Document mapping from GRIB templates to ISO 19111

[desruisseaux]

Details

Context

The Open Geospatial Consortium (OGC) and the International Organization for Standardization (ISO) normalize in ISO 19111 a conceptual model for the definition of Coordinate Reference Systems (CRS), including map projections. The Geographic Markup Language (GML), the Well-Known Text (WKT) format and the widely-used (in geospatial community) EPSG geodetic dataset are realizations of the ISO 19111 standard.

ISO 19111 provides data structures for defining custom CRS, including map projections with custom parameter values, but with one limitation: the Operation Method (in ISO 19111 terminology) must be pre-defined. In GRIB files, the equivalent of Operation Methods is templates.

A primary source of Operation Method definitions is the EPSG guidance notes, which provides all the definitions that we need for allowing the implementation of Mercator projection, Lambert Conformal Conic, etc. However some projections in use by WMO are not covered in EPSG guidance notes. An example is Pole rotation. When an Operation Method is not defined by EPSG, we need another authority for making possible to use those map projections in GML or WKT formats. For projections of particular interest to WMO, the natural authority for Operation Method definitions would be WMO.

What is needed

For making possible to use Pole Rotation and other WMO-specific projections in environments structured along ISO 19111 abstract model such as GML and WKT, we need a document that standardize the mapping of GRIB template to properties of the ISO 19111 OperationMethod class. Example of things to document are:

• Standard OperationMethod names for GRIB templates.
• Standard OperationParameter names for parameters identified by "Octets No." in GRIB.
• Formulas showing exactly how each parameter is used.

Starting point

A proposal has been drafted in OGC MetOcean working group:

• Authority codes for pole rotation.
• Problem statement (why we need that).

I understand that WMO as other priorities for now. This issue is filled as reminder that a mapping between GRIB templates and ISO 19111 Operation Methods is desired, as WMO is the only authoritative source I'm aware of for some map projections that are more relevant to the meteorological community than other communities.

Requestor

No response

[amilan17]

@desruisseaux Thank you for submitting this issue. I will put this on the agenda for the next meeting in April.

[chris-little]

@amilan17 The OGC Meteorology and Oceanography Domain WG think this is important for current and future developments of WIS2, and its development. At present, WMO machinable registers are hosted by the UK Met Office, and have been for several years. Even if WMO does not have the IT infrastructure to host such a registry, copies of register items could also be hosted in the OGC Naming Authority registry with just a documentation link to WMO as the authoritative organisation.

Action by WMO would be much appreciated. FYI @desruisseaux

[amilan17]

https://github.com/wmo-im/CCT/wiki/Teleconference.6.7.June.2023 notes:

@sebvi will take a look at this summer; @marianmajan-ibl will ask colleagues to check this too;

[amilan17]

https://github.com/wmo-im/CCT/wiki/20.to.22.September.2023 notes:

team looked at issue and we will invite you to join us during an online meeting to introduce the request and so the team can get clarifications. 

@desruisseaux

[chris-little]

Things are moving on: Another use case for the requirement of @desruisseaux. I have just noted that some scientists, modelling the atmospheres of exo-planets, are using CRSs with rotated poles and scaled grids. The planets are usually tidally locked to their star, so having a pole at the sub-solar point is convenient.

[desruisseaux]

Thanks! On my side, it would be easier at any time after October 10th.

[chris-little]

Same for me - at an OGC Technical Committee conference all next week.

[amilan17]

@chris-little and @desruisseaux will you be available to join our TT-TDCF meeting on 17 or 18 October  between 12 - 14 UTC?

[chris-little]

@amilan17 Those time are good for me.

[desruisseaux]

It works for me too.

[desruisseaux]

I do not know if the scope of the meeting would include the definitions of Coordinate Reference Systems (CRS), or if the scope would be limited to the definition of Operation Methods as stated in this GitHub issue. If the scope includes CRS, then WMO could also be an authoritative source of definitions of isobaric surfaces. This need has been expressed in opengeospatial/NamingAuthority#228.

[amilan17]

https://github.com/wmo-im/CCT/wiki/Teleconference.17.18.October.2023 notes:
Chris Little and Martin Desruisseaux introduced the request; the team discussed options;

• Chris and Martin will create a new proposal for improvements of the notes as needed
• the team can also publish the list of GRIB templates in the codes registry

[chris-little]

@amilan17 And we will ask the OGC NAming Authority (who own the OGC registers) to script some tailoring of the refernece so that the OGC standard URN references can map onto the WMO URL links.