Missing landscape classifications for SLE project

[ColinHDev]

Description of the issue

For the SLE project, OEO concepts for the following landscape classifications are needed:

• Fauna Flora Habitat (Engl.: Fauna Flora Habitat)
• Vogelschutzgebiete (Engl.: Bird sanctuaries)
• Biosphärenreservate (Engl.: Biosphere reserves)
• Landschaftsschutzgebiete (Engl.: Protected landscape areas)
• Wälder (Engl.: Forests)
• Wasserschutzgebiete (Engl.: Water protection areas)
• Überschwemmungsgebiete (Engl.: Floodplains)
• Ackerflächen mit hoher Bodenqualität (Engl.: Arable land with high soil quality)
• Ackerflächen mit geringer Bodenqualität (Engl.: Arable land with poor soil quality)

Currently, the OEO doesn't contain terms that match these concepts well.

Ideas of solution

In an earlier discussion, @stap-m proposed importing required terms from the ENVO ontology. This works in some cases, e.g. biosphere reserve (ENVO:00000376) or forested area (ENVO:00000111) but not in all of them.
The ENVO uses the landscape definitions of the International Union for Conservation of Nature (IUCN). In contrast, some of the above landscapes are defined by German laws or EU directives, e.g. the Fauna Flora Habitat or the Bird sanctuaries.

Since some of these concepts describe landscapes with similar rules, @stap-m proposed implementing them as a sector division to differentiate between German / EU laws and IUCN classifications.

Workflow checklist

• I discussed the issue with someone else than me before working on a solution
• I already read the latest version of the workflow for this repository
• The goal of this ontology is clear to me

I am aware that

• every entry in the ontology should have a definition
• classes should arise from concepts rather than from words

[stap-m]

We have region of relevance which has several subclasses which are types of regions defined by the German ROG, e.g. http://openenergy-platform.org/ontology/oeo/OEO_00360030. We should consider these in this context.

[stap-m]

@ColinHDev I drafted a concept for the region classifications, exemplyfied by the already existing region type conditionally reserved region (blue is not yet there, green could be considered for reuse). Please try a refinement (or draft a different approach). We'd need proposals for definitions and classifications, and lastly axioms.

[ColinHDev]

I hope something like this is what you had in mind 😅

• I wasn't sure whether region of relevance was the best place to put these reservoirs especially since there was already a protected area class in the ontology whose definition fits those reserves quite well. As an example, I added the classes for the Fauna Flora Habitat (Engl.: Fauna Flora Habitat) and the Vogelschutzgebiete (Engl.: Bird sanctuaries) that both are part of the "Natura 2000" area network.
• is defined by seems to be a good fit.
• plan specification seems to be fitting especially as it already has policy as a subclass.
• The word area is rarely used in the ontology, mostly spatial is used in its place. So I would keep it consistent and name it spatial legislation
• Maybe having a general legislation class can be useful for future law references in the ontology. But there wouldn't be a need for it, so it could also just be added then.

Regarding the definitions of some of the classes:

• legislation based on the definition in the Addiction Ontology: A legislation is a plan specification that is produced by a government and enables, prohibits or bans activities, or establishes rights or responsibilities.
• spatial legislation: A spatial legislation is a legislation affecting or defining a spatial region.
• ROG: The ROG is a spatial legislation providing the legal framework for spatial planning in Germany. with spatial planning law and Raumordnungsgesetz as alternative labels.

[stap-m]

• legislation based on the definition in the Addiction Ontology: A legislation is a plan specification that is produced by a government and enables, prohibits or bans activities, or establishes rights or responsibilities.
• spatial legislation: A spatial legislation is a legislation affecting or defining a spatial region.

Some thoughts:

• Good idea. ADDICTO classifies legislation as directive information entity. Why would you propose plan specification instead?
• If we added legislation we need to differentiate from the existing policy.
• I just saw, that there is spatial planning policy in OEO. It is focused on Germany, though. How would it relate to spatial legislation?

[stap-m]

Would you see the distinct legislations (ROG, BNatSchG, ...) on the individual level?

[stap-m]

• I wasn't sure whether region of relevance was the best place to put these reservoirs especially since there was already a protected area class in the ontology whose definition fits those reserves quite well. As an example, I added the classes for the Fauna Flora Habitat (Engl.: Fauna Flora Habitat) and the Vogelschutzgebiete (Engl.: Bird sanctuaries) that both are part of the "Natura 2000" area network.

Maybe we should discuss the relation between protected area and region of relevance, too...

[ColinHDev]

ADDICTO classifies legislation as directive information entity. Why would you propose plan specification instead?

The difference between legislation and policy isn't big, so having both of them as subclasses of the same parent class seemed natural. A plan specification is defined as A directive information entity with action specifications and objective specifications as parts, and that may be concretized as a realizable entity that, if realized, is realized in a process in which the bearer tries to achieve the objectives by taking the actions specified.:

• Similar to the policy, legislation could have the target specification as a part, since both policies and laws are implemented to change something to achieve a certain future state.
• But I am unsure what the action specification of a legislation could be.

If we added legislation we need to differentiate from the existing policy.

Yes. When drafting my previous suggestion, I also came across policy and felt inclined to use it as the base class for spatial legislation, since it seemed to also include types of legislation.
However when reading through the issue where it was proposed and its definition discussed, the Wikipedia page on policies was mentioned as one of the inspirations for it. There it is mentioned:

The term may apply to government, public sector organizations and groups, as well as individuals, Presidential executive orders, corporate privacy policies, and parliamentary rules of order are all examples of policy. Policy differs from rules or law. While the law can compel or prohibit behaviors (e.g. a law requiring the payment of taxes on income), policy merely guides actions toward those that are most likely to achieve the desired outcome.

I just saw, that there is spatial planning policy in OEO. It is focused on Germany, though. How would it relate to spatial legislation?

Based on Wikipedia's definition, policies aren't laws, so a spatial planning policy isn't a spatial legislation.

Would you see the distinct legislations (ROG, BNatSchG, ...) on the individual level?

I feel inclined to say yes since it makes more sense to have them as individuals, as they aren't really concepts on their own but only instances of spatial legislation.
But conditionally reserved region role and "Nature 2000" role could no longer be defined by ROG / BNatSchG "Natura 2000" since I don't think it is possible to define properties between classes and individuals. We could define both roles to be subclasses of 'is defined by' some 'spatial legislation' but we would still lose the link between the role and the exact instance of legislation that defines it.

Maybe we should discuss the relation between protected area and region of relevance, too...

I looked at the issue that introduced protected area and there hasn't really been a discussion about its introduction. The IUCN definition was cited as base for the definition and it doesn't necessarily define it as being two-dimensional, but only being a "clearly defined geographical space". Same as the definition ENVO uses for its protected area class.
I can't really wrap my head around the concept of region of relevance and what it represents. But based on the assumption it means "A region that is in some way relevant for the OEO", protected area could also be made a subclass of it?

This would be the current version:

[madbkr]

@stap-m suggested it would be interesting to mention this here:
For the tasks module it would be interesting to add these two concepts into the OEO:

exclusion zone is a two-dimensional spatial region in which specific activities are forbidden.
buffer zone is a two-dimensional spatial region meant to keep opposing parties spatially apart.

This relates to some of the concepts mentioned here.

[ColinHDev]

I have a few questions:

• Although I couldn't find an implementation of an exclusion zone in the OLS, ENVO defines a buffer zone. It uses the definition: A geographical region that serves the purpose of keeping two or more other areas (often, but not necessarily, countries) distant from one another, for whatever reason.
  • Why do our parties have to be opposing? If I think about the minimum distance wind turbines have to be away from villages, I could think of it as a buffer zone between wind turbines and towns. But in that case, I wouldn't count both "parties" as opposing one another, just having different interests. ENVO's definition makes it more universal and should be considered for our one.
• Why does it have to be a two-dimensional spatial region?
  • When thinking about the role of protected area as a subclass of two-dimensional spatial region, I found its concept hard to make sense of, since in most cases there are always some limitations.
    • A certain nature reserve is not spanning across outer space.
    • If a country could be considered an exclusion zone because it prohibits certain activities by law, it might not apply e.g. when flying above it. The US has some regulations regarding alcohol consumption, but when flying over it, these laws don't generally apply. (Source: https://simpleflying.com/which-countrys-laws-are-enforced-during-international-flights/)
  • ENVO also doesn't classify its buffer zone as a two-dimensional spatial region. It seems it doesn't even have these dimension-specific subclasses of spatial region.

[ColinHDev]

Would you see the distinct legislations (ROG, BNatSchG, ...) on the individual level?

I feel inclined to say yes since it makes more sense to have them as individuals, as they aren't really concepts on their own but only instances of spatial legislation. But conditionally reserved region role and "Nature 2000" role could no longer be defined by ROG / BNatSchG "Natura 2000" since I don't think it is possible to define properties between classes and individuals. We could define both roles to be subclasses of 'is defined by' some 'spatial legislation' but we would still lose the link between the role and the exact instance of legislation that defines it.

Since just moving the legislation classes to the individual level does not work that well, I was tasked to look into whether it makes sense to move all the roles and regions that might interact with these legislations to the individual level as well, for there is the possibility again to have e.g. conditionally reserved region role directly defined by ROG instead of defined by some spatial legislation.
I think it doesn't work when considering the addition of subclasses to protected area as shown in the above concepts. In the case of the EU bird sanctuaries and fauna flora habitat, both of them define the "Natura 2000" network of protected areas in the EU. In Germany, the "Natura 2000" is implemented by the corresponding section in the BNatSchG. So the Natura 2000 and BNatSchG should be somehow connected in the OEO, so either both of them are classes or individuals, so we can define properties between them. But Natura 2000 has to include both bird sanctuaries and fauna flora habitat and since there isn't a way for individuals to have "sub-individuals", I think using classes is the appropriate way with the above concept - Unless someone has a different idea :)

[stap-m]

• Although I couldn't find an implementation of an exclusion zone in the OLS, ENVO defines a buffer zone. It uses the definition: A geographical region that serves the purpose of keeping two or more other areas (often, but not necessarily, countries) distant from one another, for whatever reason.
• Why do our parties have to be opposing? If I think about the minimum distance wind turbines have to be away from villages, I could think of it as a buffer zone between wind turbines and towns.

I agree to your understanding of buffer zone in the context of oeo, i.e. the zone between wind turbines and villages. I think, intput from modellers would be helpful for the distiction between exclusion zone and buffer zone @Ludee @areleu and others.

• Why does it have to be a two-dimensional spatial region?

I'd see these zones rather as roles of spatial regions.

[madbkr]

@ColinHDev
I agree with your opinion on buffer zones. I was not very happy with the wording of my definition anyway.

I thought is should be a two-dimensional spatial reason because we have protected area in the OEO as a two-dimensional spatial region. I thought these would be sort of comparable but I am completely fine with changing it.

[ColinHDev]

As discussed with @stap-m, I drafted a new concept that relies more on individuals and moves the current ROG-defined region classes to the individual level.

Black represents existing classes and relations, blue represents new classes and relations and purple represents individuals and relations between them.

The implementation of this can be found at #1991.

It could still be discussed, whether protected area should be moved from two-dimensional spatial region to region of relevance.
How laws are implemented could also be discussed. The current approach defines individuals like ROG or BNatSchG as spatial legislations as they both define certain areas. To be precise, the BNatSchG in its entirety might not be classifiable as spatial legislation but maybe only §31 ff. as they define the Natura 2000 network. But I am unsure whether it is preferable to have individuals like BNatSchG "Natura 2000" or BNatSchG § 31 ff. only representing part of such legislation.

[ColinHDev]

As tasked, PR #2006 aims to implement the remaining landscape classifications that are currently missing in PR #1991.

However, it still misses Arable land with poor soil quality and Arable land with high soil quality as I couldn't come up with a good implementation in time.