I’m curious if the rgbif package can retrieve a particular piece of information for me. When you use the website, you can get to a page of information for each record (example here: http://www.gbif.org/occurrence/918957709), which includes a section called ‘Occurence Details’ and a field called ‘Remarks’. So my question is, can I pull down the information in the ‘Remarks’ field in an occ_search() call? I don’t see an argument listed for that…
Thanks!
Hi @kgturner - yep, you can most def get those fields in rgbif
e.g., try http://api.gbif.org/v1/occurrence/918957709 , which gives
{
key: 918957709,
datasetKey: "07fd0d79-4883-435f-bba1-58fef110cd13",
publishingOrgKey: "b542788f-0dc2-4a2b-b652-fceced449591",
publishingCountry: "CA",
protocol: "DWC_ARCHIVE",
lastCrawled: "2014-06-18T18:19:41.557+0000",
lastParsed: "2014-06-18T18:19:41.597+0000",
extensions: { },
basisOfRecord: "PRESERVED_SPECIMEN",
taxonKey: 3128962,
kingdomKey: 6,
phylumKey: 49,
classKey: 220,
orderKey: 414,
familyKey: 3065,
genusKey: 3127469,
scientificName: "Centaurea diffusa Lam.",
kingdom: "Plantae",
phylum: "Magnoliophyta",
order: "Asterales",
family: "Asteraceae",
genus: "Centaurea",
genericName: "Centaurea",
specificEpithet: "diffusa",
taxonRank: "SPECIES",
dateIdentified: "2009-12-31T23:00:00.000+0000",
decimalLongitude: -123.25143,
decimalLatitude: 49.26304,
stateProvince: "British Columbia",
year: 2010,
month: 2,
day: 18,
eventDate: "2010-02-17T23:00:00.000+0000",
issues: [
"COORDINATE_ROUNDED",
"GEODETIC_DATUM_INVALID",
"GEODETIC_DATUM_ASSUMED_WGS84"
],
modified: "2012-02-29T23:00:00.000+0000",
lastInterpreted: "2014-06-18T18:21:15.861+0000",
references: "http://bridge.botany.ubc.ca/herbarium/details.php?db=vwsp.fmp12&layout=vwsp_web_details&recid=195355&ass_num=V232687",
identifiers: [ ],
facts: [ ],
relations: [ ],
geodeticDatum: "WGS84",
class: "Magnoliopsida",
countryCode: "CA",
country: "Canada",
recordNumber: "RU008-16K",
rights: "http://creativecommons.org/publicdomain/zero/1.0/ & http://www.canadensys.net/norms",
rightsHolder: "University of British Columbia",
type: "PhysicalObject",
identifiedBy: "K.G. Turner",
catalogNumber: "V232687",
recordedBy: "K.G. Turner, K. Nurkowski",
bibliographicCitation: "Centaurea diffusa Lam. (UBC V232687)",
preparations: "mounted",
collectionID: "urn:lsid:biocol.org:col:15106",
nomenclaturalCode: "ICBN",
previousIdentifications: "Centaurea diffusa by K.G. Turner 2010",
verbatimSRS: "unknown",
occurrenceID: "153925",
ownerInstitutionCode: "University of British Columbia",
collectionCode: "UBC",
occurrenceRemarks: "Seed collected: 1119, 2006, A. Shipunov, Pjatigorsk City Russia, N43Ëš3'36'' E44Ëš3'0'', no alt, 1 sheet, white flowers. Voucher for a survey of phenotypic variation across the native and invaded ranges, in a common environment.",
gbifID: "918957709",
habitat: "Greenhouse Conditions, 16 hr days, water every third day",
institutionCode: "University of British Columbia",
datasetID: "http://dataset.canadensys.net/ubc-vascular-specimens",
datasetName: "University of British Columbia Herbarium (UBC) - Vascular Plant Collection",
locality: "Vancouver, University of British Columbia Horticultural Greenhouse, 6394 Stores Road University of British Columbia",
language: "en",
identifier: "153925"
}
Or in rgbif
library("rgbif")
occ_get(918957709, fields = "all")$data
name key decimalLatitude decimalLongitude issues
1 Centaurea 918957709 49.26304 -123.2514 cdround,gass84,gdativ
datasetKey publishingOrgKey publishingCountry protocol
1 07fd0d79-4883-435f-bba1-58fef110cd13 b542788f-0dc2-4a2b-b652-fceced449591 CA DWC_ARCHIVE
lastCrawled lastParsed extensions basisOfRecord taxonKey
1 2014-06-18T18:19:41.557+0000 2014-06-18T18:19:41.597+0000 none PRESERVED_SPECIMEN 3128962
kingdomKey phylumKey classKey orderKey familyKey genusKey scientificName kingdom phylum
1 6 49 220 414 3065 3127469 Centaurea diffusa Lam. Plantae Magnoliophyta
order family genus genericName specificEpithet taxonRank dateIdentified
1 Asterales Asteraceae Centaurea Centaurea diffusa SPECIES 2009-12-31T23:00:00.000+0000
stateProvince year month day eventDate modified
1 British Columbia 2010 2 18 2010-02-17T23:00:00.000+0000 2012-02-29T23:00:00.000+0000
lastInterpreted
1 2014-06-18T18:21:15.861+0000
references
1 http://bridge.botany.ubc.ca/herbarium/details.php?db=vwsp.fmp12&layout=vwsp_web_details&recid=195355&ass_num=V232687
identifiers facts relations geodeticDatum class countryCode country recordNumber
1 none none none WGS84 Magnoliopsida CA Canada RU008-16K
rights
1 http://creativecommons.org/publicdomain/zero/1.0/ & http://www.canadensys.net/norms
rightsHolder type identifiedBy catalogNumber recordedBy
1 University of British Columbia PhysicalObject K.G. Turner V232687 K.G. Turner, K. Nurkowski
bibliographicCitation preparations collectionID nomenclaturalCode
1 Centaurea diffusa Lam. (UBC V232687) mounted urn:lsid:biocol.org:col:15106 ICBN
previousIdentifications verbatimSRS occurrenceID ownerInstitutionCode
1 Centaurea diffusa by K.G. Turner 2010 unknown 153925 University of British Columbia
collectionCode
1 UBC
occurrenceRemarks
1 Seed collected: 1119, 2006, A. Shipunov, Pjatigorsk City Russia, N43˚3'36'' E44˚3'0'', no alt, 1 sheet, white flowers. Voucher for a survey of phenotypic variation across the native and invaded ranges, in a common environment.
gbifID habitat institutionCode
1 918957709 Greenhouse Conditions, 16 hr days, water every third day University of British Columbia
datasetID
1 http://dataset.canadensys.net/ubc-vascular-specimens
datasetName
1 University of British Columbia Herbarium (UBC) - Vascular Plant Collection
locality
1 Vancouver, University of British Columbia Horticultural Greenhouse, 6394 Stores Road University of British Columbia
language identifier
1 en 153925
Ok, great.
In a related question (apologies for non-pretty post), how do I get occ_search and/or occ_get to search for a list of taxon keys? The documentation says “You can pass many keys by passing occ_search in a call to an lapply-family function (see last example below).” But I don’t see this example (certainly it’s not the last one), unless it refers to
splist ← c(‘Cyanocitta stelleri’, ‘Junco hyemalis’, ‘Aix sponsa’)
keys ← sapply(splist, function(x) name_suggest(x)$key[1], USE.NAMES=FALSE)
occ_search(taxonKey=keys, limit=5, return=‘data’)
…which I’m not sure what the sapply function is supposed to do, but it doesn’t seem to work if you give it a vector of taxonkeys instead of species names. The vector of taxonkeys doesn’t work directly in the occ_search either.
This is what I want to do, ideally returning a data frame or something I can turn into a dataframe:
ghouse ← c(918957709, 466334237, 918957719, 466334233, 91895773, 466334221)
ghousedat ← occ_search(taxonKey=ghouse,
fields=c(“gbifID”,“species”, “basisOfRecord”,“year”,“eventDate”,“countryCode”,spaces`“decimalLatitude”,“decimalLongitude”, “genus”,“specificEpithet”,“collectionCode”,“institutionCode”,“locality”,“datasetKey”,“recordNumber”,“occurrenceRemarks”), return=“data”)
What I get is a list of empties:
head(ghousedat)
$918957709
[1] “no data found, try a different search”
$466334237
[1] “no data found, try a different search”
$918957719
[1] “no data found, try a different search”
$466334233
[1] “no data found, try a different search”
$918957739
[1] “no data found, try a different search”
$466334221
[1] “no data found, try a different search”
If I use occ_get, I get data, but a list of lists:
test2 ← occ_get(ghouse, fields = c(“gbifID”,“species”, “basisOfRecord”,“year”,“eventDate”,“countryCode”,
“decimalLatitude”, “decimalLongitude”, “genus”,“specificEpithet”,“collectionCode”,
“institutionCode”,“locality”, “datasetKey”,“recordNumber”,“occurrenceRemarks”))
head(test2)
[[1]]
[[1]]$hierarchy
name key rank
1 Plantae 6 kingdom
2 Magnoliophyta 49 phylum
3 Magnoliopsida 220 class
4 Asterales 414 order
5 Asteraceae 3065 family
6 Centaurea 3127469 genus
[[1]]$media
list()
[[1]]$data
datasetKey basisOfRecord genus specificEpithet decimalLongitude decimalLatitude year eventDate
1 07fd0d79-4883-435f-bba1-58fef110cd13 PRESERVED_SPECIMEN Centaurea diffusa -123.2514 49.26304 2010 2010-02-17T23:00:00.000+0000
countryCode recordNumber collectionCode
1 CA RU008-16K UBC
occurrenceRemarks
1 Seed collected: 1119, 2006, A. Shipunov, Pjatigorsk City Russia, N43°3’36’’ E44°3’0’', no alt, 1 sheet, white flowers. Voucher for a survey of phenotypic variation across the native and invaded ranges, in a common environment.
gbifID institutionCode
1 918957709 University of British Columbia
locality
1 Vancouver, University of British Columbia Horticultural Greenhouse, 6394 Stores Road University of British Columbia
So ok, what should I do?
Ah, one thing wrong here is that occurrence keys are different from taxon keys. The key 918957709 in http://www.gbif.org/occurrence/918957709 is an occurrence key, which you can pass e.g. to occ_get(), but you can’t pass to occ_search(), which in the taxonKey parameter wants taxonomic keys, like 3128962 in http://www.gbif.org/species/3128962
more on the outputs later…
Ah, ok, I’m definitely confused about the difference between taxonKey, occurrence key, and gbifID. All they all different things?
Maybe this helps in thinking about the various keys:
taxonKey, but that key can denote any taxonomic level: class, order, genus, species, etc.taxonKey can have any number of occurrences denoted by key (see below). The taxonKey should match the returned taxonKey from occ_search() function.gbifID i usually ignore, but seems to be the same as the occurrence keynubKey that’s returned in the name_*() functions, which is usually/or always the same as the taxonKey
occ_search(3128962, limit=3, fields=c('name','key','taxonKey','gbifID'))
Records found [1520]
Records returned [3]
No. unique hierarchies [1]
No. media records [1]
Args [taxonKey=3128962, limit=3, offset=0, fields=name,key,taxonKey,gbifID]
First 10 rows of data
name key taxonKey gbifID
1 Centaurea 1038571715 3128962 1038571715
2 Centaurea 920996089 3128962 920996089
3 Centaurea 859763858 3128962 859763858
I’ll see if I can make the documentation more clear on all this…
@kgturner For combining outputs of functions:
works the same for occ_get() and occ_search()
library("plyr")
out <- occ_get(c(918957709, 466334237))
rbind.fill(lapply(out, "[[", "data"))
name key decimalLatitude decimalLongitude issues
1 Centaurea 918957709 49.26304 -123.2514 cdround,gass84,gdativ
2 Centaurea 466334237 49.26304 -123.2514 cdround,gass84
res <- occ_search(taxonKey=c(2482598,2492010,2498387), limit=5, fields="minimal")
res$`2482598`$data
rbind.fill(lapply(res, "[[", "data"))
name key decimalLatitude decimalLongitude issues
1 Cyanocitta stelleri 891781350 37.73646 -122.48801 cdround,cudc,gass84
2 Cyanocitta stelleri 891046529 32.82392 -116.53233 cdround,cudc,gass84
3 Cyanocitta stelleri 891056081 37.76811 -122.47370 cdround,cudc,gass84
4 Cyanocitta stelleri 891046151 19.29372 -98.65598 cdround,cudc,gass84
5 Cyanocitta stelleri 891047537 37.86877 -122.23729 cdround,cudc,gass84
6 Junco hyemalis 891036043 45.56906 -122.54434 cdround,cudc,gass84
7 Junco hyemalis 891036414 44.32038 -73.09067 cdround,cudc,gass84
8 Junco hyemalis 891036487 37.87244 -122.25864 cdround,cudc,gass84
9 Junco hyemalis 891036897 32.89302 -117.24204 cdround,cudc,gass84
10 Junco hyemalis 891035392 44.24846 -73.07934 cdround,cudc,gass84
11 Aix sponsa 1022841740 50.55339 7.26308 cdround,cudc,gass84
12 Aix sponsa 920269599 51.76471 9.05524 cdround,gass84
13 Aix sponsa 920269677 49.82962 9.86967 cdround,gass84
14 Aix sponsa 891035006 32.74783 -97.35127 cdround,cudc,gass84
15 Aix sponsa 891035458 34.23871 -116.94964 cdround,cudc,gass84
Does that make sense? There are convenience functions in spocc to combine data outputs for users, but I guess we don’'t have these in rgbif