Proposed CMIP5-era experiment_ids for CMIP and ScenarioMIP

[durack1]

From: Taylor, Karl E.
Date: Tuesday, October 8, 2019 at 1:29 PM
To: Brian ONeill, wgcm-wip, claudia tebaldi, detlef.vanvuuren,
Jean-Francois Lamarque, veronika.eyring, Jerry Meehl, Ron Stouffer,
Cath Senior, Greg Flato
Subject: Re: WIP telco in 18 hours

Hi All,

In additional to their ScenarioMIP runs, CCCma is carrying out some RCP scenario runs
(with CMIP5 forcing) with one of their CMIP6 CanESM models.  These simulations are
not called for in CMIP6, but can provide perspective on differences between CMIP5
and CMIP6 (forcing the same model withCMIP5 and CMIP6 future forcing).  The IPCC
will likely be interested.

I think we should enable CCCma (and anyone else) to include CMIP5 RCP future
scenario runs in the CMIP6 archive (to be run with a model that also performs the
CMIP6 DECK, historical, and at least some of the ScenarioMIP simulations).  To do
this within the CMIP6 organizational framework and software infrastructure, one
option would be to ask ScenarioMIP to "sponsor" these experiments (at tier2 or
tier3 priority), and to come up with names for these experiments (or simply adopt
the names from CMIP5:  rcp26, rcp45, rcp85, rcp60)   

What do the ScenarioMIP leaders think of this idea?

A related question concerns performing *historical* simulations with both CMIP5
forcing and CMIP6 forcing to understand how different the responses are to these
different forcing estimates.  There are a couple of options:

1) Define a new experiment name (e.g., Paul has suggested "historical-cmip5"),
and this could be considered a tier3 experiment under the "CMIP" activity, under
the responsibility of the CMIP Panel)

2) Permit publication of a CMIP5-forced historical runs under the name "historical"
and distinguish them from CMIP6-forced runs using a different "forcing_index". 
In CMIP, the forcing_index is assigned an integer value (usually, but not always,
set to "1" for "standard forcing") and is generally used to distinguish similar runs,
but with different forcing estimates (e.g., two historical runs run with different
prescribed ozone, or two historical runs with different estimates of biomass
burning aerosol precursors).  

The second option would not require any special action on our part; the first
option would require registration of 4 new experiments but would make it
absolutely clear that these runs were performed with forcing different from
the standard CMIP6 forcing (for those who can't be bothered with the definition
of "forcing_index"). 

Thoughts?

best regards,
Karl

[durack1]

On 11/24/19, 6:20 PM, "Hideo SHIOGAMA" wrote:
Dear all,
Should "end_year":"2012" of hist-???-CMIP5  be "end_year":"2020"?
1850-2005 use the CMIP5 historical forcing and 2006-2020 use the CMIP5 RCP4.5 forcing.
Best wishes,
Hideo

@npgillett should 2020 become the end year for the *-cmip5 experiments? I am uncertain of what the spec was for CMIP5

[durack1]

The end_year for the proposed DAMIP *-cmip5 experiments have now been updated to 2020.

If I hear nothing from folks by the end of the week, I'll merge the existing experiment entries into master.

It would be beneficial that folks take another check of these before they are committed, the changes are visible at https://github.com/WCRP-CMIP/CMIP6_CVs/pull/837/files#diff-6642f67f4f38024deabaa41b20af5e65

[durack1]

On 12/4/19, 10:15 PM, "Hideo SHIOGAMA" wrote:

    Dear Paul et al.,

    Can you change
    "experiment":"historical well-mixed GHG-only run (CMIP5-era forcing)"
    to
    "experiment":"historical well-mixed GHG-only run (CMIP5-era historical
and RCP4.5 (2006-2020) forcing)"?

    Best wishes,
    Hideo

[durack1]

Hi all, those experiments are now registered. Can you all take a double check look at CMIP6_experiment_id.html and let me know of any errors or tweaks required promptly?

[taylor13]

@martinjuckes @davidhassell -- As agreed at the WIP meeting 2020-02-25, I am reopening this issue because I don't think the changes made to the experiment_id CV have propagated to ES-DOC and the data request.

[slavakharin]

I just noticed the description and experiment attributes for piControl-cmip5 say:

"description":"DECK: control (CMIP5-era [1850-2005] forcing)",
...
"experiment":"pre-industrial control (CMIP5-era [1850-2005] forcing)",

The year range [1850-2005] should probably be removed.

Similarly for piControl-spinup-cmip5.

No biggie though if it's stays in. Just cosmetics.

[durack1]

@slavakharin that was intentional, to indicate the modern era forcings aren’t applied

[slavakharin]

@durack1 but the forcing year range [1850-2005] implies my control run is 156 years long, while the suggested length is 500 yrs. Should this be simply

"description":"DECK: control (CMIP5-era forcing)",
"experiment":"pre-industrial control (CMIP5-era forcing)",

?

[durack1]

@slavakharin in my head the 1850-2005 is the bounds over which a mean climatological average forcing would be calculated, and explicitly excluding the 2006- observational period (as opposed to 1850-2014 for CMIP6).

If it doesn’t make sense to a collective, happy to tweak this

[slavakharin]

@durack1 I would have thought the pre-industrial control runs are performed with forcings prior 1850, not the average of 1850-2005. It is not clear to me why the piControl refers to forcings in the historical period. But I might be overthinking things.

[taylor13]

I think Slava is right. For a picontrol run, the forcing should represent pre-industrial conditions. So, I would replace in piControl-cmip5 and in piControl-spinup-cmip5 the phrase "(CMIP5-era [1850-2005] forcing)" with "(pre-industrial forcing from CMIP5)"

[durack1]

Ok so it seems 2 votes to 1. I propose the following descriptions change:

Sponsoring MIP	experiment_id	Description	Changed
CMIP	piControl	DECK: control	No
CMIP	piControl-cmip5	DECK: control (CMIP5-era pre-industrial forcing)	Yes
CMIP	piControl-spinup	DECK: spin-up portion of the control	No
CMIP	piControl-spinup-cmip5	DECK: spin-up portion of the control (CMIP5-era pre-industrial forcing)	Yes

If I get an all clear, I'll get that change implemented this week

[jflamarque]

Yes it should be modified as Karl mentioned.

Jean-François

---

Jean-François Lamarque Climate and Global Dynamics Laboratory Director National Center for Atmospheric Research P.O. Box 3000 Boulder, CO 80305 Tel: 303-4971495 https://staff.ucar.edu/users/lamar https://staff.ucar.edu/users/lamar

On Mon, Mar 16, 2020 at 2:09 PM Paul J. Durack notifications@github.com wrote:

Ok so it seems 2 votes to 1. I propose the following descriptions change: Sponsoring MIP experiment_id Description Changed CMIP piControl DECK: control No CMIP piControl-cmip5 DECK: control (CMIP5-era pre-industrial forcing) Yes CMIP piControl-spinup DECK: spin-up portion of the control No CMIP piControl-spinup-cmip5 DECK: spin-up portion of the control (CMIP5-era pre-industrial forcing) Yes

If I get an all clear, I'll get that change implemented this week

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/WCRP-CMIP/CMIP6_CVs/issues/805#issuecomment-599736419, or unsubscribe https://github.com/notifications/unsubscribe-auth/AG52JX5SV7CKCIOTZKTH4XDRH2BO7ANCNFSM4I6W27FA .

[taylor13]

But I prefer the construct: "pre-industrial forcing as specified for CMIP5" or the shortened version "pre-industrial forcing from CMIP5". I think it is confusing to say "CMIP5-era pre-industrial forcing"; which is it? CMIP5-era forcing or pre-industrial forcing? We know what we mean by it, but others may not.

[durack1]

So take 2:

Sponsoring MIP	experiment_id	Description	Changed
CMIP	piControl	DECK: control	No
CMIP	piControl-cmip5	DECK: control (pre-industrial forcing from CMIP5)	Yes
CMIP	piControl-spinup	DECK: spin-up portion of the control	No
CMIP	piControl-spinup-cmip5	DECK: spin-up portion of the control (pre-industrial forcing from CMIP5)	Yes

[durack1]

@slavakharin @taylor13 @jflamarque those changes are in place, take a peek at CMIP6_experiment_id.html and let me know of any further tweaks required.

We are still awaiting for confirmation that @martinjuckes has updated the data request, and @charliepascoe has ES-DOCs updated before this can be closed

[durack1]

Closing as noted in email earlier this week