wiki:IPSLCMIP6

Version 191 (modified by oboucher, 6 years ago) (diff)

--

General information on CMIP6

For more general information on CMIP6, see http://www.wcrp-climate.org/wgcm-cmip/wgcm-cmip6

Gateway to the CMIP6-endorsed MIPs, see https://www.wcrp-climate.org/modelling-wgcm-mip-catalogue/modelling-wgcm-cmip6-endorsed-mips

CMIP6 foundation paper: http://www.geosci-model-dev.net/9/1937/2016/gmd-9-1937-2016.html

Special GMD issue on MIPs: http://www.geosci-model-dev.net/special_issue590.html

ESDOC for CMIP6 and other : http://search.es-doc.org/

WGCM-20 has some info on CMIP6 here : https://www.wcrp-climate.org/modelling-wgcm-meetings/64-unifying-themes/unifying-themes-modelling/modelling-wgcm/modelling-wgcm-sessions/1027-wgcm20-presentations

See also: https://pcmdi.llnl.gov/projects/input4mips/ and https://docs.google.com/document/d/1pU9IiJvPJwRvIgVaSDdJ4O0Jeorv_2ekEtted34K9cA/edit?usp=sharing

The CMIP6 data request is documented at https://earthsystemcog.org/projects/wip/CMIP6DataRequest

CMIP6 citation: https://pcmdi.llnl.gov/home/CMIP6/CitationRequirements6-0.html

Climate Data Request

On peut rapporter les erreurs sur la page https://forge.ipsl.jussieu.fr/igcmg/wiki/RetourSurLesPings

Le training sur le workflow CMIP6 est disponible ici: http://forge.ipsl.jussieu.fr/igcmg_doc/wiki/Train

Supercomputing for CMIP6

Planning for CMIP6 production in 2018 : doc en lecture, commentaires possibles

Mailing list at IPSL: cmip6_dev@… available from https://listes.ipsl.fr/sympa/info/cmip6_dev

HPC consumption :

CMIP6 simulations launched during last 2 weeks through the supervisor (Limited access)

Information on new disk spaces: http://forge.ipsl.jussieu.fr/igcmg_doc/wiki/DocBenvBtgcc#SpecificfilesystemsforCMIP6

Model IPSLCM6.1 or IPSL-CM6A-LR

Deck at IPSL

Deck (entry card for CMIP)

  • Person in charge: Olivier Boucher
  • Persons performing the runs: IPSL engineers (Arnaud, Ionela, Sébastien, Sonia, Marie-Alice)
  • Models: all configurations used for the MIPs: IPSL-CM6A-LR, IPSL-CM6-MR (??), IPSL-CM5A2-(VLR)
  • piControl : 500 years at least
  • AMIP : 1979-2014 : 10 members of 35 years
  • +1%CO2/y : 140 years
  • abrupt4xCO2 : 250 years
  • DECK at IPSL

Historical (entry card for CMIP6)

  • historical : 1850-2014
  • 10 members
  • numbered from 2 to 11 and then choose which one becomes 1 with full output

Individual MIPs at IPSL

AerChemMIP (strat chemistry)

  • Persons in charge: Slimane Bekki, Marion Marchand
  • Persons performing the runs: + Tatsuo
  • Model version: IPSL-CM6-LR-REPROBUS
  • Simulations: Hincast: 1960-2010, Future: 1960-2100, Fixed CFCs: 1960-2100, Fixed GHG: 1960-2100

AerChemMIP (trop chemistry)

  • Person in charge: Didier Hauglustaine
  • Persons performing the runs: Didier Hauglustaine (+ Anne Cozic as a back up)
  • model to be used: IPSL-CM6-CHT VLR AP 96x95x39 (all chemistry NMHC_AER)
  • Simulations :
    • piControl (500 years)
      • TIER 1
        • 1.1.1 transient historical (165 years)
        • 1.1.1 transient historical 1850 NTCF emissions (165 years)
        • 2.1.1 Reference SSP3-7 (40 years)
        • 2.1.2 Perturbation SSP3-7 (40 years)
        • SSP3-7 no change in biogenic (165 years)
      • TIER2:
        • 1.1.3 transient historical perturbation 1850 biogenics (165 years)
        • 2.1.1 SSP1-2.6 impact of land use (40 years)

AerChemMIP (aerosols)

  • Persons in charge: Yves Balkanski, Olivier Boucher
  • Persons performing the runs: Yves Balkanski, CDD Crescendo, Olivier Boucher (+ Anne Cozic as a back up)
  • Create aerosols forcing : Simulation LMDZORINCA 144x142x79 NP - historical and 1 scenario (Yves)
  • model : IPSLCM6CHT 144x142x79 NP (only Aerosols) (24h/year - 377 procs)
  • Article : https://www.geosci-model-dev.net/10/585/2017/
  • simulations
    • Tier 1 Experiments
      • 1.Future policy effects on NTCFs and their climate impacts (328 years)
      • 2.1. Transient coupled ocean climate impacts
      • 2.2. ERFs, as 2.1 but specified transient SST simulations
      • 3. Historical forcing from reactive WMGHGs
      • 3.3 Perturbation: 1850 WMGHG concentrations, 2014 aerosols (not NOx) emissions. (30 years)
      • 3.4 Perturbation: 1850 WMGHG concentrations, 2014 BC emissions. (30 years)
    • Tier 2 and 3 Experiments
      • Address combinations of NTCFs to identify nonlinearities
        • Separate aerosols: coupled ocean and ERF (30 years)
        • Timeslice ERFs for aerosol, BC (30 years)
      • Quantifying the ERFs of changes in natural emissions
        • Timeslice ERFs: dust, seasalt, DMS, fires, (4*30 years)
    • piControl

C4MIP

  • Person in charge: Patricia Cadule
  • Persons performing the runs: Patricia Cadule, Laurent Bopp
  • Simulations:
    • Tier 1:
      • 1pctCO2-bgc: biogeochemically-coupled version of 1 percent per year increasing CO2 experiment (150 years)
      • esm-ssp585: update of emission-driven RCP8.5 based on SSP5 (86 years)
    • Tier 2:
      • 1pctCO2-rad: radiatively-coupled version of 1 percent per year increasing CO2 experiment (140 years)
      • 1pctCO2Ndep: 1 percent per year increasing CO2 experiment with increasing N-deposition (150 years)
      • 1pctCO2Ndep-bgc: biogeochemically-coupled version of 1 percent per year increasing CO2 experiment with increasing N-deposition (150 years)
      • hist-bgc: biogeochemically-coupled version of the simulation of the recent past with CO2 concentration prescribed (165 years)
      • ssp585-bgc: biogeochemically-coupled version of the updated emission-driven RCP8.5 based on SSP5 (86 years)
    • ssp585-bgcExt: biogeochemically-coupled version of the updated emission-driven RCP8.5 based on SSP5 - biogeochemically coupled version of the concentration-driven future extension of the SSP5-8.5 scenario (200 years)
      • ssp534-over-bgc: biogeochemically-coupled version of the RCP3.4 overshoot scenario branching from SSP58.5 (86 years)
    • ssp534-over-bgcExt: biogeochemically-coupled version of the RCP3.4 overshoot scenario branching from SSP58.5 - biogeochemically coupled version of the concentration-driven future extension of the SSP5-3.4-over scenario (200 years)
  • Other:
    • piControl: Pre-Industrial Control (500 years)
    • esm-piControl:esm-piControl (500 years)
    • 1pctCO2: 1 percent per year increase in CO2 (150 years)
    • historical: all-forcing simulation of the recent past (165 years)
    • esm-hist: all forcing simulation of the recent past with an Earth system model (165 years)
    • esm-hist-ext: post-2014 all-forcing simulation with an Earth system model (N years)
    • ssp585: update of RCP8.5 based on SSP5 (86 years)
  • Model to be used:
    • concentration driven runs: IPSL-CM6.1-LR (and potentially IPSL-CM6.1-MR for a small number of runs 1pctCO2-bgc only)
    • emission-driven runs: IPSL-CM6.1-LR and ESMIPSL (with 3D activation of CO2 transport in the atmosphere)
  • Runs to be performed:
    • concentration-driven runs (1pctCO2-bgc, 1pctCO2-rad, hist-bgc, ssp585-bgc, ssp534-over-bgc) --> Patricia Cadule and Laurent Bopp (Total : 150+140+165+86+60 = 601 years)
    • emission-driven runs (esm-piControl, esm-hist, esm-ssp585) --> Patricia Cadule (Total = 500+ + 165 + 86 = 751 years)
    • Nitrogen-deposition runs (1pctCO2Ndep, 1pctCO2Ndep-bgc) --> to be defined (Total = 150 + 150 = 300 years)

Infographics created by Jorge Martinez-Rey (jorge@…)

CDRMIP

CFMIP

  • Person in charge: Sandrine Bony
  • Persons in charge of runs: Ionela M. (avec Abderahmanne A. et Laurent F.)
  • Model: IPSL-CM6-LR (et aussi éventuellement 6.2 et 5A)
  • Tier1 (total 174 model years) + part of Tier 2 (max 1030 model years) per model

DAMIP

  • Person in charge: Guillaume Gastineau
  • Persons in charge of runs: Guillaume Gastineau + autres
  • Description: https://www.geosci-model-dev.net/9/3685/2016/
  • Model: IPSL-CM6-LR
  • Tier1: 12 historical (3 members x 4 experiments)
  • Tier2: 3 historical + 2 future
  • Tier3: 9 historical (3 members x 3 experiments) + ... + ...

DCCP

  • Person in charge: Juliette Mignot (+ Didier Swingedouw et Eric Guyliardi)
  • Person doing the runs: Juliette Mignot, Didier Swingedouw
  • Model: IPSL-CM6-VLR
  • Number of years: TBC

FAFMIP

  • Person in charge: Didier Swingedouw (didier.swingedouw@…)
  • See: http://www.met.reading.ac.uk/~jonathan/FAFMIP/ for a full description of the experimental design.
  • Model: IPSL-CM6-LR
  • Simulations: using IPSL model, it is planned to perform the three experiments (stressFAF, heatFAF, waterFAF) from Tier1 for a total of 210 years of simulations (280 years if control has to be rerun). The main difficulty is to have 3D tendencies in the ocean as diagnostics outputs of the simulations. NEMO will be hopefully equipped with a key to allow such a requirement.

GeoMIP

  • Person in charge : Olivier Boucher
  • See http://climate.envsci.rutgers.edu/GeoMIP/ for more details on GeoMIP6
  • See http://www.geosci-model-dev.net/8/3379/2015/ for experiment design and description
  • Model : IPSL-CM6-LR
  • Simulations: G1ext, G6solar, G7cirrus representing 350 years of simulations + fixed-SST counterparts, possibly G6sulfur with strat aerosol bin model
  • Planning: G1ext can be run shortly after abrupt4xCO2 (150 years), G6 and G7 exps have to wait until after scenarioMIP is underway or completed (200 years + fixed-SST exp)
  • GeoMIP at IPSL

GMMIP

  • Person in charge: Laurent Li
  • See http://www.lasg.ac.cn/gmmip/ and http://www.lmd.jussieu.fr/~li/Li_reunion_climat_GMMIP.pptx for a presentation
  • Modèle: IPSL-CM6-LR
  • La premiere simulation est une simulation AMIP etendue sur la periode 1870/2013.
  • La deuxieme simulation utilisera le modele couple, mais une partie de la SST n'est pas libre, parce qu'elle sera relaxee vers les valeurs observees sur la periode 1870/2013. Les zones de relaxation sont AMO (Atlantique du Nord) et IPO (Pacifique tropical).
  • La troisieme simulation consiste a tester l'orographie, avec les effets mecaniques et thermodynamiques separes.
  • Le nombre total d'annees est de l'ordre de 500 ans (3/5 en mode couple, le reste avec atmosphere seule).

HappiMIP

  • Unofficial MIP
  • Not enough IPSL interest for now

HighResMIP

  • Person in charge: Thomas Dubos

ISMIP6

  • Person in charge: Martin Vancoppenolle
  • Model: any IPSL model contributing to CMIP6
  • No additional specific run
  • Need to make sure all relevant diagnostics are in the standard runs

LS3MIP

  • Person in charge: Frédérique Cheruy
  • Persons doing the runs: Agnès Ducharne (runs forcés ORCHIDEE), Frédérique + Josefine (runs couplés)
  • Modèle : ORCHIDEE + IPSL-CM6-LR
  • Runs historique + scenario + ?? (~few 100 years)

LUMIP

  • Person in charge : Nicolas Vuichard
  • Person performing the runs: Devaraju Narayanappa
  • Model version: IPSL-CM6-LR + ORCHIDEE
    • Runs with IPSL-CM6-LR (total: 1000 yrs):
      • Tier1 (500 yrs): deforest-glob (80 yrs), hist-noLu (165 yrs), ssp370-ssp126Lu (85 yrs), ssp126-ssp370Lu (85 yrs), esm-ssp585-ssp126Lu (85 yrs)
      • Tier2 (500 yrs): 2 additional members of hist-noLu (2x165 yrs) and 2 additional members of ssp370-ssp126Lu (2x85 yrs)
    • Runs with ORCHIDEE offline (total: 3000 yrs)
  • More information on LUMIP from its website or the protocol description paper

OMIP

  • Person in charge : Julie Deshayes
  • Persons performing the runs : Julie Deshayes, Juliette Mignot, Christian Ethé, Laurent Bopp, CDD
  • OMIP1 : forcing COREv2
    • OMIP1 tier A avec eORCA1 resolution : eORCA1 (1deg) runs : 1 run de 310 yr, responsables : Julie et Juliette
    • OMIP1 tier B avec eORCA1 resolution (et PISCES spun up) : eORCA1 (1deg) runs : 2 runs de 310 yr (cf 2 types de concentration CO2 atmosphérique), responsables : Julie, Juliette, Laurent et Christian (le/la responsable sera identifié ultérieurement)
  • OMIP2 : forcing JRA-55do
    • OMIP2 tier A avec eORCA1 resolution : eORCA1 (1deg) runs : 1 run de 310 yr, responsables : Julie et Juliette
    • OMIP2 tier B avec eORCA1 resolution (et PISCES spun up) : eORCA1 (1deg) runs : 2 runs de 310 yr (cf 2 types de concentration CO2 atmosphérique), responsables : Julie, Juliette, Laurent et Christian (le/la responsable sera identifié ultérieurement)
  • ORCA025 : quels forcages ?
    • OMIP tier A avec eORCA025 resolution : eORCA025 (1/4 deg) runs : 1 run de 310 yr, responsable : CDD CRESCENDO qui va être embauché au LSCE
    • OMIP tier B avec eORCA025 resolution : eORCA025 (1/4 deg) runs : 2 runs de 310 yr (cf 2 types de concentration CO2 atmosphérique), responsable : CDD CRESCENDO qui va être embauché au LSCE
  • Calendrier: les runs OMIP eORCA1 ne seront lancés qu’une fois la configuration couplée avec eORCA1 figée et donc spinup pour la bio réalisé >> pas avant 2017 a priori les runs OMIP eORCA025 ne seront lancés qu’une fois la configuration forcée eORCA025 figée, a priori à la fin du contrat de Gildas Mainsant (CDD Labex IPSL physiquement au LGGE, fin de contrat decembre 2016 je crois)

PMIP

  • Persons in charge: Masa Kageyama, Pascale Braconnot , Myriam Khodri
  • 5 simulations: past1000, midHolocene, lgm, lig127k, midPliocene-eoi400
  • All simulations but past1000 are equilibrium runs: the simulations will be run to stable state, the last 100 to few hundred years will be processed and sent to ESGF.
  • Past1000: a single 1150 years (850-2000), after a short stabilisation run with 850 conditions ( a few hundred years)
  • Persons doing the runs: Pascale (midHolocene, lig127k), Myriam Khodri (past1000), Masa (lgm), Ning Tan (midPliocene-eoi400), Jean-Yves Peterschmitt
  • All PMIP4-CMIP6 simulations are Tier 1.
  • Model: IPSL-CM6-LR (Tier 1)
  • Planned timing of the simulations: past1000, midHolocene, lig127k can be started as soon as model version is frozen, lgm and midPliocene-eoi400 require more time for their set up but should start asap. The aim is to get our first results out for the PMIP conference at the end of September.
  • see https://wiki.lsce.ipsl.fr/pmip3/doku.php/pmip3:cmip6:design:index for more details

RFMIP

ScenarioMIP

SIMIP

SolarMIP

  • Person in charge : Marion Marchand, Slimane Bekki
  • Runs distributed over the other MIPs

VolMIP

  • Person in charge: Myriam Khodri
  • Persons responsible for the runs: Myriam Khodri et Nicolas Lebas
  • Model: IPSL-CM6-LR (LMDZ144x140x79, NEMO-ORCA1)
  • VolMIP: 140 simulations for a total of 1067 years

CMIP6 forcing datasets

A copy of ESGF input4MIPs data is on ciclad: /prodigfs/project/input4MIPs

Othe forcing datasets are on ciclad: /home/oboucher/CMIP6

Scripts are available on : http://forge.ipsl.jussieu.fr/igcmg/svn/TOOLS/CMIP6_FORCING/

Description paper : https://www.overleaf.com/10684852tnwyjmbqsczv#/40053679/

Greenhouse gases

Person in charge : Olivier Boucher

See http://www.climate-energy-college.net/cmip6-data for more information

We use /prodigfs/project/input4MIPs/UoM/GHGConcentrations/CMIP/UoM-CMIP-1-2-0/yr/ data

The data on curie are available from $R_IN/IGCM/ATM/GHG/CMIP6/historical

The README and extraction script is on ciclad: /home/oboucher/CMIP6/GHG and available from: http://forge.ipsl.jussieu.fr/igcmg/svn/TOOLS/CMIP6_FORCING/GHG/

The GHG to be used are CO2, CH4, N2O, HFC134a-eq (fed to CFC11) and CFC12eq (fed to CFC12).

So in the end, we have 5 files for GHG for CMIP6 runs with IPSL-CM6

  • CO2_CMIP6_0000_2014.txt
  • CH4_CMIP6_0000_2014.txt
  • N2O_CMIP6_0000_2014.txt
  • CFC11eq_CMIP6_0000_2014.txt
  • CFC12eq_CMIP6_0000_2014.txt

Note that the CFCeq files have non-zero values for pre-industrial conditions because they include some natural species.

Chemistry and aerosols

Persons in charge : Thibaut Lurton, Anne Cozic, Yves Balkanski, Olivier Boucher See http://www.globalchange.umd.edu/CEDS/ for more information

Oxidant fields extracted from IPSL-CM5A simulations on curie : /ccc/cont003/home/gen6178/oboucher/CMIP6/OXIDANTS Extraction file : /ccc/cont003/home/gen6178/oboucher/CMIP6/OXIDANTS/oxidant.sh These files need to be stored on shared spaces.

Emission fields on ciclad

  • PNNL data /prodigfs/project/input4MIPs/PNNL-JGCRI/emissions/CMIP/CEDS-2017-05-18/mon
  • VUA data /prodigfs/project/input4MIPs/VUA/emissions/CMIP/VUA-CMIP-BB4CMIP6-1-2/mon

Script on ciclad : /home/oboucher/CMIP6/AER_EMISSIONS/REGRID/regrid_v7.sh and now available from: http://forge.ipsl.jussieu.fr/igcmg/svn/TOOLS/CMIP6_FORCING/AER_TROP_EMISSIONS/ and http://forge.ipsl.jussieu.fr/igcmg/svn/TOOLS/CMIP6_FORCING/AER_TROP_CONCENTRATIONS/

The data on curie are available from xxxxx

For a description of the interpolation methods (over time) used on emission fields, see Interpolation of chemistry and aerosol input emission files

For a description of pseudo-aerosol data sets, see Pseudo aerosols

Final aerosol forcing files for CMIP6 can be found on Curie: /ccc/work/cont003/igcmg/igcmg/IGCM/ATM/AEROSOLS/CMIP6/v1/144x142/L79/aerosols????_from_inca.nc For a description of their production process, see Production of CMIP6 aerosol forcing files

Solar forcing

Person in charge: Olivier Boucher, Marion Marchand, Rong-Ming Hu, Rémy Thieublemont

See http://solarisheppa.geomar.de/cmip6 for more information

The scripts, readme and data are on ciclad /home/oboucher/CMIP6/SOLAR but can now be downloaded from: http://forge.ipsl.jussieu.fr/igcmg/svn/TOOLS/CMIP6_FORCING/SOLAR/

The solar forcing files are on ciclad /home/oboucher/CMIP6/SOLAR/OUTPUT

The data on curie are available from $R_IN/IGCM/ATM/SOLAR/CMIP6/v3.1

The time period of data file is from 1 January 1850 to 31 December 2299. We reprocessed the daily spectral solar forcing data into the data files with 2 and 6 wavebands for RRTM. We also added the variables of solar irradiance anomaly and fraction for the model inputs in a particular waveband. The netcdf files need to have the same number of days as the LMDz year that is run (either 360, 365 or 366 days). There is a flag ok_suntime_rrtm that can be activated but this only works for RRTM, flag_rrtm=1.

The average for piControl consists of the first two solar cycles after 1850 (as per CMIP6 protocol). The average for pdControl consists of a 1991-2010 average (not an official CMIP6 run).

Annual files are available for the historical and future period: solarforcing_v3.1_daily_6bands_XXXX.nc where XXXX is the year. The files are for a Gregorian calendar (365 and 366 days according to the year).

There is also a scenario with a lower solar constant in 21st century for sensitivity study. This is not an official CMIP6 scenario though.

The solar forcing file is to be renamed solarforcing.nc in the running directory.

Tropospheric and stratospheric ozone

Person in charge: David Cugnet, Marion Marchand, Slimane Bekki with help from Thibaut Lurton

See http://www.met.reading.ac.uk/ccmi/?page_id=375 for more information

Raw data are available on ciclad:

/prodigfs/project/input4MIPs/CMIP6/CMIP/UReading/UReading-CCMI-1-0/atmos/mon/vmro3/gn/v20160711

Preprocessed data are available:

  • on ciclad: /prodigfs/ipslfs/igcmg/IGCM/ATM/OZONE/UReading/historical.v20160711.v2/original
  • on curie: /ccc/work/cont003/igcmg/igcmg/IGCM/ATM/OZONE/UReading/historical.v20160711.v2/original

Two versions of the data for CMIP6 on curie are available from:

  • /ccc/work/cont003/igcmg/igcmg/IGCM/ATM/OZONE/UReading/historical.v20160711/interpol
  • /ccc/work/cont003/igcmg/igcmg/IGCM/ATM/OZONE/UReading/historical.v20160711.v2/interpol

The first version had two issues, corrected in the second:

  • one cell shift for positive longitudes due to an index error in the routine interpolating the ozone along longitude.
  • a wrong first field (december of previous year) in "climoz_LMDZ_1850_1850_clim.nc" (that has to be identical to field nr. 13).

For processing details, see: http://forge.ipsl.jussieu.fr/igcmg/wiki/InfosOzone#Informationsurlesfichiersdefor%C3%A7ageOzone-VersionCMIP6

Some of the scripts are on: http://forge.ipsl.jussieu.fr/igcmg/svn/TOOLS/CMIP6_FORCING/OZONE/

Volcanic aerosol

Persons in charge: Olivier Boucher, Myriam Khodri, Nicoas Lebas

See ftp://iacftp.ethz.ch/pub_read/luo/CMIP6/ for more information

The script can be downloaded from: http://forge.ipsl.jussieu.fr/igcmg/svn/TOOLS/CMIP6_FORCING/AER_STRAT/

The data for IPSL-CM6 on curie is taken from $R_IN/IGCM/ATM/STRATAERO/CMIP6/v1

There is a flag flag_aerosol_strato that has to be set to 2 to run with CMIP6 volcanic aerosol forcings.

The data are monthly latitude-altitude and are not interpolated temporally between months.

See http://forge.ipsl.jussieu.fr/volcan/wiki/model/LMDZ6_3par for first tests with LMDz

See http://forge.ipsl.jussieu.fr/volcan/wiki/model/IPSLCM6_3par for first tests with IPSL-CM6

Land use

Persons in charge: Philippe Peylin, Nicolas Vuichard

See https://cmip.ucar.edu/lumip for more information

Attachments (1)

Download all attachments as: .zip