European Commission logo
INSPIRE Community Forum

Correct Geometry for LandCoverUnit

Dear all,

what's the correct geometry for LandCoverUnit: Surface or Multisurface?

Within LandCover DS there is the following citation:

The attribute geometry of a LandCoverUnit is a GM_Object, which is the ISO 19107 supertype for all geometry objects. It is restricted to GM_Point or GM_Surfaces for LC needs.



Giuseppe P.

  • Lars Erik STORGAARD

    By Lars Erik STORGAARD

    Dear Giuseppe

    Perhaps this discussion in MIG-T can be helpful:


  • Stefania MORRONE

    By Stefania MORRONE

    Hi all,

    the   geometryIsKindOfGM_PointOrGM_Surface constraint  specified in the target application schema shall be fulfilled stating that “geometries shall be points or surfaces”. 

    In OCL:

    inv: self.geometry->forAll(l | l.oclIsKindOf(GM_Surface) or l.oclIsKindOf(GM_Point))

    In the frame of LIFE+IMAGINE EU project, the following encoding has been applied to geometry of LC Units

                    <gml:Surface ..">




  • Giuseppe Procino

    By Giuseppe Procino

    Hi Stefania,

    thanks you for your reply. I will use GM_Surface in according to constraint.



  • Lena Hallin-Pihlatie

    By Lena Hallin-Pihlatie

    Hi all,

    Thanks for raising this issue, i.e. whether to use GM_Surface or GM_Multisurface in the encoding for LandCoverUnits. I have no objections to your recommendation to use GM_Surface, but I'm continuing this discussion as I would like to understand the issue better and with time to verbally clarify this in the TG on LC.

    In the Implementation rules it says, regarding the geometry (GM_Object), that there is the following constraint: "Constraints of the spatial object type LandCoverUnit" meaning "Geometries shall be points or surfaces." To me it's very vaguely expressed in case this means you have to encode polygons as GM_Surface. And as Guiseppe pointed out, it is not clearly expressed in the TG on LC either.

    Could you help me to get the full picture on this?  If I recall correcly, for example Corine Land Cover data is maintained as (ESRI) multipart polygons, which to my understanding would easily be transformed and shared as GM_MultiSurface. Will it not have concequences on the id management if these CLC multipolygons have to be split into simple features (polygons) before the transformation to GM_Surface?

    I'd also like to ask the former members of the TWG on LC what your intentions were? Did you mean to restrict the GM_Objects only to GM_Surface and in that case what was you reasoning behind it?



  • Stefania MORRONE

    By Stefania MORRONE

    Hi all,

    I'm not a domain expert, but it seems to me that while IR just say that "Geometries shall be points or surfaces" (very vaguely expressed, I agree), the TG are clear when they specify  (section that  'The term 'surface' is used instead of 'polygon' for conformity with ISO 19107 standard. A GM_Polygon can not exist on its own and shall be part of a GM_Surface. The generic 2D geometry object for 2D is a surface (GM_Surface), according to ISO Standard. Conceptually, the difference is that a surface can be an aggregation of patches .... The attribute geometry of a LandCoverUnit is a GM_Object, which is the ISO 19107 supertype for all geometry objects. It is restricted to GM_Point or GM_Surface for LC needs ".




This discussion is closed.

This discussion is closed and is not accepting new comments.

Land Cover & Use

Land Cover & Use

Join this group to share your knowledge, learn and collaborate in solving issues related to the Land Cover and Land Use themes