EnvO biome hierarchy is confusing and inconsistently axiomatized about nmdc-schema

[cmungall]

Like searching on env_broad_scale of "Terrestrial biome" may not retrieve "subalpine biome" samples

I suggest making an ENVO issue

[turbomam]

These challenges appear to hold for both the soil samples and the surface and benthic water samples

[turbomam]

For now, we are leaning towards using a static env_broad_scale annotation of "Terrestrial biome" and a static env_medium annotation of soil

[turbomam]

For env_local_scale I would like to use the terms that EnvO considers equivalent to the NLCD terms asserted by NEON.

There are all "area of..." terms. I understand that means getting into shadow classes, but I don't think there are consistent axioms from the areas to something else. I will double check.

[turbomam]

The missing terms and inconsistent axioms may mean that users searching though the NMDC data portal may not retrieve all Biosamples that have the qualities the expect.

Like searching on env_broad_scale of "Terrestrial biome" may not retrieve "subalpine biome" samples

[turbomam]

[turbomam]

Add a USDA soil classification in addition to FAO

[aclum]

@turbomam @cmungall There was some work over the summer with envo, did this get resolved?

[turbomam]

Yes, terrestrial biome and many other classes with "biome" in the label are classified as a biome by inference now.

'subalpine biome' is not classified as a 'terrestrial biome' at this point. I'm not sure whether it should be either, but I'm certainly not an expert. Do we want to say that an aquatic biome can't be a 'subalpine biome'?

[turbomam]

I think there's still work do do on EnvO biome modeling, but that may not all effect us

For example, a variety of classes are asserted to be subClasses of biome but should probably be inferred.