Geofeature defs

[nicholascar]

This is a major update to SWEET Ontology Realm Geologic ontology and a couple of related ontologies.

Within SWEET Ontology Realm Geologic, it provides a class hierarchy beneath the class soreag:GeologicFeature of about 50 classes, all with definitions drawn from literature which is cited. It also provides cited definitions for a few classes in the SWEET Ontology Realm Geologic Basin, SWEET Ontology Realm Geologic Continental & SWEET Ontology Realm Geologic Orogen ontologies.

It tidies up the top-level hierarchy of SWEET Ontology Realm Geologic too by adding in explicit subclassing triples for classes for which these are inferred by reasoning. This helps with hierarchical display in tools such as Protege.

All changes are attributed to the Geological Survey of Queensland, Nicholas Car and John McKellar equally since John & I did the work for the GSQ.

I was able to provide pleasant diffs for:

• realmGeolBasin.ttl
• realmGeolContinental.ttl
• realmGeolOrogen.ttl
• rela.ttl

since changes were small but I wasn't able to do so for the main realmGeol.ttl file since Protege reformatting moved things around (e.g. new spaces).

[nicholascar]

@dr-shorthair @lewismc @smrgeoinfo any chance of a review here? There are a lot of classes, sure, but they are all done in the same style so if you review a few and like them, you should be happy with the rest.

[dr-shorthair]

SedimentaryBasin sub-class-of Province is not convincing to me. That implies that every individual SedimentaryBasin is also a Province. Is this true @smrgeoinfo ?

[johnmckellar]

Yes, every individual sedimentary basin is now regarded by GSQ as a ‘province’. The implication is that sedimentary basins are sedimentary provinces. This was not GSQ’s previous position, where the term ‘province’ was applied only to older sedimentary basins that had been strongly tectonised and/or metamorphosed and whose outlines were indefinite (e.g., see Jell, 2013: https://spaces.hightail.com/space/Cdb1o8ieg1).

Provinces, overall, are now taken to embrace the terms: craton, shield, orogen, orogenic collage, sedimentary basin (sedimentary province, as indicated above), tectonised/metamorphosed province, metallogenic (mineral province), igneous (petrographic) province and large igneous province.

In adopting this approach, GSQ is now in full alignment with Geoscience Australia, in that organisation’s approach to Australian geological provinces: http://www.ga.gov.au/applications/provexplorer/australian-geological-provinces

[dr-shorthair]

OK - my 1970's geology education may need an update ;-)

[dr-shorthair]

But I guess that begs a question - while it may reflect the consensus in Australia, is that shared throughout the SWEET community?

[johnmckellar]

We can look further afield, at the USGS, albeit via Wikipedia:
https://en.wikipedia.org/wiki/Geologic_province#/media/File:World_geologic_provinces.jpg

[geoderekh]

I have expressed to John before about the Australian usage of terms and the differences to the way terms are used in the broader community. My view is that Australian geologists seem to like smaller spatial subdivisions when defining orogens or provinces, to an extent that is somewhat decoupled from international usage.
Could there be trouble in paradise?

[johnmckellar]

Yes, the USGS regard regions of extended crust as a province (“Extended crust = A province of thinned crust (>50 %) due to extension—e.g. The Basin & Range, North America”): https://earthquake.usgs.gov/data/crust/definitions.html And they give the example of the B & R.
I did not include extended crust in the list of provinces, but, clearly, it must be added, if we decide that this is the way forward, and I believe this should be the case. So, the B&R Province is regarded as an ‘Extended Crust Province’, with its many sedimentary basins. But, every basin in the B&R (Extended Crust) Province need not be considered an individual province. It is all a matter of scale and how the ontology is applied. A common-sense approach must prevail. It seems to me that there can be no ontological Utopia, and that anything that we develop has to cater for, or adapt to, the geological terminology that exists, as varied and confusing as it is. Also, some of the geological terminology may need to change too. All we can do is aim to bring it together with reasonably well defined terms and associated altLabels, as suggested.

Moreover, in taking the B&R a step or two further, the ranges in the province are known as the Great Basin Ranges, and, importantly, the basins are collectively known as the Great Basin, which is largest region of contiguous endorheic drainage in North America. So, herein lies another question: Should the Great Basin itself be identified as a province: a sedimentary basin province, a geographic province, a hydrogeologic (or hydrographic) province? Thus, we would have one province type overlapping another. As for the individual basins or watersheds, the list is long: https://en.wikipedia.org/wiki/List_of_Great_Basin_watersheds

Where do we go from here and how do we/you deal with this (given that the B&R Province, in definition by the USGS, is an existing province)? Are they basins within a basin (The Great Basin)? Or, are they to be regarded as subbasins? That the latter is the case is suggested by the following: https://www.fs.fed.us/rm/boise/research/gis/documents/GBRIHUC4methods3.pdf

But, then you have:
https://www.nps.gov/grba/planyourvisit/the-great-basin.htm where the following is indicated:
‘The Great Basin includes most of Nevada, half of Utah, and sections of Idaho, Wyoming, Oregon, and California. The term "Great Basin" is slightly misleading; the region is actually made up of many small basins.’

So, terminology and usage thereof are just as confused in the USA as they are in Australia (and undoubtedly elsewhere). (And, certainly, let’s definitely not go down the subprovince route here, which has been the case for metamorphosed/tectonised provinces and subprovinces in Queensland. But, they, nonetheless, exist in the literature.)

The Great Basin, as a present-day, internal-drainage, sedimentary-basin and hydrogeologic system, is comparable in its enormity to the present-day Lake Eyre Basin in central Australia, although the latter has quite different tectonics and lacks basin-and-range structural architecture.

As a side note, my wife and I have explored parts of the B&R some years ago, and, in particular, Death Valley. It was amazing to see geology in action: the springs, dunes, salt deposits, scree deposits and fault scarps. The region has little rainfall, but, when it does rain, it causes a huge amount of erosion and sedimentation. We were totally blown away by then-recent images of car almost completely buried by conglomerate, and a fairly large concrete toilet block (both male and female) completely washed well away from its (concrete) foundations.

To precis: We are stuck with provinces, like it or not. We just need to work out how we deal with them. GA and the USGS consider basins to be provinces. Is the B & R extended crust province sufficient to deal with the basins/subbasins therein? Should the Great Basin also be considered to be a province (sedimentary-basin, hydrogeologic or whatever province), with province types overlapping—is this functional or not? Does the Great Basin comprise basins or subbasins?

[smrgeoinfo]

Perhaps trying to shoehorn the concepts you need into SWEET is the fundamental problem. Rob Raskin was not a geologist, and that is reflected in the SWEET concept hierarchies. If I understand the use case, the challenge is to categorize parts of the Earth at a regional scale, and there are different ways of doing this-- deformation process perspective (thrust belt, extended terrane, craton), sedimentation perspective (foreland basin, forearc basin, intracratonic basin, passive margin embankment), crustal architecture (continent crust, lithosphere, ocean crust), plate tectonics (subduction complex, spreading center, transform margin), magmatic activity (flood basalts, volcanic arc, plutonic arc..).

Seems like the key thing about calling something a 'province' has to do with regional extent, and some unifying tectonic history or processes.

[smrgeoinfo]

Bottom line-- I don't think 'sedimentary basin' should be a kind of province. Brainstorming here... Distinction could be: a 'province' is a composite of various features related to a shared geologic/tectonic history in a connected (at the time of the defining history) region, not denoting any particular process. A sedimentary basin is an individual feature characterized by processes of subsidence and sedimentation. A sedimentary basin can be partOf a province, but a basin by itself would not be considered a province.

[johnmckellar]

Hi Stephen, I thank you and the others for your and their comments—and please except my apologies for my slow reply.

The USGS have used the Glossary definition to define geologic provinces as "any large area or region considered as a whole, all parts of which are characterized by similar features or by a history differing significantly from that of adjacent areas", and they divide geologic provinces into six main categories: shield, platform, orogen, basin, large igneous province, and extended crust.

Your comment: "Seems like the key thing about calling something a 'province' has to do with regional extent, and some unifying tectonic history or processes.”

Let me give an example here to demonstrate why sedimentary basins meet your criteria (those cited directly above): I refer to the Sydney Basin - Gunnedah Basin - Bowen Basin system of eastern Australia, with these basins being generally referred to individually, or, if collectively, simply as the Sydney-Gunnedah-Bowen basin system. Such is the geographic scale of this system (which involved extension/transtension, thermal relaxation, and foreland basin phases), it could be classed as a superbasin. However, it is more massive in geographic coverage than this, as the Bowen Basin was connected to the Galilee Basin to the west (in central Queensland) over the Springsure Shelf, and the Galilee Basin was connected, at times, in its southwest, over the at-times-active Canaway Fault/Ridge, to the Cooper Basin in south-western Queensland and north-eastern South Australia. This entire system occupied much of eastern, central Australia, and north-eastern South Australia during the Permian – Middle Triassic.

The entire system was not only massive (regionally extensive, a criterion of yours, above), but it had quite-complex, unifying tectonic histories and processes (another criterion of yours for a province). These basins formed largely in the late Pennsylvanian, and were certainly underway by the early Permian. Their initial formation followed the change in rotation of Pangaea, at about 300 Ma, from clockwise to anticlockwise. In eastern Australia, associated dextral/right-lateral transtension (resulting from the anticlockwise rotation) led to the formation of magmatic grabens and half-grabens. This represented the initial birth of the system. Magmatic grabens and half-grabens also formed at the same time in the Northern Hemisphere, such was the global nature of this event, with the loop in the pole path featuring in all apparent polar wander paths from the major continents.

In eastern Australia/eastern Gondwana, following a period of thermal subsidence, continuing anticlockwise rotation gave rise to dextral transpression and commencement of the Hunter-Bowen Orogeny, with its episodic compressional pulses that are marked by unconformity in the sedimentary basins (representing the uplift of half-graben infill) and which have been dated by Ar-Ar step-heating analysis in South Africa, where the orogeny is termed the Cape Orogeny. However, this deformation extended right along the Panthalassan margin of southern Gondwana. Each country has its own name for it, but overall it is embraced by the term, Samfrau Geosyncline (or Orogenic Zone) of DuToit (1937).

These compressional pulses deformed the eastern margin of Australia, forming a mountainous thrust belt which gave rise to foreland loading of the sedimentary basin system. Deformation was more pronounced in the north, where the Bowen Basin was more greatly affected by the deformation, to such an extent that Jell (2013; Geology of Queensland, GSQ book publication) actually referred the Bowen Basin to the New England Orogen (i.e., he considered the basin to be part of the orogen; not everyone may agree with this). A major compression at the end of the Middle Triassic uplifted the whole system in eastern Australia and terminated sedimentation. This termination extended further afield into the Galilee and Cooper basins.

The tectonic model for the Bowen Basin involves integration of quite-complex tectonic, structural and depositional histories. The more distal Galilee and Cooper basins, as inferred, had comparable histories, although sedimentation in the former began earlier, in its north-east, likely in the latest Mississippian, because of local foreland loading resulting from the formation of a thrust system following mid-Carboniferous Kanimblan orogenesis.

If you cannot advance another criterion or further criteria to exclude sedimentary basins from geologic provinces, then, as the basinal system described above very clearly meets your criteria for a geologic province, then you must accept the fact that sedimentary basins are indeed geologic provinces, a view, indicated above, that is shared by GA and the USGS.

Your comment: "A sedimentary basin is an individual feature characterized by processes of subsidence and sedimentation".

According to Miall (2000), there are three crustal stress environments in which sedimentary basins are created: extensional basins (in which the axis of maximum shear stress is vertical), contractional basins, and shear basins (where, for the latter two, the axis of maximum shear stress is vertical). These different basin types form in various plate-tectonic settings. Tectonism and sedimentation are intimately related (Miall, 2000), even if at least remotely (Dickson, 1975), and, in the latter case, are associated with mantle thermal processes and dynamic topography or epeirogenesis, the influence of which is now known to be of considerable regional significance, involving either thermal expansion or contraction of the crust (Mail, 2000). Downwelling convection (asthenosphere-flow) currents in regions of cold mantle (generally, if not entirely, corresponding with geoid lows) explain the connection between epeirogeny and cratonic-basin development. Ingersol & Busby (1995) listed five subsidence mechanisms that generate accommodation space for sediment, including asthenospheric-flow downwelling, and, although a number of basin classifications exist, all sedimentary basins can be related to their plate-tectonic setting. As such, sedimentary basins are tectonic entities. Even epeirogenesis causes crustal deformation (tectonic deformation), although this is not as readily as apparent as that on a convergent margin, especially in a foreland setting.

Your comment: "I don't see what the distinction is between soreag:TectonicEntity and soreag:GeologicProvince is anyway".

Yes, Stephen, I agree with you—there is no distinction!! What I did was to exclude from provinces other features that I didn’t classify as provinces.

So, here lies a potential resolution: Even though I have argued above that sedimentary basins are geologic provinces that meet your criteria for being so, let’s not refer to them as provinces, but just as sedimentary basins. This should please both you and Simon.

Sedimentary basins can be classed as tectonic entities, along with other tectonic entitiies that include geologic provinces. I suggest that we eliminate Province as a class and place all under tectonic entities.

Tectonic entities could then embrace:
lithosphere, crust (continental, oceanic) [note that the lithosphere is a tectonic entity as it is subdivided into tectonic plates, which are indisputably tectonic entities; and it will link to other ontologies (layers of Earth), as I think I have seen the inference of this made by Nick?]
tectonic plate
mid-ocean ridge
subduction zone
volcanic arc
present-day continent, ancient continent, supercontinent
craton, shield, platform, supercraton
orogen, orogenic collage
sedimentary basin,subbasin, superbasin
tectonised/metamorphosed province,
igneous (petrographic) province, large igneous province
extended crust province
etc., as other tectonic entities can be further added

Can you accept these proposals? And, can we just define the above provinces under tectonic entities?

[dr-shorthair]

Trough sub-class-of SedimentaryBasin is not convincing to me. That implies that every individual Trough is also a SedimentaryBasin . Is this true @smrgeoinfo ?

[johnmckellar]

Considering basin type or status, we may have the hierarchy of superbasin, basin, sub-basin or trough. However, not every basin has a higher-order superbasin—in fact, most don’t. The preeminent superbasin in Australia is the Great Australian Superbasin, which, from a hydrogeological point of view, is intimately associated with the Great Artesian Basin (but with the latter also encompassing older aquifers of underlying basinal systems).
And, not every basin has to have a sub-basin or a trough. In the context here, a trough is essentially a sub-basin. But, for example, in the continental realm, and for the principal Phanerozoic troughs of Australia, most, if not all, form parts of larger sedimentary basins. And, indeed, for the extensive, Sydney-Gunnedah-Bowen basin system of eastern Australia (and also considering the Cooper Basin and, to some extent, the Galilee Basin of central Australia), the formation of troughs at ≈300 Ma represents the initiation of basin formation and subsequent growth. These troughs represent the initial structural formation of a series of half-grabens through dextral (right-lateral) transtension as a thermally-elevated Pangaea switched from a clockwise to an anticlockwise rotation. This was thus a global event, and the formation of such troughs (largely magmatic grabens and half-grabens) not only occurred in the Southern Hemisphere (in Gondwana), but also in the Northern Hemisphere (in Laurussia/Laurasia).

The main troughs in the Bowen Basin fall in the basin, but there are a few that sit on its western margin (viz., the Katanga and Arbroath troughs), but they are regarded as part as part of the basin system, and also to form part of the ‘(latest Carboniferous –) Early Permian East Australian Rift System’. They could be considered to be small basins in their own right, but considering their overall geological and tectonic context, this would be stretching things a little too far.

In the oceanic realm, a trough is regarded as an elongate depression that is differentiated from an oceanic trench by being shallower, narrower and shorter. Unlike (subduction-zone) trenches, they form through a number of geological mechanisms, including rifting. In this realm, they form part of oceanic basins, the floor of which receives sediment of one kind or another.

So, to answer the question, troughs form parts of sedimentary basins, whether continental or oceanic, and receive sediment.

[dr-shorthair]

Partonymy (mereonomy) is not the same as subsumption.
I think there is a flaw here.

[johnmckellar]

Hi Simon, your point is a little ambiguous!! If you have a problem with a trough being part of a sedimentary basin (and, as such, being a sedimentary basin in its own right), then you must also have a problem with a sub-basin (also a sedimentary basin in its own right) being part of a sedimentary basin. Issue is raised with the former, but perplexingly not with the latter. As an example, John & Fielding (1993; https://www.publish.csiro.au/AJ/AJ92014) refer to the Denison Trough a being a deep elongate sub-basin located along the western margin of the Bowen Basin.

[dr-shorthair]

That's why I would like to see a diagram. If 'Sedimentary Basin' is a specific ranking within a hierarchy, with specified smaller things having part-of relationships with specified bigger things, then sub-classing is incorrect, and a 'has-part' relationship between individuals would be required. OTOH if 'Sedimentary Basin' is a general class, with other ones being special cases, subsumption is correct. That was why my original question was framed as 'implies that every individual Trough is also a SedimentaryBasin .'

[smrgeoinfo]

To start, note that there are three Troughs in SWEET currently: sorepsg:Trough (geometric object), soreagb:Trough (sedimentary basin), sophfd:Trough (geometric object with low pressure).

[smrgeoinfo]

Basins have parts. they might be called 'sub-basins' or 'troughs'. Some basins are called troughs. If someone can propose an unambiguous, logically coherent definition to distinguish a sub-basin from trough, or a trough from a basin, Then we have some distinct concepts for the ontology. If not, perhaps its best to stick with 'Sedimentary basin' and 'sub-basin' with various skos:altLabels. ?

[johnmckellar]

This is a difficult one!! Most, if not all of the broadly continental troughs that I know of, are parts of basins, and are essentially subbasins. They are generally interconnected structural entities, grabens and half-grabens, that formed through extension/transtension.

But there is one that I know of where this is not the case, and it has been renamed as a syncline—but the old name still sticks. Where they have formed as faulted structures, they can be defined by their steeply dipping sides. However, I cannot be absolutely sure that all continental troughs are subbasins and not basins in their own right.
In the oceanic realm, as previously indicated, troughs form through a variety of mechanisms and can be narrow basin or geological rifts. As such, they receive sediment and form subbasins within their larger oceanic basins, also sedimentary basins.

So, considering the above, I think that the suggestion by Stephen that troughs may be either subbasins (parts of basins) or basins in their own right should prevail. Note that, apart from low-pressure, meterologic troughs, there are also glacial troughs, another name for U-shaped glacial valleys, which, of course, are also sedimentary basins (and subbasins).

Precis: Let’s use basins and subbasins, with troughs as an altLabel for both subbasins and basins, but I think glacial trough can stand alone as an altLabel for a glacial valley, under a geomorphological ontology that we need not concern ourselves with(?).

[smrgeoinfo]

sounds like can be resolved with basin and subBasin

[johnmckellar]

Yes

[dr-shorthair]

Is every Supercraton also a GeologicProvince? Is this true @smrgeoinfo ?

[johnmckellar]

A supercraton consists of two or more (largely Archaean) cratons (Blewett, 2012: https://press.anu.edu.au/publications/shaping-nation), and as cratons are tectonic provinces, a supercraton consists of as many provinces as there are cratons in the supercraton in question. And, if the truth be known, it we dig a little deeper in a craton, we would undoubtedly find provinces within provinces (if they are still recognisable as such, considering their age and complex, multiphase tectonic history). I will leave Nick to deal with how best to approach the multi-province aspect.

[smrgeoinfo]

question remains: is a supercraton a craton in which case it should be a subClass, and if a craton is a GeologicProvince, then a supercraton is a GeologicProvince.

As I read the proposed definitions a craton is a "part of Earth's continental crust that has attained stability and has been little deformed for a prolonged period" so it sounds like a supercraton might or might not be a craton. Western North America basement (east of Cordillera) is an amalgamation of Archean through Early Proterozoic terranes, amalgamated by Early Proterozoic time, and certainly behaving as a craton through the Early Paleozoic, so perhaps a 'supercraton'? This 'craton' has seen considerable tectonic activity on the western margin since then, so is it still a 'craton' or has the craton changed extent? is it still a super craton?

[johnmckellar]

A supercraton would be a geological province, but, Stephen, what are you suggesting to be the subclass—the craton or the supercraton? We should not be here to question what others, in published literature, have delineated a supercration to be, and if their usage of the term is valid.
I am no expert on cratons. I am just referring to the Glossary and other sources. But, in terms of the very old ages of the features that we are discussing here, I imagine that a craton can form and be stable and little deformed for a long period of (geological) time, thus satisfying the criteria for being termed a craton. But, then, if after such a long period, this stability is interrupted by tectonism or whatever, causing, say, an amalgamation of cratons, should the amalgamating cratons then cease to be termed cratons? Should the term supercraton be discarded as the once-stable cratons that comprised it have ceased to be cratons (because of tectonic amalgamation), and a new craton (not a supercraton) is formed (which then needs to be stable and little deformed for a succeeding interval)? Again, should we not comply with the published literature.

I also assume that a craton, although stable, would not be isolated from any major tectonic activity, as tectonism even at plate margins can affect a whole plate. Does not the qualification ‘little deformed’ cater for this? I welcome any definition to satisfy all queries. In achieving this, it seems that we may deviate from the published definitions and develop our own. If this is the case, we can overcome issues of copyright and links to copyrighted material (see my comments re this elsewhere here).

[smrgeoinfo]

I agree that to the extent possible its good to define terms to be logically consistent with common usage. Problems arise when common usage is inconsistent. I think the role of information resources like SWEET is to provide some way to disambiguate usages (much like wikipedia...) with clear definitions of the various possible interpretations of a term, and identifiers to distinguish the meanings.

As far as Craton/supercraton, perhaps best if they are both kinds of 'tectonic feature', (or 'geologic feature', 'geologic province', 'tectonic entity', all possible labels for what seems to be the same concept), and Supercraton 'hasPart' Craton. This seems compatible with a quick google survey of usage in the literature, e.g. Bleeker and Davis 2004, Smirnov et al 2013, Salminen et al 2018 and the definition (originally?) proposed by Bleeker 2003: "large ancestral landmasses of Archean age with a stabilized core that on break-up spawned several independently drifting cratons" (which is of course close to the proposed definition in the PR :) )

[smrgeoinfo]

So short answer to @dr-shorthair original question, yes I think every supercraton could be considered a 'geologic province' (pending refinement of definition of 'geologic province')

[johnmckellar]

Both kinds of 'tectonic entity' is OK with me, rather than classifying under 'province', but along with provinces which are also tectonic entities, as suggested in my previous post??

[dr-shorthair]

Is every OrogenicCollage also a GeologicProvince? Is this true @smrgeoinfo ?

[johnmckellar]

Again, a good question and the answer is similar to that posed for supercratons: An orogenic collage consists of a number of orogens (more than one) of different age, e.g., the Tasmanides of eastern Australia. Some may consider the Tasmanides (Tasman Orogenic Zone) to represent a single orogen, but most workers regard the Tasmanides as embracing three separate orogens ranging in age from Neoproterozoic to late Middle Triassic (if not to the Norian, i.e., middle Late Triassic—taking into account what the South Africans consider for their Cape Orogeny, and disregarding early Late Triassic dextral transtension in eastern Australia).

So, an orogenic collage consists of the number of provinces that constitute it. Again, over to Nick on how to deal with this.

[smrgeoinfo]

The problem is with the proposed definition of 'Geologic Province'-- it is extensional, i.e. list of things that are Provinces: "cratons/shields (in the continental realm), orogens, sedimentary basins, and tectonised/ metamorphosed and/ or mineralised regions, as well as large igneous provinces" I think there needs to be an intentional definition-- i.e. what are the properties of a part of the Earth that make it a 'Geologic Province'?

My proposal would be something like:
A geologic province is a part of the Earth defined by a shared geologic history.

ScopeNote:
A given part of the Earth might be included in various geologic provinces based on different temporal intervals in its history. For example, western Arizona is part of the North American Craton province in the Early to Middle Paleozoic; it is part of the SW North America Megashear province during the middle Mesozoic; it is part of the SW North America transpressional linkage province in the Late Cretaceous; it is part of the low-angle subduction hanging wall province in the Early Tertiary; it is part of the crustal extension province in the Oligocene and Miocene, and part of the Eastern California Shear zone province in the Late Miocene and Pliocene.

[johnmckellar]

OK with me & fits USGS & GA, especially the former in their definition!!

[smrgeoinfo]

sounds like this can be resolved?

[johnmckellar]

Yes

[johnmckellar]

Hi Simon, We will soon find out!! Cheers, John [removing email details - Nick]

[nicholascar]

@dr-shorthair what's your response to @johnmckellar's comments

@smrgeoinfo @lewismc can you please review? I'm keen to keep momentum here! Let's get PRs merged as they can always be revised.

I have another PR of a comprehensive sphere (atmosphere, hydrosphere) updates ready...

[dr-shorthair]

My concern is primarily around the position of Geologic Province in the hierarchy, and also whether they can be nested. I'm looking for some feedback from non-Australians as to whether this conception is local or general.

Is there a diagram of the subsumption hierarchy? That would also help in the evaluation.

[nicholascar]

First 1/3 of hierarchy

[nicholascar]

Second 1/3 of hierarchy:

[nicholascar]

Third 1/3 of heirarchy:

[nicholascar]

@dr-shorthair wrote:

I'm looking for some feedback from non-Australians as to whether this conception is local or general.

Note that John says "Much of the proposed structure of the Features ontology is based on the Guide for Geological Nomenclature in Sweden" (as per #187 (comment))

[dr-shorthair]

Yes, but presenting a single tree is misleading as it does not tease out the relationships between individuals. Most of this is fine, but I'm concerned that in some places there is a conflation of

• FeatureA is an instance of class X and therefore is also an instance of its superclass Y
• FeatureA (an instance of class X) is a part-of FeatureB (an instance of class Y)

These are very different relationships.

[johnmckellar]

In putting this in the vernacular, so a very simple beginner such as I can understand, I take it that you are saying that, for a term (in this case ‘trough’) that can have different applications and thus meanings (and our languages are full of such terms), we need to explain, and inter-relate them all (and indeed compare them where no relationships exist)?? Does not the context for usage of the term provide this, here in the hierarchy/tree? Trough type can be further subdivided geologically or tectonically in the hierarchy? What do you think Nick, Derek?

[nicholascar]

tease out the relationships between individuals

@johnmckellar and I have already discussed adding some more properties to capture the various rankings of subclasses within a superclass (LithostratigraphicUnit subclasses for instance) and I agree, we will likely need to look into breaking some other subclass relations into part/whole.

If you can suggest sensible part/whole relations, I'll add them into the PR. Do you have some obvious ones in mind?

[dr-shorthair]

dcterms:isPartOf
dcterms:hasPart
To start with anyway

[smrgeoinfo]

LOTS of comments.
POsting for now since I see NIck and Simon are online...
definition should include sufficient conditions.
definition should follow Aristotle-- genera and differentia.

move typical assertions, and other notes to skos:scopeNote.

random comments (others in line)
definitions from Glossary of Geology-- do we need permission?

Why are soreagcont:StableContinent, soreagcont:Supercontinent, soreagcont:ActiveContinent not subclasses of soreag:Continent ??

Why is isn't soreag:LargeIgneousProvince a subclass of soreag:IgneousProvince?

I worked on a set of vocabularies like this in the context of Dave Soller's US National Geologic database project a while back. A couple of observations-- Definitions in the AGI Glossary are for geologists, NOT for information modeling. They are variously circular, ambiguous, or just not definitions (an X is an X...). Releasing the glossary we generated for that project got hung up on intellectual property questions-- AGI glossaries are copyrighted, and there were questions about the legality of using their definitions and attaching URIs, making them publicly available. The Glossary (at that time) was a cash-cow for AGI, and AFAIK they haven't produced an online public version yet with URIs for the terms in the glossary.

I haven't gotten all the way through Nick's contributions-- a massive effort, kudos!. It brings up the fundamental question again-- is SWEET supposed to be a glossary with 'fuzzy', geologist-friendly definitions (or none at all), or is it supposed to be an ontology that can be used for semantically precise annotation of data? If the first, ignore most of my comments. If the second--there's a LOT of work that needs to be done.

[smrgeoinfo]

Anticline is defined as a fold with older rocks in the core. The challenge for writing a definition is how to unambiguously specify what 'core' means in a way that might be computable. This could be done using the stratigraphic younging direction (geospatially up or down) of the normal to the intersection between folded layers and the hinge surface. Of course, this would seem unnecessarily verbose to most geologists.... So the question is still-- is SWEET a glossary (online version of the glossary of geology/wikipedia), or a computable ontology.

It would be more useful to add as a comment (not either necessary or sufficient condition) that anticlines are typically antiformal.

Suggestion for computable precision:
An anticline is a fold (link to definition of fold) in which the younging of folded layers (link to definition of layer) in a direction normal to the intersection of the folded layers and hinge surface (link to definition of hinge surface) is downward relative to the Earth Surface reference frame (link to definition).
Comment: Anticlines typically have an Antiformal shape (link to definition of antiform)

[smrgeoinfo]

anticlinorium--
Is an Anticlinal Structure an anticline??? What does regional extent mean???
Britannica definition is somewhat more useful "An anticlinorium is a large anticline on which minor folds are superimposed" https://www.britannica.com/science/anticlinorium

A quick look at usage of the term is the geoscience literature (https://scholar.google.com/scholar?hl=en&q=anticlinorium) suggests that 'large scale' could be translated to km-scale wavelength, but for now this will probably need to be suggested as a typical rather than necessary or sufficient value, thus indicated in a comment, not the definition.

suggested definition:
An anticlinorium is a large anticline in which the folded layers are co-axially folded with a wavelength less that the wavelength of the anticlinorium.
Comment: wavelength of anticlinioria typically on the order of km in common usage.

[smrgeoinfo]

antiform--
"The term is usually used when the folded layers do not possess a stratigraphic order, when the stratigraphic order of the folded layers is not known, or when the fold core also contains the stratigraphically younger rock." should be skos:scopeNote-- it is not necessary or sufficient.

[smrgeoinfo]

For this definition to be machine actionable, need definition of 'duplex', 'branch line', 'trailing branch line', 'leading branch line', 'stack of horses', 'thrust sheet'.
As a glossary, having these terms is probably useful for geologists, but SWEET is just providing a URI for a definition from Neuendorf et. al.; is that useful, or legal?

[nicholascar]

Out of scope for this first PR @smrgeoinfo! If we go down this path, we will never complete this job. Let's get this definition in, unless it's wrong, and see future PRs make this more machine actionable by recursively defining terms.

We do agree to adding in links to open definitions (Wikipedia etc) in place of closed ones where possible so will try to amend this PR for that.

[nicholascar]

As to legality: It's surely legal to cite things, as you would in any academic publication. Something about fair use...

[smrgeoinfo]

Agree that defining all the dependent terms would explode the problem, and best to recognize, but leave for now.
As far as the legal question, I'm no expert but suspect there might be a problem only if def's from AGI are quoted verbatim.

[johnmckellar]

Where we are modifying, this is not a problem. We can cite: Modified from ... But, if we are not, we can just change the wording/phraseology to suit, just like Wikipedia's Glossary of geology has done. Maybe, we can still cite After ... I am getting some advice on the legality of this.

[smrgeoinfo]

as an ontology (as opposed to glossary or dictionary for people) need definition of 'structural complex', 'roof thrust', 'floor thrust', 'imbricate thrust'. I"m not going to try suggesting that in this review.

[nicholascar]

as an ontology

Agree, in the long term.

I"m not going to try suggesting that in this review.

Great, thanks! we'll proceed with this for now and try to build up steam for recursive definitions. (I think there's more work needed on the spheres and layers first through)

[johnmckellar]

Having our own glossary is the only way out, after talking to our head librarian. Cecelia suggested the AusGeoRef Glossary, but it is the AGI Glossary of Geology. AusGeoRef is a cooperative 'venture' between GA and AGI. Agree—leave recursive definitions for now, but lots of work down the track.

[smrgeoinfo]

'body of rocks' or Geologic Unit?

[johnmckellar]

The International Code, both in online and hard-copy/printed versions (e.g., Murphy & Salvador), refer to 'body of rocks' throughout.

[smrgeoinfo]

A formal geologic unit (link to def) that is a proper part of a formation (link to def)

[smrgeoinfo]

sopropm:Flexure is a SolidProperty/StrengthProperty. Obviously not applicable here.
What is a flexure???
Isn't a monocline a kind of fold???

Suggest def:
A structural feature consisting of a pair of folds that bound a relatively short, more steeply dipping fold limb (link to def) between two more extensive generally gently dipping longer limbs.

Scope note: e.g. Kaibab monocline (link)(

[geoderekh]

I agree flexure is problematic because it is used to describe folding mechanisms which are independent of the geometry of a fold (monocline or otherwise). However, I think of a monocline as a single structural feature rather than a closely spaced antiform-synform pair in sub-horizontal rocks as this is the way it is represented on most maps.

Revised suggest def:
A step-like fold consisting of a relatively short steeply dipping limb (link to def) bounded by the axial surfaces of a pair of folds between two more extensive and gently dipping longer limbs.
Scope note: e.g. Kaibab monocline (link)

[smrgeoinfo]

Better! Maybe could be simplified further...
A step-like fold consisting of a relatively short steeply dipping limb (link to def) between two more extensive and gently dipping longer limbs.

[johnmckellar]

Are you happy with this @geoderekh

[geoderekh]

yup!

[smrgeoinfo]

shouldn't this be a subclass of shear zone?

[johnmckellar]

Do we start to develop subclasses of shear zones @geoderekh? We have the end members of brittle shear and ductile shear and what else in between?

[dr-shorthair]

If you have different kinds of shear-zone in your data, then OK. But don't make this a stamp-collecting exercise ...

[smrgeoinfo]

My vote to avoid stamp collecting would be to deprecate 'mylonite zone' and avoid the problem.

[johnmckellar]

Agreed

[geoderekh]

We could make a whole ontology about the spectrum existing between faults/brittle SZs and ductile SZs, but I dont think that is in scope here. Having two end members with a bunch of altLabels where relevant.

[smrgeoinfo]

suggest def:
A ductile shear zone (link to def--have to add subtype of shear zone) in which the deformed rocks are mylonitic rock (link to def).

subclass of shearzone/ductileshearzone

Scope note -- rest of the text in current definition

[cmungall]

It looks like you are deleting classes rather than deprecating/obsoleting them?

This means that older URLs to soreagb:Basin will no longer resolve, and worse, there will be a referential integrity problem with any external data or triples that reference it, without an easy way to fix this (perhaps I am missing something and there as external set of replacement triples being maintained).

Note that this kind of issue would be solved by opaque IDs, which allows labels to freely change so long as the definition remains the same but when I raised this in #49 the consensus was that terminology in earth sciences was fixed and there was no drift... time to revisit?

[nicholascar]

Hi Chris, we'll wind this one back and retain the original Basin class! Bad practice from us to break URIs.

Not that I really think there actually are any URIs pointing to it out there, do you know of any?

[smrgeoinfo]

Definition: A large area of exposed basement rock (link to def) in a craton (link to def), commonly with a very gently convex surface, surrounded by sediment covered platforms.

Scope Note: The rocks of virtually all shield areas are Precambrian. (Neuendorf et al., 2011)

is a shield always a part of a craton?, or is it a kind of craton??

[geoderekh]

I like the revised def.
Cratonic shields and cratonic platforms are not so fundamentally different, with the distinction being the amount of younger cover on them, or from another perspective, whether the peneplain is exposed (shield) or subsurface (platform).
I can't give a professional opinion here, or even an educated guess, but I have a feeling is that shields and platforms are parts of a craton rather than kinds of cratons.

[johnmckellar]

From what I can gather, a shield is always part of as craton, but if part of the craton is exposed over a large area, that part of it is termed a shield. Please correct me if I am wrong. Online, you will see reference to the Australian Shield (= Western Australian Shield), but, as far as I can determine, GA does not embrace this usage, and all Australian cratonic regions are referred to by them only as ‘cratons’.

The revised definition is little different to the original definition, the exception being the exclusion of the scope note; and all of it is referred to the Glossary (Neuendorf et al.). I am quite happy with this, but the problem is with the referencing, an issue also raised elsewhere here, several times. From a copyright perspective, we can get away with referencing a definition, in this case for a ‘shield’, to the Glossary. But, to provide a direct link for definition of terms (for all terms in a definition), as has been suggested (and clearly a necessity), becomes a problem. We cannot provide links to Neuendorf et al., as it is not freely available. And, from what we have here in all of this feedback, is that their definitions have issues. Apart from directly linking to free, reputable-source definitions (that we like), do we need to develop our own glossary (which also includes and directly acknowledges the liked definitions)? When you look at the comments here, this is what we are doing (i.e., developing our own definitions)!! To base definitions directly on Neuendorf (or other published glossaries) would breach copyright fair-dealing, without a paid agreement/subscription in place. This, however, brings us to the Glossary of geology — Wikipedia:
https://en.wikipedia.org/wiki/Glossary_of_geology

And Wikipedia has also been calling out for monetary contributions too; and, besides, it’s not ‘officially’ published. When I compare the two (obviously not fully), Wikipedia appears to have taken the Glossary definitions and just changed the wording/phraseology, to overcome copyright (at least for the time being).

In developing vocabs, we have been instructed to follow published national and international standards and published definitions (not make up our own). But, this seems to be almost unworkable, without falling foul of copyright laws and having agreements in place with copyright owners. I can see where those of you that don’t want definitions are coming from, but, without definitions, we have disorder. Maybe, considering the scale of what we are doing, it is possible to develop mutually-beneficial relationships (whatever they could be) with publishers, but, then, as also pointed here, the Glossary is not up-to-date either (as likely would be the case for other glossaries/dictionaries). Considering the latter point, I have been disturbed, confused and frustrated by the distinctly different definitions provided for some terms by various reputable sources; and have been astonished by the lack of definitions provided in major references on a number of subjects, e.g., a recently published book on Coal Geology (that I was looking through) that defines coal, but not coal geology!! This is not an isolated instance.

Precis: Revised definition OK, with SKOS note. But published definitions are largely associated with copyright issues and we seem to have issues with them in any case. It further seems to me that we want to develop our own definitions—big job!!

[smrgeoinfo]

An orogen is a part of the Earth’s crust deformed by tectonic processes involving large-scale faulting, regional metamorphism, and associated magmatism, producing significant vertical displacements of rock bodies.

ScopeNote: usually caused by convergence (± accretion and/or collision) between two tectonic plates or major crustal block

[smrgeoinfo]

.. position of Geologic Province in the hierarchy,...>
Is there a diagram of the subsumption hierarchy? That would also help in the evaluation.

It seems to me that GeolgoicProvinces are defined by shared history and ?perhaps by spatial contiguity during that history. Thus not necessarily a subclass of soreag:TectonicEntity. I don't see what the distinction is between soreag:TectonicEntity and soreag:GeologicProvince is anyway.

[smrgeoinfo]

As @cmungall points out, probably best not to change the URI for soreagb:Basin, just change the label to disambiguate better from soreahb:Basin. The cows have left the barn at this point using opaque URIs in SWEET....

[nicholascar]

OK

[smrgeoinfo]

not a definition.
suggest: a sub-basin is a sedimentary basin that is part of a larger sedimentary basin.

This restriction could(should) be added:
soreagb:Subbasin sorel:partOf soreagb:SedimentaryBasin

ScopeNote: A sedimentary basin may be subdivided into two or more subbasins that have separate depocentres and are separated from one another by some tectonic/structural element, such as a fault or basement high. (Definition created here: Paul Blake, John McKellar)

[smrgeoinfo]

define the concept. Currently what it says is 'a superbasin is a term'. I don' think that's what we want.
suggest def:
A sedimentary basin that includes several interconnected sedimentary basins of the same or similar age.

question-- is being interconnected a necessary/sufficient property of parts of a superbasin. That's the implication of the proposed definition.
2. a basin hasPart subbasin; also a superbasin hasPart basin. What differentiates soreagb:SedimentaryBasin with sub-basins from a soreagb:Superbasin that includes basins?

ScopeNote-- rest of the stuff in the current definition

[smrgeoinfo]

I'd suggest that the restriction to 'widespread, if not worldwide' is unnecessary. Stratigraphic events can be observed in any stratigraphic section. A different class for 'global' or 'widespread' stratigraphic events should be defined.

Deeper problem-- use of term 'event' will have the connotation of a temporal feature. I think the intention here is that the 'stratigraphic event' is the physical manifestation of an event that occurred during deposition of a stratigraphic unit, so that 'stratigraphic event' is a physical thing that can be observed in outcrop. In that understanding, then 'stratigraphic event' must be a kind of 'stratigraphic unit'.

Given the restriction to 'widespread, if not worldwide' spatial extent, perhaps what is intended here is something like 'marker bed', which would be another kind of stratigraphic unit?

[smrgeoinfo]

not a definition.
is a stratigraphic feature a kind of geologic unit? See discussion of 'stratigraphicEvent' in the comments.

NOTE: The current SWEET ontolgoy does not seem to have a 'GeologicUnit' concept, which is central to the GeoSciML conceptual model. soreag:LithodemicFeature, soreag:StratigraphicFeature, soreag:GeophysicalFeature all appear to me to be kinds of GeologicUnit. From GeoSciML v3.2: "a body of material in the Earth whose complete and precise extent is inferred to exist (NADM GeologicUnit, Stratigraphic unit in sense of NACSN or International Stratigraphic Code), or a classifier used to characterize parts of the Earth (e.g. lithologic map unit like 'granitic rock' or 'alluvial deposit', surficial units like 'till' or 'old alluvium')."

[smrgeoinfo]

Seems like InnerCore and OuterCore are subclasses of GeosphereLayer, with a partOf Relation to Core, not a subclass relation.

[nicholascar]

Not one of our mistakes- we haven't touched properties like this.

I actually do have another PR waiting in the wings for a tidy-up of the Solid Earth, Hydrosphere & Atmosphere and their layers (all using isPartOf relations) but I will wait for this PR to get through before lodging that one.

[smrgeoinfo]

Restrictions on craton in realmGeolContinental.ttl seem problematic:

  rdfs:subClassOf [      rdf:type owl:Restriction ;
      owl:allValuesFrom matrRockIgneous:Kimberlite ;
      owl:onProperty sorelch:hasSubstance ;     ]

would seem to imply that a craton may only consist of Kimberlite (or a subClass of Kimberlite? That doesn't seem correct.
also there are three separate owl:allValuesFrom restrictions on owl:onProperty sorelph:hasPlanetaryStructure, with different classes as the target. I suspect this will lead to logical validation failure. Without diving into the science question of are these actually the only planetary structures allowed on a craton, I think the restriction should be (in Turtle encoding):

rdfs:subClassOf [       rdf:type owl:Restriction ;
      owl:allValuesFrom [
          rdf:type owl:Class ;
          owl:unionOf (
              soreagb:IntracratonicBasin
              :ContinentalPlatform
              :Shield     ) ;  
        ] ;
      owl:onProperty sorelph:hasPlanetaryStructure ;
    ] ;

[nicholascar]

We didn't contribute any restrictions like this, only definitions and sources, so we've not examined the logic of them.

Please can we leave such discussion to a separate PR/Issue? There are enough things to deal with in this one already. Yes, I think the sort of change you suggest @smrgeoinfo is good but let's have that as another PR.

[smrgeoinfo]

I agree.

[dr-shorthair]

dcterms:isPartOf
dcterms:hasPart
To start with anyway

In SWEET context probably better to use sorel:partOf and sorel:contains

[dr-shorthair]

Spelling? Usually Techtonostratigraphic

[nicholascar]

I think the "tech" is always me! Likely it's a smart spell checker that knows I often write about tech!

[dr-shorthair]

tectonic

[dr-shorthair]

Yikes. Tectonostratigraphy, please.

[dr-shorthair]

tectonic

[dr-shorthair]

tectonostratigraphic

[dr-shorthair]

tectonostratigraphic

[dr-shorthair]

tectonostratigraphic

[johnmckellar]

Thanks everyone for the feedback, especially Stephen & Simon, albeit bordering on deluge. I will reply to all next week, as some research and deliberation are required, once I resurface with the assistance of my floaties.

[geoderekh]

"little deformed over a prolonged period" is not an ideal part of a definition for craton as it is ambiguous and subjective, not to mention that many cratons are highly deformed at their margins.
Also a craton includes all of the lithosphere, not just the crust. Age, stability, thickness are some of the key criteria here.
Suggested def: A relatively old, stable, and thick part of the Earth's lithosphere that forms the nucleus of a continent.

[smrgeoinfo]

restrict to continental lithosphere?
A relatively old, stable, and thick part of the Earth's continental lithosphere that forms the nucleus of a continent.

sort of implicit in 'nucleus of continent' I guess, avoids possible interpretation it could be oceanic lithosphere forming nucleus of continent?

[geoderekh]

I think so. The thickness also implies continental, but there is ambiguity in just 'relatively thick lithosphere' which could be continental lithosphere, cratons, or oceanic plateaux. So the restriction is good.

I now wonder if this def implies that all continents have cratonic nucleii (do they??)

[dr-shorthair]

AFAICT there are two things being addressed in this proposal and reviews:

• textual definitions, which optimally should reflect the community usage and consensus, else if this is not yet agreed then definitions must be accompanied by their provenance

• relationships between classes.

I was primarily concerned that the latter was expressed purely as a subsumption hierarchy, even though some of the relationships should be related to partonymy.

SWEET does have a set of high-level relations available though I don't think they are used much. To support the geological features hierarchy I would expect to see these appear in relationships like

soreag:SubBasin rdfs:subClassOf [
                                        a owl:Restriction ;
                                        owl:allValuesFrom soreag:Basin ;
                                        owl:onProperty sorel:partOf ;
                                     ] ;
.

[dr-shorthair]

I'm now finding it difficult to keep on top of all the separate discussions here.
I suggest breaking it out into multiple issues.
The PR was very big and now looks like it should be broken up into a few pieces, matched with issue threads.

[nicholascar]

I'm now finding it difficult to keep on top of all the separate discussions here.
I suggest breaking it out into multiple issues.
The PR was very big and now looks like it should be broken up into a few pieces, matched with issue threads.

I'm keen not to break the PR up. With the decision to focus on textual definitions first and to tidy the subsumption and part/whole relations second and to stick to those relations only, the main work here, going on at points above, is to address the definitional challenges and that's all a single conceptual thing. I think we are close to satisfying the points above too.

I will make a series of changes to the PR shortly with the updating commits referencing comments here. If those are accepted, I'll then close comments above which will neaten the PR and perhaps make it possible to pass.

[pbuttigieg]

I really think we should set up a Semantic Harmonization Cluster for geological features. It helped us work through many of the same convoluted issues for the cryosphere (@rduerr). Happy to bring in the ENVO semantic analysis approach, which helped unpack things.

[dr-shorthair]

The deep-thinkers in geological ontologisation (Steve Richard, Boyan Brodaric, Mark Jessell, others) are mostly working in the Loop3D project, the results of which are unfortunately not open, yet (I think about 12 months overdue now). This is frustrating, but I think it would be risky to start another process in parallel.