- Related Research Areas
Direct assimilation of satellite radiance data has significantly benefited numerical weather prediction (NWP). The number of radiance data sensing the atmosphere will continue to increase significantly during the next decade. Therefore, improving the assimilation of radiances become increasingly important. The PI proposes to improve the assimilation of radiance data from NASA EOS and NOAA weather satellites for global NWP by developing ensemble capabilities on the variational data assimilation system, GSI, which both NASA and NOAA use for their operational data assimilation. The improved data assimilation system, called hybrid ensemble-GSI, combines the advantages of the variational and ensemble-based data assimilation methods. Compared to GSI, the hybrid incorporates flow-dependent ensemble covariance to estimate the background error covariance so that the background forecast and the observations are more appropriately weighted during the assimilation. Compared to the traditional ensemble Kalman filter (EnKF), the hybrid method, by using mathematically and physically appropriate model space covariance localization, avoids inappropriate assumptions of explicit vertical location of the satellite radiances. The hybrid is also more robust when a small ensemble is used. The objectives of the research proposal are to (1) develop the hybrid system based on GSI using extended control variable method, (2) select effective ensemble generation method for the hybrid system, (3) examine the effectiveness of the new system, as compared to the operational GSI, and the EnKF, to assimilate satellite radiance data, and (4) study the impact of the radiance data on top of other observations revealed by the three data assimilation schemes. Two months of 6 hourly assimilation cycles and forecasts up to 7-day lead times will be conducted. We will verify common meteorological parameters on multiple pressure levels, and for different regions of the globe including Northern Hemisphere, Southern Hemisphere and Tropics. We will also evaluate the hurricane track forecast for both the Atlantic and Pacific basins for the selected summer month. This proposal also contains a significant education component built on and integrated into the existing education program in the National Weather Center (NWC) at Norman, Oklahoma, (which houses the PI’s home institution,) for the purpose of training the next generation satellite data assimilation experts. Besides training the graduate research assistant and the postdoc who will work on the proposed research, the PI will develop and teach new course material on satellite data assimilation in her current graduate level atmospheric data assimilation class; mentor undergraduate students, especially female students, selected nationwide, on research using NASA satellite data through NWC REU(Research Experience for Undergraduate students) program; and participate NWC outreach AVID (Advancement Via Individual Determination) and Upward Bound programs, for high school students, especially under-represented groups on the concepts of remote sensing and numerical weather forecast and to emphasize the importance of knowledge on STEM in a science career. The proposed educational activities will be closely tied with the research component. The research proposal directly addresses one major goal in Weather Focus Area of NASA Earth Science Research Program, which is to effectively assimilate satellite data, including for example, data from NASA EOS sensors, into research and operational weather forecast models in order to improve and extend U.S. and global weather prediction. The education proposal directly addresses NASA SMD (Science Mission Directorate) Earth Science education goals to inspire the next generation of Earth explorer.
Project PI: Xuguang Wang/University of Oklahoma
NWC 5341,School of Meteorology, College of Atmospheric & Geographic Sciences ,The University of Oklahoma,National Weather Center,Website Admin 120 David L. Boren Blvd., Suite 5900 Norman, Oklahoma 73072.
Phone: (405) 325-3426
Nothing to see here at the moment. Check back later.
Log in to start a discussion.
- Any registered users can join
- Anybody can view this project
- Any registered users can leave comments
- Anybody can view comments
- Joined 4 years, 6 months ago
Visit our help center