Consistency Evaluation of Precipitable Water Vapor Derived From ERA5, ERA-Interim, GNSS, and Radiosondes Over China

As an important component of atmosphere, water vapor is highly involved in the global water cycle and energy exchange. To date, Global Navigation Satellite System (GNSS) and radiosonde techniques have been widely used to detect the water vapor content in the atmosphere. Recently, European Centre for Medium-Range Weather Forecasts (ECMWF) has released the latest reanalysis data set, namely, ECMWF Re-Analysis (ERA5), whose temporal and spatial resolutions have a significant improvement over its previous-generation product of ERA-Interim. This study aims to assess the consistency of precipitable water vapors (PWVs) derived from ERA5, ERA-Interim, GNSS, and radiosondes during the entire year of 2017. The GNSS-derived PWVs were obtained at 41 Crustal Movement Observation Network of China stations, while the radiosonde-derived PWVs were acquired at the adjacent 41 radiosonde stations. The nationwide PWVs derived from the ERA5 and ERA-Interim show root-mean-square (RMS) errors of 1.8 and 2.1 mm with respect to the GNSS PWVs and 2.7 and 2.8 mm with respect to the radiosonde PWVs, respectively. Besides, the RMS errors exhibit significant regional and seasonal differences. The nationwide relative RMS of the ERA5- and ERA-Interim-derived PWVs are 11.1% and 13.4% with respect to the GNSS PWVs and 16.2% and 17.8% with respect to the radiosonde PWVs, respectively. The relative RMS values show significant difference between the east and west sides of Hu line across China. Furthermore, the nationwide PWVs are obtained using GNSS data sets at over 200 Crustal Movement Observation Network of China stations to compare with the ERA5-derived PWVs in China. Results indicate that the spatial distribution of the PWVs derived from the two data sources is quite consistent, suggesting a great application prospect of the ERA5 products in China.