Spatial variability of stratiform precipitation estimated from the X-band radar measurements

Understanding errors and uncertainties of the spaceborne microwave radiometer rainfall retrieval algorithms requires knowledge of the variability of rainfall parameters on spatial scales smaller than resolutions of these radiometers. This paper addresses the issue of determining parameters that describe this variability in terms of radar reflectivites and rainfall rates. The high resolution (112 m) polarimetric X-band radar measurements were used for deriving statistics. The random component due to fluctuating nature of radar echos was removed from observed radar reflectivity measurements which were corrected for the partial attenuation using differential phase shift data. The radar measurements were converted then to rainfall rates using the algorithms specifically tailored for the X-band polarimetric radar data. The variability of rainfall fields was assessed in terms of the relative standard deviations of bot radar reflectivity and rainfall rate for wide intervals of these parameters divided in several bins. The spatial scale of variability studies varied from 500 m to several kilometers. The smallest spatial scales exhibit the smallest relative variability and there is a monotonic increase when the spatial scale gets larger. The probabilistic results obtained in this study can be satisfactorily approximated by the gamma function which is convenient for the modeling of the microwave radiometer retrievals. Results are presented for stratiform rains and snow.