change MODIS TVF to VIIRS GVF

[angehung5]

As described in #133, annual MODIS TVF is updated by using monthly VIIRS GVF. Just a quick check, everything looks ok.

Note that only example SE files and point files are updated. Grabbing data from HPSS took more time than I expected. Also I want to upgrade my gridding method and need some time to dig into. But I don't want to hang it too long so I just made the example files for repo. Still working on daily global files.

Data table in Readme, and the canopy fraction section in the gap-filling description document (a compact version) are also updated .

[drnimbusrain]

@angehung5 Thanks! A couple comments/changes.

[drnimbusrain]

@angehung5 Is the VIIRS GVF still have the same link/reference as VCF? Seems like this should be updated for GVF product reference.

Also, the link to your global extension and processing document is outdated, and should be changed to your updated ARL tech memo. However this is your NOAA Google doc and should be made public in a sharepoint document.

[angehung5]

Sorry my bad, the link is updated now.
The document is not necessarily outdated. A short description of GVF is added. I think it should be enough for now, will update it with the ARL tech memo once it gets published.

[angehung5]

@drnimbusrain Daily global files are available on Hopper. Transferring to Hera now.

The global data processing script is also updated with upgraded gridding method. Now it takes ~ 1 min to create a global input file (using climatological canopy data).
@zmoon Could you please take a look at the Lint error? It says python/global_data_process.py:15:1: F401 'monet' imported but unused which is not the case. Also I believe the environmen.yml file needs to be updated as well.

[drnimbusrain]

@angehung5 Think you should add the monet package to the environment yaml file here and see if that fixes it.

[angehung5]

@angehung5 Think you should add the monet package to the environment yaml file here and see if that fixes it.

Updated, but seems it is not the issue...

[drnimbusrain]

@angehung5 Think you should add the monet package to the environment yaml file here and see if that fixes it.

Updated, but seems it is not the issue...

Right seems still issues. I assumed you tested and it worked properly with the usage of monet here. Also not sure what the new gfortran CI issues are here.

[zmoon]

not sure what the new gfortran CI issues are here.

This is the message, for posterity. In the first run, with the default namelist.

...

 **********************************************************************
 *** SUBROUTINE: CANOPY_WRITE_NCF
 ***   ERROR DEFINING VARIABLE time
 ***   IN FILE output/southeast_us_2022-07-01-11-00-00.0000.nc
 ***   NetCDF: Name contains illegal characters
 **********************************************************************
Error: Process completed with exit code 2.

/ is a common culprit for this error. Trailing whitespace also not allowed. More info.

[drnimbusrain]

not sure what the new gfortran CI issues are here.

This is the message, for posterity. In the first run, with the default namelist.

...

 **********************************************************************
 *** SUBROUTINE: CANOPY_WRITE_NCF
 ***   ERROR DEFINING VARIABLE time
 ***   IN FILE output/southeast_us_2022-07-01-11-00-00.0000.nc
 ***   NetCDF: Name contains illegal characters
 **********************************************************************
Error: Process completed with exit code 2.

/ is a common culprit for this error. Trailing whitespace also not allowed. More info.

Yes, trailing whitespace is strange here, did not see this error before.

[drnimbusrain]

@zmoon Please see latest commit to develop, where the fortran build/test CI seemingly passed successfully without that error: https://github.com/noaa-oar-arl/canopy-app/actions/runs/8994540709/job/24708145727

However, there are other python script errors showing even though it says they complete successfully?

[zmoon]

format "('''', a, '''')" will put single quotes around the string

[drnimbusrain]

@angehung5 I think that other gfortran issue with file name is propagating from your branch, not something wrong in upstream develop? I have never seen it before.

[angehung5]

@drnimbusrain I don't know...The branch was updated before I made any changes on the python script.
Just have two branches (mine and develop) run on Hopper, both are working fine. And canopy_ncf_io_mod.F90 from two branches are identical.

[drnimbusrain]

@drnimbusrain I don't know...The branch was updated before I made any changes on the python script. Just have two branches (mine and develop) run on Hopper, both are working fine. And canopy_ncf_io_mod.F90 from two branches are identical.

@angehung5 Yes, I understand. I test the latest develop branch with debug options and do not find this error that the CI/ build-test gfortran error reported. Hopefully, Zach can help here.