18184
Reporter: rdmpage
Type: Bug
Summary: Display grid-based data as grid not point
Description: The GBIF portal consistently distorts data that has been collected on a grid. By summarising a grid as a single point locality the user gets a misleading impression of precision. For example the recently added dataset "BrioCAT: Banco de datos de los briòfitos de Cataluña "http://www.gbif.org/dataset/77a24e14-534b-4864-bb95-ae0966ce89ce very obviously is grid-based, yet each occurrence is a point in GBIF. Visiting the original data site http://biodiver.bio.ub.es/biocat/index.jsp you can see the UTM grid-based sampling. This is, I suspect, a common method of sampling especially for plant data. It would be great if GBIF could represent this data with more fidelity, because at the moment it's deeply misleading.
Priority: Major
Resolution: Duplicate
Status: Closed
Created: 2016-01-29 10:06:38.873
Updated: 2017-10-10 14:56:07.472
Resolved: 2017-10-10 14:56:07.452
Author: rdmpage
Created: 2016-01-29 12:40:24.809
Updated: 2016-01-29 12:40:24.809
Just to follow up, I guess part of the problem here is that the data coming from the provider (via IPT) is in the form of point localities, so we've already lost the fact that it's a polygon. Hence:
(a) can IPT include data that is sampled at grid level? In other words, can we support polygons as well as points at the level of data entry and/or export?
(b) a more accurate visual representation in the GBIF portal would be to show the polygon in which each occurrence is reported, rather than a point locality.
Author: kbraak@gbif.org
Created: 2016-01-29 15:04:19.657
Updated: 2016-01-29 15:04:19.657
Hi Rod. Thanks for reporting this issue.
To answer a), yes, it is possible to specify a location by polygon in DwC for example using http://rs.tdwg.org/dwc/terms/#footprintWKT. In the dataset metadata, a polygon can also be used to describe the geographic extent of the dataset.
Your point b) makes perfect sense. We'll continue to investigate this further. Thanks.