search

tdwg / dwc-for-biologging

Darwin Core recommendations for biologging data
Creative Commons Attribution 4.0 International
13 stars 3 forks source link

Associate acceleration measurements to deployment, not to separate event for each #6

Open peterdesmet opened 5 years ago

peterdesmet commented 5 years ago

@sarahcd, just looked at your "Mahoney-data-DwC-A-test-2". This is the structure I could detect:

event_id events assoc occ assoc event mof assoc occ mof
F53:capture1 1 1 0 0
F53:recovery1 1 1 0 0
IdCoy_P3_1:capture1 1 1 0 0
IdCoy_P3_1:recovery1 1 1 0 0
IdCoy_P3_1:capture2 1 1 0 0
IdCoy_P3_1:recovery2 1 1 0 0
F53:deployment1 1 1361 0 1361 * 4
IDCoy_P3_1:deployment1 1 10809 0 0
IDCoy_P3_1:deployment2 1 15300 0 0
F53:deployment1:xxx 76869 0 76869 * 2 0

I would simplify this to:

event_id events assoc occ assoc event mof assoc occ mof
F53:capture1 1 1 0 0
F53:recovery1 1 1 0 0
IdCoy_P3_1:capture1 1 1 0 0
IdCoy_P3_1:recovery1 1 1 0 0
IdCoy_P3_1:capture2 1 1 0 0
IdCoy_P3_1:recovery2 1 1 0 0
F53:deployment1 1 1361 76869 * 2 1361 * 4
IDCoy_P3_1:deployment1 1 10809 0 0
IDCoy_P3_1:deployment2 1 15300 0 0

All acceleration measurements are done in F53:deployment1. Just like you group all occurrences under that event (with each occ. having a more precise eventDate), I would group all acceleration measurements under that same event (with each meas. having a more precise measurementDeterminedDate). That way, the Event core is uncluttered and gives a quick overview of captures, recoveries and deployments.

Note, since acceleration measurements are measured independently from GPS positions (correct me if untrue), it is not possible to flatten the Events into the occurrences, which is an approach I like, but have to reconsider 😄 .

sarahcd commented 4 years ago

Thanks @peterdesmet! This update will be made in Mahoney use case V3, which I'm working on now. (Although I'll be grateful for a check before we close this.) As you suspect, the acceleration measurements because they were not collected on the same schedule as GPS fixes. This will be pretty common for deployments of multi-sensor tags (and tags with no location sensors) and is one reason I chose this sample dataset. Thanks for pointing out measurementDeterminedDate. I agree it doesn't seem ideal to spread the measurement timestamps across two tables and terms, I don't have a better suggestion, but maybe something to discuss further.

One last related question: Can I use BasisOfRecord in the event or FOM tables? Currently in my example I don't define the measurements without occurrences as MachineObservation.

peterdesmet commented 4 years ago

Can I use BasisOfRecord in the event or FOM tables? Currently in my example I don't define the measurements without occurrences as MachineObservation.

No, you can't. There is no way to differentiate between machine or human measurements in MOF. I think you will have to rely on the definition in measurementTypeID