Publishing trait data for plants - is it possible? Species-level traits from literature & traits measured in the lab

Hi everyone!

I have a question about the possibility of publishing the species trait data on GBIF. With my colleagues, we have a project on establishing the Ukrainian Database of Plant Traits. Firstly, we collected basic species-level traits from literature (e.g. life span, life form, plant height and phenology published in floras - for all species of Ukrainian flora). Then we also measured other traits in the lab for the samples collected in the field (e.g. seed mass, seed shape, SLA - not for all species, but for certain previously defined). Now we are thinking about publishing these data on GBIF.

First of all, I was really surprised that I haven’t found good examples of such data published, especially regarding the first part of our project - descriptive traits for the species level. So I started to ask myself - is it possible to publish such data with Darwin Core in GBIF and how to do it?

I can imagine our data published as two datasets:

  1. Checklist of species with annotations or supplementary data for life span, life form, plant height and other traits. But I do not really have an idea how to structure all these supplementary data. And I do not see appropriate terms in Darwin Core, e.g. for the flowering time which we have for all species indicated per each month of the year, or even for life forms and life span. Or I just cannot find them?

  2. Sampling event dataset with measurements of traits for certain species in certain localities (seed mass, seed shape, SLA).

But of course, the main question is - are such data needed and appropriate in GBIF at all? If people are not really publishing them and there are no terms to make these data standardized, it is a bit confusing.

I would be grateful for any feedback and advice on this matter!

Its very good data and the process of collecting every bit if the information is going to be laborius. I feel this is the future requirement for the scientists to go ahead and analyze the ecosystem or a landscape in a meaningfull way in the climate change scenario.

This subject is beyond the level of collecting occcurrence points. Future ecologist will hopefullly have a open mindset for sharing such trait data as it will be very essential in the days to come.

I feel that incase you do not find much of info for getting the data into darwin core format. You can initiate it, and make a format. Connect with the mentor’s in Gbif, who will guide you on this matter.

Wishing you good luck.

Sangeetha

But let me ask, why we cannot use MeasurementOrFact extension linked to the Taxon Core, as it seems to be possible according to List of Darwin Core terms - Darwin Core? In such a way, you @driadash can publish two dataset, one for event+(occurrence+MoF), and another one as taxon+Mof. I guess this might look cumbersome, but works. @DagEndresen what do you think on that point?

@Sangeetha @oleh.prylutskyi thank you for your comments!

I wonder if @DagEndresen and @rukaya could also comment on these data formats and the possibility of linking the MeasurementOrFact extension (categorical traits !) with a checklist. But anyway, this solution does not look very elegant and probably explains why I did not find any similar information in the existing datasets on GBIF.

@driadash for the generic species traits (e.g. life span, life form, plant height and phenology), you could publish a checklist with a Taxon Description extension (http://rs.gbif.org/extension/gbif/1.0/description.xml) like in this dataset: Validated red lists of Flanders, Belgium. The information will be ingested by GBIF like here: Lutra lutra and aggregated on the corresponding taxon page like here: Lutra lutra (Linnaeus, 1758) (scroll down to the DESCRIPTION category).

For the measured traits, you can use a MeasurementOrFact extension for occurrences.

Hi all! I wasn’t actually aware of the Taxon Description extension, but that looks nice and like it will fit well with what you want to do. We would just have to install it on your IPT :slight_smile:

Then for the other dataset you can use MoF extension with event + occurrence as @mgrosjean and @oleh.prylutskyi have suggested.

I am wondering whether it is not more effective to publish the trait measurements in e.g. TRY TRY Plant Trait Database or the Open Traits network https://opentraits.org/ and link the trait data to the checklist data at GBIF and vice versa?

1 Like

@nl-bif thank you! we are aware of TRY, but we decided to check if there are some options to publish these data on GBIF. At the moment, we decided to publish a data paper with our database first, and then contribute data to relevant networks.