Abstract: Radiosonde data from October 2021 to September 2022 were used to evaluate the data quality of two types of ground-based microwave radiometers (MWRs),MWP967KV MWR at Xingtai station and QFW-6000A MWR at Tangshan station,in Hebei province.The analysis focused on temperature and humidity retrieval errors at different altitudes,seasons,day-night transitions,and precipitation conditions.The frequency and average intensity of temperature inversions were calculated,and the sensitivity of MWR temperature profile data was analyzed.The results show that the quality of MWR temperature and humidity data is significantly influenced by altitude,season,and diurnal factors,with notable differences between the two instrument types.Under non-precipitation conditions,temperature data quality at 4-8 km and relative humidity data quality at 1-4 km were relatively better; during precipitation,the data quality of both temperature and humidity deteriorated for both instruments.Both MWR types were insensitive to slight temperature changes and their temperature profiles unable to effectively reflect weak temperature inversions,suggesting the need for retrieval algorithm optimization to improve inversion detection performance.