Determine policies for env_medium

[mslarae13]

https://github.com/microbiomedata/issues/discussions/837

Criteria for completion

• Determine NMDC policy for env_medium
• Determine NMDC policy for env_medium in SOIL
• Draft ADR for environment trial methods / policies

Not part of this issue:

• The ADR will change once https://github.com/microbiomedata/issues/issues/844 & https://github.com/microbiomedata/issues/issues/845 are complete.
• Finalized policy will be recorded & ADR will be updated with tasks https://github.com/microbiomedata/issues/issues/846
• How many env_medium are not a child of environmental material? (from what's in NMDC mongo or postgres or NCBI) (see other issue)
  • @sierra-moxon can review
  • @sierra-moxon could help & review with this

Policy for env_medium parent class recommended is environmental material.

Using OAK, structurally do these searches for subclasses & use axiomatic (rules/ children queries that leave out specific strings / text.)

[mslarae13]

TBD, discussion, maximum # of terms to include? -- Probably not an issue here unless it's a LOT (what is a LOT IDK yet)

• UI improvements are a renewal goal

[mslarae13]

Does environmental material need limited before it's considered the top level for env_medium?

• using environmental material eliminates anthropogenic litter

Decision: yes, use environmental material, and do an evaluation of what was excluded. Top level may be changed.

• Evaluate lost material entity terms. Which terms weren't lost because they're also in environmental material. Do we need to bring any back in?

[mslarae13]

env_medium for soil

• environmental material - ??
• Chris recommends using lexical search. Find all terms with the word "soil" in their label or definition. By axiom it's hard.