Produce Ozone ancillaries for the ACCESS-ESM1.6 piControl experiments

[atteggiani]

Ozone forcing for ACCESS-ESM1.6 piControl:

Forcing dataset definitions:

See https://input4mips-cvs.readthedocs.io/en/latest/dataset-overviews/ozone/

Input file (when FZJ-CMIP-ozone-1-0 is available at NCI):

${INPUT4MIPS_DIR}/CMIP7/CMIP/FZJ/FZJ-CMIP-ozone-1-0/atmos/monC/vmro3/gn/v20250904/vmro3_input4MIPs_ozone_CMIP_FZJ-CMIP-ozone-1-0_gn_185001-185012-clim.nc

Independent variable definitions:

        double lon(lon) ;
                lon:units = "degrees_east" ;
                lon:bounds = "lon_bnds" ;
                lon:long_name = "longitude" ;
        double lat(lat) ;
                lat:units = "degrees_north" ;
                lat:bounds = "lat_bnds" ;
                lat:long_name = "latitude" ;
        double plev(plev) ;
                plev:units = "hPa" ;
                plev:bounds = "plev_bnds" ;
                plev:long_name = "pressure" ;
        double time(time) ;
                time:units = "days since 1850-01-01" ;
                time:long_name = "time" ;
                time:calendar = "noleap" ;
                time:climatology = "climatology_bounds" ;
       double climatology_bounds(time, nv) ;

Forcing variable definition:

        float vmro3(time, plev, lat, lon) ;
                vmro3:units = "mole mole-1" ;
                vmro3:long_name = "vmro3" ;
                vmro3:cell_methods = "time: mean" ;

Ancillary file relative to ESM1.6 config directory:

As per ACCESS-ESM1.6 ancillary directory structure:

modern/pre-industrial/atmosphere/forcing/global.N96/${VERSIONING}/ozone_1850_ESM1.anc

where VERSIONING is a date in YYYY.MM.DD format.

Ancillary fields:

Processing:

Processing examples:

ACCESS-ESM1.5 CMIP6 ancillaries

UKESM CMIP7 ancillaries

UKESM CMIP6 ancillaries

Where used:

[MartinDix]

input4MIPS supplies two versions of the pre-industrial ozone, a single year mean similar to CMIP6 and also a 20 year file designed to allow simulation of the stratospheric quasi-biennial oscillation (QBO) in the PI control.

The analysis of the QBO simulated in CMIP by Richter et al https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2019JD032362 doesn't include CM2 or ESM1.5 so I've created equivalent plots

The stratospheric resolution of ESM1.6 is too low to simulate anything like a QBO so we might as well just use the single year file. Not surprisingly ACCESS-CM2 looks very like HadGEM3-GC31-LL.

[MartinDix]

Comparing 1850 from /g/data/qv56/replicas/input4MIPs/CMIP6/CMIP/UReading/UReading-CCMI-1-0/atmos/mon/vmro3/gn/v20160711/vmro3_input4MIPs_ozone_CMIP_UReading-CCMI-1-0_gn_185001-189912.nc and
/g/data/qv56/replicas/input4MIPs/CMIP7/CMIP/FZJ/FZJ-CMIP-ozone-1-0/atmos/monC/vmro3/gn/v20250904/vmro3_input4MIPs_ozone_CMIP_FZJ-CMIP-ozone-1-0_gn_185001-185012-clim.nc

[penguian]

See https://github.com/ACCESS-Community-Hub/CMIP6-Input-Scripts/blob/main/scripts/mrd599/src/python/ozone_ancil.py for the production of the ancillary file.

[penguian]

Should the units be for mass mixing ratio or volume mixing ratio?

[penguian]

See gadi:/g/data/p66/txz599/data/ancil/CMIP6/ozone_2014_2101_ssp585_ESM1.anc for a possible input file for https://github.com/ACCESS-Community-Hub/CMIP6-Input-Scripts/tree/main/scripts/txz599/ACCESS-ESM_tools/scripts/climate_stabilisation_ozone

The input file is in FieldsFile format, and when loaded into Iris, the resulting cube contains

Mass Fraction Of Ozone In Air (1)	time	model_level_number	latitude	longitude
Shape	1056	38	145	1

...

Attributes
STASH	m01s00i060
source	'Data from Met Office Unified Model'
um_version	'7.3'

Looking up STASH code m01s00i060 at https://reference.metoffice.gov.uk/um/stash/_m01s00i060 yields

Name	OZONE **

[penguian]

From examining the gadi:/g/data/p66/txz599/data/ancil/CMIP6/ozone_2014_2101_ssp585_ESM1.anc file, the conclusion is that the ozone concentration to be written to the ancillary file is

• A zonal average.
• In units of mass mixing ratio, derived by dividing the standard ozone molar mass of 48.0 by the standard dry air molar mass of 28.97, giving approximately 1.6569. Code at https://code.metoffice.gov.uk/trac/ancil/browser/contrib/branches/dev/jonnywilliams/r4049_CMIP6_ozone_forcing_new/OzoneConc/interpolate_height_hist.pro uses 1.65 as the ratio.

Note that the actual ozone molar mass in the stratosphere is more than 48, due to heavier isotopes of oxygen.
See https://agupubs.onlinelibrary.wiley.com/doi/epdf/10.1029/2005JD006342 and https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2006JD007855

[penguian]

See https://code.metoffice.gov.uk/trac/ancil/ticket/2641 for the ANCIL ticket that describes CMIP7 ozone forcing and ancillary processing for UKESM 1.3 (the version of UKESM that the UK Met Office is using for the CMIP7 Fast Track).

[penguian]

See PCMDI/input4MIPs_CVs#378 where Steven Hardiman and Jane Mulcahy are asking for temperature datasets to match the published CMIP7 ozone forcing datasets. We can use the spinup files for piControl, but not for the historical forcings, because we would need a month-for-month match on atmospheric pressure data.

[hegglinm]

@penguian PI and historical zmta files can now be found at https://fz-juelich.sciebo.de/s/6GXCfmMcQsysxng

The data will be uploaded to ESGF soon as well. Please let me know if you find anything strange and keep an eye on updates on ESGF.