List of CMIP7 branded names

General
1

Description of request:

Could someone please share with me the batch_config_esm1-6_cmip7_baseline.yml ?

I am trying to CMORize some CMIP7 fast track ACCESS-ESM1.6 results (abrupt-lig127k), but can’t access/find the example baseline or any list of the new CMIP7 branded names, which would make things easier for me.

Thanks!

Environment:

What executed:

Actual results:

Expected results:

Additional info:

2

Hi, @LaurieM , is this what you looking for? ACCESS-MOPPy/src/access_moppy/mappings/cmip6_to_cmip7_compound_name_mapping.json at main · ACCESS-NRI/ACCESS-MOPPy · GitHub

3

I think so! thanks!

4

Hi Rhaegar,

However, even with the correct name, it does not currently work because of the structure and name of the outputs (which work for the ACCESS-ESM1.5 but not 1.6).

Namely, for the atmosphere, Moppy looks for: Patterns searched: [‘output[0-9][0-9][0-9]/atmosphere/netCDF/*.pa-*_mon.nc’]., while the outputs are [‘output[0-9][0-9][0-9]/atmosphere/access-esm1p6.um7p3.3d.fld_sXXX.1mon.mean.0101.nc’];

Similarly, for the ocean, the pattern searched is ‘ocean-*-surface_temp-1monthly-*-ym_*.nc’ while the ACCESS-ESM1.6 outputs are “access-esm1p6.mom5.Xd.VAR.1mon.mean.0101.nc”.

Similarly, for sea-ice the Patterns searched: is [‘output[0-9][0-9][0-9]/ice/iceh-1monthly-mean_*.nc’], and the ACCESS-ESM1.6 outputs are [‘output[0-9][0-9][0-9]/ice/access-esm1p6.cice5.2d.siXXX.1mon.mean.0101.nc’]

Note that even for the ACCESS-ESM1.5, I could not CMORize any of the sea-ice data (into CMIP6+) because of the mapping of the variable names.

Thanks for your help!

5

Hi Laurie,

For the situation you mentioned, Moppy provides an option to define your own file pattern instead of relying on the automatic file discovery process.

If you’re using the Python API, I’d suggest using glob to retrieve the file paths you need and passing the resulting file list to ACCESS_ESM_CMORiser. For example:

import glob

files = glob.glob(
    "{PATH}/output[0-9][0-9][0-9]/atmosphere/access-esm1p6.um7p3.3d.fld_sXXX.1mon.mean.0101.nc"
)

cmip7_cmoriser = ACCESS_ESM_CMORiser(
    input_data=files,
    compound_name="atmos.rsds.tavg-u-hxy-u.mon.GLB",
    experiment_id="piControl-spinup",
    source_id="ACCESS-ESM1-6",
    variant_label="r1i1p1f1",
    grid_label="gn",
    activity_id="CMIP",
    cmip_version="CMIP7",
)

If you’re using batch processing, you can override the file pattern directly in batch-config.yml:

input_folder:{PATH TO THE FOLDER}
# Optional: per-variable file-pattern overrides.
file_patterns:
  atmos.tas.tavg-h2m-hxy-u.mon.glb: "/output*/atmosphere/access-esm1p6.um7p3.3d.fld_sXXX.1mon.mean.0101.nc"

Our default file discovery patterns were developed based on the archive structure in p73. I assume you’re working with data in jq44, which appears to use a different file naming pattern that isn’t supported by the default discovery process yet.

Defining a custom file pattern should be a suitable workaround for now.

6

Thanks Rhaegar. I will try that.

Note that I am running exp. with the ACCESS-ESM1.6 for CMIP7 production following the guidelines in: https://access-esm1p6-configs--904.org.readthedocs.build/

I am thus not sure why I don’t have the default output/archive structure.

7

Yeah, that’s because we currently have two different file patterns for ACCESS-ESM1.6 model output. The one you’re using, which follows the guidelines, is probably the newer pattern that Moppy doesn’t support yet. We’ll add support for it in the next Moppy release.

8

Can you please let me know when the next Moppy released is out?

I included the file_patterns in the yml file, and it propagated to the cmor*sh and cmor*py but it still does not look for the correct file pattern, and I am not sure why: Error processing atmos.pr.tavg-u-hxy-u.mon.glb: No files found for variable ‘atmos.pr.tavg-u-hxy-u.mon.glb’. No ‘mon’ files for ‘pr’ found in 100 output director(y/ies) (output000…output099). Patterns searched: [‘output[0-9][0-9][0-9]/atmosphere/netCDF/.pa-_mon.nc’].