[Wrf-users] FW: Using Diffusion TKE equation at 1 km

[Kemp, Eric M.]

Dear wrf-users:

I am attempting to model TKE with the WRF-ARW model using a horizontal resolution of 1 km and fairly high vertical resolution (~25 m in lowest 2 km, ~250 m in next 8 km, then ~500 m).  The IC and
LBCs are provided by the NCEP NAM model.  I am currently testing the TKE diffusion option (km_opt = 2) with
diffusion calculated in physical space (diff_opt=2) and the PBL scheme turned off (bl_pbl_physics = 0).
Unfortunately these runs are producing NANs in the output file for multiple variables.  I have tested with both surface layer formulations (sf_sfclay_physics = 1 and 2), but I see no improvement.  I have however had success when running with the Smagorinsky 2D deformation scheme (km_opt=4) and TKE provided by the MYJ PBL scheme (bl_pbl_physics=2).  The time step is 3 and 1/3 seconds.

My question:  Can the diffusion TKE option be used at 1 km?  Are there other settings that should be changed to use it?

Thanks,

Eric Kemp
Northrop Grumman Corporation/TASC

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.ucar.edu/pipermail/wrf-users/attachments/20070426/3a6d8be8/attachment.html