The 28^th BMB-NCAR Project Meeting

Time: 2:00 pm~3:30 pm, 13 July, 2006

Participants: Bill Kuo, Xiang-yu Huang, Yong-run Guo and Min Chen

1. Dr Kuo gave a brief review about the working status of 3 projects of BMB-NCAR collaboration. He pointed out that there are just a few months left for the end of BMB-NCAR GPS project. One important part of this project is the real-time operational application of GPS data in NWP. However, the task hasn’t been completed on time due to some real problems such as the unstable transmission network, etc. The current goal is to assimilate real-time GPS data within 1.5hours into the operational numerical forecast system of BMB before the end of July. Director Zhang suggested another setup of WRF model with 10 km resolution for the assimilation of GPS data. As Zhong and Fan has given great efforts on the current setup of WRF model with 27-9-3km, we have to well coordinate GPS and 3DVAR projects.

2. Dr Huang mentioned that the current setup of WRF model has existed. The 27 and 9 km domains are designed for operational runs and the 3 km is for experiments. Zhong has done a lot of valuable work on the bug fix of WRFVAR codes and the data quality control for the assimilation of GPS and AWS data. So the 27km domain should have no problem for the operational run.

3. Min introduced some information about the unreasonable cold bias of the current WRF operational forecast. Besides the two suggestions (a. repeat the tests of the single 27 domain and check the results; b. do some experiments without wrfvar;) provided by Drs Guo and Huang, another advice is that we can use MM5’s 9 km background error covariance to replace that of WRF, which could be problematic.

4.Dr Kuo suggested that we could have 3 options for the operational assimilation of GPS data on NWP of BMB.

1) To assimilate all available data for WRF’s 27 km domain, and then generate initial conditions for 9 km domain via interpolation from 27 km. About the configuration of physics package, Zhong’s results of sensitive experiments under the guidance of Wei Wang can be regarded as a reference. Or the similar physics package to MM5 can be used.

2) In August, GPS and all of other available data can be assimilated only on the 27km domain and on both 27 and 9 km domains after August.

3) Parallel to the operational MM5 run, WRF will run operationally with the assimilation of all available data, and also run with assimilation of all available data except GPS. But the latter run is not for real-time.

5. A telephone conference will be held at 6:00 pm 17 July (next Tuesday).