4.4 Using a Spectrum of OSSEs to Explore Synergies Among Active and Passive Measurements for Spaceborne Estimates of Atmospheric Winds

Monday, 29 January 2024: 5:15 PM
Key 9 (Hilton Baltimore Inner Harbor)
Derek J. Posselt, PhD, JPL, Pasadena, CA; and S. M. Hristova-Veleva, S. C. Tucker, Ph.D., X. Zeng, H. Nguyen, I. Yanovsky, L. Wu, and A. Ouyed

Measurements of the three-dimensional distribution of horizontal winds (hereafter 3D winds) are crucially important for a large number of research and applications activities. In particular, space-based observations of 3D winds have been identified as a major gap in the current observing system. The recent completion of, and future plans for future, space-based Doppler wind lidar (DWL) has yielded promising new results; however, DWL is limited to observations in clear air and thin clouds, and only provides observations along a 2D curtain. Lidar wind measurements can be complemented by wide swath observations from passive infrared (IR) and microwave (MW) sounders, which can provide water vapor images that can be tracked in time to provide atmospheric motion vectors (AMVs). It is likely that an optimal estimate of 3D winds may be provided by leveraging the synergy between DWL and IR and/or MW sounders, but questions remain about the optimal way in which to leverage the information from each platform, as well as the volume of measurements that may be required. Studies are needed to quantify measurement capability and uncertainty globally and regionally and for cases of particular interest to operational agencies (e.g., NOAA for high-impact weather and climate).

We have developed a set of tools that are capable of evaluating various potential observation architectures for a future three dimensional winds observing system. These include: optical flow algorithms for producing AMVs from sequences of water vapor images; a lidar forward model that provides coverage and uncertainty for DWL measurements; retrieval algorithms that provide estimates of measurement resolution and uncertainty; and machine learning techniques that are capable of correcting errors in AMVs using coincident DWL measurements. In this presentation, we provide results for coverage statistics and uncertainties for current measurements and one or more candidate future observing systems.

- Indicates paper has been withdrawn from meeting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