nielsklazenga
commented
9 years ago The Darwin Core documentation at https://code.google.com/p/darwincore/wiki/Event#habitat now recommends recording habitat information using the Environment Ontology (ENVO, http://environmentontology.org/). In addition to:
ENVO has the following concepts that can be used:
- environmental material (http://purl.obolibrary.org/obo/ENVO_00000154)
- environmental feature (http://purl.obolibrary.org/obo/ENVO_00000195) and
- biome (http://purl.obolibrary.org/obo/ENVO_00000570),
The environmental feature concept comes close to the substrate element proposed in issue #14.
It may be possible to deliver the Darwin Core habitat concept (http://rs.tdwg.org/dwc/terms/habitat) as well as the ENVO concepts, so we could deliver the detailed habitat description using the Darwin Core concept and the categorised data using the ENVO concepts. However, I believe it is the intention to eventually include environmentalMaterial, environmentalFeature and biome in Darwin Core.
ben3000
DNAdl8
AaronWilton
The HISPID Habitat group contains the following elements:
Issue #14 furthermore proposes a 'Substrate' field in the sense that I would interpret the word, the current HISPID concept being more the underlying geology.
According to the HISPID 3 documentation Habitat is an aggregated field, that may be split up into the six other elements in herbarium databases. It also says that the translation of collectors' historical descriptions of habitat is prone to error and misinterpretation and that institutions that do store the discrete concept must combine those before delivery, if they want to deliver the Habitat field. That has not been the practice, as almost everyone is delivering Habitat and some herbaria deliver additional habitat field as well.
Under Collection notes in HISPID 3, the point is made much more clearly than under the Habitat group:
Of the seven concepts above, Habitat has been mapped to dwc:habitat, Topography to locationRemarks and Associated Species to associatedTaxa (but see #20), The other concepts cannot be mapped to Darwin Core.
All HISPID Habitat fields, except Aspect, are free text fields. I don't think there is any point in splitting up the Habitat field if the resulting fields are also free text. In fact, splitting up the field would just make it harder for users to find the data, as different herbaria, or even different people at the same herbarium, or the same people in different circumstances, might put the same information in different fields. Just keeping everything in the one field will also avoid the errors and misinterpretations that the HISPID documentation warns us for.
I think the discrete concepts could be useful, but only if they had vocabularies on them – and would be interpreted fields – and would be used in addition to the detailed habitat description in the habitat field. For the habitat information that I record with my bryophyte collections I can currently only use the Habitat field (I make notes about light and moisture etc., in addition to vegetation, topography, substrate etc.). I would happily populate the discrete fields, if I were told the values I could use and if it didn't mean I had to give up on the detail and context in my habitat descriptions.
I actually think that it might be best if our implementation of the dwc:habitat element also would have a vocabulary on it (for example the IUCN Habitat classification, http://rs.gbif.org/vocabulary/iucn/habitat.xml, or a more detailed Australian one, as long as there is some classification we can refer to) and that what we currently deliver under this concept might be best placed somewhere else, perhaps locationRemarks, but I haven't given that too much thought yet (and I do tend to change my mind every now and then), so I am not going to pursue this. I might actually not really pursue any of this – except that all habitat information should be delivered in habitat, which is already in HISPID – but I just put it out there, as it needs to be discussed.