iobis / env-data

ENV-DATA related issues and documentation
2 stars 0 forks source link

Absence environmental/habitat data #24

Open rubenpp7 opened 3 years ago

rubenpp7 commented 3 years ago

Hi,

I have received a dataset with many MoF records documenting the absence of a certain habitat feature such as the following:

image

All these "measurementTypes" = 0 for all the records, meaning that these features were searched for but not present for any event.

wardappeltans commented 3 years ago

Hi Ruben, Are there biological/biodiversity records linked to those habitat measurements?

If yes than it is relevant for OBIS, the question on which vocab can be moved to https://github.com/nvs-vocabs/OBISVocabs

rubenpp7 commented 3 years ago

Hi Ward,

These MoFs are linked to a specific eventID that apart from those habitat MoFs is also linked to some occurrences, or in human-readable language (having that all these habitat features = absent), it would be like saying:

"We found NO "Tree Branches" using the gear that we used to find this Nephrops norvegicus and at that specific location and time"

So I guess yes, these habitat MoFs are linked to the occurrences, I'll be happy to move my question to OBISVocabs

albenson-usgs commented 3 years ago

To me, the absence of structural/habitat categories ("Artificial Woods", "1000 to 2000 depth band", "Bathyal Biological Zone") seems ineffective to include. The list of what's not there could be endless and probably better ways to determine these things than to have them listed in an emof table- who would use it? To be clear the presence of these things does seem to fit in emof, just not the absence.

pieterprovoost commented 3 years ago

While this seems a bit superfluous at first sight, I do believe that there can valuable information in there. I imagine there's a publication that goes with this dataset that uses these habitat features to do multivariate statistics. Including the same features in the dataset would improve the reproducibility of the research. But I would at least condense the biological zone measurement types into something like this:

measurementType measurementValue
biological zone supralittoral
biological zone upper infralittoral