7.5 Impact of Assimilating AMV Winds Using a Variational Feature Track Correction (VarFTC)

Tuesday, 30 January 2024: 2:45 PM
Key 9 (Hilton Baltimore Inner Harbor)
Ross N. Hoffman, NOAA, Cambridge, MA; and H. Liu, K. E. Lukens, K. Garrett, and K. Ide

Handout (6.2 MB)

Atmospheric Motion Vector (AMV) winds are assimilated in operational NWP systems and show positive impact on global weather forecasts. Still there exist issues in characterizing the errors of these observations and how to best assimilate them. First, AMVs may have biases due to height assignment errors. Second, AMVs may have additional wind speed biases. Third, AMVs are representative of layer motion with depth up to a few hundred hPa.

To reduce the impact of AMV errors on numerical weather prediction, Hoffman et al. (2022, doi:10.1002/qj.4207) proposed a variational feature track correction (VarFTC) method in which an observation operator optimally averages the NWP background winds in the vertical. Here, a prototype VarFTC observation operator is implemented in the NCEP gridpoint statistical interpolation (GSI) data assimilation (DA) system. The parameters describing the vertical averaging are determined offline based on previous DA cycles. The VarFTC operator reduces observation minus background (OmB) standard deviation of AMVs by up to 10%. The AMV observation error estimates are reduced accordingly in the GSI when using the VarFTC operator.

The European Space Agency (ESA) Aeolus mission provided high vertical resolution, highly accurate global 3D wind profiles observed from space. Because of these characteristics the Aeolus winds are expected to provide anchoring information, i.e., information to help constrain the VarFTC. In addition, it was demonstrated that the Aeolus winds improved FV3GFS global forecasts particularly in the upper troposphere and lower stratosphere of the Tropics and Southern Hemisphere (Garrett et al., 2022, doi:10.1002/qj.4331).

Global Observing System Experiments (OSEs) are performed comparing the VarFTC operator to the operational AMV observation operator with and without the assimilation of Aeolus Lidar wind profiles. The OSEs show that the VarFTC operator improves NCEP operational global forecast skill. The presence of Aeolus winds enhances the positive impact of the FTC operator.

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner