incorrect classification: mangrove island is-a ecosystem

[cmungall]

I'm not sure if this is intentional, this is in both feature and ecosystem hierarchy

[cmungall]

it looks like misplaced parentheses in the ecosystem definition (one of the perils of boolean disjunctive logical defs, hard for humans to reason over)

('environmental system'  and ('has part' some 'cellular organisms'))  <--- premature close
or 
('has part' some 'collection of organisms')
or
('determined by' some 'cellular organisms')
or
('determined by' some 'collection of organisms')
or
('determined by' some 'plant anatomical entity')
or
('determined by' some 'anatomical entity')

Even with the parentheses fixed, it varies a lot from the OWL definition

An environmental system which includes both living and non-living components.

In general these should mirror each other

[pbuttigieg]

It's expected - a mangrove island is an ecosystem. It's also an island. Only the "island" superclass is asserted.

Agreed, the ecosystem equivalent axiom should be cleaned up.

Edited the axion to:

'environmental system'
 and ( 
  'has part' some 'cellular organisms'
or 
  'has part' some 'collection of organisms'
or 
  'determined by' some 'cellular organisms'
or 
  'determined by' some 'collection of organisms' 
or 
  'determined by' some 'plant anatomical entity'  
or
  'determined by' some 'anatomical entity' 
)

Even with the parentheses fixed, it varies a lot from the OWL definition

An environmental system which includes both living and non-living components.

In general these should mirror each other

Does it? The various predicates were meant to capture the wide range of living things that would make something an ecosystem, especially I don't know of a compact way to say "living thing" with OBO resources.

However, anatomical parts may no longer be living, it's true: is there a way to assert this condition?
'determined by' some ('anatomical entity' and 'part of' some 'cellular organisms')?

I don't specify abiotic things in the equivalence axiom as it's sort of implicit in the environmental system. Should this be done for completeness? Wouldn't this need an assertion of "non-living"?

[cmungall]

I assumed the ecosystem was meant to parallel the feature hierarchy, with linkages via determined by rather than subclass.

Why is island placed under ecosystem but (say) coral reef is not?

[pbuttigieg]

I assumed the ecosystem was meant to parallel the feature hierarchy, with linkages via determined by rather than subclass.

Yes, in general. But having a "mangrove island ecosystem" class is sort of superfluous, as a mangrove island must have living things as parts anyway.

Why is island placed under ecosystem but (say) coral reef is not?

Island isn't, mangrove island is - and only by inference. It's asserted as a subclass of island.
It's placed there as it links to mangrove swamp which has ecological communities as parts.

Coral reef is not linked to organisms/PCO classes yet, so it isn't classified as an ecosystem (should be done)

[cmungall]

Note that the OWL definition for ecosystem has been modified since this ticket was made 4 years ago.

Mangrove island no longer classifies under ecosystem:

This resolves my original comment.

However, @pbuttigieg in the comments you stated that mangrove island should be classified as an ecosystem:

Island isn't, mangrove island is - and only by inference. It's asserted as a subclass of island.
It's placed there as it links to mangrove swamp which has ecological communities as parts.

I still find the distinction odd - are there really islands without living components? Nevertheless it seems the original point of the issue is addressed. However, the ontology is still out of sync with how @pbuttigieg believes it should be.

How should we proceed? Should I close this issue, and we can continue to discuss general issues around ecosystem classification on an existing issue, such as:

• Create design patterns for biome and ecosystem #1209