[WRF-Chem] real.exe runs successfully but wrfbdy_d01 only contains one timestamp

[hyewon cho]

Hello,

I am currently running a WRF-Chem simulation, and after executing real.exe, I noticed that the wrfbdy_d01 file contains only one time step in the Times variable.

My namelist.input is configured with interval_seconds = 21600 (6-hour intervals), and my met_em files are also available at 6-hour intervals. For example:


ls -lh WPS/met_em.d01.*
met_em.d01.2024-12-20_00:00:00.nc
met_em.d01.2024-12-20_06:00:00.nc
met_em.d01.2024-12-20_12:00:00.nc

Since real.exe should process multiple met_em files and generate a wrfbdy_d01 file with multiple time steps, I expected to see multiple time records in Times. However, when checking the boundary file, only one timestamp (00:00:00) is present:

ncdump -v Times wrfbdy_d01
Times = "2024-12-20_00:00:00" ;

What I Have Tried
1. Checking if real.exe reads all met_em files correctly
→ The log indicates that real.exe does read multiple met_em files, such as:
d01 2024-12-20_00:00:00 Input data is acceptable to use: met_em.d01.2024-12-20_00:00:00.nc
d01 2024-12-20_06:00:00 Input data is acceptable to use: met_em.d01.2024-12-20_06:00:00.nc

his suggests that real.exe is correctly detecting and reading multiple input times.

2. Verifying the time steps in wrfbdy_d01
ncdump -v Times wrfbdy_d01
→ Output:
Times = "2024-12-20_00:00:00" ;

- If real.exe is reading multiple met_em files, why does wrfbdy_d01 contain only the first time step (00:00:00)?
- Could the missing TSLB variable in met_em be affecting boundary condition generation?
- I enabled multi_bdy_files = .true., but only one boundary file was created. Could there be a missing configuration step?

Any insights or suggestions would be greatly appreciated! Thank you.

 

[ehsantaghizadeh]

Hello,

I am currently running a WRF-Chem simulation, and after executing real.exe, I noticed that the wrfbdy_d01 file contains only one time step in the Times variable.

My namelist.input is configured with interval_seconds = 21600 (6-hour intervals), and my met_em files are also available at 6-hour intervals. For example:


ls -lh WPS/met_em.d01.*
met_em.d01.2024-12-20_00:00:00.nc
met_em.d01.2024-12-20_06:00:00.nc
met_em.d01.2024-12-20_12:00:00.nc

Since real.exe should process multiple met_em files and generate a wrfbdy_d01 file with multiple time steps, I expected to see multiple time records in Times. However, when checking the boundary file, only one timestamp (00:00:00) is present:

ncdump -v Times wrfbdy_d01
Times = "2024-12-20_00:00:00" ;

What I Have Tried
1. Checking if real.exe reads all met_em files correctly
→ The log indicates that real.exe does read multiple met_em files, such as:
d01 2024-12-20_00:00:00 Input data is acceptable to use: met_em.d01.2024-12-20_00:00:00.nc
d01 2024-12-20_06:00:00 Input data is acceptable to use: met_em.d01.2024-12-20_06:00:00.nc

his suggests that real.exe is correctly detecting and reading multiple input times.

2. Verifying the time steps in wrfbdy_d01
ncdump -v Times wrfbdy_d01
→ Output:
Times = "2024-12-20_00:00:00" ;

- If real.exe is reading multiple met_em files, why does wrfbdy_d01 contain only the first time step (00:00:00)?
- Could the missing TSLB variable in met_em be affecting boundary condition generation?
- I enabled multi_bdy_files = .true., but only one boundary file was created. Could there be a missing configuration step?

Any insights or suggestions would be greatly appreciated! Thank you.

May I ask if you have resolved your problem?

 

[kwerner]

@hyewon cho,

The wrfbdy_d01 file only prints the time stamp for the beginning of each time period, and it does not include the final time. It's hard to say if the namelist you attached is the one you're using because you mentioned you set interval_seconds = 21600, but it's set to 3600 in the namelist.input file. Since the met_em* files are available every 6 hours, and you're only running real.exe for 6 hours, it would be expected to just include the initial time in the wrfbdy_d01 file.