Issue with v3 v4 data for WRF4.2.1 and Ndown.exe

[pgoddard]

Hello,

I did post about this topic a couple weeks ago, but I thought I would provide some more details to whomever can help. (old post: https://forum.wrfforum.com/viewtopic.php?f=6&t=11825&sid=250b82cf7a8e3452db7a298899517410)

I am running WRF 4.2.1, and WPS 4.2. I recently updated the WRF preprocessors as well as I now use v4 for the geostatic data. I am using the EDGAR-HTAP v2 data from https://edgar.jrc.ec.europa.eu/htap_v2/ and the anthro_emiss executable from https://www.acom.ucar.edu/wrf-chem/download.shtml

I can run the real.exe for my coarse grid with the namelist option force_use_old_data set equal to false. However, when I try to run wrf.exe I get an error (also see the attached d01_rsl.error.0000, I also attached the namelist.input used at the time of executing wrf.exe):

d01 2014-08-27_00:00:00 Input data is acceptable to use: wrflowinp_d01
d01 2014-08-27_00:00:00 Input data processed for aux input 4 for domain 1
mediation_integrate: med_read_wrf_chem_emissions: Open file wrfchemi_00z_d01
HOURLY EMISSIONS UPDATE TIME 0.0 0.0
mediation_integrate: med_read_wrf_chem_emissions: Read emissions for time 2014-08-27_00:00:00
d01 2014-08-27_00:00:00 Error trying to read metadata
d01 2014-08-27_00:00:00 File name that is causing troubles = wrfchemi_00z_d01
d01 2014-08-27_00:00:00 You can try 1) ensure that the input file was created with WRF v4 pre-processors, or
d01 2014-08-27_00:00:00 2) use force_use_old_data=T in the time_control record of the namelist.input file
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE: <stdin> LINE: 332
---- ERROR: The input file appears to be from a pre-v4 version of WRF initialization routines
-------------------------------------------
Rank 0 [Mon Mar 22 14:59:39 2021] [c0-0c2s1n2] application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0
forrtl: error (76): Abort trap signal


I think I am doing the necessary steps to use only WRF v4 data and pre-processors, but if there is any step someone thinks I may have missed, I would be grateful for any troubleshooting ideas.

I can run wrf.exe if I set the namelist option force_use_old_data set equal to true. And the simulation runs properly and provides output for my coarse domain. However, when I run my inner domain with ndown.exe inner domain, the wrf.exe will halt with the following error:

SIMULATION START DATE = 2014-08-27_00:00:00
Timing for processing wrfinput file (stream 0) for domain 1: 9.64600 elapsed seconds
Max map factor in domain 1 = 1.04. Scale the dt in the model accordingly.
D01: Time step = 30.00000 (s)
D01: Grid Distance = 9.000000 (km)
D01: Grid Distance Ratio dt/dx = 3.333333 (s/km)
D01: Ratio Including Maximum Map Factor = 3.466527 (s/km)
D01: NML defined reasonable_time_step_ratio = 6.000000
---- WARNING : Older v3 input data detected
-------------- FATAL CALLED ---------------
FATAL CALLED FROM FILE: <stdin> LINE: 684
---- Error : Cannot use moist theta option with old data
-------------------------------------------
Rank 0 [Mon Mar 22 17:44:11 2021] [c4-0c0s6n1] application called MPI_Abort(MPI_COMM_WORLD, 1) - process 0
forrtl: error (76): Abort trap signal

I do not know if this is associated with the v3 – v4 error from above. But, looking at the wrf_bdy files before and after the ndown.exe step, the T_BTYE variable looks quite different. The range of this temperature tendency changes from ~ -50 to 50 k/s to ~ 100 to 5230 k/s. The latter values are noticeably incorrect (I attached a screen shot of the variable from wrf_bdy before and after ndown.exe).

I did see this post, https://forum.wrfforum.com/viewtopic.php?f=7&t=11255, which looks like a similar issue to mine. But, from what I can tell, the issue was resolved(?) by WRF version 4.2.1.

For the second part of my question, I also attached the namelist.input and error output, noted by “d02” in the file names



Thank you for your time and help,
Paul

 

[jordanschnell]

Hi Paul,

I am unsure about the first part of your question - since you are creating pre-v4 data, you must set the force_use_old_data = T. As for your ndown error, it likely is coming from this namelist option: "use_theta_m" in the &dynamics namelist, which is by default = 1; i.e. the error "Error : Cannot use moist theta option with old data". Try setting the option to 0.

Jordan

 

[pgoddard]

Hi Jordan,

Thank you for your help. Sorry if I was not clear with my first question. I was just curious to which step of mine may be using v3 data -- even though I think I am using the most up-to-date version of the preprocessors and geo-static data.

Though, I posed the first question in case it was associated with the ndown error with respect to the moist theta option. But, if setting use_theta_m equal to 0 resolves my issue with ndown, then I did need to worry about the first question.

I will try a ndown in the coming days and let you know how it goes.

Thanks again,
Paul

 

[Abhi_nav]

Hi Paul,

I am trying to run WRF Chem with EDGAR-HTAP inventory with anthro_emiss utility. I am able to create files for my domain wrfchemi_<hour>_d01.nc however, when I run the model I get the following:

mediation_integrate: med_read_wrf_chem_emissions: Read emissions for time 2018-12-01_03:00:00
d01 2018-12-01_03:00:00 Error trying to read metadata
d01 2018-12-01_03:00:00 Input data is acceptable to use: wrfchemi_00z_d01
**WARNING** Time in input file not being checked **WARNING**
d01 2018-12-01_03:00:00 Input data is acceptable to use: wrffirechemi_d01_2018-12-01_03:00:00
**WARNING** Time in input file not being checked **WARNING**
Timing for main: time 2018-12-01_03:02:30 on domain 1: 16.75999 elapsed seconds
Timing for main: time 2018-12-01_03:05:00 on domain 1: 97.65652 elapsed seconds

The model continue to run despite the error but I'm unsure if my output is correct. Kindly provide your suggestions. Thanks

 

[Catrainy]

Hello,
I encountered the same problem as you. My wrf.exe also continued to run, but there was still a fatal error occured at the end. So I would like to ask you sincerely, do you know the answer of the question? I would appreciate it if you would help me!

Hi Paul,

I am trying to run WRF Chem with EDGAR-HTAP inventory with anthro_emiss utility. I am able to create files for my domain wrfchemi_<hour>_d01.nc however, when I run the model I get the following:

mediation_integrate: med_read_wrf_chem_emissions: Read emissions for time 2018-12-01_03:00:00
d01 2018-12-01_03:00:00 Error trying to read metadata
d01 2018-12-01_03:00:00 Input data is acceptable to use: wrfchemi_00z_d01
**WARNING** Time in input file not being checked **WARNING**
d01 2018-12-01_03:00:00 Input data is acceptable to use: wrffirechemi_d01_2018-12-01_03:00:00
**WARNING** Time in input file not being checked **WARNING**
Timing for main: time 2018-12-01_03:02:30 on domain 1: 16.75999 elapsed seconds
Timing for main: time 2018-12-01_03:05:00 on domain 1: 97.65652 elapsed seconds

The model continue to run despite the error but I'm unsure if my output is correct. Kindly provide your suggestions. Thanks

 

[Abhi_nav]

Hello,
I encountered the same problem as you. My wrf.exe also continued to run, but there was still a fatal error occured at the end. So I would like to ask you sincerely, do you know the answer of the question? I would appreciate it if you would help me!

The error of metadata file seems to be a trivial issue with no impact on the run in my case. I suppose it is because WRFv4 thinks the file is made for older version though I may be wrong. You should check what caused the fatal error in your case maybe do a grep fatal rsl.error.* or grep Error rsl.error.* to get what was the issue. All the best.

 

[onishi]

Hello
I am not sure if this discussion is still active.
For the sake of records, I share my experience and a solution I found.

When I moved from v3 to v4, I had the same problem

ERROR: The input file appears to be from a pre-v4 version of WRF initialization routines

Then I found a solution to run a WRF-Chem v4+ without using force_use_old_data = .true. in namelist.
It was simply to add the following global attribute in wrfchemi input files (netcdf format)

TITLE = "Created For WRF V3. and V4." ;

In share/input_wrf.F, it looks for a global attribute "TITLE" and checks if it contains "V4." (attention to a "V" capital and a period "." at the end). If it does not find "TITLE" attribute or find "V4." in it, it returns "FALSE" from "is_this_data_ok_to_use" subroutine, and stops the simulation with the error message in question.

This is still the case in v4.4.1 that I use for now.

I hope this may help.

All the best
Tatsuo

 

[randy004/037]

Hello
I am not sure if this discussion is still active.
For the sake of records, I share my experience and a solution I found.

When I moved from v3 to v4, I had the same problem



Then I found a solution to run a WRF-Chem v4+ without using force_use_old_data = .true. in namelist.
It was simply to add the following global attribute in wrfchemi input files (netcdf format)



In share/input_wrf.F, it looks for a global attribute "TITLE" and checks if it contains "V4." (attention to a "V" capital and a period "." at the end). If it does not find "TITLE" attribute or find "V4." in it, it returns "FALSE" from "is_this_data_ok_to_use" subroutine, and stops the simulation with the error message in question.

This is still the case in v4.4.1 that I use for now.

I hope this may help.

All the best
Tatsuo

Hi tatsuo, can you tell me how to add T"ITLE = "Created For WRF V3. and V4." tinto wrfchemi file?

 

[onishi]

Hello
I use my own python script to create wrfchemi files.
If you use python, it is just "setncattr" method from "Dataset" constructor of "netCDF4" Package that creates the global attribute.
If you just want to add a global attribute to an existing netcdf file, you can also use NCO operator "ncatted" as follows,

ncatted -O -h -a TITLE,global,c,c,"Created For V3. and V4." wrfchemi_file

Please check the man page of ncatted for more details.

All the best