Height issue in 3D Tiles

[ibgreen]

@Beilinson regarding your comments in #2445

I don't have a clear opinion on this at the moment, but it is worth noting that

• during I3S tile to 3D tiles conversion, loaders.gl goes a step further and adds compensation for local sea level variations using the geoid calculated via spherical harmonics and EGM reference data, accounting for the actual difference between mean sea level height and WGS84 ellipsoid height.
• This has been done to meet Esri's exacting conversion standards and seems to be working well for that purpose.
• That does seem to be different from what you are pointing out.
• Whether the same compensation is appropriate for 3D tiles visualization / would solve the problem is not completely clear.

Also, just out of curiousity:

• Do you have other 3D data that misaligns? Rendering 3D tiles over flat maps (unless for e.g. a very flat coastal city) is usually problematic anyway, as inland locations will hover a mile or so above the "sea level" base map, far outweighing any issue from this effect..