Remove/filter by data from botanical gardens?
Hello,
Someone brought up recently whether or not they could exclude observations from botanical gardens in GBIF occurrence searches. I think they meant just occurrences from within actual gardens, not observations from more natural areas in a database provided by a botanical garden.
Is something like this already flagged as an issue in the occurrence issue enum? I didn't see it myself. I guess one could find the data providers that are botanical gardens and exclude those if they are in the search results.
Thanks for any guidance.
Cheers, Scott Chamberlain
Hi Scott,
There is actually no easy way of doing this. Filtering out BasisOfRecord=LivingSpecimen would get you some of the way there, and looking into the establishmentMeans a step further, but the some of the data just don’t have fields to filter them out.
Cheers, Tim
On 07 Oct 2014, at 03:00, Scott Chamberlain myrmecocystus@gmail.com wrote:
Hello,
Someone brought up recently whether or not they could exclude observations from botanical gardens in GBIF occurrence searches. I think they meant just occurrences from within actual gardens, not observations from more natural areas in a database provided by a botanical garden.
Is something like this already flagged as an issue in the occurrence issue enum? I didn't see it myself. I guess one could find the data providers that are botanical gardens and exclude those if they are in the search results.
Thanks for any guidance.
Cheers, Scott Chamberlain _______________________________________________ API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
… plus LivingSpecimen is used a lot for culture collections (e.g. fungi, algae) and all germplasm / seed records with mostly correct wild location data. You would filter out a few correct native occurrences by using basis of record.
And establishmentMeans is not yet part of our API search filters - so you would have to filter records yourself locally.
Markus
On 07 Oct 2014, at 16:47, Tim Robertson trobertson@gbif.org wrote:
Hi Scott,
There is actually no easy way of doing this. Filtering out BasisOfRecord=LivingSpecimen would get you some of the way there, and looking into the establishmentMeans a step further, but the some of the data just don’t have fields to filter them out.
Cheers, Tim
On 07 Oct 2014, at 03:00, Scott Chamberlain myrmecocystus@gmail.com wrote:
Hello,
Someone brought up recently whether or not they could exclude observations from botanical gardens in GBIF occurrence searches. I think they meant just occurrences from within actual gardens, not observations from more natural areas in a database provided by a botanical garden.
Is something like this already flagged as an issue in the occurrence issue enum? I didn't see it myself. I guess one could find the data providers that are botanical gardens and exclude those if they are in the search results.
Thanks for any guidance.
Cheers, Scott Chamberlain _______________________________________________ API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
Hi. Thanks for the tips! I will see if some combination of these can be useful in removing botanical garden occurrences.
Scott
On Tue, Oct 7, 2014 at 8:02 AM, Markus Döring mdoering@gbif.org wrote:
… plus LivingSpecimen is used a lot for culture collections (e.g. fungi, algae) and all germplasm / seed records with mostly correct wild location data. You would filter out a few correct native occurrences by using basis of record.
And establishmentMeans is not yet part of our API search filters - so you would have to filter records yourself locally.
Markus
On 07 Oct 2014, at 16:47, Tim Robertson trobertson@gbif.org wrote:
Hi Scott,
There is actually no easy way of doing this. Filtering out
BasisOfRecord=LivingSpecimen would get you some of the way there, and looking into the establishmentMeans a step further, but the some of the data just don’t have fields to filter them out.
Cheers, Tim
On 07 Oct 2014, at 03:00, Scott Chamberlain myrmecocystus@gmail.com
wrote:
Hello,
Someone brought up recently whether or not they could exclude
observations from botanical gardens in GBIF occurrence searches. I think they meant just occurrences from within actual gardens, not observations from more natural areas in a database provided by a botanical garden.
Is something like this already flagged as an issue in the occurrence
issue enum? I didn't see it myself. I guess one could find the data providers that are botanical gardens and exclude those if they are in the search results.
Thanks for any guidance.
Cheers, Scott Chamberlain _______________________________________________ API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
Also not something that can be done via the API, but botanical gardens will also tend to be fairly large locality outliers (large numbers of extremely tightly clustered points from relatively diverse taxa). Often, I've found, many gardens are all geo-referenced to a single reference locality, so you don't even need to do a bounding box filter, you can just filter out specific points.
- Alex
On 10/07/2014 10:47 AM, Tim Robertson wrote:
Hi Scott,
There is actually no easy way of doing this. Filtering out BasisOfRecord=LivingSpecimen would get you some of the way there, and looking into the establishmentMeans a step further, but the some of the data just don’t have fields to filter them out.
Cheers, Tim
On 07 Oct 2014, at 03:00, Scott Chamberlain myrmecocystus@gmail.com wrote:
Hello,
Someone brought up recently whether or not they could exclude observations from botanical gardens in GBIF occurrence searches. I think they meant just occurrences from within actual gardens, not observations from more natural areas in a database provided by a botanical garden.
Is something like this already flagged as an issue in the occurrence issue enum? I didn't see it myself. I guess one could find the data providers that are botanical gardens and exclude those if they are in the search results.
Thanks for any guidance.
Cheers, Scott Chamberlain _______________________________________________ API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
Alex,
Thanks, that's a good idea. I was thinking that there are a finite number of botanical gardens, so I could simply collect coordinates for each of them, and a reasonable bounding box around each one perhaps.
Scott
On Tue, Oct 7, 2014 at 10:03 AM, Alex Thompson godfoder@acis.ufl.edu wrote:
Also not something that can be done via the API, but botanical gardens will also tend to be fairly large locality outliers (large numbers of extremely tightly clustered points from relatively diverse taxa). Often, I've found, many gardens are all geo-referenced to a single reference locality, so you don't even need to do a bounding box filter, you can just filter out specific points.
- Alex
On 10/07/2014 10:47 AM, Tim Robertson wrote:
Hi Scott,
There is actually no easy way of doing this. Filtering out
BasisOfRecord=LivingSpecimen would get you some of the way there, and looking into the establishmentMeans a step further, but the some of the data just don’t have fields to filter them out.
Cheers, Tim
On 07 Oct 2014, at 03:00, Scott Chamberlain myrmecocystus@gmail.com
wrote:
Hello,
Someone brought up recently whether or not they could exclude
observations from botanical gardens in GBIF occurrence searches. I think they meant just occurrences from within actual gardens, not observations from more natural areas in a database provided by a botanical garden.
Is something like this already flagged as an issue in the occurrence
issue enum? I didn't see it myself. I guess one could find the data providers that are botanical gardens and exclude those if they are in the search results.
Thanks for any guidance.
Cheers, Scott Chamberlain _______________________________________________ API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
Hmmm...I don't see establishmentMeans anywhere in the returned data from the /occurrence . I do see documentation on it, endpoint https://code.google.com/p/darwincore/wiki/Occurrence#establishmentMeans, but is it included in data output anywhere? Maybe called something else?
S
On Tue, Oct 7, 2014 at 10:42 AM, Scott Chamberlain scott@ropensci.org wrote:
Alex,
Thanks, that's a good idea. I was thinking that there are a finite number of botanical gardens, so I could simply collect coordinates for each of them, and a reasonable bounding box around each one perhaps.
Scott
On Tue, Oct 7, 2014 at 10:03 AM, Alex Thompson godfoder@acis.ufl.edu wrote:
Also not something that can be done via the API, but botanical gardens will also tend to be fairly large locality outliers (large numbers of extremely tightly clustered points from relatively diverse taxa). Often, I've found, many gardens are all geo-referenced to a single reference locality, so you don't even need to do a bounding box filter, you can just filter out specific points.
- Alex
On 10/07/2014 10:47 AM, Tim Robertson wrote:
Hi Scott,
There is actually no easy way of doing this. Filtering out
BasisOfRecord=LivingSpecimen would get you some of the way there, and looking into the establishmentMeans a step further, but the some of the data just don’t have fields to filter them out.
Cheers, Tim
On 07 Oct 2014, at 03:00, Scott Chamberlain myrmecocystus@gmail.com
wrote:
Hello,
Someone brought up recently whether or not they could exclude
observations from botanical gardens in GBIF occurrence searches. I think they meant just occurrences from within actual gardens, not observations from more natural areas in a database provided by a botanical garden.
Is something like this already flagged as an issue in the occurrence
issue enum? I didn't see it myself. I guess one could find the data providers that are botanical gardens and exclude those if they are in the search results.
Thanks for any guidance.
Cheers, Scott Chamberlain _______________________________________________ API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
Hi Scott
Occurrence record fields are elastic - we support the full DwC, but can only render from GBIF.org what is shared by the data publisher.
So that field will only appear on records where it is present - here is an example of a single record: http://api.gbif.org/v1/occurrence/675943831
That record above is the interpreted view of the verbatim content we observed from the source: http://api.gbif.org/v1/occurrence/675943831/verbatim
It is also returned in a search (same dataset as record above): http://api.gbif.org/v1/occurrence/search?datasetKey=b6015b60-6f96-43a9-88e5-...
I hope this helps, Tim
On 07 Oct 2014, at 20:48, Scott Chamberlain scott@ropensci.org wrote:
Hmmm...I don't see establishmentMeans anywhere in the returned data from the /occurrence . I do see documentation on it, endpoint https://code.google.com/p/darwincore/wiki/Occurrence#establishmentMeans, but is it included in data output anywhere? Maybe called something else?
S
On Tue, Oct 7, 2014 at 10:42 AM, Scott Chamberlain scott@ropensci.org wrote: Alex,
Thanks, that's a good idea. I was thinking that there are a finite number of botanical gardens, so I could simply collect coordinates for each of them, and a reasonable bounding box around each one perhaps.
Scott
On Tue, Oct 7, 2014 at 10:03 AM, Alex Thompson godfoder@acis.ufl.edu wrote: Also not something that can be done via the API, but botanical gardens will also tend to be fairly large locality outliers (large numbers of extremely tightly clustered points from relatively diverse taxa). Often, I've found, many gardens are all geo-referenced to a single reference locality, so you don't even need to do a bounding box filter, you can just filter out specific points.
- Alex
On 10/07/2014 10:47 AM, Tim Robertson wrote:
Hi Scott,
There is actually no easy way of doing this. Filtering out BasisOfRecord=LivingSpecimen would get you some of the way there, and looking into the establishmentMeans a step further, but the some of the data just don’t have fields to filter them out.
Cheers, Tim
On 07 Oct 2014, at 03:00, Scott Chamberlain myrmecocystus@gmail.com wrote:
Hello,
Someone brought up recently whether or not they could exclude observations from botanical gardens in GBIF occurrence searches. I think they meant just occurrences from within actual gardens, not observations from more natural areas in a database provided by a botanical garden.
Is something like this already flagged as an issue in the occurrence issue enum? I didn't see it myself. I guess one could find the data providers that are botanical gardens and exclude those if they are in the search results.
Thanks for any guidance.
Cheers, Scott Chamberlain _______________________________________________ API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
Tim,
Thanks, I should have thought of that!
Cheers, Scott
On Tue, Oct 7, 2014 at 2:30 PM, Tim Robertson trobertson@gbif.org wrote:
Hi Scott
Occurrence record fields are elastic - we support the full DwC, but can only render from GBIF.org what is shared by the data publisher.
So that field will only appear on records where it is present - here is an example of a single record: http://api.gbif.org/v1/occurrence/675943831
That record above is the interpreted view of the verbatim content we observed from the source: http://api.gbif.org/v1/occurrence/675943831/verbatim
It is also returned in a search (same dataset as record above):
http://api.gbif.org/v1/occurrence/search?datasetKey=b6015b60-6f96-43a9-88e5-...
I hope this helps, Tim
On 07 Oct 2014, at 20:48, Scott Chamberlain scott@ropensci.org wrote:
Hmmm...I don't see establishmentMeans anywhere in the returned data from the /occurrence . I do see documentation on it, endpoint https://code.google.com/p/darwincore/wiki/Occurrence#establishmentMeans, but is it included in data output anywhere? Maybe called something else?
S
On Tue, Oct 7, 2014 at 10:42 AM, Scott Chamberlain scott@ropensci.org wrote:
Alex,
Thanks, that's a good idea. I was thinking that there are a finite number of botanical gardens, so I could simply collect coordinates for each of them, and a reasonable bounding box around each one perhaps.
Scott
On Tue, Oct 7, 2014 at 10:03 AM, Alex Thompson godfoder@acis.ufl.edu wrote:
Also not something that can be done via the API, but botanical gardens will also tend to be fairly large locality outliers (large numbers of extremely tightly clustered points from relatively diverse taxa). Often, I've found, many gardens are all geo-referenced to a single reference locality, so you don't even need to do a bounding box filter, you can just filter out specific points.
- Alex
On 10/07/2014 10:47 AM, Tim Robertson wrote:
Hi Scott,
There is actually no easy way of doing this. Filtering out
BasisOfRecord=LivingSpecimen would get you some of the way there, and looking into the establishmentMeans a step further, but the some of the data just don’t have fields to filter them out.
Cheers, Tim
On 07 Oct 2014, at 03:00, Scott Chamberlain myrmecocystus@gmail.com
wrote:
Hello,
Someone brought up recently whether or not they could exclude
observations from botanical gardens in GBIF occurrence searches. I think they meant just occurrences from within actual gardens, not observations from more natural areas in a database provided by a botanical garden.
Is something like this already flagged as an issue in the occurrence
issue enum? I didn't see it myself. I guess one could find the data providers that are botanical gardens and exclude those if they are in the search results.
Thanks for any guidance.
Cheers, Scott Chamberlain _______________________________________________ API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
API-users mailing list API-users@lists.gbif.org http://lists.gbif.org/mailman/listinfo/api-users
participants (5)
-
Alex Thompson -
Markus Döring -
Scott Chamberlain -
Scott Chamberlain -
Tim Robertson