 28% (in ZH) and 38% (in ZDR).; The first part of this thesis focuses on the development of an algorithm that corrects for rain-path attenuation. The second part of this thesis describes a methodology that estimates drop size distribution (DSD) from the attenuation-corrected radar measurements. Two algorithms that estimate the three-parameter 'normalized' Gamma DSD model are developed for X-band radar polarimetric observations and compared against S-Pol radar and disdrometer spectra observations. The constrained-gamma method is so named because of the constrained mu-Λ relation and the "beta" beta is so named because of the estimation of the mean axis ratio of drops.; From the statistical analysis and comparisons of disdrometer spectra observations and S-Pol DSD retrievals, it is found that the beta-method introduces errors from the use of KDP, while the constrained-method works reasonably well at low and high rain rates and provides relatively accurate retrieval of the DSD parameters. Error statistics show that the beta-method introduces an additional 20% and 30% error in NW and mu while for the estimation of D 0 both algorithms have similar performance.