CZO variables vocabulary and mapping to higher-level concepts

[emiliom]

@izaslavsky and @aufdenkampe: I had some loose conversations about this with @horsburgh in late November, but I'm creating an issue out of this topic to give it more visibility and a platform for further discussions.

A discussion on categories/types to facilitate faceted searching would be very helpful for http://viz.criticalzone.org, in the near term. Variables (or series) and variable names in particular, as reflected in the CZOCentral WOF endpoints, are too unwieldy at this time to be used as-is in faceted searching. Options for higher-level concepts include:

• CUAHSI Concepts Map. Currenty browseable via the Startree applet at http://water.sdsc.edu/cuahsi_ontology/index.html and http://central.criticalzone.org/startree.aspx. Its applicability in the CZO context seems somewhat limited, but it's a concrete place to start and it's also what HydroDesktop uses.
• New ODM2-based development. Long-term solution, but it would require non-trivial work with an unclear time frame.

I'd be very interested in hearing a recap of the CZO vocabulary work done up to date, and status (eg, current adoption in display files and WOF services, vs decision to stop recommending it in favor of the CUAHSI vocabularies).

Potential linkages to IGSN vocabularies would be helpful, too.

Discussing this at one of our team conference calls would be ideal, but links to existing, relevant documents would be a good starting point too. There's the CZO Shared Vocabulary Registry, but that doesn't provide much background or status.

[emiliom]

BTW, I created this within the CZOData/CZOWOFWaterML1 repo b/c I'm coming at it from a WOF harvesting perspective, and b/c there currently isn't a more relevant github repo in CZOData.

[horsburgh]

@emiliom - The CZO Shared vocabulary registry that you provided the link to should be deprecated. We made the decision some time ago to move as many of those terms as possible over to the CUAHSI system and just use the CUAHSI system (rather than having two systems that were essentially the same, but with slightly different terms in some cases). I did the work to move those terms to the CUAHSI system with the following caveats:

1. DataLevel is a CV in the CZO system, but not in the CUAHSI system. It was originally, but we got feedback that it shouldn't be from some of the CZO people. Odd that it ended up being a CV again in the CZO system.
2. The CZO system had a Method CV which only had one term ("unknown"). I'm assuming it wasn't being used. There is not an equivalent CV in the CUAHSI system.
3. The CZO system had a QualifierCode CV. This is not a CV in the CUAHSI system. We decided a long time ago that we couldn't pre-suppose what qualifiers people would want to use.4.

The paper that describes the CUASI system is at: http://dx.doi.org/10.1016/j.envsoft.2013.10.012

Agreed that we should discuss a plan forward very soon.

[emiliom]

Thanks, @horsburgh! Very helpful.

If DataLevel and QualifierCode were part of the CZO CV's, that means they're potentially being used in display files, and presumably (somehow, maybe as notes?) in the CZOCentral WOF services. So we need to keep them within our sights. Is the message communicated to the CZO's that these 2 CV's are no longer used and available, or are they a remaining, available add-on to the CUAHSI CV's? BTW, I don't think these two CV's are important for the CZO Data Visualization Portal, specially compared to more central ones like variable names and site types.