Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[ERROR] in the co2 cmorized data #315

Closed
adagj opened this issue Mar 31, 2022 · 17 comments
Closed

[ERROR] in the co2 cmorized data #315

adagj opened this issue Mar 31, 2022 · 17 comments
Labels
Erratum error error in the dataset Retract

Comments

@adagj
Copy link
Collaborator

adagj commented Mar 31, 2022

Hi,
after a disussion with @DirkOlivie and @tjiputra, we wonder if the cmorized varible co2 should be retracted fro the non-esm simulations.

Describe the error
For piControl co2 is ~430.9*1e-6 mol/mol (/projects/NS9034K/CMIP6/CMIP/NCC/NorESM2-LM/piControl/r1i1p1f1/AERmon/co2/gn/latest/). Should be ~282e-6mol/mol
The numbers are almost the same for the abrupt-4xCO2 simulation (/projects/NS9034K/CMIP6/CMIP/NCC/NorESM2-LM/abrupt-4xCO2/r1i1p1f1/AERmon/co2/gn/latest/ on NIRD). We know that the piControl simulation and the 4xCO2 simulation have different co2 concentration, so that can't be correct.

For the PAMIP simulations forced by year 2000 emissions and GHG forcing (e.g. /projects/NS9034K/CMIP6/PAMIP/NCC/NorESM2-LM/pa-pdSIC-ext/r1i1p1f1/AERmon/co2/gn/latest/ on NIRD), the co2 is 282.5e-6 mol/mol in the cmorized output. It should be at year 2000 level so more like ~370e-6 mol/mol.

Too me it seems like the CO2 variable in NorESM raw data always has values around 430 kg/kg so it is probably not really used, and then we should probably not cmorize it.

Way to correct
(i) For the simulations with interactive carbon cycle (esm simulations), the cmorized fields are correct and make sense.

(ii) For the simulations without interactive carbon cycle but still coupled to the ocean, the cmorized fields are not useful (and should possibly not be cmorized).

(iii) It looks to me the conversion from [kg/kg] to [mol/mol] in:
/projects/NS9034K/CMIP6/CMIP/NCC/NorESM2-LM/piControl/r1i1p1f1/AERmon/co2/gn/latest/
(as well as abrupt-4xCO2 and probably other simulations) was not done correctly.

@adagj adagj added the error error in the dataset label Mar 31, 2022
@YanchunHe
Copy link
Collaborator

We have discussed this before, refer to #181

The unit conversion is done by a factor of 1e6*29./(12.+2.*16.). from kg/kg to mol/mol.

If the co2 in some of the experiments should be retracted, please list those experiments and tables, so that Alok can retract them.

@oyvindseland
Copy link
Collaborator

I think the title of this issue is somewhat misleading. The number is correct for PI conditions but the number is not connected to the CO2 value that is actually used for the model in a non-carbon cycle setting. As it is more confusing than describing the question is if that variable can be retracted for all non-concentration driven simulations. Jerry can likely provide a list of simulations that should retain the value. If retracting the field is to much work, an errata should probably be published

@monsieuralok
Copy link
Collaborator

@YanchunHe what should be done regarding this issue? and which datasets/variables/experiments are affected?

@YanchunHe
Copy link
Collaborator

Hi Ada and Øyvind,

@adagj @oyvindseland ?

Could any of you indicate in more details what to proceed by Alok?

Could you list the all the datasets (experiments, etc) that should be reported with Erratum and/or retracted?

Thanks!

@YanchunHe
Copy link
Collaborator

Could anyone make a summary how to proceed?

@oyvindseland
Copy link
Collaborator

Hi I think the best way forward is to retract the variable from all non-CO2 prognostic simulations @tjiputra for list of CO2 prognostic, with the explanation in errata that the cmorized value is not the value used for the simulations.

@monsieuralok
Copy link
Collaborator

@YanchunHe Please could you put list here for all variables and frequency .

@monsieuralok
Copy link
Collaborator

@YanchunHe please could you look at this issue what needs to be done?

@YanchunHe
Copy link
Collaborator

YanchunHe commented Aug 29, 2023

Hi @monsieuralok , after confirmed with @tjiputra, datasets from the following experiments will be retracted:

NorESM2-LM

  • 1pctCO2
  • 1pctCO2-bgc
  • 1pctCO2-cdr
  • 1pctCO2-rad
  • hist-GHG
  • hist-aer
  • hist-nat
  • hist-piAer
  • hist-piNTCF
  • hist-spAer-all
  • historical
  • midHolocene
  • piControl
  • ssp126
  • ssp245
  • ssp370
  • ssp370-lowNTCF
  • ssp585

NorESM2-MM:

  • 1pctCO2
  • abrupt-4xCO2
  • historical
  • ssp126
  • ssp245
  • ssp370
  • ssp585

@JorgSchwinger
Copy link
Collaborator

  • esm-hist

  • esm-pi-CO2pulse

  • esm-pi-cdr-pulse

  • esm-ssp585

These are emission driven simulations, for these the co2 variable should be okay and shouldn't be retracted??? @tjiputra?

@tjiputra
Copy link
Collaborator

@JorgSchwinger, @YanchunHe, @monsieuralok,
yes, this is what i told Yanchun today. Probably just a typo.

@YanchunHe
Copy link
Collaborator

@JorgSchwinger, @YanchunHe, @monsieuralok,
yes, this is what i told Yanchun today. Probably just a typo.

Oh, yes, sorry that I made a mistake forgetting to remove these esm-* from the list. I have corrected from the above list.

@monsieuralok the above list is now the correct one.

@monsieuralok
Copy link
Collaborator

monsieuralok commented Oct 3, 2023

@YanchunHe if I am correct in NorESM2-MM piControl should also be there ? Should be remove for all version or did we published any corrected version or planning to publish as will open an errata issue? Example data version 20230616 is having correct datasets or not?

@YanchunHe
Copy link
Collaborator

@YanchunHe if I am correct in NorESM2-MM piControl should also be there ? Should be remove for all version or did we published any corrected version or planning to publish as will open an errata issue? Example data version 20230616 is having correct datasets or not?

Yes, even if they are version v20230616, the co2 should be extracted, except for those experiments started with esm-*.

I have a look, and there are following experiments with v20230616 that need to be retracted (plus those earlier version as show above):

  • ssp245
  • historical
  • piControl
  • amip
  • amip-p4K
  • 1pctCO2

While the following with can be kept:

  • esm-ssp534-over
  • esm-hist

Thanks!

@oyvindseland
Copy link
Collaborator

Just to make sure. You can retract one dataset without changing the rest of the datasets? I presume that if you want to republish one dataset you need to republish the whole set, but given the instability of the server I hope that republishing of all major datasets can be avoided.

@monsieuralok
Copy link
Collaborator

@oyvindseland yes only co2 datasets will be retracted; not the whole experiments
@YanchunHe Here is errata issue:- https://errata.es-doc.org/static/view.html?uid=2dc98710-b977-ff15-8602-828bd7f8d844

@YanchunHe
Copy link
Collaborator

Thanks @monsieuralok

According to @tjiputra, the CO2 concentration used in the concentration-driven experiments are provided as below:

Atmospheric_CO2_SSP585.csv
Atmospheric_CO2_SSP370.csv
Atmospheric_CO2_SSP245.csv
Atmospheric_CO2_SSP126.csv
Atmospheric_CO2_hist.csv

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Erratum error error in the dataset Retract
Projects
None yet
Development

No branches or pull requests

6 participants