Abstract: To take full advantage of FY-4A temperature profile data to understand the evolutions of weather processes and nowcasting, based on the atmospheric temperature profile of radiosonde, microwave radiometer and FY-4A satellite from 1 January 2021 to 31 March 2022, 897 samples are matched and their deviation characteristics are evaluated. The results show that the correlation coefficient between FY-4A satellite temperature and that of microwave radiometer is 0.9891, and the correlation coefficient between FY-4A satellite temperature and that of radiosonde is 0.9820. The mean temperature of FY-4A satellite is 0.51℃ smaller than that of the radiosonde below 10 km height, and the standard deviation is 0.50℃. The mean temperature of FY-4A satellite is 0.53℃ larger than that of the microwave radiometer below 10 km, and the standard deviation is 0.75℃. FY-4A temperature is consistent with the mean deviation trend of the radiosonde at 0000 UTC and 1200 UTC. Compared with 0000 UTC, the deviation sample of FY-4A at 1200 UTC is less discrete. When there is precipitation, the temperature deviation of microwave radiometer and FY-4A gradually increases above 600 m height, and the deviation reaches the maximum (about 9.35℃) near 1500 m height. In the range of 3000-8500 m height, the deviation ranges from 1.35℃ to 5.10℃, and the standard deviation ranges from 1.41℃ to 4.99℃. In the case of precipitation, the deviation values and standard deviations of FY-4A temperature and radiosonde are small. Although the deviation values and standard deviations of FY-4A temperature and radiosonde are different at different heights in the whole layer, the deviation is between -0.31 and 3.60℃. When there are clouds, the mean deviation between FY-4A temperature and that of microwave radiometer is -0.40℃, and the mean standard deviation is 3.79℃. The overall mean deviation between FY-4A temperature and radiosonde is 0.31℃, and the mean standard deviation is 2.66℃. Both the deviation and standard deviation between FY-4A temperature and that of microwave radiometer are larger than those between FY-4A and radiosonde when there are clouds. The deviation and standard deviation of microwave radiometer temperature and that of radiosonde with FY-4A are small in clear sky. The above conclusions can provide reference for the further use of FY-4A satellite data, as well as for the quality control of FY-4A satellite data and its application in weather analysis and forecast.