search

geoschem / geos-chem

GEOS-Chem "Science Codebase" repository. Contains GEOS-Chem science routines, run directory generation scripts, and interface code. This repository is used as a submodule within the GCClassic and GCHP wrappers, as well as in other modeling contexts (external ESMs).
http://geos-chem.org
Other
169 stars 165 forks source link

The missing files: Regions_land.dat and Regions_ocean.dat #2484

Closed rippleofheart closed 1 month ago

rippleofheart commented 1 month ago

Your name

Yuhao Mo

Your affiliation

The University of Melbourne

What happened? What did you expect to happen?

This morning, I wanted to use GEOS-Chem(Version 14.4.3) to generate the CO2 emission in the global bio-ecosystem. I decided to do a test. I chose the grid resolution: 2.0*2.5degree, set up the environment, entered these commands 'cmake ../CodeDir -DRUNDIR=..', 'make -j', 'make install' according to the 'QuickStart Guide'. And I changed the tag_bio_and_ocean_CO2 from 'false' to 'true' in geoschem_config.yml. I kept the default start date and end date (20190701--20190801) and saved. Then I used the 'dryrun' to help me download the datasets from WashU. Everything went well and no regions_land.dat and regions_ocean.dat occurred in my run directory.

However, when I entered the command './gcclassic'. The error came out: READ_REGIONS: Reading Regions_land.dat 0%????6?`??@,?0%???mm???@?H2O2
At line 396 of file /data/gpfs/projects/punim0711/YuhaoMo/test8_CO2_simple_bio_225_72/CodeDir/src/GEOS-Chem/GeosCore/co2_mod.F90

Then, an empty file Regions_land.dat occurred in my run directory, as shown in the figures. I think it should not be empty because I checked co2_mod.F90, the code is for reading the Regions_land.dat. The reason why it has an empty Regions_land.dat is that, in my opinion, the co2_mod.F90 read it with 'a', so it generated an empty file automatically.

I thought I would be able to get a good result on CO2 emissions from the ecosystem, but I didn't expect to get it wrong because I was missing this data. Who can help me solve the problem? I searched the Internet and have no idea where is the datasets and how to generate it.

Best Regards, Yuhao MO bug1

What are the steps to reproduce the bug?

  1. Run the command './createRunDir.sh' in 'run' directory in 'GCClassic'
  2. Choose 10. CO2 by typing '10' and press Enter
  3. Choose 'MERRA-2' by typing '1' and press Enter
  4. Choose 'Horizontal resolution: 2.0 *2.5' by typing '2' and press Enter(Other resolution maybe don't support the bio CO2)
  5. Choose 'Number of levels: 47' by typing '2' and press Enter
  6. Write a path name
  7. Write a directory name and decide not to track run directory with git Set up the right environments by source .bash file(I wrote it by myself but I don't think it is the source of bugs) Go to the build directory in run directory you created just now and :
  8. Enter the command: cmake ../CodeDir -DRUNDIR=..
  9. Back to parent directory(Run directory), and edit the geoschem_config.yml by command 'nano -m geoschem_config.yml '
  10. Find the 'tag_bio_and_ocean_CO2' and change the 'false' to 'true'
  11. Save it and quit
  12. Go to the 'build' directory and enter command 'make -j'
  13. Enter command 'make install'
  14. Back to the run directory by 'cd ..' and enter command './gcclassic --dryrun | tee log.dryrun'
  15. Enter './download_data.py log.dryrun --washu'
  16. Finally enter './gcclassic', you will see the bug.
    • READ_REGIONS: Reading Regions_land.dat 0?cH?G6?`??@,?0?cH?mm???@?H2O2
      At line 396 of file /data/gpfs/projects/punim0711/YuhaoMo/test13_CO2_simple_bio_225/CodeDir/src/GEOS-Chem/GeosCore/co2_mod.F90 Fortran runtime error: Bad value during integer read

Please attach any relevant configuration and log files.

geoschem_config.txt summarize_build.txt HISTORY.txt HEMCO_Config.txt log.txt

What GEOS-Chem version were you using?

14.4.3

What environment were you running GEOS-Chem on?

Other (please explain below)

What compiler and version were you using?

0penMPI/4.1.4; GCC/11.3.6; netCDF-C++4/4.3.1; netCDF-Fortran/4.6.6; flox/2.6.4; CMake/3.26.3

Will you be addressing this bug yourself?

Yes

In what configuration were you running GEOS-Chem?

GCClassic

What simulation were you running?

CO2

As what resolution were you running GEOS-Chem?

2 × 2.5

What meterology fields did you use?

MERRA-2

Additional information

Run the code on HPC from University of Melbourne.

yantosca commented 1 month ago

Thanks for writing @rippleofheart. The tagged CO2 simulation is deprecated and slated for removal in 14.6.0. I am not sure if anyone has used it recently.

rippleofheart commented 1 month ago

Thanks for writing @rippleofheart. The tagged CO2 simulation is deprecated and slated for removal in 14.6.0. I am not sure if anyone has used it recently.

Thanks for your reply. I am wondering where I can do the CO2 simulation(CO2 emission) from the bioecosystem by the current GEOS-Chem. Do I need to use the former version?

yantosca commented 1 month ago

@rippleofheart: You can take a look at the Regions files (archived in our older geos-chem-unittest repo):

These files only ever worked with the 2 x 2.5 degree grid.

rippleofheart commented 1 month ago

@rippleofheart: You can take a look at the Regions files (archived in our older geos-chem-unittest repo):

These files only ever worked with the 2 x 2.5 degree grid.

Thank you for your reply! I debugged the GEOS-CHEM for 1 week and I found it could not generate the 0.625*0.5 degree CH4 flux between 2003 to 2018, too. Some files are missing. I am so tired to solve the problem and I have to change to CAMS, sorry.

Thank you for your answer again. Best regards,

Morry

yantosca commented 1 month ago

Thanks @rippleofheart. I will close out this issue now.