Evaluation of GFSv16 for Near‐Real‐Time Data Impact Studies During the Atmospheric River Reconnaissance Program 2022

Abstract Landfalling Atmospheric Rivers (ARs) are crucial for the water resources on the U.S. West Coast, but can also cause hazardous weather and water events there. Yet, accurately predicting AR impacts on precipitation remains a challenge. This study investigates the influence of dropsonde data from AR Reconnaissance (AR Recon) on improving weather forecasts in winter 2022, specifically AR‐related precipitation forecasts for the U.S. West Coast. Using operational Global Forecast System version 16 (GFSv16) simulations with and without assimilated dropsonde observations, we evaluated the impact of dropsondes on forecasting key meteorological fields like geopotential height, horizontal wind and water vapor transport, and precipitation. The results demonstrate that assimilating dropsonde data can improve the forecast accuracy of landfalling AR characteristics and associated precipitation, although mixed results were obtained in water vapor transport forecasts. Forecast improvement was achieved for medium to strong AR events, especially when dropsondes targeted the AR cores, showing a large error reduction in integrated vapor transport (approximately 20 kg m −1 s −1 ) and precipitation (by >5 mm d −1 or ∼20%–40%) for strong AR conditions. The analysis indicates that the effectiveness of dropsonde data is influenced by AR characteristics, such as intensity and track, as well as the complex terrain of the U.S. West Coast. These findings highlight the value of targeted field campaigns in improving AR precipitation forecasts and underscore the need for continued research to address remaining uncertainties, ultimately enhancing water resource management and extreme weather preparedness on the West Coast.