Unknown reason for difference between idealized runs in x- and y- directions

[littleLi]

To whom it may concern,

I am trying to implement a new surface drag method in WRF, so I want to check the symmetry in x- and y- directions.

The point I started at is the default WRF (no new code implemented), 2d idealized bell shaped mountain case (WRF/test/em_hill2d_x). I changed the default namelist.input to higher resolution, and changed the related part of code in dyn_em/module_initialize_ideal.F, namelist.input, and input_sounding to get a symmetric run in y-direction. Surprisingly, the two results are not the same.

Then I tried to test the asymmetry I saw above, I changed the run from 2d to 3d. The mountain is still bell shaped, and totally symmetric in x- and y- directions. The only difference between my two runs are the input_sounding files, the winds are kept at 10m/s and potential temperatures are kept at 273K at all levels, the only difference is the wind is from west/south. Picture attached is drawn by ncview and you can see the obvious asymmetry between the two runs (U field in the left pic and V field in the right one, same color bar).

I am not sure what to do now, do you have any idea why there's the asymmetry between two directions?

Besides the picture, namelist.input is attached here. 3D test runs last for 5 minutes, resolutions are 40m, and runs are set to be open boundaries.

Thanks,

Li

 

[Ming Chen]

Your namelist.input looks fine and I don't find anything wrong. Now my question is, how did you change the code and the sounding?
I would suggest that you change those stuff one by one:
(1) Just change the sounding and see whether you can get the expected results;
(2) Change the code and use the default sounding, see what results you can get.
This will help to narrow down to the possible reasons for the asymmetric output.