Splitting ACCESS-OM2 into multiple model names?

[adele-morrison]

From the COSIMA meeting today:

Currently all resolutions of ACCESS-OM2 show up in the Catalog as the ACCESS-OM2 model. Usually we only use this name for the 1deg config. Do people think it's worth going through all the metadata.yaml files for ACCESS-OM2-01 and ACCESS-OM2-025 experiments and renaming the model from just ACCESS-OM2? It would make it easier for people to search and find output that is relevant for them if they can filter by model name better.

[adele-morrison]

I don't know why there's a bug label on this sorry, I can't edit the labels.

[marc-white]

Sticking my initial thoughts in-meeting here:

• Would adding extra 'things' to the Catalog name potentially confuse people, or is this pretty standard nomenclature?
• It seems like resolution would be a convenient thing to be able to search on at the access-nri-intake-catalog level (i.e., when you're searching for experiments). Are there any other similar things you'd want to search on at the experiment level that you can't search on currently?

[adele-morrison]

Currently there is an ACCESS-OM2-025 model name that only contains the CMIP 1/4deg runs. But the non-CMIP 1/4 deg runs are labelled ACCESS-OM2. Confusingly the 1deg CMIP runs are labelled ACCESS-OM2 along with all the 1/10th runs etc.

[dougiesquire]

Just noting that the model name is whatever is specified in the model field in the metadata.yaml file. So changing these is as simple as updating the metadata.yaml files.

[adele-morrison]

Yep, got that @dougiesquire. I was checking that we aren't missing some reason why this wasn't done in the first place? If not, I suggest we go ahead and rename all higher res ACCESS-OM2 model names in the metadata.yamls.

[adele-morrison]

Ok, I renamed the model name to ACCESS-OM2-01 in the metdata.yaml of all the ACCESS-OM2-01 experiments I have write access to.

@aekiss, @hrsdawson @mauricehuguenin @pedrocol @willaguiar can you do the same please? Anyone else we need to tag?

[adele-morrison]

I guess once that's done we'll need to rerun all the Intake stores for those experiments.

[dougiesquire]

I guess once that's done we'll need to rerun all the Intake stores for those experiments.

Yes, a new version of the catalog will need to be built

[anton-seaice]

Is this the field:

model: access-om2-01

https://github.com/ACCESS-NRI/access-om2-configs/blob/83885a45017b3a34ea4fad15d155449c7217ad70/metadata.yaml#L28

It looks like its already been changed in the release config ?

[adele-morrison]

Is this the field: model: access-om2-01

Yes that's it. Is it case sensitive? All of the existing ones are in capitals model: ACCESS-OM2-01

[anton-seaice]

Oh - we appear to have not done that consistently:

model: ACCESS-OM2 is in some of the 025deg configs

https://github.com/ACCESS-NRI/access-om2-configs/blob/da5f7965b41b516c2ca0eaa4b235d81d03e19f89/metadata.yaml#L33

Hopeing @marc-white can confirm about the case-sensitivity ?

[aekiss]

• It seems like resolution would be a convenient thing to be able to search on at the access-nri-intake-catalog level (i.e., when you're searching for experiments). Are there any other similar things you'd want to search on at the experiment level that you can't search on currently?

It could be handy to be able search on the items in the keywords field in metadata.yaml such as forcing dataset (JRA55do, ERA5) and repeat-year/interannual.

[marc-white]

Oh - we appear to have not done that consistently:

model: ACCESS-OM2 is in some of the 025deg configs

https://github.com/ACCESS-NRI/access-om2-configs/blob/da5f7965b41b516c2ca0eaa4b235d81d03e19f89/metadata.yaml#L33

Hopeing @marc-white can confirm about the case-sensitivity ?

Oh - we appear to have not done that consistently:

model: ACCESS-OM2 is in some of the 025deg configs

https://github.com/ACCESS-NRI/access-om2-configs/blob/da5f7965b41b516c2ca0eaa4b235d81d03e19f89/metadata.yaml#L33

Hopeing @marc-white can confirm about the case-sensitivity ?

I've not yet been able to find anywhere where access-nri-intake-catalog or intake-esm where it does case modification on the way in, so I think it's likely to be case-sensitive (although it doesn't help that every example given has fully capitalized model names anyway).

[aekiss]

I've fixed these ones

/g/data/cj50/access-om2/raw-output/access-om2-01/01deg_jra55v140_iaf*/metadata.yaml
/g/data/ik11/outputs/access-om2-01/01deg_jra55v140_iaf_cycle4_jra55v150_extension/metadata.yaml

[aekiss]

We'll also need to fix metadata.yaml in any actively running control dirs, and fix the capitalisation in https://github.com/ACCESS-NRI/access-om2-configs