Protected area names - where to add?

[nina14213]

Based on my search of the Darwin Core (DwC) GitHub issues and documentation, there is currently no dedicated field in Darwin Core for recording protected area information, which creates challenges for biodiversity data publishers who need to document whether specimens or observations were collected within protected areas.

We have tried several imperfect approaches to encode protected area information in our datasets:

• dwc:locationRemarks - Comments field where protected area information is noted, e.g., "Białowieski Park Narodowy" or "Cisy Staropolskie imienia Leona Wyczółkowskiego" (names according to Protected Planet: https://www.protectedplanet.net/country/POL)
• dwc:verbatimLocality - Contains this information but in a less normalized way, e.g., "Puszcza Białowieska. Białowieski P.N. Oddz. 316-342. Gniazdo mysikrólika (Regulus regulus)" (English: "Białowieża Forest. Białowieża National Park, Section 316-342. Goldcrest nest (Regulus regulus)")

The main problems with these approaches are:

• Lack of standardization - No consensus on which field to use means data is inconsistent across datasets
• Mixed information - Protected area names are combined with other locality details, making extraction and querying difficult
• No controlled vocabulary - Records are difficult to filter as they are present in many columns and their names vary (abbreviated vs. full names, different languages)

I'd like to discuss adding dedicated terms that would significantly improve data quality and usability:

• dwc:protectedAreaName - for standardized protected area names,
• dwc:protectedAreaID - for linking to i.e. World Database on Protected Areas (WDPA) identifiers, which provide globally unique IDs for each protected area.

This standardization would benefit both scientists and policymakers by enabling easy filtering of taxa within specific protected areas, allowing analysis of biodiversity patterns across protected area networks, and assessing their conservation needs.

[ben-norton]

@nina14213 We (the Mineralogy Extension TG) have proposed a new Darwin Core term, namedPlace. tdwg/dwc#795 Does this satisfy your use case?
We are also leveraging localityID for linking to external resources for geologically significant locations. An example is the Red Cloud Mine on Mindat: https://www.mindat.org/loc-3348.html
These solutions have been user-tested in the geology community and yielded successful results.

[debpaul]

Greetings @nina14213 TIL there exists the WDPA with identifiers, thank you! I'll be mulling over how we can encourage those with projects using TaxonWorks, to use these.

IF you find that the namedPlace suggestion would work, then @ben-norton wouldn't we want @namedPlaceID to go with? (I realize that's tricky b/c of the scope of what kind of identifiers would be wanted / used could likely be unweildy and overload a term).

@tucotuco the "namedPlace" discussion is not a new one (I recall the same from many years ago). Your thoughts? Or can you point us to a past discussion?

I also see INSPIRE Geographical Names has a NamedPlace concept. See Technical Doc.

[baskaufs]

FYI, the existing Darwin Core term dwciri:inDescribedPlace is intended to link a location to the lowest level in some standardized hierarchy of places. However, it is expected to have an IRI value from some place like GeoNames or the Getty Thesaurus of Geographic Names.

[ben-norton]

The namedPlace term is specifically designed to capture locations that are not in a traditional hierarchy but are critically important to earth sciences (some minerals are actually named after the mine where they were originally discovered, such as Franklinite).
@debpaul A namedPlaceID would be great, but I don't think it's nearly as important as the namedPlace field. Our user testing indicated that locationID meets the need for storing a persistent URL to a resource that describes the namedPlace (although the resource is an HTML page, not a machine-actionable resource, but that's a separate discussion).

[debpaul]

FYI, the existing Darwin Core term dwciri:inDescribedPlace is intended to link a location to the lowest level in some standardized hierarchy of places. However, it is expected to have an IRI value from some place like GeoNames or the Getty Thesaurus of Geographic Names.

@baskaufs would the ID resource @nina14213 offers (WDPA) work for dwciri:inDescribedPlace?

[baskaufs]

I am not familiar with WDPA. I did a brief Google search and the documentation says "he WDPA ID is the globally unique identifier for each protected area in the WDPA. It is specific to a protected area in a specified geographical space, and does not change over time unless the designation for that protected area changes. " However, it does not say whether the globally unique identifier is an IRI or something else like a UUID. I tried to find an example, like https://www.protectedplanet.net/2013 but it doesn't expose the GUID. So I'm not sure. @nina14213 may know more. If the GUID is an IRI, one could use dwciri:inDescribedPlace. If it is not, then one would need something else.

[debpaul]

@baskaufs I'm looking. Next bit I see is
"WDPA Site Identifiers (WDPAID, WDPA_PID), a persistent identifier for each protected area, provided by UNEP-WCMC. Agency partners are encouraged to track WDPA Site Identifier values in source datasets to improve the efficiency and accuracy of PAD-US and WDPA updates."

From https://www.usgs.gov/data/protected-areas-database-united-states-pad-us-30-world-database-protected-areas-wdpa

The screenshots below from https://app.ibat-alliance.org/pdf/wdpa_manual.pdf

[MasoncBSU]

Hi, This is Mason Croft the PAD-US Technical Specialist. I manage PAD-US and the PAD-US Submissions to the WDPA.

The short answer is no, WDPA codes are not an actual Identifier. The WDPA fields are just regular integer fields.

PAD-US is an aggregation of voluntarily provided already existing datasets. Most of the source data doesn't contain WDPA IDs or the other Conservation Measure fields (GAP Status Code, IUCN Category, etc.) that PAD-US contains. WDPA codes are assigned using scripts and manual review with each PAD-US subset WDPA submission. I am working on a series of tools to make this process smoother and less painful, but it is not fully automated yet.

[debpaul]

Thanks @MasoncBSU that's really helpful to understand. In essence that WDPA ID is not a globally unique ID, it seems it might? be unique inside your world, but maybe not permanent? Good to know for planning standards needs and expectations.

[MasoncBSU]

Correct, it is not a global unique database ID.

I think WCMC (the creators of the WDPA) faces many of the same issues globally that we face with PAD-US. The WDPA isn't a centralized enterprise database. It is individually created and curated by various countries and entities around the world each managing their own parts. With each one submitting their own piece as a shapefile or file geodatabase, then WCMC brings it all together.

The codes are meant to persist and should stay the same between updates, unless significant changes happen to the management of the Protected Area, but they are not actually assigned by the database.

The PAD-US portion is a subset, because not all of the areas that we track in PAD-US meet the IUCN definition of a protected area, which is necessary to be included in the WDPA. PAD-US uses a broader definition of "protection" meaning areas with a natural, biological, historical, cultural, or recreational purpose. The IUCN definition is narrower than that.

[tucotuco]

It is true that there is no dedicated field for the name of a protected area. The common wisdom for existing solutions favors putting the information in the higherGeography term. That terms includes an example with the protected area "Parque Nacional Nahuel Huapi". But that isn't the same as a dedicated field.

There are a couple of discussions related to this issue, this one about higherGeography and this one about a proposed new term, namedPlace.