Volcano subtypes

[GoogleCodeExporter]

Include active vs inactive? GAZ uses these as qualifiers. The qualifiers would 
be applicable to all subtypes of volcano. Term explosion? We could use class 
expressions in GAZ.

Include classification by lava type? E.g. 
 * andesitic volcano (using ENVO:01000233 - need def for this)
 * dacitic volcane 

Add "volcano complex", "group/chain of stratovolcanoes" etc (see previous 
tracker item on ranges/clusters). Not to be confused (I think) with "complex 
volcanoes"?

Add term for
https://en.wikipedia.org/wiki/Submarine_volcano

Original issue reported on code.google.com by cmung...@gmail.com on 22 Dec 2013 at 7:08

[pbuttigieg]

@cmungall

Include active vs inactive? GAZ uses these as qualifiers. The qualifiers would
be applicable to all subtypes of volcano. Term explosion? We could use class
expressions in GAZ.

Active and inactive sound something like qualities (cf. gkoutos/pato#53 although they are probably more general than environmental qualities). PATO has them:

• active
• inactive

We can add the top-level classes ("active volcano", "inactive volcano") to be filled out by the reasoner (e.g. 'volcano' and 'has quality' some 'active').

Include classification by lava type? E.g.

• andesitic volcano (using ENVO:01000233 - need def for this)
• dacitic volcane

I can add a classification by lava type too, adding the requisite lava classes in the material hierarchy.

Add "volcano complex", "group/chain of stratovolcanoes" etc (see previous
tracker item on ranges/clusters). Not to be confused (I think) with "complex
volcanoes"?

Could we use the field class for the complexes, groups, and chains (#88, #90)?

Add term for
https://en.wikipedia.org/wiki/Submarine_volcano

Cooking...

[cmungall]

re: pato terms; I think there may be issues with ascribing function to volcanoes...

[pbuttigieg]

Agreed, it's probably more a disposition. Shall we consider that for environmental quality?

[cmungall]

What are the desired semantics, something like?

active volcano = volcano and capable_of some eruption
inactive volcano = volcano and not capable_of some eruption

(I'm not a vulcanologist, this may be too simplistic?)

We could just encode this directly without any need for introduction of anything beyond the process hierarchy. However, it may be useful to introduce a simple way of indicating state, e.g.

active = inheres_in some (capable_of ...)
inactive = inheres_in some (not (capable_of ...))

1. We may wish to reuse the same pattern in different places (e.g. geysers)
2. We may wish to defer on the precise modeling of activity, but still have a way of saying what we think we mean
3. Even if we decide on a precise model we may decide to refactor this later, and it's better to refactor in a single place (the quality hierarchy) than in n places
4. NOT axioms are ignored by EL reasoners like Elk, but we can get a weaker form of EL reasoning by using binary or n-state categories like this with deferred semantics. We do this in ontologies like CL for acellular, binucleate, enucleate, etc

So yes, I think an EnvoQ. Not sure on the relevant level of specificity. I think a generic activity as the disposition to do something may be too generic. "eruptive activity" sounds weird.

[pbuttigieg]

Inconsistent defs for active vs. inactive see: https://en.wikipedia.org/wiki/Volcano#Active

Most scientists consider a volcano active if it has erupted in the last 10,000 years (Holocene times) – the Smithsonian Global Volcanism Program uses this definition of active

It's a combination of capability and periodicity, but the capability idea is a good place to start. I agree with the points in your numerated list.

I think a generic activity as the disposition to do something may be too generic. "eruptive activity" sounds weird.

It does. 'Capability' rather than 'activity' is the more natural construct here, it seems. We would call an active volcano capable of some eruption (~ capability of erupting), as you constructed above. An "active" volcano is actually an erupting volcano (i.e. participant in some eruption process).

No activity on gkoutos/pato#53.

[pbuttigieg]

It seems that, for example, dacite lava can reform (that is derive from) dacite (rock) as well as the converse. One should probably distinguish between igneous rock formation processes (which go through a lava phase) and processes where igneous rock is melted into its corresponding lava.
Would one ever have dacite lava with 'derives into' some dacite and 'derives from' some dacite simultaneously? I'm more a fan of the process way, as it seems more precise and useful once inputs and outputs are defined. Also, I can't find a straightforward def for dacitic lava - defs seem to always point to the rock type (e.g. http://volcanoes.usgs.gov/images/pglossary/dacite.php).

Back to the original post:
Do you have proposed defs for the dacitic and andesitic volcano types?
I'll add the classes now, but still looking for good defs (i.e. is it a 'composed primarily of' [dacite, andesite] rock type, or 'has part' [dacitic, intermediate] lava? Has part is probably incorrect, but it may work for now).

I suppose rhyolite and its associated lava types should be in there for completeness.