Deployment date range issues

[albenson-usgs]

What should deployment date range be populated with if you don't know when the deployment ends? Should it have a date range? If so, what do you put as the end date? What happens if the animal is not recaptured?

[jdpye]

In my world, things like estimated battery life are necessarily fuzzy and shouldn't be taken as hard end-dates. Other tags are hard-programmed to be shut off at date X.

Further to not knowing about an end date at all, what should we do if we have a general idea but not a specific one? Especially looking for an option that doesn't upset ISO standards.

[albenson-usgs]

Once a decision is made on this, the wiki needs to be updated with that information in the Data Guidelines section under eventDate.

[peggynewman]

It seems sensible that the guidelines point out the way in which it was determined that the deployment ended, or even if it did at all.

[sarahcd]

If we want/need to include an end date for a track, realistically that date will commonly represent many different things:

• the removal of the tag by a human,
• a scheduled automated release,
• the last available data point,
• the end of the time the data can be reliably associated with the live animal,
• the time that the animal is captured or confined (even if the tag remains attached),
• the end of a segment of data that is meaningful to the data owner (e.g. the end of a foraging trip, migration, or experimental run, like a homing flight)

When dealing with bio-logging data, typically a specific end datetime needs to be chosen to define which data records to include/exclude even if, say, the likely date of death remains fuzzy.

I agree we need a guideline and examples for this. One idea is to recommend creating an event and occurrence for the end of each track with

• an ISO compliant timestamp
• describe the type of deployment end in occurrenceRemarks (Occurrence table) and eventRemarks (Event table)

We could also propose a controlled list. In Movebank we have deployment-end-type with options captured, dead, equipment failure, fall off, released, removal, unknown, described at http://vocab.nerc.ac.uk/collection/MVB/current/MVB000084/ I don't think these options are complete, so if we came up with an improved list I could update our vocabulary too. In this case we'd want a place for a controlled value and a place for other notes.

[jdpye]

Love the idea of a controlled list for these, I'll see if I have anything that falls outside of the stated examples, to help extend the vocab.