Data below -60 deg lat should be attributed to Antarctica
15936
Reporter: omeyn
Assignee: cgendreau
Type: Bug
Summary: Data below -60 deg lat should be attributed to Antarctica
Priority: Major
Resolution: Fixed
Status: Resolved
Created: 2014-06-17 13:37:14.843
Updated: 2015-12-10 10:42:55.988
Resolved: 2015-12-10 10:42:55.835
Description: This comes in as a issue from Anton, Node manager for SCAR. From email:
"Following the iso standard (https://www.iso.org/obp/ui/#iso:code:3166:AQ) all below 60° South should be Antarctic (Territories south of 60° south latitude). Currently I have the impression that a distance of 200 nm was used. This means that a large proportion of our efforts to get data published are not easily visible as Antarctic data.
It is very important that this should be addressed.”
Additionally wikipedia reports: "The ISO 3166-1 alpha-2 "AQ" is assigned to the entire continent regardless of jurisdiction. Different country calling codes and currencies[95] are used for different settlements, depending on the administrating country."
This is a major deviation from our current country + eez geo lookup.
The count of records with lat <= -60 and countrycode not null and countrycode != 'AQ' is: 33650.
The query for lat <= -60 and lng not null gives following country codes and their respective counts:
AQ 890349
NULL 140444
US 11077
PE 8191
AU 7412
AR 2703
CO 1455
GS 1017
CL 449
EC 319
NZ 282
BO 166
FK 162
NI 79
GB 51
PG 31
BR 31
RU 28
CR 27
MX 23
EH 16
PY 16
AG 14
ZW 14
TZ 11
TF 9
FR 8
CA 8
ZA 7
VE 5
NA 4
ID 4
PA 4
BS 3
CN 3
NO 3
BW 2
ZZ 2
CU 2
SJ 1
ES 1
IT 1
GE 1
SH 1
IS 1
AO 1
SB 1
PR 1
JP 1
NC 1
DM 1
]]>
Author: trobertson@gbif.org
Created: 2014-06-17 14:09:49.582
Updated: 2014-06-17 14:09:49.582
We need the thoughts of [~mraymond] here. Data is attributed to a single publishing or location country - we are not set up to do otherwise without serious effort.
Should a record < -60 degrees be attributed to Antarctica, meaning it won't be counted in the effort of e.g. Argentina?
I suspect not (e.g. considering who funds GBIF).
Data search exists for sub 60 degrees.
We could consider doing "data for Arctic / Antarctic" and data about... analytics reports for example but not the real time country page counts.
Author: omeyn@gbif.org
Comment: Seems reasonably safe to add logic that says if geo lookup returns null and lat <= -60 then AQ, which would take care of our nulls, and probably is correct, but doesn't take away anything claimed by a country.
Created: 2014-06-17 14:16:47.774
Updated: 2014-06-17 14:16:47.774
Author: mraymond
Comment: I need to discuss this with you briefly to be sure I have understood correctly but this sounds reasonable to me. We can also point Anton to the page for SCAR: http://www.gbif.org/node/51e4e8ff-bff0-4996-8b27-9107d61b3478 which shows their data publication efforts. It might be that what he really wants is just the analytic pages for organization Participants, but doing one for Arctic/ Antarctic could be a good intermediate solution.
Created: 2014-06-17 16:50:23.31
Updated: 2014-06-17 16:50:23.31
Author: omeyn@gbif.org
Comment: After discussion here we've decided to go with the nulls -> AQ logic I proposed above.
Created: 2014-06-18 13:53:42.438
Updated: 2014-06-18 13:53:42.438
Author: cgendreau
Comment: Just to be sure, all parties are still ok with the proposed solution?
Created: 2015-11-12 12:34:53.243
Updated: 2015-11-12 12:34:53.243
Author: cgendreau
Created: 2015-12-10 10:42:50.138
Updated: 2015-12-10 10:42:50.138
Proposed solution applied.
All records below -60 deg have been reinterpreted.