New term proposal: site

[cmungall]

Checklist

• I have reviewed the existing MIxS checklists and extensions that the term does not exist.

Name

sample site

Title

sample_site

Description

A standardized identifier that references an environmental sampling or monitoring site. Examples could be NEON or LTER sites, or stable identifiers for individual plots that are the sites for multiple studies.

A non-exhaustive list of potential identifier systems that could be used:

• https://deims.org, e.g. NEON ABBY site: https://deims.org/0e39569d-ff81-4b78-9c34-b07309504d80
• GAZ (for higher level locations)

Note that DEIMS has extensive well-defined JSON metadata for each site e.g.
https://deims.org/api/sites/0e39569d-ff81-4b78-9c34-b07309504d80

Note that where bounding boxes are provided it should be possible to do this from a lookup

Short description of pattern or range.

CURIE

Pattern or range syntax

CURIE

Example

deims:0e39569d-ff81-4b78-9c34-b07309504d80

Unit

No response

Requirement

No response

Multivalued

No

Extensions(s)

all environmental

[cmungall]

bioregistry request for ID prefix for DEIMS: biopragmatics/bioregistry#1381

[timalamenciak]

This is a good approach and solves a challenge we've run into with the research site property on Wikidata. Many ecological research sites in Canada are so remote that they would not have named locations. The approach here seems to be registering the sites with DEIMS then linking to the identifier, which is good. It may also be worth simply including the decimal degrees centroid of the research site as a permissible value.

I wonder if this should be multivalued? One study may involve multiple distinct research sites, and thus multiple DEIMS ids.

[only1chunts]

@timalamenciak the geo location decimal degrees should be included as values in the term "geographic location (latitude and longitude)" (aka "lat_lon").
The expected usage of MIxS is at the sample level rather than the project level, so we wouldn't usually expect multiple values for something like a sampling site, however, in order to make the term as useful as possible to everyone (even those wishing to apply it to a project level metadata object) I dont see any reason why it couldn't be multi-valued.