The Gauging and Modeling of Rivers in the Sky

Most of the horizontal water vapor flux outside of the tropics occurs within atmospheric rivers (ARs). These synoptic features are behind extreme precipitation and flooding in coastal mountainous regions and they can affect the atmospheric dynamics and predictability. For their impacts to be skillfully predicted, it is important for weather forecasting systems to be able to represent AR properties. In order to evaluate a model’s ability to capture AR characteristics, observations of ARs are needed, and the only approach currently available is by probing ARs with research aircraft in a targeted field campaign.

Using the European Centre for Medium-Range Weather Forecasts Integrated Forecasting System (IFS) and dropsonde observations taken during the 2018 AR Reconnaissance field campaign over the Northeast Pacific Ocean, results show that although the AR structure is modeled well, the short-range water vapor flux forecasts in the IFS have a root mean square error of 60.0 kgm^-1s^-1 (21.9% of mean observed flux). The errors are shown to be most related to uncertainties in the winds near the top of the planetary boundary layer. It is suggested that as well as implementing model improvements, the initialization of the forecasts could improve through the assimilation of additional space-based observations, such as from AEOLUS and the Meteosat Third Generation Infrared Sounder, and from dropsondes in future AR Recon field campaigns. In conclusion, the findings highlight a possible barrier in the forecasting of high-impact weather and suggests a research area where efforts should be focused to improve weather forecasting systems.