The Seventh BMB-NCAR Project Meeting

Time: 03:30~05:00 pm, Sep 16, 2005

Participants: Dr. Huang, Dr. Guo, Shui-Yong Fan

1. Discuss the project progress

· Paralell runs

Shui-Yong found the results of OPR 3dvar are the same on three platforms, bluesky, big linux and pc linux. It was agreed that EXP00 could be used as the standard reference.

EXP00: run OPR (the BMB operational system) on NCAR’s bluesky to get a standard reference.

Shui-Yong also completed EXP02.

EXP02: change KMA winter BES to KMA summer BES and tune some parameters including RF_pass, var_scaling, len_scaling and check_rh.

Shui-Yong showed the verification results of EXP00 and EXP02. The scores of EXP00, EXP01 and EXP02 were compared. The scores of EXP00 were considered to be reasonable. EXP02 has in general marginally better scores than those of EXP00 and EXP01, especially when the forecast length is longer than 24h.

Shui-Yong has made an experiment list to record every experiment.

· Gen_be work

Shui-Yong has completed the coding of gen_be for mm5 and is now debugging the code.

· Verification

Dr. Huang, Dr. Guo, Min Chen and Shui-Yong have discussed observation quality control in verification during the past two weeks.

Dr. Guo suggested to compute vertically integrated BIAS and RMS scores.

Dr. Huang suggested to look into the standard output files to check, e.g., initial and final cost functions, the ratio J_min/N_obs, some innovation statistics, etc.

2. Make a task list

· Clean up all the diagrams and make them easier to compare. Write some text for the results obtained so far.

· Do the single-obs tests using the EXP02 configuration.

· Re-run 3DVAR of EXP02 and make statistics of the initial and final cost functions, and some innovation statistics

· Run EXP03, which has the same configuration as EXP02 except ncar3dvar is replaced by wrfvar_2.1.

· Continue the stage0 work.

3. Next meeting

Next meeting may be at 2:00 pm, 23 Sept, if Shui-Yong has completed some work by that time. Otherwise, 2:00 pm, 10 Oct.