Refactoring biome

[pbuttigieg]

Related to #208 #807 #864

@cmungall and I discussed how ensure that the biome/ecosystem hierarchies are a) well-founded and have semantic substance 2) coordinated with the astronomical body part hierarchy

Firstly, all biomes will be made subclasses of their corresponding ecosystem type (#807). They maintain a link to the ABP through the determined by axiom that links ecosystems to ABPs.

1. will be addressed by adding processes of system development and change (e.g. ecological succession) and using these to axiomatise ecosystem states consistently (e.g. biomes would be in a form of ecological equilibrium). These processes can be used directly in annotation too, simplifying annotation and allowing systems the ability to infer the ecosystem+state term (e.g. biome, transitional ecosystem, or pioneer ecosystem).

2. This would driven by design patterns: selected ABPs will have a auto-generated ecosystem cognate, selected ecosystems would then have a biome form, causally linked to successional processes.

https://en.wikipedia.org/wiki/Ecological_succession

[pbuttigieg]

PS: in some cases, the ABP is the ecosystem (populated by inference: tundra ecosystem would have an asserted subclass axiom: part of some astronomical body.) - and no additional ABP class is needed.

[pbuttigieg]

We'll also have to leverage biosphere to be the link between ecosystems and ecosystem-level ABPs.
The biosphere is an ABP, and certain ecosystems (like forests, for the moment) are only parts of a biosphere.

We need a defined class to bring these into the ABP hierarchy, something like ecological ABP.

[kaiiam]

Would this mean that ENVO would have ecosystem and astronomical body part as the "main" hierarchies (in addition to environmental material) rather than the current: biome, and environmental feature?

[pbuttigieg]

Would this mean that ENVO would have ecosystem and astronomical body part as the "main" hierarchies (in addition to environmental material) rather than the current: biome, and environmental feature?

Feature has been queued for deprecation for quite a while - figuring out how to distribute the classes globbed under it is tricky.

I don't think it's actually correct to say that ENVO has any "main" hierarchies - that's just an artifact of initialisation, the microwave background radiation of the first push.

[kaiiam]

I don't think it's actually correct to say that ENVO has any "main" hierarchies - that's just an artifact of initialisation, the microwave background radiation of the first push.

Fair enough, it's just how one might (and often from my experiences how people do) perceive ENVO, similar to how GO has MF, CC, BP as it's main hierarchies.

[pbuttigieg]

xref #1321