search

WCRP-CMIP / CMIP6_CVs

Controlled Vocabularies (CVs) for use in CMIP6
Creative Commons Attribution 4.0 International
155 stars 78 forks source link

Should the ocean grid of HadGEM3 and UKESM1 models be eORCA1 instead of ORCA ? #494

Closed jypeter closed 6 years ago

jypeter commented 6 years ago

@matthew-mizielinski @durack1 I have just talked to my local ORCA specialist (@oliviermarti) and read #457, because I was wondering why UK Academic Community PMIP4 models stated Ocean grid: 360×292 (ORCA1) x L75 and why the CV for the same models say ORCA1 [...] 360 x 330 longitude/latitude;

Olivier told me that if you mention a 360 x 330 grid, you should then use the eORCA1 grid name instead of ORCA1 . So maybe this should be corrected for the appropriate UK models

And the additional question is: will the UKESM1-0-LL and HadGEM3-GC31-LL used for PMIP4 use 360x292 or 360x330 ?

durack1 commented 6 years ago

@alistairsellar @jonseddon @markelkington can you folks chime in on this query? Once I have guidance we can update all relevant registered entries

matthew-mizielinski commented 6 years ago

@jypeter, you are probably correct on this issue it hadn't occurred to me that the extended ORCA grid should be referred to differently. I suspect you are correct in that we should replace ORCA with eORCA in all of the HadGEM3 and UKESM1 descriptions. I'll confirm this with our ocean modelling teams and get back to you asap.

matthew-mizielinski commented 6 years ago

And the additional question is: will the UKESM1-0-LL and HadGEM3-GC31-LL used for PMIP4 use 360x292 or 360x330 ?

As far as I am aware all of our model configurations use the extended ORCA grid, and as such I would expect the PMIP runs to use the same grid. However, the UK PMIP leads have ultimate control of this.

matthew-mizielinski commented 6 years ago

I've had confirmation from our ocean model reps and yes we should state that we are using the eORCA grids.

@durack1, I'll get you some code to tweak the model descriptions tomorrow.

Thanks for raising this @jypeter.

matthew-mizielinski commented 6 years ago

@durack1, The following code should make the required changes.

models = [
    'HadGEM3-GC31-HH', 'HadGEM3-GC31-HM', 'HadGEM3-GC31-LL', 
    'HadGEM3-GC31-LM', 'HadGEM3-GC31-MH', 'HadGEM3-GC31-MM', 
    'UKESM1-0-LL', 'UKESM1-0-MMh']

for model_source_id in models:
     source_id[model_source_id]['model_component']['ocean']['description'] = \
         source_id[model_source_id]['model_component']['ocean']['description'].replace('ORCA', 'eORCA')
     source_id[model_source_id]['model_component']['seaIce']['description'] = \
         source_id[model_source_id]['model_component']['seaIce']['description'].replace('ORCA', 'eORCA')
durack1 commented 6 years ago

@matthew-mizielinski thanks for the code snippets, makes my job a lot easier! Take a peek at CMIP6_source_id.html and let me know if any further tweaks are required.

markelkington commented 6 years ago

Hi Matthew

What is the significance of the eORCA – I assume this isn’t just a resolution issue as we are defining ORCA025 and ORCA1 for ES-DOC. I’ll need to get these realigned so that the file headers are correct. Do you have a link to some additional information on this, or somebody in the NEMO team that can explain it to me.

Mark

From: Matthew Mizielinski [mailto:notifications@github.com] Sent: 05 April 2018 10:49 To: WCRP-CMIP/CMIP6_CVs Cc: Elkington, Mark; Mention Subject: Re: [WCRP-CMIP/CMIP6_CVs] Should the ocean grid of HadGEM3 and UKESM1 models be eORCA1 instead of ORCA ? (#494)

@jypeterhttps://github.com/jypeter, you are probably correct on this issue it hadn't occurred to me that the extended ORCA grid should be referred to differently. I suspect you are correct in that we should replace ORCA with eORCA in all of the HadGEM3 and UKESM1 descriptions. I'll confirm this with our ocean modelling teams and get back to you asap.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHubhttps://github.com/WCRP-CMIP/CMIP6_CVs/issues/494#issuecomment-378880906, or mute the threadhttps://github.com/notifications/unsubscribe-auth/AFTIJySF9jZziA-Jl1Ti9aXgj9dqgTI6ks5tleiBgaJpZM4TFWDd.

oliviermarti commented 6 years ago

ORCA grids have a southern limit at about 78°S. eORCA grids are identical to ORCA grids, except that they add points under the Antartic ice shelves up to about 86°S.

I don't know about any official documentation about that. Could NEMO system team provide a reference ?

Olivier