Why does EPSG 3006 not validate?

[AndersEkmanLm]

Currently, SWEREF 99 TM (EPSG 3006) does not pass validation. This poses a bit of a problem, since SWEREF 99 is the official reference frame for Sweden, and it would be practical to be able to use it in Swedish inspire services. As I understand it, other countries have similar problems with their official systems.

I am trying to understand why the validator does not allow EPSG 3006, and would appreciate some help.

EPSG 3006 fulfills all the requirements listed in 6.1.1.1 and 6.1.1.2 of the TG DS Template. However, it is not included in table 2 (p. 24) of the same document. Judging from this issues, this seems to be the trouble.

When I read the document, I don’t interpret table 2 to be meant as a complete list of all allowed systems (it only contains a small subsets of all the systems that would be allowed under 6.1.1.1 and 6.1.1.2). Is that a misreading on my part? Or is perhaps this just a technical limitation in the validator – it only allows systems in table 2, even though the technical guidelines would allow more?

Or in other words, what would it take to allow usage of EPSG 3006? An update to the validator, or an update of the technical guidelines, or perhaps something else?

[carlospzurita]

Dear @AndersEkmanLm

The validator uses a static list of allowed reference systems to execute validations for datasets, extracted from the table you mention in the data specification templates. Other reference systems can be allowed, but only for regions outside of continental Europe.

We are going to discuss this issue internally and come back to you with more feedback on how to proceed. In the meatime, can you please share with us a data example or link? It can be useful for us to reproduce the validation.

Kind regards.

[MarcoMinghini]

Dear @sMorrone and @heidivanparys, what is your opinion about this issue? Currently, the Validator returns an error because it only allows the CRSs listed in the table mentioned above, so this is a more general discussion not strictly related to the Validator. Thank you.

[heidivanparys]

This issue is related to https://github.com/INSPIRE-MIF/helpdesk-registry/issues/80

In short: we are waiting for the Commission to complete the procedure for an amendment to COMMISSION REGULATION 1089/2010 to allow for a CRS registry to be established.

For more background information, see https://github.com/INSPIRE-MIF/helpdesk-registry/issues/80.

So far, there is not much we can do, I think. I don't see this item on the agenda for the 13th MIG meeting though, perhaps it could be requested to add a status on this to the agenda?

[MarcoMinghini]

Thank @heidivanparys, I will ask to add an agenda topic on this, although I am not sure there are relevant news.

So I understand that, from the pure validation perspective, the test (based on the current version of the IR) is correct, i.e. the analysis from @AndersEkmanLm is not correct.

[heidivanparys]

So I understand that, from the pure validation perspective, the test (based on the current version of the IR) is correct, i.e. the analysis from @AndersEkmanLm is not correct.

From a pure validation perspective, the test is indeed correct as I see it, it tests TG Requirement 2 from the TG DS Template:

The identifiers listed in Table 2 shall be used for referring to the coordinate reference systems used in a data set.

Or in other words, what would it take to allow usage of EPSG 3006? An update to the validator, or an update of the technical guidelines, or perhaps something else?

It would take:

1. Amendment to IR 1089/2010 to allow for a CRS registry to be established
2. Creation of a CRS registry, including governance structure with a control body.
3. Update of the validator to validate against the CRS registry instead of against the current static list.
4. Approval of the control body to add EPSG:3006 to registry (i.e. confirmation of the control body that EPSG:3006 fulfills the requirements in the implementing rules).

[Ana-SGA]

Dear @heidivanparys, @MarcoMinghini,

we have a similar problem like @AndersEkmanLm - data validation fails for the official Croatian reference system EPSG 3765 http://www.opengis.net/def/crs/EPSG/0/3765.

But also we have a problem with the language, since the Croatian language (hrv) isn't included in the enumeration that the Validator uses, example for WMS service attached.

How can we solve this?

[heidivanparys]

@Ana-SGA The list of languages used in the validator, at least for the test referred to, is present in https://inspire.ec.europa.eu/schemas/common/1.0/common.xsd, line 733:

    <xs:simpleType name="euLanguageISO6392B">
        <xs:restriction base="languageISO6392B">
            <xs:enumeration value="bul"/>
            <xs:enumeration value="cze"/>
            <xs:enumeration value="dan"/>
            <xs:enumeration value="dut"/>
            <xs:enumeration value="eng"/>
            <xs:enumeration value="est"/>
            <xs:enumeration value="fin"/>
            <xs:enumeration value="fre"/>
            <xs:enumeration value="ger"/>
            <xs:enumeration value="gre"/>
            <xs:enumeration value="hun"/>
            <xs:enumeration value="gle"/>
            <xs:enumeration value="ita"/>
            <xs:enumeration value="lav"/>
            <xs:enumeration value="lit"/>
            <xs:enumeration value="mlt"/>
            <xs:enumeration value="pol"/>
            <xs:enumeration value="por"/>
            <xs:enumeration value="rum"/>
            <xs:enumeration value="slo"/>
            <xs:enumeration value="slv"/>
            <xs:enumeration value="spa"/>
            <xs:enumeration value="swe"/>
        </xs:restriction>
    </xs:simpleType>

That file was last updated in 2011, and Croatian became an official EU language in 2013, according to https://europa.eu/european-union/about-eu/eu-languages_en. So https://inspire.ec.europa.eu/schemas/common/1.0/common.xsd would have to be updated. The correct way to do this is to create an issue in https://github.com/INSPIRE-MIF/application-schemas/. Please let me know if you need help with that.

[Ana-SGA]

Dear @heidivanparys,

thanks for the suggestion regarding the Croatian language (https://github.com/INSPIRE-MIF/application-schemas/issues/7).

Can anyone tell for the coordinate reference system - will there be an amendment to the Regulation 1089/2010 to allow for the establishment of the CRS Registry (hopefully including EPSG:3765 also)? Or "just" INSPIRE Validator changes are required?

Who can propose addition to the agenda for MIG? Is it possible to change it this late?

[AndersEkmanLm]

Thank you for your replies.

I have a few follow up questions. I understand that your interpretation of the technical guidelines is that until an official Inspire registry for coordinate reference systems is set up, only systems listed in table two can be used.

This may very well be true – I am no expert in this area – but I do not understand how that follows from the text of the technical guidelines, and I would appreciate some help with understanding that.

The heading for section 6.1.1.4 where table two is found is “Identifiers”. In my understanding, it regulates what types of identifiers must be used, and not what CRSs. If this section should be interpreted to mean that only systems in table two may be used, where does it specifically say so? How were the requirements under sections 6.1.1.1 and 6.1.1.2 boiled down to the short list in table 2?

Under 6.1.1.4, it also says:

1. Coordinate reference system parameters and identifiers shall be managed in one or several common registers for coordinate reference systems.
2. Only identifiers contained in a common register shall be used for referring to the coordinate reference systems listed in this Section.

I do not take “common registers” to mean some official Inspire registry. On the contrary, I would interpret e.g. the EPSG registry to be such a registry. It contains all the information needed to verify that SWEREF 99 - and the map projections belonging to it, e.g. SWEREF 99 TM (EPSG::3006) - fulfills all the requirements in the technical guidelines.

In general, where in the technical guidelines or elsewhere can I find the requirements that there must be an official Inspire register?

Under A.2.1 and A.2.2 there are explicit tests for checking that systems conform to sections 6.1.1.1 and 6.1.1.2. Why would that be needed if all you have to do is check against a whitelist?

While it may not be listed in the right places, it is a matter of fact that SWEREF 99 fulfills the requirements under section 6.1.1.1 and 6.1.1.2. It is unfortunate that Swedish governmental authorities cannot use the official Swedish reference systems until a control body confirms what is already well known.

[heidivanparys]

It seems that the INSPIRE requirements regarding the coordinate reference systems are based on previous work done by EuroGeographics and the Joint Research Centre, documented in the document "Map Projections for Europe".

The INSPIRE Data Specification on Coordinate Reference Systems explicitly refers to "Map Projections for Europe":

The requirements and recommendations related to Map projections are based on the results from the “Map Projections for Europe” workshop. These are:

• Lambert Azimuthal Equal Area (ETRS89-LAEA) for pan-European spatial analysis and reporting, where true area representation is required;
• Lambert Conformal Conic (ETRS89-LCC) for conformal pan-European mapping at scales smaller than or equal to 1:500,000;
• Transverse Mercator (ETRS89-TMzn) for conformal pan-European mapping at scales larger than 1:500,000.

and

Between 14-15 December 2000 the “Map Projections for Europe” workshop was organised to propose the map projections to be used for representation of data in plane coordinates in general applications. The use of the following projections was recommended:

• Lambert Azimuthal Equal Area (ETRS89-LAEA) for spatial analysis and display;
• Lambert Conformal Conic (ETRS89-LCC) for conformal pan-European mapping at scales smaller or equal to 1:500,000;
• Transverse Mercator (ETRS89-TMzn) for conformal pan-European mapping at scales larger than 1:500,000.

These recommendations have been used by the European Commission (EC) for geo-referencing the data internally within the EC. For the representation of data in plane coordinates in general pan-European applications in continental Europe in the frame of INSPIRE, these projections are either mandated or recommended. For regions outside of continental Europe, for example for overseas MS territories, the MS shall define a map projection they consider most suitable for the application. The ETRS89-LAEA projection in INSPIRE is recommended for spatial analysis and reporting.

The following discussions also refer to "Map Projections for Europe":

• https://wayback.archive-it.org/12090/20210120074117/https://inspire.ec.europa.eu/forum/discussion/view/198388/axis-order-in-crs-used-for-inspire-data
• https://wayback.archive-it.org/12090/20161216062535/http://inspire-forum.jrc.ec.europa.eu/pg/forum/topic/12663/crs-specifications-of-inspire-order-of-coordinate-axis/#annotation-2539

So my assumption is that when the INSPIRE requirements regarding the coordinate reference systems were written, stating

1.3.2. Two-dimensional Coordinate Reference Systems

— Two-dimensional geodetic coordinates (latitude and longitude) based on a datum specified in 1.2 and using the parameters of the GRS80 ellipsoid. — Plane coordinates using the ETRS89 Lambert Azimuthal Equal Area coordinate reference system. — Plane coordinates using the ETRS89 Lambert Conformal Conic coordinate reference system. — Plane coordinates using the ETRS89 Transverse Mercator coordinate reference system.

the authors really meant the coordinate reference systems as described in "Map Projections for Europe", page 110 and onwards. That would also explain the axis order (see also https://github.com/INSPIRE-MIF/helpdesk-registry/issues/80): the descriptions in that document use axis order N, E.

Note that the information source for the CRSs with the codes as listed in the TGs in the EPSG registry is stated as "European Commission Joint Research Centre "Map Projections for Europe". http://www.ec-gis.org", see e.g. https://epsg.org/crs_3044/ETRS89-UTM-zone-32N-N-E.html.

The summary of "Map Projections for Europe" also makes it clear that

the goal was to define suitable map projections for spatial data for use by the Commission in its activities

and it makes sense that the Commission then uses those recommendations in its initiatives.

Note that the implementing rules for the interoperability of spatial data sets and services will be amended, see also this presentation from the last, 13th, MIG-T meeting. The implementing rules will be in public consultation soon:

[heidivanparys]

@carlospzurita I think this issue should be moved to the general INSPIRE helpdesk.

[AndersEkmanLm]

Thank you @heidivanparys for the clarifying information. It was very helpful in understanding the background of the requirements. As I understand your and others interpretation, the list in table 2 is to be taken as the full allowed set, and it is based on the ETRS-TMzn recommended in "Map Projections for Europe". This I would not have understood on my own, so thanks for all the pointers and explanations. We will continue to discuss how to handle this internaly at Lantmäteriet.

[heidivanparys]

It would take:

1. Amendment to IR 1089/2010 to allow for a CRS registry to be established

Please note that the proposal for amendment is open for feedback, see https://ec.europa.eu/info/law/better-regulation/have-your-say/initiatives/13182-Revision-of-INSPIRE-regulations-interoperability-of-spatial-data-sets-and-services_en

[fabiovinci]

The INSPIRE MIG-T approved the change proposal during the 76th MIG-T Meeting. In the next step, the proposal will be assessed by the MIG through a 2-week scrutiny period starting on 6/12/2023.

[arantzaetxebarria]

We would like to inform you that the proposal has been successfully published. https://inspire.ec.europa.eu/crs

Regards