Simulation of the brightness temperature of spaceborne microwave radiometer by using COBRA data

One of the important issues for the rain retrievals with a spaceborne microwave radiometers is the drop size distribution (DSD) models, because the relationship between the brightness temperature (Tb) and the rain rate is generally calculated from the prefixed rain structure model with a fixed DSD model (typically Marshall-Palmer DSD). For this reason, a rain model with the information of DSD, phase, density, fall velocity and so on is required to evaluate the retrieval error. Multi-parameter radar is suitable to establish such kind of rain model. In this study, brightness temperatures of low frequency channels of a spaceborne microwave radiometer are simulated by using ground based multi-parameter radar (COBRA: CRL Okinawa bistatic multiparameter radar) data to construct the basis of the rain model and are compared with the simultaneous observation by TRMM/TMI. Since the target frequencies are 10 and 19 GHz, liquid phase precipitation is mainly considered. Two approaches are tried to simulate the Tb field: 1) use of the variable Z-R relationship with a fixed exponent and 2) the DSD data from multiparameter analysis by COBRA. In the first approach, the optimum coefficient in the Z-R relationship is searched in each pixel of TMI to match the observed Tb for each frequency. The result shows that the spatial distributions of the optimum coefficient at the both frequencies agree well with each other, supporting the validity of the chosen DSD model (Z-R relationship). On the other hand, the preliminary result with the DSD parameters estimated from COBRA data gives larger amplitude of Tbs than the observed Tbs indicating that some parameter tunings are needed in the DSD estimation by COBRA.