WRF & MPAS-A Support Forum

We know for sure that WRFV3.9.1.1 has trouble to run with too many processors. Based on your grid numbers specified in namelist, I guess 16 processors will be enough for running this case. Too many processors will cause the module to hang, or to crash with segmentation fault.

Luwen One option is to add your own codes to solve_em.F, right after the call of all RK steps. Please find the piece of code below: CALL after_all_rk_steps ( grid, config_flags, & 4594 moist, chem, tracer, scalar, & 4595 th_phy, pi_phy, p_phy, rho_phy, & 4596 p8w, t8w, dz8w, & 4597 REAL(curr_secs,8)...

Manh, CFL violation indicates the model integration is not stable, which can be attributed to many reasons. (1) The max time step is 6 times of grid interval. Please check to make sure it is correct. (2) If you turn off SST update, can the model run successfully? (3) If you run with SST from the dat...

Where did you download the clWRF codes? Can you also tell me whether there is any contact person who is rresponsible for clWRF? Thanks.

Bastien,
One option is that, you may run ungrib and set the time interval to 3 minutes. ungrib has the capability to interpolate data to the time you specify. This will give you input data for ndown run. Please try and let me know whether it works for you.

I guess the job is not done successfully. Otherwise you should see 'SUCCESS' at the end of the rsl files. Note that the error message doesn't necessarily show in rsl.out.0000 and rsl.error.0000. It may appear in any of your RSL files. Please take a look at all those files and there should be some in...

You're namelist.input looks fine. What is the error message in your rsl and/or log file?

EFFC shouldn't be a scalar variable. You need to understand the physical meaning of EFFC and how it works in the code. More important, you also need to understand how Registry works. For example, what does the code do with the definition of "ikjftb" for a state variable, and what variable should be ...

This is because UA is the wind at current time step, while u_phy is calculated from UA of the previous time step. To confirm this, please save wrfout at each time step, then compare UA with U_PHY. Let me know whether you can get identical results. Thanks.

Please use Vtable.ECMWF to ungrib ERA5 data. Then try again. Let me know if you still have issues.
ERA5 is relatively new and variables are saved in different files. Please make sure you download all required files.

No, it is not working that way. SKINTEMP is a a required variable in WRF and must be provided.

David,
The utility avg_tsfc.exe is designed to calculate T2 over water points. This utility is used to overcome some issues related to unrealistic T2 over water. It doesn't affect T2 and SKINTEMP over land points. Your results are correct and expected.

Please set smooth_option = 0. Again, please let me know if you still cannot run REAL.

This time the error comes from &domai. Please set feedback = 0, then try again.
Let me know if you still have problems.

Your namelist.input looks fine. The topography in D02 doesn't show any large gradient, which means that EPSSM is not an issue. This error indicates that something might be wrong in the model physics. Can you do the following tests and see how the model works: (1) Turn off adaptive time step, then ru...

We have never done that before. But I think it is practical to modify the code to make SCM run over ice. Please pay attention to how the model process snow/ice process and make sure you provide all the necessary information. Good luck!

Hi, DZBOT, DZSTRETCH etc. are designed for setting vertical eta levels. For ideal cases, vertical levels are calculate based on ZTOP, which corresponds to real attitude. In wrfout, you can find variables PH and PHB, (PH + PHB)/9.8 will give you the real altitude of each model level. You can do verti...

Yes you are right. SCM is designed to run over land points.

Please provide more information about this failed case:
(1) which version of WRF did you ru?
(2) How did you compile and run WRF?
(3) what data did you use to drive WRF?
(4) Please upload your namelist.input for me to take a look.

EFFC is a local variable used in the Morrison scheme. To output this variable, you need to define it as a state variable in Registry.EM_COMMON, then pass this variable from the physics to the microphysics driver, and from the microphysics driver to the solver (i.e., solve_em.F). If you are familiar ...

Also, please let me know which version of WRF you are running and send me your namelist.input to take a look.

There is no error message in your rsl files. Can you check all the rsl files and find possible error messages?

I have posted an answer to this question. Please see
viewtopic.php?f=39&t=9830

Please change the nameless options from: time_step = 30, 6, time_step_fract_num = 0, 0, time_step_fract_den = 1, 1, num_metgrid_levels = 30, 30, num_metgrid_soil_levels = 4, 4, to: time_step = 30, time_step_fract_num = 0, time_step_fract_den = 1, num_metgrid_levels = 30, num_metgrid_soil_levels = 4,...

Hi, I just run a test case with the option output_diagnostics = 1. wrfxtrm files look just fine. Note that I turn off sst_update and spectral nudging. This test indicates that the option of output_diagnostics works as expected. If you get those weird results, it is possibly because something is wron...