Determining the optimal sampling interval of atmospheric water vapor with GNSS methods

Global Navigation Satellite System (GNSS) estimates of precipitable water vapor (PW) provide a continuous and all-weather way to monitor water vapor through the atmospheric column. These estimates can be assimilated into numerical weather prediction (NWP) models to improve their predictive capability, and can also be used to observe spatial and temporal variations of atmospheric moisture associated with rapidly changing weather events. Networks of GNSS stations can now continuously stream data, potentially providing PW estimates with temporal resolution of less than a minute. This study attempts to determine the optimal time sampling of GNSS derived column water vapor products. PW estimates with too fine of a temporal resolution will oversample the same atmospheric column while increasing the observational noise of each individual data point. Estimates with too coarse of a resolution will under-sample the moisture field when conditions are rapidly evolving. Answering these questions is important if GNSS observations are to be combined with other observational systems and used in very short-term forecasting applications.