Abstract
The effect of rainfall inhomogeneity within the sensor field of view (FOV) affects significantly the accuracy of rainfall retrievals causing the so-called beam-filling error. Observational analyses of Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) and Precipitation Radar (PR) data suggest that the beam-filling error can be classified in terms of the mean rain rate and the rainfall inhomogeneity parameter or coefficient of variation (CVR, standard deviation divided by mean). The dependence of the beam-filling error on the rain rate and CVR has been confirmed quantitatively using a single channel at 19.4 GHz. It is also found significantly different beam-filling errors for the two different regions, the East and West Pacific, where the spatial and vertical distributions of rainfalls are different. It is also observed that the vertical distribution of rainfall is related to the spatial variability of rainfall (CVR) and similarly to the spatial variability of TMI 85.5 GHz brightness temperature (CV Tb). Based on these findings, this study exploits the CV Tb to correct the beam-filling error in a direct inversion from a rainfall (R) and brightness temperature (T b) curve at a single frequency, and to reduce the retrieval error in the context of a Bayesian-type inversion method for multi-frequency rainfall retrievals. Both the experiments suggest that the spatial variability of the high-frequency radiometer data appears to contain useful information for retrievals.
Acknowledgement
This work is supported by the Korea Foundation for International Cooperation of Science & Technology (KICOS) through a grant provided by the Korean Ministry of Education, Science and Technology (MEST) (K20701010357-07B0100-10 710) and Korea Research Foundation programme (KRF-C00306).