Run WRF with my own data

[antonleontiev]

Hello! I want to make some modifications to the initial grib2 data. I use my own data of water vapor, but after running WPS and WRF I obtain the same result for cases of original data and my data.
I change only 1 message at the initial grib2 file, which corresponds to precipitable water (as I understand this is a field, which I should use: 313:147918503:d=2018022818WAT:entire atmosphere (considered as a single layer):3 hour fcst
There is a procedure, which I used:
1. Insert my own data calculated to the specific area instead of original grib2 data using wgrib2 software. As a result I have an area with my data, surrounded by the original data.
2. Then I use these modified files as an input for WPS, it produces met_em files, which are used as input for real.exe. After running the wrf.exe and comparing result with result of original grib2 files, there is no difference between them, the difference exactly equal to 0.
Are there mistakes? What should I do to fix them?

 

[Ming Chen]

I am a little bit confused by your message. Please clarify which variable you did change, precipitable water or vapor?
If you change the precipitable water, bot that it has no impact on model results. This is because this variable is not included in wrf input and wrfbdy.
If you change qvapor, then RH should be changed correspondingly.

 

[antonleontiev]

Thank you for your reply!
I changed precipitable water in original grib2 file, which was used as input for WPS. As I understand from your answer, I should change qvapor and relative humidity? Is it possible to add precipitable water to the list of variables?

 

[Ming Chen]

Precipitable water is not a variable included in WPS. It is not used for running WRF.
You need to change vapor and RH for your sensitivity test.

I am not sure why you want to add precipitable water. If you need this variable for the purpose of verification, you can add it to your variable lists.