A Method for Retrieving Thermodynamic Atmospheric Profiles Using Microwave Radiometers of Meteorological Observation Networks

Ground-based microwave radiometers (MWRs) are used to provide temperature and relative humidity (RH) profiles mainly at boundary layer with temporal resolution of 1-2 min. However, accuracy of MWR away from radiosondes locations is not satisfactory. In this study, we proposed temperature and RH retrieval models based on fifth-generation reanalysis (ERA5) data that allow the replacement of the original model based on adjacent radiosondes. This study focuses on how to resolve the unrepresentativeness of ERA5 data. A quality control (QC) scheme was proposed after analyzing the characteristics of ERA5 data, which was confirmed to have positive effects for retrieval results. Three retrieval models for MWR at Nanjiao site based on ERA5 data (ERA5-retrieval), co-located radiosondes (RS-retrieval 1), and adjacent radiosondes (RS-retrieval 2) were developed. An extensive validation that was carried out showed ERA5-retrieval improved the accuracy of temperature and RH profiles compared with the original method RS-retrieval 2 and LV2 product from MWR. Furthermore, the proposed retrieval method was applied to six MWR sites without radiosondes co-located in Beijing and illustrated the feasibility and accuracy of the proposed method. In overall, the retrieval approach proposed based on ERA5 data and data QC can improve both temperature and RH accuracy for MWR sites away from radiosondes.