4DVAR physics minutes 07/20/06
Participants: Hans, Jimy, Qingnong, XIaoyan
Contents |
Hans
Hans has discussed with Dale and Bill to get Jimy involved in the phys tl/ad work. It is up to Jimy to find out whether he has time to work 20% on the project (0.1 FTE for half a year).
Although Hans has no FTE left on the task, he promised to help Xiaoyan to run 4dvar. He will also follow the progress and coordinate the work.
He stressed that the goal for 4dvar in the AFWA project and NPOESS project is an operational implementation on a 45 km innerloop. Simple and efficient phys tl/ad routines are needed.
Jimy (0.1 FTE AFWA)
Jimy will find out whether he has time to use 0.1 FTE before next Feburary.
He will provide nonlinear code to Qingnong and Xiaoyan in the coming months:
- Surface coupling
- K(z)
- A simple condensation routine
If there are still time left, he will look into
- A simple PBL
- A simple convection scheme.
Qingnong (0.15 FTE AFWA + 0.125 FTE NPOESS = 0.275 FTE)
Qingnong agreed to make sure he has time for the 0.275 FTE (plus 0.1 FTE on the parallelization work). (Qingnong - could you please provide an estimate on how much of your total 0.375 FTE have already been spent on para and phys?)
He will help Xiaoyan in gradient test and adjoint test.
He will develp tl and ad of Jimy's nonlinear code.
Xiaoyan (0.25 FTE NPOESS)
Xiaoyan will discuss with Qingnong on her work plan for the coming months and make sure she has 3 months full time on the phys tl/ad work.
She will help Qingnong analyze nonlinear code and prepare them for TAF. She will also learn TAF from Qingnong.
Her major tasks will be testing: nonlinear test (simply compare nonlinear runs), gradient test, adjoint test and 4dvar test. Qingnong and Hans will help her.
Remarks
- Hans agreed to give the latest version of wrf 4dvar to Jimy and Xiaoyan within the week, and provide help to get everyone started.
- Jimy and Qingnong agreed to press on the work before they go to the WRF-GRAPES workshop in China in September.