Problem with surface pressure and its extrapolation from mslp

[GabrieleBentiv]

While I was checking the correctness of the initialization I noticed a weird aspect about the surface pressure field in the wrfinput file generated by real.exe.
Surface pressure seems to depend only on the value of the msl pressure I provide in the metgrid file, despite the fact that I also provide a surface pressure field. The presence of the surface pressure field is required as removing or replacing the field with only zeros results in a fatal error, but it does not appear to influence the output.

I tried looking into the code but I couldn't locate the part where this field is extrapolated from mslp. What I did find where a few options in the README.namelist files (both in run folder and outside) suggesting an opposite behaviour from what I am seing (namely: psfc_from_msl = False and use_surface = True) but maybe they are not the ones actually being considered. I did not put them in the namelist.input so I believe I am keeping those default values. What am I missing? Thank you in advance for any input.

In the link you can find both the metgrid file of the first date and the wrfinput resulting from real.exe:
Proton Drive

(I am using WRF v4.5)

 

[kwerner]

Hi, Apologies for the long delay in response while our time has been out much over the past few weeks, due to conferences and holidays. Thank you for your patience. If you are still experiencing this issue, can you also attach the namelist.input file you used? Thanks!

 

[GabrieleBentiv]

Hi, this is the namelist in question:

 

[kwerner]

Thanks for sharing all the files. I used ncview to look at the two fields (PMSL and PSFC) in the met_em file, and compared it to the PSFC field in the wrfinput file, but it looks like PSFC in wrfinput is nearly identical to PSFC in the met_em file. It's possible I'm misunderstanding. I'm attaching a screenshot that shows the three fields. Can you let me know what I'm missing? Thanks.

 

[GabrieleBentiv]

You are right and maybe the file I chose is not the most telling to get what my problem is.

My question essentially is: is there a possibility that real.exe might ignore PSFC (even though it’s required in the met_em file) and instead compute the surface pressure from PMSL, or is the model simply doing the correct thing (based on the namelist settings)?

 

[kwerner]

I believe the model is doing the correct interpolations and you should be okay to proceed forward.