Assimilating GNSS Airborne Radio Occultation Bending Angles With MPAS–JEDI During a Sequence of Atmospheric Rivers

Abstract

Global Navigation Satellite System (GNSS) airborne radio occultation (ARO) observations, collected during Atmospheric River Reconnaissance (AR Recon) campaigns, enhance AR sampling beyond traditional dropsondes. Assimilating ARO bending angles requires a forward operator accounting for ARO geometry and proper integration in a data assimilation system. This study implements a modified ARO bending angle operator in the Joint Effort for Data Assimilation Integration (JEDI) system. Using the Model for Prediction Across Scales–Atmosphere (MPAS–A) coupled with JEDI, we evaluate the impact of assimilating ARO bending angles in a case study of an 11-flight AR sequence over the United States West Coast in January 2023 through 6-hourly cycling observation system experiments. Observation errors follow methods for spaceborne RO, yielding promising results; however, more suitable error characterization is needed and may benefit from consideration of the regional environment. Background humidity, temperature, and wind fields show improvements over the northeast Pacific, fitting dropsondes temperature more closely. While impacts on short-term forecasts are modest, clear improvements are found in 1–7 day forecasts across all variables, reflecting a better-constrained synoptic steering flow. ARO assimilation enhances intensity and location of integrated vapor transport as well as precipitation forecasts, particularly at longer lead times. These findings indicate that ARO integration into MPAS–JEDI has reached a level of maturity suitable for broader implementation in operational and research forecast models. Encouraging results from dense AR sampling suggest that global forecasting benefits could be achieved if ARO technology were installed on more aircraft and applied across a wider range of cases.