Apparent Short-Period GNSS-ZTD Disturbance Correlated With Precipitation Events

In recent years, Global Navigation Satellite System (GNSS)-zenith total delay (ZTD)/precipitable water vapor (PWV) has been attempted for exploring or predicting precipitation events. However, current GNSS precipitation prediction models take little account of short-period disturbances. In this study, GNSS is used to show the short-period tropospheric disturbance during strong convection for the first time. Taking Hong Kong as an example, several short periods before, during, and after precipitation events using GNSS-ZTD in April 2019 are analyzed using a bandpass filter. It is verified that the gravity waves-like disturbances are strongly correlated with heavy precipitation, which mainly occurs in the preprecipitation (3–6 h) and midprecipitation periods. The probability of precipitation is 50%, 27.14%, and 22.86% for the early, middle, and late periods of precipitation events. The preprecipitation disturbance demonstrated the ability of GNSS-ZTD used for heavy precipitation prediction. This also provides a new parameter for real-time precipitation forecasting in various machine/deep learning models.