[tdwg-content] A proposal to improve Darwin Core for invasive species data

[Quentin Groom]

Dear Rich and Chuck,
yes, the use case is critical. To me Rich's observations are correct, but I
find them too strict for the multitude of uses that DwC is applied to. For
example, a while ago I digitised an old Flora (1831), georeferencing all
the locations in the process. The resulting DwC file, as Rich would point
out, associates a taxon with a location. There is no alternative, because
the only dates available are the publication date and biographic date of
the author's activity (
http://www.gbif.org/dataset/bfaa049a-90cd-412d-9660-5380591ba4a5).

Another example is the vast amount of plant surveying that occurs in
Europe. This is done, mostly by amateurs, where they list all the taxa in a
map grid square (e.g. 1 km2; 4km2 or 100km2). These surveys are used as the
basis for distribution atlases. These surveys have more in common with
checklists, they are not observations of organisms, but of taxa in a
location. In this case there is often a date associated with a survey, so
the event is more explicit. Only, in cases of people observing rare taxa or
using smartphone apps do they observe a organism in a location.

In previous conversations I've had with non-European zoologists they have
been surprised by this way of working. However, these only amount to a tiny
fraction of all the observations collected. I imagine in countries outside
Northern Europe there are not enough active observers to cover every 4km2
of a country, so people observe individual organisms and often ignore the
common taxa that are "everywhere".

As Rich puts it, there is a "continuum from Organism to Taxon". I think DwC
has to accommodate this continuum but that might mean that the use cases
need to be made explicit in the definitions of terms.

The assessment of whether something is native to an area can't be reduced
down to locations and dates. Early records are useful, as are fossils and
microfossils, but other data are used to assess native status. Webb (1985)
http://archive.bsbi.org.uk/Wats15p231.pdf suggested eight criteria, but
there are others. Again this is an association between a taxon and a
location that is hard to avoid.

Thanks for taking the time on these proposals!
Regards
Quentin





Dr. Quentin Groom
(Botany and Information Technology)

Botanic Garden Meise
Domein van Bouchout
B-1860 Meise
Belgium

ORCID: 0000-0002-0596-5376 <http://orcid.org/0000-0002-0596-5376>

Landline; +32 (0) 226 009 20 ext. 364
FAX:      +32 (0) 226 009 45

E-mail:     quentin.groom at plantentuinmeise.be
Skype name: qgroom
Website:    www.botanicgarden.be


On 25 June 2016 at 20:54, Chuck Miller <Chuck.Miller at mobot.org> wrote:

> It's the problem of different uses cases again. The use cases need to be
> more explicit. Recording a taxonomic checklist and recording an occurrence
> of an organism are different use cases that happen to share some attributes
> but also differ.  Without clear understanding of the use case, how can
> anyone know what the term being used means precisely.
>
> Are some Darwin Core terms actually unique to particular use cases?
>
> How about a binary term like IsInvasive (True, False or Null) - for
> checklists only?
>
> Regards,
> Chuck
>
>
>
> On Jun 25, 2016, at 8:25 PM, Richard Pyle <deepreef at bishopmuseum.org>
> wrote:
>
> I think we may be talking about different things; or maybe the same thing
> but from fundamentally different perspectives.
>
>
>
> In the context of “checklist-style” dwc:Occurrence instances that
> represent direct associations of a dwc:Taxon instance to a dwc:Location
> instance, then all we really need to do is clarify the definitions of
> dwc:establishmentMeans and dwc:occurenceStatus in how they should be used
> to represent properties relevant to the presence of a Taxon at a Location;
> and also whether a new term is needed to represent properties that don’t
> really fit into either of those terms.  If I understand correctly, this is
> the basis for your proposal; and I agree we at least need clarification of
> dwc:establishmentMeans, and perhaps also a new term.
>
>
>
> My reply was in the context of moving away from such “checklist-style”
> instances of dwc:Occurrence, which I’ve always felt overloaded the concept.
> I think we should be more explicit about what we mean by Occurrence
> instances.  Taxa don’t occur at locations.  Organisms occur at locations,
> and their occurrence is fixed in 4-dimentional space (i.e., time being one
> of the dimensions); and hence via dwc:Events.  Organisms are assigned to
> Taxa via dwc:Identification.  My comment was based on how we can use this
> more explicit approach to capture information  on Nativeness/etc.
>
>
>
> When we were discussing the concept of an “Organism”, we all seemed to
> agree that it could apply to a single individual organism, or a colony of
> individuals (e.g., a coral head, or ant colony), or a defined group of
> multiple individuals (e.g., wolf pack or whale pod).  We also discussed
> whether a “population” could be within the scope of an instance of
> “Organism”.  I don’t recall where we ended up on that as the “upper limit”
> to an organism (i.e., some would argue that a population is the lower limit
> of a taxon, rather than the upper limit of an Organism).  The basic problem
> is that there is a continuum from Organism to Taxon, and somewhere along
> that continuum we need to draw the line when assigning instances to one
> class or the other.
>
>
>
> In any case, DwC currently does not have a class that represents the
> intersection of a Taxon and a Location – which is the logical thing one
> might want to apply the status of Native vs. Invasive, etc.  The question
> is, can we meaningfully (and without stretching definitions of our existing
> classes and intended data representations too far) represent the
> native/invasive/etc. status of a Taxon at a Location using the existing
> Occurrence and other DwC classes. If we accept than dwc:Occurrence is an
> instance of dwc:Organism + an instance of dwc:Event, and dwc:Event is an
> instance of dwc:Location + Date/Time and other properties, and we also
> accept that an dwc:Organism is what dwc:Identification instances apply to,
> then we can logically apply the Native/Invasive/etc. status to such
> Occurrence instances.
>
>
>
> For example, suppose I defined an Instance of “Organism” as the population
> of Aus bus within the state of Hawaii (i.e., the dwc:Organism instance has
> scope “population”, and an dwc:Identification instance applies the
> dwc:Taxon “Aus bus” to that population-level organism).  We can create one
> or more dwc:Occurrence records to represent the presence of that Organism
> at one or more dwc:Event instances involving Hawaii as the dwc:Location,
> and whatever other dwc:Event properties that are relevant to the entire
> population of that Organism within Hawaii (such as a date when the
> population was first recorded from Hawaii via observation or representative
> specimens).  If this is an acceptable instance of dwc:Occurrence (i.e., the
> population of Aus bus in Hawaii), then it seems appropriate to me to apply
> values of “Native” or “Invasive” or whatever via dwc:establishmentMeans as
> applied to that Occurrence.
>
>
>
> Doing it this way allows you to monitor it (as a population) over time via
> multiple dwc:Occurrence instances associated with the same population-level
> dwc:Organism instance.
>
>
>
> The problem isn’t a field for “Nativeness”, the problem is what dwc class
> should that field be applied to.  It seems to me that dwc:Occurrence is
> appropriate, as long as the associated dwc:Organism can be defined as
> representing a population.
>
>
>
> Of course, one can still simply use instances of dwc:Occurrence without
> any implied Organism or Event at all (ala the checklist approach of
> anchoring a Taxon to a Location, without implied Organism or Event
> instances), but I would like to think that our community is moving away
> from that approach towards a more explicit approach for documenting
> biodiversity.
>
>
>
> Maybe I’m attacking this issue from the wrong angle, in which I apologize
> for cluttering the conversation.
>
>
>
> Aloha,
>
> Rich
>
>
>
>
>
> *From:* Quentin Groom [mailto:quentin.groom at plantentuinmeise.be
> <quentin.groom at plantentuinmeise.be>]
> *Sent:* Friday, June 24, 2016 9:27 PM
> *To:* deepreef at bishopmuseum.org
> *Cc:* Steve Baskauf; TDWG Content Mailing List
> *Subject:* Re: [tdwg-content] A proposal to improve Darwin Core for
> invasive species data
>
>
>
> Dear Rich,
>
> Thanks for taking a look at the proposal!
>
>
>
> perhaps you could clarify your comment regarding the upper limit of
> "population". Is this what I would call the use of Darwin Core for
> observations verses checklists?
>
>
>
> You can only tell if an organism is invasive if you monitor it over time
> therefore this term in inappropriate for a single observation. Also,
> whether something is invasive is a different concept to whether something
> is native or alien, both can be invasive. I seems to me that the currently
> suggested vocabulary for dwc:establishmentMeans is conflating two concepts.
> Also, none of these terms have much to do with how the organism became
> established.
>
>
>
> Currently, there are no fields where nativeness can be properly described
> and all we need is 1. This is needed for calculating essential biodiversity
> variables, for horizon scanning and invasive species monitoring. This is in
> contrast to occurrenceStatus where we currently have three ways to declare
> absence.
>
>
>
> I would have preferred to deprecate the term dwc:establishmentMeans,
> because its definition doesn't match its suggested vocabulary. However, I
> chose to retain it for the sake of stability. It has already been suggested
> that a better term would be introductionMeans.
>
>
>
> I don't understand your way of indicating nativeness or how it would work.
> For many species nativeness is a concept based upon limited available
> evidence. It isn't something that can be pinned down to a particular event
> or location. If Darwin Core was only ever used for single observations I
> would agree that we don't need a term for nativeness, but as it is also
> used for checklists and we have to accommodate terms that relate to a taxon
> over a large area.
>
>
>
> Regards
>
> Quentin
>
>
>
>
>
>
>
>
> Dr. Quentin Groom
>
> (Botany and Information Technology)
>
>
>
> Botanic Garden Meise
>
> Domein van Bouchout
>
> B-1860 Meise
>
> Belgium
>
>
>
> ORCID: 0000-0002-0596-5376 <http://orcid.org/0000-0002-0596-5376>
>
>
>
> Landline; +32 (0) 226 009 20 ext. 364
>
> FAX:      +32 (0) 226 009 45
>
>
>
> E-mail:     quentin.groom at plantentuinmeise.be
>
> Skype name: qgroom
>
> Website:    www.botanicgarden.be
>
>
>
>
>
> On 25 June 2016 at 08:13, Richard Pyle <deepreef at bishopmuseum.org> wrote:
>
> Hi All,
>
>
>
> When we discussed the scope of the class “Organism”, I believe we
> considered the upper limit of “population” – but I can’t remember whether
> we accepted that upper limit.  If so, then the “Organism” instance
> participating in an particular Occurrence instance could logically be
> qualified as “invasive” or “native” (or whatever), in which case it seems
> more appropriate to apply terms such as “native”, “introduced”, “invasive”,
> etc. to dwc:establishmentMeans
>
>
>
> I realize this is squishy, but we don’t really have another class within
> DwC-space to which the property of “native”, “introduced”, “invasive”, etc.
> can be applied.  Moreover, I don’t think there SHOULD be such a class,
> because it short-circuits the basis for the presence of Taxon X at Location
> Y (i.e., this should be established via
> Taxon-->Identification-->Organism-->Occurrence-->Event-->Location).
>
>
>
> I realize this is an extraordinarily convoluted way of saying “Taxon X is
> native to Location Y”; but ultimately that’s what we want…. Right?
>
>
>
> Aloha,
>
> Rich
>
>
>
> *From:* tdwg-content [mailto:tdwg-content-bounces at lists.tdwg.org] *On
> Behalf Of *Steve Baskauf
> *Sent:* Sunday, June 19, 2016 1:45 PM
> *To:* Quentin Groom
> *Cc:* tdwg-content at lists.tdwg.org
> *Subject:* Re: [tdwg-content] A proposal to improve Darwin Core for
> invasive species data
>
>
>
> Getting caught up on this thread after a holiday.
>
> Some previous discussion on dwc:establishmentMeans in 2010 was at:
> http://lists.tdwg.org/pipermail/tdwg-content/2010-October/001730.html
> http://lists.tdwg.org/pipermail/tdwg-content/2010-October/001731.html
>
> The opinion expressed in that thread was that dwc:establishmentMeans was a
> property of an organism at a particular place and time (i.e. an
> Occurrence.): how did a particular organism come to be in that place at
> that time.  In that view, an organism might be at a location because it was
> a representative of a native species, or because it was managed at that
> location by humans.  In that perspective, it would not make sense to use
> the value "invasive" with dwc:establishmentMeans because that is more of a
> property of a species at a location rather than an individual organism at
> that location and time.
>
> Steve
>
> Quentin Groom wrote:
>
> I've been working on a proposal to improve Darwin Core for use with
> invasive species data.
>
>
>
> The proposal is detailed on GitHub at
> https://github.com/qgroom/ias-dwc-proposal/blob/master/proposal.md.
>
>
>
> The proposal is for a new term "origin" and suggested vocabularies for
> establishmentMeans and occurrenceStatus.
>
>
>
> I'd welcome your feedback on the proposal.
>
>
>
> From my perspective it provides some needed clarity on
> the establishmentMeans and occurrenceStatus fields, but also adds the
> origin that is needed for invasive species research and for conservation
> assessments.
>
>
>
> I'm not sure of the best way to discuss this, but if you have concrete
> proposals for changes you might raise them as issues on GitHub, as well as
> mentioning them here.
>
>
>
> Regards
>
> Quentin
>
>
>
>
>
>
>
>
> Dr. Quentin Groom
>
> (Botany and Information Technology)
>
>
>
> Botanic Garden Meise
>
> Domein van Bouchout
>
> B-1860 Meise
>
> Belgium
>
>
>
> ORCID: 0000-0002-0596-5376 <http://orcid.org/0000-0002-0596-5376>
>
>
>
> Landline; +32 (0) 226 009 20 ext. 364
>
> FAX:      +32 (0) 226 009 45
>
>
>
> E-mail:     quentin.groom at plantentuinmeise.be
>
> Skype name: qgroom
>
> Website:    www.botanicgarden.be
>
>
>
>
>
> --
>
> Steven J. Baskauf, Ph.D., Senior Lecturer
>
> Vanderbilt University Dept. of Biological Sciences
>
>
>
> postal mail address:
>
> PMB 351634
>
> Nashville, TN  37235-1634,  U.S.A.
>
>
>
> delivery address:
>
> 2125 Stevenson Center
>
> 1161 21st Ave., S.
>
> Nashville, TN 37235
>
>
>
> office: 2128 Stevenson Center
>
> phone: (615) 343-4582,  fax: (615) 322-4942
>
> If you fax, please phone or email so that I will know to look for it.
>
> http://bioimages.vanderbilt.edu
>
> http://vanderbilt.edu/trees
>
>
>
>
>
> _______________________________________________
> tdwg-content mailing list
> tdwg-content at lists.tdwg.org
> http://lists.tdwg.org/mailman/listinfo/tdwg-content
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.tdwg.org/pipermail/tdwg-content/attachments/20160625/a77c5d4c/attachment-0001.html>