AOD550 WRF-SOLAR: wrf.exe is shifting the auxinput15's timestep +1 hour !

[saud04]

Dears,

I'm using WRF V4.4 and run it as WRF-SOLAR.
I created an auxiliary netCDF file for AOD550 and ANGEXP, and modify namelist.input to read the auxinput15.
When wrf.exe starts running it read the auxiliary file successfully, but the issue is: when I checked WRF output I found that AOD550 and ANGEXP are shifted +1 hour.
To be clear: WRF read AOD550 and ANGEXP from auxinput15 at time 2022-11-02_00:00:00 and write them in wrfout at time 2022-11-02_01:00:00 and keeps 2022-11-02_00:00:00 in wrfout ZERO!

could you please help me to fix it.

Files:
1- auxinput15 ( File upload and sharing. Large file transfers. Free online cloud storage. )
2- namelist.input (attached)
3- wrfout after extracting AOD550 and ANGEXP ( File upload and sharing. Large file transfers. Free online cloud storage.)

Your support is appreciated.

 

[Ming Chen]

Hi,
I would suggest that you (1) write all your AOD and ANGEXP data to a single file ( just like the style of wrflowinp) and (2) set frames_per_auxinput15 = 1,
then rerun this case. Please let me know whether this can fix the issue.

 

[saud04]

Hi,
thanks for the response.
1) all AOD and ANGEXP are in a single file, have a look at it please ( ncr_d01_25 ), what is the style of wrflowinp?
2) when I set frames_per_auxinput15 = 1, wrf.exe keeps the first AOD time step in wrf_out zero and write the remaining AOD time steps in wrf_out = the first AOD time step in auxinput15! so I think I should keep frames_per_auxinput15 = 25.

NOTE: wrf.exe always keeps the first AOD time step in wrf_out = zero, and write the first AOD time step in auxinput15 into the second AOD time step in wrf_out, regardless the start time in the namelist.input, I tried to set start time to (2022-11-02_06:00:00) and I got the same result!

could you please attached any auxinput15 file or wrflowinp?

thanks again,

 

[Ming Chen]

Hi,
I looked at your file ncr_d01_25.nc and it looks all fine to me. I am perplexed by the model behavior ....

Please see the uploaded wrflowinp file for your reference. Please remember to rename wrflowinp_d01.txt as wrflowinp_d01.nc.

This is the file for updating surface data.

 

[saud04]

Hi,
I looked at your file ncr_d01_25.nc and it looks all fine to me. I am perplexed by the model behavior ....

Please see the uploaded wrflowinp file for your reference. Please remember to rename wrflowinp_d01.txt as wrflowinp_d01.nc.

This is the file for updating surface data.

Hi Chen,

Thank you for replying. I've tried multiple solutions, but it always ends up with the same result!
Could you take a look at the attached namelist.input, is it correct or missing something?

It seems that wrf.exe read the first time step in auxiliary file as the second time step! do you think it is an index ID issue in auxiliary?

FYI: I created the auxiliary file by the following steps:
1- Extract variables (Times, AOD5502D and ANGEXP2D) from met_em files to a new file using NCO.
2- Calculate AOD550 and ANGEXP.
3- Insert the AOD550 and ANGEXP values to the auxiliary file using NCO.

 

[saud04]

update:
I noticed that wrf.exe read the first AOD550 and ANGEXP from wrfinput (which is equal zero), then read the remaining times steps from the auxiliary file!
How to fix it?

BTW, I modified the following lines in Registry.EM_COMMON before the compilation:
state real aod5502d ij misc 1 - i{15}rh "AOD5502D" "Total aerosol optical depth at 550 nm" ""
state real angexp2d ij misc 1 - i{15}rh "ANGEXP2D" "Aerosol Angstrom exponent" ""

 

[viviq18]

update:
I noticed that wrf.exe read the first AOD550 and ANGEXP from wrfinput (which is equal zero), then read the remaining times steps from the auxiliary file!
How to fix it?

BTW, I modified the following lines in Registry.EM_COMMON before the compilation:
state real aod5502d ij misc 1 - i{15}rh "AOD5502D" "Total aerosol optical depth at 550 nm" ""
state real angexp2d ij misc 1 - i{15}rh "ANGEXP2D" "Aerosol Angstrom exponent" ""

Hi Chen,

Thank you for replying. I've tried multiple solutions, but it always ends up with the same result!
Could you take a look at the attached namelist.input, is it correct or missing something?

It seems that wrf.exe read the first time step in auxiliary file as the second time step! do you think it is an index ID issue in auxiliary?

FYI: I created the auxiliary file by the following steps:
1- Extract variables (Times, AOD5502D and ANGEXP2D) from met_em files to a new file using NCO.
2- Calculate AOD550 and ANGEXP.
3- Insert the AOD550 and ANGEXP values to the auxiliary file using NCO.

hi
I'm sorry to bother you, but I'd like to ask you how to operate NCO to extract the AOD variables (while keeping the global attributes) to get the auxiliary files. What the command I need in NCO?

 

[saud04]

hi
I'm sorry to bother you, but I'd like to ask you how to operate NCO to extract the AOD variables (while keeping the global attributes) to get the auxiliary files. What the command I need in NCO?

I used CHIMERE model, then I extracted & calculated AOD550 from the output of CHIMERE, cause CHIMERE output has AOD at different levels.
Feel free to ask if you need more help.

 

[saud04]

hi
I'm sorry to bother you, but I'd like to ask you how to operate NCO to extract the AOD variables (while keeping the global attributes) to get the auxiliary files. What the command I need in NCO?

ncks -v lat,lon,optdaero,Times infile outfile
(optdaero from CHIMERE output)

 

[Hao Lyu]

Hi saud04
Sorry to bother you, but I want to know if you successfully run the WRF solar? Could you give me an example of your auxinput15 file if you have run it successfully? By the way, where did you get your AOD data? Is it from a geostationary satellite data or something else?

 

[little_grey]

Hi,
thanks for the response.
1) all AOD and ANGEXP are in a single file, have a look at it please ( ncr_d01_25 ), what is the style of wrflowinp?
2) when I set frames_per_auxinput15 = 1, wrf.exe keeps the first AOD time step in wrf_out zero and write the remaining AOD time steps in wrf_out = the first AOD time step in auxinput15! so I think I should keep frames_per_auxinput15 = 25.

NOTE: wrf.exe always keeps the first AOD time step in wrf_out = zero, and write the first AOD time step in auxinput15 into the second AOD time step in wrf_out, regardless the start time in the namelist.input, I tried to set start time to (2022-11-02_06:00:00) and I got the same result!

could you please attached any auxinput15 file or wrflowinp?

thanks again,

Hi! Sorry to bother you! I have the same problem, my wrf.exe keeps the first AOD time step in wrf_out = zero, and writes the first AOD time step in auxinput15 into the second AOD time step in wrf_out. Have you solved this problem?