Extended Abstract (200K)P1.10
Ongoing Japanese long-term Reanalysis project (JRA-25); Assimilation of NOAA polar-orbiter satellite sounder data
Masami Sakamoto, Japan Meteorological Agency, Tokyo, Japan; and S. Kobayashi, K. Kato, T. Matsumoto, H. Koide, K. Onogi, T. Ose, and H. Hatsushika
In 2001 we started a 26-year (1979 - 2004) reanalysis under a collaboration project between Japan Meteorological Agency (JMA) and Central Research Institute of Electric Power Industry (CRIEPI). JMA's operational spectral forecast model and 3D-VAR assimilation technique are hired with horizontal resolution of T106 and vertical L40. Basically, ERA-40's conventional and satellite sounding observation data, provided by European Centre for Medium-Rage Weather Forecasts (ECMWF), is adopted. Newly reprocessed Geostationary Meteorological Satellite's (GMS) Atmospheric Motion Vector (AMV) and snow depth data over Chinese territory have been added as original data. Tropical Cyclone wind retrieval data, which was developed by Dr. Michael Fiorino in Lawrence Livermore National Laboratory, are also adapted. The final reanalysis data production was started in April 2004. So far, our system reproduces more relevant distribution of precipitation than other major reanalyses.
3D-VAR assimilation technique allows various types of data to be directly assimilated. JMA has never experienced direct assimilation of Non-advanced (up to NOAA-14) TIROS Operational Vertical Sounder (TOVS) operationally. As the first attempt at TOVS direct assimilation, we developed simple, but effective quality control (QC) and automatic bias correction scheme on the basis JMA's operational Advanced TOVS (ATOVS) assimilation method. To balance horizontal distribution of assimilated data, time window length for TOVS observation was optimized. In some cases when single Stratospheric Sounding Unit (SSU) was available, we find that our assimilation system shows rather stable behavior in stratosphere than ERA-40.
Supplementary URL: http://www.jreap.org/index-e.html
Poster Session 1, Poster Session 1
Monday, 10 January 2005, 2:30 PM-4:00 PM
Previous paper Next paper