The way I've come to understand coverage good practice, value in corner should be used (this is the point for which we provide the exact information), but the Elevation Data Spec goes the way of Bitmaps, mandating value in center:
IR Requirement Annex II Section 6.2.1.2 Constraints of the spatial object type RectifiedGridCoverage
Grid points of a RectifiedGridCoverage shall coincide with the centres of cells of the geographical grids defined in Section 2.2 of Annex II of the Commission Regulation on Interoperability of Spatial Data Sets and Services at the same resolution level.
I discussed this with Emmanuel Devys from IGN France, he confirmed my confusion:
IMHO, the elevation posts are supposed to be at the grid points and not in the middle of the cell. Elevation posts are NOT surfaces. If the encoding is GeoTIFF (quite common), the PixelIsPoint is the option that should be used. Therefore other spec. elements working for pixels (with square area cells) should be avoided for clarity – IMHO). I am curious to get the justification for such a choice in the EL spec.
The way I've come to understand coverage good practice, value in corner should be used (this is the point for which we provide the exact information), but the Elevation Data Spec goes the way of Bitmaps, mandating value in center:
IR Requirement Annex II Section 6.2.1.2 Constraints of the spatial object type RectifiedGridCoverage
I discussed this with Emmanuel Devys from IGN France, he confirmed my confusion:
Which option should be implemented?