Is it possible to have XMP metadata indexed? I am working on a project to organize 50,000 wildlife photos taken with camera traps. They have metadata stored in the images using the Darwin Core XMP standard ( http://rs.tdwg.org/dwc/ ). If I understand correctly, currently exiv2extractor is used to index basic exif metadata, but not any XMP metadata that may be stored in the same images.

What would be needed to easily make this extra metadata available via desktop search?

Which version of KDE are you using? Different versions of KDE use different methods to extract their metadata - one uses Strigi, while the other uses it's own plugins.

At the moment I am not actually running KDE, I need to get my hands on a extra computer, my current one is running out of capacity

I did some research and it looked to me that exiv2extractor only extracts exif metadata, not XMP. Am I mistaken in that?

I have a sample image that someone with a bit of time might be able to test: http://www.vcn.bc.ca/~apater/DwC-SampleImage.jpg

Gwenview at least is capable of displaying EXIF, IPTC and XMP metadata, as I suspect will be Digikam as well.
In terms of how much metadata Dolphin will be able to display however.

I missed where gwenview shows this metadata. DigiKam shows other XMP metadata, but not DwC. This will change in version 4.0, according to with Gilles Caulier.

In any case, I thought it was nepomuk that took care of indexing metadata, what I don't know is if it currently indexes this particular variant of XMP metadata, DwC. And if it does, will it then be visible from Dolphin?

If Nepomuk indexes the data, then it should be visible in Dolphin.
In terms of making the metadata visible in Gwenview, you need to select the fields you are interested in via "More.." in the metadata sidebar.

Thanks for your help, bcooksley. I finally got KDE running in a virtualbox, so I can answer some of questions.

Gwenview can see and display DwC XMP metadata, although in it's raw XMP format.
DigiKam cannot, but will in the next release. ( https://bugs.kde.org/show_bug.cgi?id=328018 )
Nepomuk does not currently index DwC XMP metadata, at least not according to nepomuksearch.

All of these use the exiv2 library, if I am not mistaken. Full DwC support should be upcoming in that library.
http://dev.exiv2.org/issues/937

So, would a seperate DwC datafeed for nepomuk still be needed?

It will depend on how the Nepomuk indexer utilises Exiv2 to extract file metadata - if it only extracts certain fields then it will need adjusting to show the XMP metadata.