Support for JRA55-do v1.5.0 and v1.5.0.1 to get forcing up to ~present day

[aekiss]

We are currently using JRA55-do v1.4.0 forcing, which covers 01Jan1958 - 31Dec2018, but v1.5.0 and 1.5.0.1 are now available https://climate.mri-jma.go.jp/pub/ocean/JRA55-do/, and it could be useful to support them.

JRA55-do v1.5.0 covers 01Jan1958 - 15Jul2020 and has these improvements relative to v1.4 (see sec C.11 here):

• fixes the sign error in some cyclones in previous JRA55 versions JRA55 cyclones with incorrect sign  #186. These corrections affect 1967-1987 in the North Atlantic and 1959 in the Northeast Pacific.
• exclusively uses surface meteorological variables (specific humidity, air temperature, and wind vector) based on delayed-mode analysis of COBE-SST; in contrast, v1.4.0 used (presumably less accurate) real-time analysis of COBE-SST for parts of December in 2016, 2017, 2018 and 2019.

JRA55-do v1.5.0.1 is updated daily, running from 01Jan2020 to 4 days behind the present day (although data for the last 30 days are preliminary, and replaced day-by-day by final, higher-quality data based on delayed-mode analysis of COBE-SST). This is currently a preliminary release, but it could be good to start working with it as it will be useful to be able to simulate near-present-day conditions.

[aekiss]

JRA55-do v1.5.0 is already available on NCI at /g/data/qv56/replicas/input4MIPs/CMIP6/OMIP/MRI/MRI-JRA55-do-1-5-0 and has the same directory structure as v1.4.0, so it looks very straightforward to switch (much easier than the switch from 1.3.1 to 1.4.0).

As far as I know v1.5.0.1 is not yet available on NCI.

[aekiss]

I've made a request for NCI to download v1.5.0.1 and automatically keep it up to date.

[aekiss]

JRA55-do v1.5.0.1 is not from the same source as v1.5.0, so NCI can't put it within the ESGF replication under qv56.

So instead I've downloaded it to
/g/data/ik11/inputs/JRA-55/JRA55-do-1-5-0-1
and included a script that can update it whenever we like: https://github.com/COSIMA/JRA55-do-1-5-0-1

[aekiss]

I've created a jra55v150 branch in each of the 3 IAF configs
https://github.com/COSIMA/1deg_jra55_iaf/tree/jra55v150
https://github.com/COSIMA/025deg_jra55_iaf/tree/jra55v150
https://github.com/COSIMA/01deg_jra55_iaf/tree/jra55v150

and done a quick test run at 1deg only.

[aekiss]

I've also created a jra55v1501 branch in each of the 3 IAF configs
https://github.com/COSIMA/1deg_jra55_iaf/tree/jra55v1501
https://github.com/COSIMA/025deg_jra55_iaf/tree/jra55v1501
https://github.com/COSIMA/01deg_jra55_iaf/tree/jra55v1501

I did a quick test run at 1deg only. This apparently works fine for 2020-01-01 to 2021-12-31 but doesn't run for 2022 because it can't handle part-year forcing files due to 31 Dec being hard-coded in forcing.json, which won't match files like /g/data/ik11/inputs/JRA-55/JRA55-do-1-5-0-1/mirror/atmos/3hr/rsds/gr/latest/rsds_input4MIPs_atmosphericState_OMIP_MRI-JRA55-do-1-5-0-1_gr_202201010130-202203312230.nc

To resolve this we'll probably need to use the new libaccessom2 {{year}}{{month}}{{start_day}}-{{year}}{{month}}{{end_day}} features added to support ERA5 (#242).
e.g. see https://github.com/COSIMA/1deg_era5_iaf/blob/master/atmosphere/forcing.json

However we may need to generalise this, because when we update these forcing files the ending date will typically not be the end of the month.

[aidanheerdegen]

How real-time do you want? As recent as possible?

[aidanheerdegen]

In which case I'd suggest a script to update the forcing.json

[aekiss]

Yes that could work.

[aidanheerdegen]

I have never used it but it should work to call it automatically with something like

userscripts:
    init: update_forcing_json.sh

This is called here
https://github.com/payu-org/payu/blob/master/payu/experiment.py#L111

before setup, so any modifications to config files will persist and be copied into work and properly accounted for by runlog (git)

Let me know if you need any assistance with this.

[aekiss]

01deg_jra55v140_iaf_cycle4_jra55v150_extension uses JRA55-do v1.5.0 (2019 only) and v1.5.0.1 (1 Jan 2020 onwards) to extend 01deg_jra55v140_iaf_cycle4 from 1 Jan 2019 to March 2022. This will be continued into the future, about 4 months behind real time. It includes the same diagnostics as the final year of cycle 4 (including BGC). It's available via the COSIMA Cookbook and from
/g/data/ik11/outputs/access-om2-01/01deg_jra55v140_iaf_cycle4_jra55v150_extension
and metadata is here.

[aekiss]

I've just heard that JRA55-do updates will be discontinued sometime in 2023, and an ERA-5 based model forcing dataset will be created to take its place.

From Alistair:

As you probably know, JRA55-do will no longer be updated starting sometime this year. Moving forward, as was just announced by Gustavo Marques at the DRAKKAR meeting, there is a plan to develop a new ocean forcing dataset based on ERA-5. NCAR will collaborate with GFDL and ECMWF to produce this product and NCAR is currently looking for some funds to start the effort. ERA-5 meets all the criteria for a new upstream product and ECMWF has just completed a back-extension to 1940. ECMWF plans to routinely update ERA-5 until after ERA-6 is made available, which is scheduled for 2026. It is not clear how long it will take to produce the adjustments and new ocean-forcing dataset but there is a sense of urgency to minimize the gap once JRA55-do is discontinued.

[aekiss]

JRA55-do updates will terminate at the end of January 2024: https://climate.mri-jma.go.jp/pub/ocean/JRA55-do/

[aekiss]

As discussed here, JRA55 has been superseded by the new JRA-3Q.

Does anyone know if there will be a JRA3Q-do based on that?

[access-hive-bot]

This issue has been mentioned on ACCESS Hive Community Forum. There might be relevant details there:

https://forum.access-hive.org.au/t/jra-3q-as-a-replacement-for-jra55/1759/2

[aekiss]

There won't be a JRA3Q-do

[aekiss]

Development of a new ERA5-based ocean forcing dataset is underway, and will be called CFORCE. For more details see Gokhan Danabasoglu's talk and slides from the 2024 CESM Ocean Model Working Group meeting.

[aekiss]

Heads up: /g/data/ik11/inputs/JRA-55/JRA55-do-1-5-0-1/mirror contained unfinalised data for 2023-12-10T15:00:00 onwards, so has been renamed mirror-DO-NOT-USE.

You should use /g/data/ik11/inputs/JRA-55/JRA55-do-1-5-0-1/mirror-2024-04-11 instead. This contains only finalised data and extends to the end of January 2024 (which is the end of the JRA55-do dataset).

[aidanheerdegen]

Will the NCI dataset by updated to the end of January 2024?

[aekiss]

Do you mean /g/data/qv56/replicas/input4MIPs/CMIP6/OMIP/MRI/MRI-JRA55-do-1-5-0?

This version (1.5.0) only extends to 15 July 2020. Only v1.5.0.1 goes to Jan 2024 and NCI can't include this with ESGS replication on qv56.

[aidanheerdegen]

Thanks. If this is deemed important enough then it could be an NCI supported dataset.

It is already on the list of requested datasets for ACCESS-NRI to support. Hooray!

[aekiss]

The cycle 4 2019-2023 extension is complete:
/g/data/ik11/outputs/access-om2-01/01deg_jra55v140_iaf_cycle4_jra55v150_extension

see https://cosima.org.au/index.php/2020/07/29/data-available-0-1-1958-2018-access-om2-iaf-run/ and https://forum.access-hive.org.au/t/access-om2-control-experiments/258#iaf-11

[access-hive-bot]

This issue has been mentioned on ACCESS Hive Community Forum. There might be relevant details there:

https://forum.access-hive.org.au/t/cosima-twg-meeting-minutes-2024/1734/17