E3SM-3-0 Registration for CMIP6 (source_id)
Model description:
Registration PR to CMIP6:https://github.com/WCRP-CMIP/CMIP6_CVs/issues/1250
CMIP6 officially closed registration, we will update local version of CMIP6_CVs and publish to CMIP6-E3SM-Ext repository on ESGF.
Metadata definition:
TBA
source_id registration of E3SM-3-0
"E3SM-3-0":{
"activity_participation":[
"CMIP"
],
"institution_id":[
"E3SM-Project"
],
"label":"E3SM 3.0",
"label_extended":"E3SM 3.0 (Energy Exascale Earth System Model)",
"model_component":{
"aerosol":{
"description":"MAM5 w/ prognostic stratospheric aerosols (atmos physics grid)",
"native_nominal_resolution":"100 km"
},
"atmos":{
"description":"EAM (E3SMv3.0, cubed sphere spectral-element; 5400 els., 30x30 per cube face. Dynamics: degree 3 (p=3) polynomials within each spectral els., 112 km ave. resolution. Physics: 2x2 finite volume cells within each spectral els., 1.5 degree (168 km) average grid spacing; 80 vertical layers w/ top at 60 km)",
"native_nominal_resolution":"100 km"
},
"atmosChem":{
"description":"Troposphere: UC Irvine interactive chemistry module. Stratosphere: linearized interactive ozone (LINOZ v3) (atmos physics grid)",
"native_nominal_resolution":"100 km"
},
"land":{
"description":"ELM (E3SMv3.0, 0.5 deg lat/lon grid, dynamic vegetation growth model), MOSART (E3SMv3.0, 0.5 deg lat/lon grid)",
"native_nominal_resolution":"50 km"
},
"landIce":{
"description":"none",
"native_nominal_resolution":"none"
},
"ocean":{
"description":"MPAS-Ocean (E3SMv3.0, new IcoswISC30E3r5 mesh, nearly uniform 30 km mesh w/ Ice-Shelf Cavities (wISC) around Antarctica and has 64 vertical layers.)",
"native_nominal_resolution":"30 km"
},
"ocnBgchem":{
"description":"none",
"native_nominal_resolution":"none"
},
"seaIce":{
"description":"MPAS-Seaice (E3SMv3.0, MPAS-Ocean grid; the CICE Consortium’s Icepack as the default sea ice column physics)",
"native_nominal_resolution":"30 km"
}
},
"release_year":"2024",
"source_id":"E3SM-3-0"
},
Comments
Hello, @Chris Golaz , @Shaocheng Xie , @Luke Van Roekel , @Peter Thornton , @Darin Comeau @Renata McCoy, the infrastructure team is gearing up tools and workflow for v3.LR data publication. I have updated based on E3SM V3.0 Description and texts from Shaocheng and Chris’s manuscript, could you help review the information above in source_id registration, which will be converted as part of the meta-data of those CMIP files. Thanks a lot!
@Jill Chengzhu Zhang : thank you for working on this. The meta data looks fine to me. But it might be good for @Wuyin Lin to double check.
Thank you, Chris!
@Jill Chengzhu Zhang this looks good to me - thanks!! The string limit may be a problem (1024 char, if I remember correctly), you should try to count the characters in the full string and see if anything needs to be cut.
Thanks for the reminder. I will work with @Anthony Bartoletti to fit this within the CMOR limit.
@Renata McCoy I found the previous issue about char limit: https://github.com/WCRP-CMIP/CMIP6_CVs/issues/1127#issuecomment-1190819161
We can further trim the texts if needed when putting together the description texts and merge into CMIP_CVs
The formatted entry (employing actual space characters for indents) comes to 2397 chars. When I remove all indenting, it still comes out to 1661 chars:
"E3SM-3-0":{
"activity_participation":[
"CMIP"
],
"institution_id":[
"E3SM-Project"
],
"label":"E3SM 3.0",
"label_extended":"E3SM 3.0 (Energy Exascale Earth System Model)",
"model_component":{
"aerosol":{
"description":"MAM5 w/ prognostic stratospheric aerosols (atmos physics grid)",
"native_nominal_resolution":"100 km"
},
"atmos":{
"description":"EAM (E3SMv3.0, cubed sphere spectral-element; 5400 els., 30x30 per cube face. Dynamics: degree 3 (p=3) polynomials within each spectral els., 112 km ave. resolution. Physics: 2x2 finite volume cells within each spectral els., 1.5 degree (168 km) average grid spacing; 80 vertical layers w/ top at 60 km)",
"native_nominal_resolution":"100 km"
},(etc etc etc)
The entry should need no formatting for submission - it can all reside on a single line. But we are still almost 600 chars over the limit.
I have no idea what to summarize or eliminate here. Can some of these physics descriptions be given as reference URLs? Must every realm be itemized?
@Anthony Bartoletti have you try to form a single line description as what is for v2 here: https://github.com/PCMDI/cmip6-cmor-tables/commit/3f9e63acd05cf4eec6a81b4e79e9f7d8084020e7#diff-82b29ceb3b677cfd5c4e9240e6d172a2a8e9145acc5d29d0a17d5fa0644c165aR1126 ?
"source":"E3SM 2.0 (2022): \\naerosol: MAM4 with new features in resuspension, marine organics, secondary organics, and dust (same grid as atmos physics)\\natmos: EAM (v2.0, cubed sphere spectral-element grid; 5400 elements, 30x30 per cube face. Dynamics: degree 3 (p=3) polynomials within each spectral element, 112 km average resolution. Physics: 2x2 finite volume cells within each spectral element, 1.5 degree (168 km) average grid spacing)\\natmosChem: Troposphere specified oxidants (except passive ozone with the lower boundary sink) for aerosols. Stratosphere linearized interactive ozone (LINOZ v2) (same grid as atmos physics)\\nland: ELM (v1.0, same grid as atmos physics, satellite phenology mode), MOSART (v1.0, 0.5 degree latitude/longitude grid)\\nlandIce: none\\nocean: MPAS-Ocean (E3SMv2.0, EC30to60E2r2 unstructured SVTs mesh with 236853 cells and 719506 edges, variable resolution 60 km to 30 km; 60 levels; top grid cell 0-10 m)\\nocnBgchem: none\\nseaIce: MPAS-Seaice (E3SMv2.0, EC30to60E2r2 unstructured SVTs mesh with 236853 cells and 719506 edges, variable resolution 60 km to 30 km; 5 ice categories, 7 ice layers, 5 snow layers)"That is just the “source” entry, (1167 chars). Creating a single line is not very different from removing the indents. It just deletes the newline chars at the end of each line.
The full entry for “E3SM-2-0” is as follows:
"E3SM-2-0":{"activity_participation":["CMIP","RFMIP","ScenarioMIP"],"cohort":["Registered"],"institution_id":["E3SM-Project"],"license_info":{"exceptions_contact":"@llnl.gov <- e3sm-data-support","history":"","id":"CC BY 4.0","license":"Creative Commons Attribution 4.0 International (CC BY 4.0; https://creativecommons.org/licenses/by/4.0/)","url":"https://creativecommons.org/licenses/by/4.0/"},"source_id":"E3SM-2-0","source":"E3SM 2.0 (2022): \naerosol: MAM4 with new features in resuspension, marine organics, secondary organics, and dust (same grid as atmos physics)\natmos: EAM (v2.0, cubed sphere spectral-element grid; 5400 elements, 30x30 per cube face. Dynamics: degree 3 (p=3) polynomials within each spectral element, 112 km average resolution. Physics: 2x2 finite volume cells within each spectral element, 1.5 degree (168 km) average grid spacing)\natmosChem: Troposphere specified oxidants (except passive ozone with the lower boundary sink) for aerosols. Stratosphere linearized interactive ozone (LINOZ v2) (same grid as atmos physics)\nland: ELM (v1.0, same grid as atmos physics, satellite phenology mode), MOSART (v1.0, 0.5 degree latitude/longitude grid)\nlandIce: none\nocean: MPAS-Ocean (E3SMv2.0, EC30to60E2r2 unstructured SVTs mesh with 236853 cells and 719506 edges, variable resolution 60 km to 30 km; 60 levels; top grid cell 0-10 m)\nocnBgchem: none\nseaIce: MPAS-Seaice (E3SMv2.0, EC30to60E2r2 unstructured SVTs mesh with 236853 cells and 719506 edges, variable resolution 60 km to 30 km; 5 ice categories, 7 ice layers, 5 snow layers)"},It runs to 1569 characters. The “source” section begins about 30% of the way in, and runs for 1148 chars.
The “E3SM-3-0” entry given at the top of this page has no “source” section.
My take from this comment is that the description texts from each entry will be merged into the description for source, which has a 1024 char limit, or i’m totally off?
I honestly have no idea. When Renata remarked on your proposed “E3SM-3-0” entry above - I assumed the char limit applied to all of it. If only to any given string, we are in much better shape.
THe longest tring in your entry is the “atmos” description:EAM (E3SMv3.0, cubed sphere spectral-element; 5400 els., 30x30 per cube face. Dynamics: degree 3 (p=3) polynomials within each spectral els., 112 km ave. resolution. Physics: 2x2 finite volume cells within each spectral els., 1.5 degree (168 km) average grid spacing; 80 vertical layers w/ top at 60 km)
and it is only 304 chars long. If we need to manually copy each description (or measure them separately) to see if the sum to 1024, we need a new system …
How much of each “entry: description” counts?
With:
"land":{
"description":"ELM (E3SMv3.0, 0.5 deg lat/lon grid, dynamic vegetation growth model), MOSART (E3SMv3.0, 0.5 deg lat/lon grid)",
"native_nominal_resolution":"50 km"
},
Do I measure from “ELM”, or from “description”:”ELM”, or from “land”:{ “description”: “ELM” …?
@Jill Chengzhu Zhang , @Anthony Bartoletti please see section “Proposed shortened string:” in https://e3sm.atlassian.net/wiki/spaces/DOC/pages/3439689981 to see the full description string that will be created from all entries, this needs to fit within 1024 char limit.
@Renata McCoy @Jill Chengzhu Zhang OK, so each realm (atmos, land, etc) section is concatenated along with their description, but the “description =” is eliminated, and the entries are delimited by newline (“\n”) characters. My first attempt to approximate (with real newlines for clarity) looks like:
"aerosol": "MAM5 w/ prognostic stratospheric aerosols (atmos physics grid)", "native_nominal_resolution":"100 km"
"atmos": "EAM (E3SMv3.0, cubed sphere spectral-element; 5400 els., 30x30 per cube face. Dynamics: degree 3 (p=3) polynomials within each spectral els., 112 km ave. resolution. Physics: 2x2 finite volume cells within each spectral els., 1.5 degree (168 km) average grid spacing; 80 vertical layers w/ top
at 60 km)", "native_nominal_resolution":"100 km"
"atmosChem": "Troposphere: UC Irvine interactive chemistry module. Stratosphere: linearized interactive ozone (LINOZ v3) (atmos physics grid)", "native_nominal_resolution":"100 km"
"land": "ELM (E3SMv3.0, 0.5 deg lat/lon grid, dynamic vegetation growth model), MOSART (E3SMv3.0, 0.5 deg lat/lon grid)", "native_nominal_resolution":"50 km"
"landIce": "none", "native_nominal_resolution":"none"
"ocean": "MPAS-Ocean (E3SMv3.0, new IcoswISC30E3r5 mesh, nearly uniform 30 km mesh w/ Ice-Shelf Cavities (wISC) around Antarctica and has 64 vertical layers.)", "native_nominal_resolution":"30 km" "ocnBgchem": "none", "native_nominal_resolution":"none"
"seaIce": "MPAS-Seaice (E3SMv3.0, MPAS-Ocean grid; the CICE Consortium’s Icepack as the default sea ice column physics)", "native_nominal_resolution":"30 km"
But this is still 1273 characters. I did not attempt to apply other string substitutions, as I’m unaware of what is appropriate. If we need to lose another 250 characters, it is going to be a challenge.
Is there a well-defined term for new IcoswISC30E3r5 mesh, nearly uniform 30 km mesh w/ Ice-Shelf Cavities (wISC) around Antarctica and has 64 vertical layers`? (etc).
@Anthony Bartoletti if following “Proposed shortened string:” noted by @Renata McCoy in the one-line description for source, you should exclude "native_nominal_resolution":"30 km" , after that, I think we are within 1024
OK (is that only because they are “native”? There are 100km, 50km, and 30kn values. I will toss all of them. But the “Proposed shortened string:” does not make this explicit, Their “Original” string never mentioned “native” in the first place. Ours does, so it is unclear that it could be removed.)
So, it appears that what you want is this:
"E3SM-3-0":{
"activity_participation":[
"CMIP"
],
"institution_id":[
"E3SM-Project"
],
"label":"E3SM 3.0",
"label_extended":"E3SM 3.0 (Energy Exascale Earth System Model)",
"model_component":{
"aerosol":{
"description":"MAM5 w/ prognostic stratospheric aerosols (atmos physics grid)",
},
"atmos":{
"description":"EAM (E3SMv3.0, cubed sphere spectral-element; 5400 els., 30x30 per cube face. Dynamics: degree 3 (p=3) polynomials within each spectral els., 112 km ave. resolution. Physics: 2x2 finite volume cells within each spectral els., 1.5 degree (168 km) average grid spacing; 80 vertical layers w/ top at 60 km)",
},
"atmosChem":{
"description":"Troposphere: UC Irvine interactive chemistry module. Stratosphere: linearized interactive ozone (LINOZ v3) (atmos physics grid)",
},
"land":{
"description":"ELM (E3SMv3.0, 0.5 deg lat/lon grid, dynamic vegetation growth model), MOSART (E3SMv3.0, 0.5 deg lat/lon grid)",
},
"landIce":{
"description":"none",
},
"ocean":{
"description":"MPAS-Ocean (E3SMv3.0, new IcoswISC30E3r5 mesh, nearly uniform 30 km mesh w/ Ice-Shelf Cavities (wISC) around Antarctica and has 64 vertical layers.)",
},
"ocnBgchem":{
"description":"none",
},
"seaIce":{
"description":"MPAS-Seaice (E3SMv3.0, MPAS-Ocean grid; the CICE Consortium’s Icepack as the default sea ice column physics)",
}
},
"release_year":"2024",
"source_id":"E3SM-3-0"
},
This will result in ~999 chars, producing something like:
"aerosol": "MAM5 w/ prognostic stratospheric aerosols (atmos physics grid)",\n "atmos": "EAM (E3SMv3.0, cubed sphere spectral-element; 5400 els., 30x30 per cube face. Dynamics: degree 3 (p=3) polynomials within each spectral els., 112 km ave. resolution. Physics: 2x2 finite volume cells within each spectral els., 1.5 degree (168 km) average grid spacing; 80 vertical layers w/ top at 60 km)",\n "atmosChem": "Troposphere: UC Irvine interactive chemistry module. Stratosphere: linearized interactive ozone (LINOZ v3) (atmos physics grid)",\n "land": "ELM (E3SMv3.0, 0.5 deg lat/lon grid, dynamic vegetation growth model), MOSART (E3SMv3.0, 0.5 deg lat/lon grid)", \n "landIce": "none",\n "ocean": "MPAS-Ocean (E3SMv3.0, new IcoswISC30E3r5 mesh, nearly uniform 30 km mesh w/ Ice-Shelf Cavities (wISC) around Antarctica and has 64 vertical layers.)",\n "ocnBgchem": "none",\n "seaIce": "MPAS-Seaice (E3SMv3.0,MPAS-Ocean grid; the CICE Consortium’s Icepack as the default sea ice column physics)",Thanks Tony, this should fit within the cmor char limit. In your future test, could you replace this with the place=-holder string currently used in cmip6-cmor-tables/Tables/CMIP6_CV.json.
OK, I can edit our local “cmip6-cmor-tables/Tables/CMIP6_CV.json“ with this entry. and run more e2c testing.
This will replace the “placeholder”:
"E3SM-3-0":{
"activity_participation":[
"CFMIP",
"CMIP",
"DAMIP",
"RFMIP",
"ScenarioMIP"
],
"cohort":[
"Published"
],
"institution_id":[
"E3SM-Project"
],
"license_info":{
"exceptions_contact":"@llnl.gov <- e3sm-data-support",
"history":"2022-08-23: initially published under CC BY 4.0",
"id":"CC BY 4.0",
"license":"Creative Commons Attribution 4.0 International (CC BY 4.0; <https://creativecommons.org/licenses/by/4.0/)>",
"source_specific_info":"",
"url":"<https://creativecommons.org/licenses/by/4.0/>"
},
"source_id":"E3SM-3-0",
"source":"E3SM 2.0 (2022): (details TBD)"
},(We don't need "license_info" ?)
But I’ll wait to push it (if I am even authorized…)
I don’t know if cmore needs “license_info”, can launch a small test to verify.
My initial test (CMORizing) failed with the CV, as it was missing the compressed “source” section. Comparing to other entries, I surmise the correct entry will be:
"E3SM-3-0":{
"activity_participation":[
"CMIP"
],
"institution_id":[
"E3SM-Project"
],
"label":"E3SM 3.0",
"label_extended":"E3SM 3.0 (Energy Exascale Earth System Model)",
"model_component":{
"aerosol":{
"description":"MAM5 w/ prognostic stratospheric aerosols (atmos physics grid)"
},
"atmos":{
"description":"EAM (E3SMv3.0, cubed sphere spectral-element; 5400 els., 30x30 per cube face. Dynamics: degree 3 (p=3) polynomials within each spectral els., 112 km ave. resolution. Physics: 2x2 finite volume cells within each spectral els., 1.5 degree (168 km) average grid spacing; 80 vertical layers w/ top at 60 km)"
},
"atmosChem":{
"description":"Troposphere: UC Irvine interactive chemistry module. Stratosphere: linearized interactive ozone (LINOZ v3) (atmos physics grid)"
},
"land":{
"description":"ELM (E3SMv3.0, 0.5 deg lat/lon grid, dynamic vegetation growth model), MOSART (E3SMv3.0, 0.5 deg lat/lon grid)"
},
"landIce":{
"description":"none"
},
"ocean":{
"description":"MPAS-Ocean (E3SMv3.0, new IcoswISC30E3r5 mesh, nearly uniform 30 km mesh w/ Ice-Shelf Cavities (wISC) around Antarctica and has 64 vertical layers.)"
},
"ocnBgchem":{
"description":"none"
},
"seaIce":{
"description":"MPAS-Seaice (E3SMv3.0, MPAS-Ocean grid; the CICE Consortium’s Icepack as the default sea ice column physics)"
}
},
"release_year":"2024",
"source_id":"E3SM-3-0",
"source":"E3SM 3.0 (2025): \naerosol: MAM5 w/ prognostic stratospheric aerosols (atmos physics grid),\natmos: EAM (E3SMv3.0, cubed sphere spectral-element; 5400 els., 30x30 per cube face. Dynamics: degree 3 (p=3) polynomials within each spectral els., 112 km ave. resolution. Physics: 2x2 finite volume cells within each spectral els., 1.5 degree (168 km) average grid spacing; 80 vertical layers w/ top at 60 km),\n atmosChem: Troposphere: UC Irvine interactive chemistry module. Stratosphere: linearized interactive ozone (LINOZ v3) (atmos physics grid),\n land: ELM (E3SMv3.0, 0.5 deg lat/lon grid, dynamic vegetation growth model), MOSART (E3SMv3.0, 0.5 deg lat/lon grid), \n landIce: none,\n ocean: MPAS-Ocean (E3SMv3.0, new IcoswISC30E3r5 mesh, nearly uniform 30 km mesh w/ Ice-Shelf Cavities (wISC) around Antarctica and has 64 vertical layers.),\n ocnBgchem: none,\n seaIce: MPAS-Seaice (E3SMv3.0,MPAS-Ocean grid; the CICE Consortium’s Icepack as the default sea ice column physics)"
},The “source” entry runs to 983 chars, well below 1024.
I am going to re-try my test (day.pr).
Hello, @Chris Golaz , @Shaocheng Xie , @Luke Van Roekel , @Peter Thornton , @Elizabeth Hunke @Renata McCoy , I’m working on registering E3SM v3 LR simulation with CMIP6, I copied over the component model description from v2 LR, could you please help make updates for our v3 LR model data? The registration has character limits, so we need to keep it concise. Thank you all.