How can I keep the initial input_sounding profile fixed during an ideal simulation (em_seabreeze2d_x)?

[kaee]

Hello.


I am running an ideal WRF case (em_seabreeze2d_x).


I understand that ideal.exe only generates the initial condition (wrfinput_d01) from input_sounding, and afterwards wrf.exe evolves the atmosphere normally.


My goal is not to change the initial sounding. Instead, I want to keep the initial vertical sounding (temperature and/or wind profile) fixed during the entire simulation, so that it does not evolve with time.


I found some old discussions suggesting that solve_em.F should be modified, but I am not sure where this should be done or whether solve_em.F is really the correct place.


Has anyone implemented something similar?

• Is solve_em.F the correct file to modify?
• Or should the restoring be implemented somewhere else (e.g., RK update routines, physics routines, or another module)?

Any suggestions or examples would be greatly appreciated.


Thank you.

 

[Ming Chen]

I am not sure I completely understand your question. The ideal case of em_seabreeze2d_x runs with full-physics and uses periodic lateral boundary. The input sounding has no wind. It only provides initial condition. Once the model starts, all variables are updated by WRF phsyics/dynamics.

 

[kaee]

I am not sure I completely understand your question. The ideal case of em_seabreeze2d_x runs with full-physics and uses periodic lateral boundary. The input sounding has no wind. It only provides initial condition. Once the model starts, all variables are updated by WRF phsyics/dynamics.

Thank you. I think I understand your explanation. If I understand correctly, input_sounding is used only to generate the initial condition, and it is not referenced again during the simulation.

In that case, if I want to keep some variables from the initial sounding (for example, the upper-level wind or potential temperature profile) close to their initial values, is modifying solve_em.F essentially the only approach? Or is there another part of the WRF code that would be more appropriate?

 

[Ming Chen]

Hi Kaee,

Once wrf.exe starts, all variables including upper-level winds and potential temperature will be evolving based on the model physics/dynamics. In this case, if you want to manually change some variables to make them stay at their initial values, you will need to set all the tendencies related to these variable to zero. From this perspective, the answer is yes. You can modify solve_em.F to set tendencies to zero.