Notes from the TDWG 2016 Meeting in Santa Clara, Costa Rica

These are notes on the Darwin Core session held on Wednesday December 7, 2016 at the TDWG 2016 Annual Conference, held at the Instituto Tecnologica de Costa Rica (TEC) and the Instituto Nacional de Biodiversidad (INBio) in Santa Clara de San Carlos, Alajuela. A video recording of the workshop is available at this link. The presentation given at the workshop is here

There was a discussion of the nature of an occurrence. With regards to the terms in the proposal, as one person put it “What is this a property of?”. The general consensus was that an occurrence is not such a clean concept as a observation in space and time. It encompasses a range of types of observations, including checklists that include observations made over a spatial extent and a period of time. An occurence may be observations of a single individual, but also of groups of individuals or a population. This means that some DwC terms have different meanings and usages depending on the context of the observation.

The issue of the degree of establishment came up several time. Users need help determining if an observation should be used in species distribution modelling, particularly if that observation is from a living collection, such as a zoo or a garden. This information is sometimes indicated in dwc:basisOfRecord and at other times in dwc:establishmentMeans. By restricting the usage of dwc:establishmentMeans to introduction pathway it limits the ability of Darwin Core for expressing the degree of establishment of an alien organism.

The framework of Blackburn et al. (2011) was mentioned as a possible solution, though this is currently not part of the proposal. Blackburn et al. (2011) recommends a hierarchical framework to describe how well establish an organism is. Another proposal was to have a flag "for distribution" in Darwin Core that indicates to the user that the observation can be used for species distribution modelling.

The changes to dwc:establishmentMeans were controversial, because although the definition has not changed in the proposal, the usage has changed considerably. As Darwin Core currently doesn’t have versioning it will be confusing to use the same identifier for what is, in effect, a different meaning for dwc:establishmentMeans. The recommendation from the workshop was that the proposal gets reviewed at by the TDWG Technical Architecture Group, particular with respect to whether dwc:establishmentMeans should be deprecated in favour of a dwc:pathway term.

The issue of completing dwc:establishmentMeans for native plants was raised. Leaving a blank is unsatisfactory. It was proposed, without decent, to add “natural dispersal” to the vocabulary. It was also suggested that the term "unknown" is added to each of the vocabularies.

Several attendees suggested that it would be useful to have examples of how to use the new dwc:establishmentMeans vocabulary in the context of real data. Also, there need to be guidelines on how to use the vocabulary in the context of point observations and checklists.

Other points that were raised

dwc:organismScope could be used more widely to express the scope of an observation, particularly when it referred to checklists that cover an areas and/or time span.

References

Blackburn, T. M., Pyšek, P., Bacher, S., Carlton, J. T., Duncan, R. P., Jarošík, V., ... & Richardson, D. M. (2011). A proposed unified framework for biological invasions. Trends in ecology & evolution, 26(7), 333-339. http://doi.org/10.1016/j.tree.2011.03.023

Contributors to the discussion

if I have missed anyone out feel free to add yourselves

• Aaron Wilton
• Annie Simpson
• Arthur Chapman
• Dimi Brosens
• Joel Sachs
• John Wieczorek
• Lee Belbin
• Niels Klazenga
• Paula Zermoglio
• Peter Desmet
• Pier Luigi Buttigieg
• Ramona Walls
• Rob Guralnick
• Steve Baskauf