organismQuantity values not interpreted - excluded

[damianooldoni]

We published two occurrence datasets containing abundancy values in field organismQuantity. In both datasets these values are excluded (not interpreted). For more info, see riparias/anb-plants-occurrences#5 and riparias/vmm-macrophytes-occurrences#27

Any idea? This issue could be similar to #2759 and #4020.

Thanks in advance for taking care of it and congrats for the great work you all are doing!

[ManonGros]

As far as I know, our system expects a decimal value for this field. I don't think it can handle characters.

[mdoering]

Even though organismQuantity goes together with the unit organismQuantityType it is clearly not purely numerical. If we indeed interpret it that way we should change that, DwC gives valid examples that are not numerical.

[ManonGros]

Right now users can search based on organism quantity and quantity type (less than 2 square meters, more than 1 sequence read, etc.)
I don't think this is something that could be done if the system also took into account non-numerical values. Could it?

Isn't it useful to be able to filter for quantities?

[mdoering]

It clearly is, but there are non numerical values that are also useful to search for, e.g. the given example value r for the Braun Blanquet scale.

[MattBlissett]

I see there are ACFOR and DAFOR scales, we could at least support those: https://en.wikipedia.org/wiki/Abundance_(ecology)

We'd need to maintain the numeric (range) filters, and also allow string filters.

[mdoering]

it would be great if we could split the field into a numerical one and another one where we parse values into known scales depending on the quantityType given - or even by inspecting distinct values. We could default to numerical. But treat known non numerical scales differently.

[damianooldoni]

Thanks @mdoering for letting us know that GBIF is working on this. Any idea when this issue will be tackled in production?

[mdoering]

Thanks @mdoering for letting us know that GBIF is working on this. Any idea when this issue will be tackled in production?

@timrobertson100 @fmendezh ?

[CecSve]

I see there are ACFOR and DAFOR scales, we could at least support those: https://en.wikipedia.org/wiki/Abundance_(ecology)

We'd need to maintain the numeric (range) filters, and also allow string filters.

Would a controlled vocabulary for the strings help in interpretation?