Information in the Search Interface is incomplete compared to the xls docs

[mpmoine]

Dear ES-DOC support,

When comparing contents between what is published on ES-DOC search interface and what is in the xls documents we pushed on Github, it turns out that some information is missing.

For example,

on https://explore.es-doc.org/cmip6/models/cnrm-cerfacs/cnrm-esm2-1 :

CNRM-CERFACS > CNRM-ESM2-1 :: Atmosphere > Grid > Discretisation > Vertical > Coordinate Type
--
Description | Type of vertical coordinate system
Values | -- Awaiting modelling group input --

Despite the information is in the pushed xsl on:

https://github.com/ES-DOC-INSTITUTIONAL/cnrm-cerfacs/blob/master/cmip6/models/cnrm-esm2-1/cmip6_cnrm-cerfacs_cnrm-esm2-1_atmos.xlsx

hybrid sigma-pressure: sigma system near terrain and isobaric above)

Thanks for considering this issue.

Marie-Pierre.

[asladeofgreen]

Thanks @mpmoine. Can confirm that this property appears to have been skipped by the spreadsheet processor. Am analysing the cause and will let you know when fix will be published.

[asladeofgreen]

@mpmoine It seems that the last specialisation in a realm is not being processed, i.e. the iterator terminates early. Should be fairly simple fix.

[mpmoine]

@momipsl

Thanks Mark for solving this issue. Can you confirm that you will re-process all the realms we published (the above issue in atmos realm was just an example, we may have the same troubles for other realms)

[asladeofgreen]

@mpmoine I have just reprocessed and republished all the CMIP6 model documents. I can confirm that all realms have been re-processed.

https://explore.es-doc.org/cmip6/models/cnrm-cerfacs/cnrm-esm2-1

[mpmoine]

@momipsl

Thanks for this fix.

Now, for the example I pointed above, we see the correct value in the Search Interface:

CNRM-CERFACS > CNRM-ESM2-1 :: Atmosphere > Grid > Discretisation > Vertical > Coordinate Type
--
Description | Type of vertical coordinate system
Values | Hybrid sigma-pressure: sigma system near terrain and isobaric above 

Unfortunatelly, there is still a lot of occurences of Awaiting modelling group input. It is difficult to make an exhaustive list.

To take examples, it is still the case in land > vegetation or in atmos > dynamical core

[asladeofgreen]

@mpmoine We republished last week and I have just taken a look at the CNRM-CERFACS documents. In particular I looked at the atmosphere > dynamical core and it is clear that some fields are not being extracted from the spreadsheet. I will investigate this issue during the course of the week.

[asladeofgreen]

So I have just been analysing the source of the problem. When looking at the atmos spreadsheet for the of the CNRM-CM6-1 model I observe that the list of dynamical core >> prognostic variables has not been entered correctly.

The prognostic variables is derived from an extensible controlled vocabulary. The grey data input box is a drop-down, indicating that the user can select from a list of entries (as defined in the vocabulary). As the vocabulary is extensible then the user also has the option of specifying 'other' and entering a value in the column to the right.

In the spreadsheet it appears that someone has copied and pasted incorrectly so that the controlled vocabulary list is no longer visible. This has caused the spreadsheet processor to error and abort processing of the dynamical core.

The solution is partly for the spreadsheet to be corrected and possibly for the tooling to take into consideration this scenario.