Assimilation of high-resolution GNSS tropospheric delays and its effects on a severe convective event nowcasting

Abstract

A crucial factor limiting convective weather nowcasting is the lack of timely updated and accurate atmospheric water vapor observations. The Global Navigation Satellite System (GNSS) can accurately sense water vapor with high temporal resolutions, which is adequate to observe many meso- and small-scale variations associated with convective weather. In this contribution, an hourly cycling data assimilation system is established to investigate the influence of assimilating GNSS zenith total delays (ZTD) on severe convective weather nowcasting. The contributions of assimilating ZTD with different temporal resolutions are discussed in detail by validating with the radiosonde observations. The results demonstrate that the assimilation of ZTD significantly improves the moisture distribution of the middle and lower troposphere. Furthermore, model simulations become wetter or drier as the frequency of ZTD assimilation increases. Verification of the precipitation forecasts is performed by comparing them with the radar-estimated precipitation. The results indicate that assimilation of GNSS ZTD improves the accuracy of precipitation forecast in the nowcasting range of 0–6 h. Compared to the control experiment, the hourly ZTD assimilation experiment reveals the highest precipitation forecast skill scores, followed by the experiments of assimilating ZTD every three and six hours, indicating that the rapid update of water vapor information could contribute to improving the precipitation nowcasting in a rapidly developing convective system.