v3.2-v4. changes Mapped feature

[smrgeoinfo]

in v3.2 resolutionScale is a MD_RepresentativeFraction from ISO19139'
in v4. its resolutionRepresentativeFraction with an integer value.
I think we should use the v4 approach for the owl implementation

v4 adds an exposure property specified by an ExposureTerm. This is not in v3.2, but I'd vote to keep in the OWL

In v3.2 MappedFeature has samplingFrame that is an association to an abstract SF_SpatialSamplingFeature. in v4. we have mappingFrame specified by MappingFrameTerm. In this case, I'd suggest MappedFeature with 'hasMappingFrame' property to a 'MappingFrame' class, and the implementation of that class (a URI, a Class for a vocab term, a SamplingFEature instance) could be specified by a conformance class (application scheme).

MappedFeature observationMethod is specified by a Category; in the current OWL draft it is specified by a 'Lineage' class, the properties of which are not specified. Seems like the closest thing to the intention of this property would be something like sosa:Procedure but we don't want to use that because its domain is sosa:Observation, which would imply mappedFeature is observation (and event), when it is closer to being an Observation.result.

v3.2 Mapped feature has a metadata association to ISO19115 MD_Metadata, this was removed in v4. I'd suggest OWL implementation with 'hasMetadata' property to a Metadata class, and implementation of metadata is left to conformanceClass/profile specification. The observationMethod property is really a shortcut convenience property for when a full metadata (provenance) description is not linked.