Monday, 21 January 2008: 9:15 AM
Medium-range ensemble forecasting with a local ensemble transform Kalman filter and JMA bred- and singular-vector EPS
219 (Ernest N. Morial Convention Center)
A local ensemble transform Kalman filter (LETKF) system is now under development at the Japan Meteorological Agency (JMA) for a possible future operational choice. 51-member ensemble prediction experiments using the LETKF system are performed and compared with the operational bred-vector (BV) system and the singular-vector (SV) system which will be in operations in November 2007. The analysis ensemble mean for the BV and SV systems is provided by the operational four-dimensional variational (4D-Var) data assimilation system, different from the analysis by LETKF. A typhoon case has indicated advantages of LETKF. This presentation includes several verifications of ensemble predictions by the LETKF, BV, and SV systems, as well as latest progress up to the time of conference.