Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

Cannot find CESM2 CMIP6 experiment output

Eric

Eric
Member
Hello,

The CESM2 official website gives a list of CESM2 EXPERIMENTS, DATA & DIAGNOSTIC OUTPUT files and their locations:
/glade/collections/cdg/timeseries-cmip6/
However, I went to this directory and did not find most of the output files listed on the website, for example, b.e21.BHIST_BPRP.f09_g17.CMIP6-esm-hist.001 and b.e21.BSSP585_BPRPcmip6.f09_g17.CMIP6-esm-ssp585.001 (The former is lost and the latter only has atm output)
Can you tell me where to get these cmip6 experiment outputs? Thank you!
 

strandwg

Moderator
Staff member
Hello,

The CESM2 official website gives a list of CESM2 EXPERIMENTS, DATA & DIAGNOSTIC OUTPUT files and their locations:
/glade/collections/cdg/timeseries-cmip6/
However, I went to this directory and did not find most of the output files listed on the website, for example, b.e21.BHIST_BPRP.f09_g17.CMIP6-esm-hist.001 and b.e21.BSSP585_BPRPcmip6.f09_g17.CMIP6-esm-ssp585.001 (The former is lost and the latter only has atm output)
Can you tell me where to get these cmip6 experiment outputs? Thank you!

That directory name isn't correct - would you tell me the URL of the web page that you were looking at?

All CMIP6 data are available via ESGF.
 

Eric

Eric
Member
Hello Gary,

For b.e21.BHIST_BPRP.f09_g17.CMIP6-esm-hist.001, I was looking at https://esgf-node.llnl.gov/search/c...experiment_id=esm-hist&variant_label=r1i1p1f1 as instructed in the CESM2 website. It retunrs 188 results but I don't understand the name of these returns. For example, the first returned result is CMIP6.CMIP.NCAR.CESM2.esm-hist.r1i1p1f1.AERmon.co2.gn
What does AERmon.co2.gn mean?

For b.e21.BSSP585_BPRPcmip6.f09_g17.CMIP6-esm-ssp585.001, I was looking at https://esgf-node.llnl.gov/search/c...periment_id=esm-ssp585&variant_label=r1i1p1f1 No results were returned.

By the way, can I access these CMIP6 output from the glade storage system?

Thank you!!
 

katec

CSEG and Liaisons
Staff member
Eric, the name of the variable in each file is determined by the model that produced it, the length of time the variable was averaged over, and then the CMIP name of the variable. AERmon means Monthly atmospheric chemistry and aerosol data. CO2 means carbon dioxide. This webpages gives an overview of the data source and averaging codes: Data Request Section: miptable And this website gives an overview of the CMIP standardized variable naming: Variable List

It looks like the data from BSSP585_BPRPcmip6 has not been published to the ESGF yet. Can you get in contact with the scientists involved in that experiment to ask how to access the data?
 

strandwg

Moderator
Staff member
The esm-ssp585 run was apparently never published into ESGF.

I'm correcting that now.
 

Eric

Eric
Member
Thanks Kate and Gary for your help! The information is very helpful.
Gary, I noticed that in /glade/collections/cdg/data/CMIP6/ScenarioMIP/NCAR/CESM2/ssp585, there are four folders:
r4i1p1f1
r10i1p1f1
r11i1p1f1
Do you know what is the difference between these folders? What do r, i, p, and f mean?

Thank you!
 

strandwg

Moderator
Staff member
Thanks Kate and Gary for your help! The information is very helpful.
Gary, I noticed that in /glade/collections/cdg/data/CMIP6/ScenarioMIP/NCAR/CESM2/ssp585, there are four folders:
r4i1p1f1
r10i1p1f1
r11i1p1f1
Do you know what is the difference between these folders? What do r, i, p, and f mean?

Thank you!

The "r" values differentiate between the ensemble members and correspond to the same numbers for the historical runs.

The "i", "p" and "f" are used to denote different initializations, different physics schemes, and different forcings. They are all set to "1" for these runs as all three are the defaults for CESM2 CMIP6 runs.
 

Eric

Eric
Member
Hi Gary,

I noticed the SSP585 data that you uploaded to ESGF is under /CMIP6/C4MIP in glade system and is r1i1p1f1. There are also SSP585 simulation results under /CMIP6/ScenarioMIP, r4, r10, and r11. Can I regard these as different members of a ssp585 simulation ensemble? Can I use these data that is not published in ESGF for scientific analysis? Similar condition exists for other simulations, for example, BHIST. Only r1i1p1f1 is published in ESGF while other members are available in glade. Can I use them for scientific analysis?

There are three folders under each variable's directory: files, latest, and v20210607. Do you know what the difference between data under different folders is?

Thank you!
 

katec

CSEG and Liaisons
Staff member
So the SSP data generated by the C4MIP experiment is not equivalent to the ScenarioMIP experiments. The C4MIP model setup is different than the ScenarioMIP setup, and thus they CANNOT be considered part of the same ensemble. The ScenarioMIP SSP simulations were done twice, because an error in the forcing was found after they were first completed. The original data was retracted and the new simulation data published. Only the data currently available on ESGF should be used for scientific analysis. (Unless you have specific notice from the PIs for a specific MIP) This goes for all CMIP experiments. The ESGF data is the final form, analyzed and confirmed result data from NCAR's CMIP experiments.

And, just to answer your last question, the files in most of the "latest" and "v20210607" sub-directories are actually the same files. If you type "ls -al" it will show you that the filenames in the list are just links back to the original data in the "files" directory.
 

Eric

Eric
Member
Hi Kate,

Thanks for the response! So the data on ESGF is the same as the data under glade/collections/cdg/data/CMIP6/, right? If so I would not need to download them from ESGF. And it seems only data of SSP585 is published to ESGF. Are other SSP scenarios' simulation results available for scienctific analysis? Thanks!!
 

strandwg

Moderator
Staff member
Hi Gary,

I noticed the SSP585 data that you uploaded to ESGF is under /CMIP6/C4MIP in glade system and is r1i1p1f1. There are also SSP585 simulation results under /CMIP6/ScenarioMIP, r4, r10, and r11. Can I regard these as different members of a ssp585 simulation ensemble? Can I use these data that is not published in ESGF for scientific analysis? Similar condition exists for other simulations, for example, BHIST. Only r1i1p1f1 is published in ESGF while other members are available in glade. Can I use them for scientific analysis?

There are three folders under each variable's directory: files, latest, and v20210607. Do you know what the difference between data under different folders is?

Thank you!

As Kate notes, there are two kinds of SSP585 runs. Under ScenarioMIP, concentrations are specified; under C4MIP, emissions are specified. The SSP585 experiments have two different names - "ssp585" from ScenarioMIP, "esm-ssp585" from C4MIP. There are three ensemble members for ScenarioMIP "ssp585" - r4, r10 and r11, and there's one C4MIP "esm-ssp585" run, r1.

All of the data on glade are published into ESGF.

Other SSP scenarios are available as part of ScenarioMIP - SSP1.26, SSP2.45, and SSP3.70. The specifics of the directory structure are part of how ESGF operates and aren't particularly important.
 

Eric

Eric
Member
Hello,

I saw that on this webpage: Data Request Section: miptable there is a variable called fco2antt {Amon}: Carbon Mass Flux into Atmosphere Due to All Anthropogenic Emissions of CO2
However, I didn't find this variable in the CESM2 CMIP6 published data. Can you help me find this variable? Thanks!
 
Top