Optimization and impact assessment of Aeolus HLOS wind assimilation in NOAA's global forecast system

Abstract

The European Space Agency Aeolus mission launched the first‐of‐its‐kind space‐borne Doppler wind lidar in August 2018. The Aeolus Level‐2B (L2B) Horizontal Line‐of‐Sight (HLOS) wind observations are integrated into the NOAA Finite‐Volume Cubed‐Sphere Global Forecast System (FV3GFS). Components of the data assimilation system are optimized to increase the forecast impact from these Aeolus observations. Three observing‐system experiments (OSEs) are performed using the Aeolus L2B HLOS winds for the period of August 2–September 16, 2019: a baseline experiment assimilating all observations that are operationally assimilated in NOAA's FV3GFS but without Aeolus; an experiment adding the Aeolus L2B HLOS winds on top of the baseline configuration; and an experiment adding the Aeolus L2B HLOS winds on top of the baseline but also including a total least‐squares (TLS) regression bias correction applied to the HLOS winds. The variances of the Aeolus HLOS wind random errors (i.e., observation errors) are estimated using the Hollingsworth–Lonnberg (HL) method. Results from both OSEs demonstrate positive impact of Aeolus L2B HLOS winds on the NOAA global forecast. The largest impact is seen in the tropical upper troposphere and lower stratosphere where the Day 1–3 wind vector forecast root‐mean‐square error (RMSE) is reduced by up to 4%. Additionally, the assimilation of Aeolus impacts the steering currents ambient to tropical cyclones, resulting in a 15% reduction in track forecast error in the Eastern Pacific basin Day 2–5 forecasts, and a 5% and 20% reduction in track forecast error in the Atlantic basin at Day 2 and Day 5, respectively. In most cases, the additional TLS bias correction increases the positive impact of Aeolus data assimilation in the NOAA global numerical weather prediction (NWP) system when compared to the assimilation of Aeolus without bias correction.

The impact of Aeolus winds on surface wind forecast over tropical ocean and high latitude regions

“…The OSEs with the ECMWF model demonstrated that Aeolus winds are able to improve wind vector and temperature forecasts, especially in the upper troposphere and/or lower stratosphere over tropical and polar regions (Rennie et al, 2021). Similar results were also found from the OSEs with NOAA's Global Forecast System, the DWD model and the Environment and Climate Change Canada global forecast system (Cress et al, 2022;Garrett et al, 2022;Laroche and St-James, 2022). Moreover, regarding the weather and climate events, Aeolus is able to improve the track forecasts for tropical cyclones in the Eastern Pacific basin and Atlantic basin (Garrett et al, 2022).…”

“…Similar results were also found from the OSEs with NOAA's Global Forecast System, the DWD model and the Environment and Climate Change Canada global forecast system (Cress et al, 2022;Garrett et al, 2022;Laroche and St-James, 2022). Moreover, regarding the weather and climate events, Aeolus is able to improve the track forecasts for tropical cyclones in the Eastern Pacific basin and Atlantic basin (Garrett et al, 2022). In addition to this, Aeolus benefits the forecasts of the West African Monsoon and the changes in the El Niño-Southern Oscillation (ENSO) state over the Eastern Pacific by capturing the large-scale atmospheric circulation (Cress et al, 2022).…”

“…For evaluating the contribution of Aeolus observations to NWP, the observing system experiments (OSEs) with and without Aeolus data assimilation have been performed with global NWP models at many institutions, including the ECMWF, National Oceanic and Atmospheric Administration (NOAA), Deutscher Wetterdienst (DWD), Météo-France, UK Met Office, etc. (Cress et al, 2022;Garrett et al, 2022;Forsythe and Halloran, 2022;Pourret et al, 2022;Rennie and Isaksen, 2022). The OSEs with the ECMWF model demonstrated that Aeolus winds are able to improve wind vector and temperature forecasts, especially in the upper troposphere and/or lower stratosphere over tropical and polar regions (Rennie et al, 2021).…”

Investigation of dynamical scenarios leading to particularly high impact of Aeolus on NWP forecasts

“…The study introduced the experimental setup, the data quality and consistency and provided an overview of the systematic changes in the analysis and impact on forecast error. The results overall confirms the promising statistical improvements demonstrated in the global impact studies of Garrett et al (2022); Laroche and St-James (2022); Pourret et al (2022) and Rennie et al (2021).…”

“…Their assimilation into global NWP models led to a significant impact on the analysis and forecasts over Europe and around the development of tropical cyclones, as well as their interaction with the midlatitude waveguide (Pu et al, 2010;Weissmann and Cardinali, 2007;Weissmann et al, 2012). Since the launch of the Aeolus satellite in August 2018, several observing system experiment (OSE) studies have been conducted to investigate the impact of the Rayleigh and Mie HLOS wind observations in various NWP models (Garrett et al, 2022;Laroche and St-James, 2022;Pourret et al, 2022;Rennie et al, 2021). Most of these studies concentrated on global forecast error statistics showing overall large improvements.…”