High-frequency (HF) radar data, derived from a pair of newly developed radar stations in the Pearl River Estuary (PRE) of China, were validated through comparison with in situ surface buoys, ADCP measurements, and model simulations in this study. Since no in situ observations are available in the radar observing domain, a regional high-resolution ocean model covering the entire PRE and its adjacent seas was first established and validated with in situ measurements, and then the HF radar data quality was examined against the model simulations. The results show that mean flows and tidal ellipses derived from the in situ buoys and ADCP were in very good agreement with the model. The model–radar data comparison indicated that the radar obtained the best data quality within the central overlapping area between the two radar stations, with the errors increasing toward the coast and the open ocean. Near the coast, the radar data quality was affected by coastlines and islands that prevent HF radar from delivering high-quality information for determining surface currents. This is one of the major drawbacks of the HF radar technique. Toward the open ocean, where the wind is the only dominant forcing on the tidal currents, we found that the poor data quality was most likely contaminated by data inversion algorithms from the Shangchuan radar station. A hybrid machine-learning-based inversion algorithm including traditional electromagnetic analysis and physical oceanography factors is needed to develop and improve radar data quality. A new radar observing network with about 10 radar stations is developing in the PRE and its adjacent shelf, this work assesses the data quality of the existing radars and identifies the error sources, serving as the first step toward the full deployment of the entire radar network.

The accuracy of temperature and relative humidity (RH) profiles retrieved by the ground-based microwave radiometer (MWR) is crucial for meteorological research. In this study, the four-year measurements of brightness temperature measured by the microwave radiometer from Huangpu meteorological station in Guangzhou, China, and the radiosonde data from the Qingyuan meteorological station (70 km northwest of Huangpu station) during the years from 2018 to 2021 are compared with the sonde data. To make a detailed comparison on the performance of machine learning models in retrieving the temperature and RH profiles, four machine learning algorithms, namely Deep Learning (DL), Gradient Boosting Machine (GBM), Extreme Gradient Boosting (XGBoost) and Random Forest (RF), are employed and verified. The results show that the DL model performs the best in temperature retrieval (with the root-mean-square error and the correlation coefficient of 2.36 and 0.98, respectively), while the RH of the four machine learning methods shows different excellence at different altitude levels. The integrated machine learning (ML) RH method is proposed here, in which a certain method with the minimum RMSE is selected from the four methods of DL, GBM, XGBoost and RF for a certain altitude level. Two cases on 29 January 2021 and on 10 February 2021 are used for illustration. The case on 29 January 2021 illustrates that the DL model is suitable for temperature retrieval and the ML model is suitable for RH retrieval in Guangzhou. The case on 10 February 2021 shows that the ML RH method reaches over 85% before precipitation, implying the application of the ML RH method in pre-precipitation warnings.