Quantitative Evaluation Of Microwave Brightness Temperature Data Of FY3-D Meteorological Satellite

Atmosphere temperature and humidity are important parameters that affect global climate and climate change,and play an increasingly important role in global climate and meteorological research.The Fengyun(FY3)-D satellite was successfully launched on 15 November 2017 and carried the Microwave Humidity and Temperature Sounder(MWHTS)and Microwave Temperature Sounder(MWTS)on board which can realize the all-weather vertical detection of global atmosphere humidity and temperature profiles,and is of great value for improving the accuracy of global numerical weather prediction.The key of microwave atmosphere detection is quantification.Therefore,accuracy is the most important index for microwave radiometer observation data.In this paper,the quantitative evaluation research for MWHTS/MWTS microwave radiometer observation data is carried out based on two mainstream satellite remote sensing data evaluation schemes.The first scheme is to make cross comparison with the observation of same type microwave radiometer,the ATMS on S-NPP satellite is selected as the reference sensor.The quality of the MWHTS/MWTS's brightness temperature data is evaluated by making cross comparison with ATMS observation data at the same frequency.However,the result of the evaluation method is affected by the data consistency in time and space,which contributes to strong limitations.Therefore,the second scheme is more commonly used—comparing the brightness temperature observed by satellite microwave radiometer with the brightness temperature simulated by RT forward model,which can realize data consistency in time and space.The second scheme will also bring a new problem,that is,the difference between observation brightness temperature and simulated brightness temperature(referred as “O-B” or OMB)is actually caused by the mixture of three complex factors--input atmospheric parameter error?RT model calculation error and microwave sounder observation error,but there is no way to make error separation between above factors in mainstream research.In order to evaluate the brightness temperature of FY3-D satellite more accurately,this paper will study the separation of error factors based on the mathematical model and statistical data of observed and simulated brightness temperature difference,and estimate the observation brightness temperature error of microwave radiometer separately.The brightness temperature observed by MWTS and ATMS were compared to demonstrate the root mean square error(RMSE)for every channel is less than 1.1K.The brightness temperature observed by MWHTS and ATMS were compared to suggest that RMSE for channel 13 and channel 15 is less than 0.6K,while RMSE for other comparable channels is bigger than 1.5K.Furthermore,the difference between MWTS observation and radiative transfer(RT)model simulations,referred as “O-B”,suggest that RMSE for channel 1 to 3 and channel 11 is less than that,RMSE for other channels is bigger than that for corresponding ATMS channels.The “O-B” differences for MWHTS suggest that RMSE for channel 3 to 6 which is near the center of 118.75 GHz oxygen absorption line is less than 1K,and RMSE for other channels is similar to that for corresponding ATMS channels.In conclusion,for channel 1 to 11 and channel 11,the quality of brightness temperature observed by MWTS is better than corresponding ATMS channels,while for other channels,the quality of brightness temperature observed by MWTS is a little weaker than corresponding ATMS channels and that the quality of brightness temperature observed by MWHTS is better than corresponding ATMS channels except for channel 1.Furthermore,the observation error of microwave radiometer obtained by error separation accords with theoretical expectation,which verifies the rationality of error separation model to a certain extent.