Investigating the benefits of 3D passive winds in global weather forecast

Passive 3D winds or atmospheric motion vectors (AMVs) can be derived by tracking the humidity features in the retrieval moisture from hyperspectral infrared sounder radiance. To investigate the impact of these types of AMVs in the global weather forecasts, the Quantitative Observing System Assessment Program (QOSAP) carried out an OSSE using a two-month period of the ECMWF Cubic Octahedral O1280 (ECO1280) Nature Run (NR). Both observations assimilated operationally in June/July 2020 and the 3D passive winds are simulated. Analyses and forecasts are generated using NOAA Global Forecast System (GFS) model. The AMVs are simulated by tracking humidity features in the NR. A cloud mask, which is the NR cloud top pressure grid, is then applied to filter out wind simulations where clouds would obscure. The winds are then sampled based on three polar satellite constellation with local equator crossing times of 5:30, 9:30 and 13:30. The AMVs are produced at eight pressure levels (200, 250, 300, 400, 500, 600, 700, 850 hPa). Forecast metrics in the northern and southern hemispheres have shown that the assimilation of this AMV type has a positive impact.