Quality Control Of Ground-based Microwave Radiometer Observations And Parameters Retrieval Study Based On Cloudy Samples

This paper focuses on two aspects of the data quality control and inversion application of the ground-based microwave radiometer.First,we compare two radiative transfer modes,the MWGRND and the monoRTM,and judge the working state of the radiometer using the simulation data based on the radiative transfer mode MWGRND.Then distinguish and select the "clear sky" and "cloud sky" sample and establish the correction scheme,for the purpose of preliminary quality control on the observed brightness temperature.Finally,the second time quality control of the”cloud sky,sample observation brightness temperature are carried on for the purpose of retrieving cloud parameters and atmospheric temperature and humidity profile.Firstly,we get the data of the atmospheric stratification by reanalysis the data of NCEP FNL 6 hours,and obtained the whole sample after time matching between NCEP FNL and the ground-based microwave radiometer.Inputing the NCEP FNL profile data to the radiation transmission mode MWGRND,we calculate the simulated brightness temperature under clear air condition,and analyze the consistency between the observed data and the simulated data of the whole sample for judging the working state of the microwave radiometer during the observation period.Secondly,based on the whole sample consistency analysis,we drawing the method of selecting the "clear sky" and "cdoud sky",namely "the minimum brightness temperature group approach" and "the maximum brightness temperature group approach".Based on the "clear sky,samples,we build the system deviation correction scheme,and carry out the preliminary quality control on the whole sample.Finally,we analysis the relationship between downward radiation brightness temperature and cloud parameters and temperature and humidity profile based on the”cloud sky,samples selected by the adjusted brightness temperature data.The "cloud sky,brightness temperature are consist of two parts including the "atmospheric radiation contribution1,and the "cloud radiation contribution",it is necessary to carry out the second time quality control for different purposes of inversion.Retrieval of the cloud parameters are based on the "cloud radiation contribution" part using multiple linear regression method.Retrieval of the temperature profile and humidity profile required”cloud effect correction",excluding the"cloud radiation contribution" and based only on the "atmospheric radiation contribution" part using a neural network inversion method.