# occ_search

0th

Percentile

##### Search for GBIF occurrences

Search for GBIF occurrences

##### Usage
occ_search(
taxonKey = NULL,
scientificName = NULL,
country = NULL,
publishingCountry = NULL,
hasCoordinate = NULL,
typeStatus = NULL,
recordNumber = NULL,
lastInterpreted = NULL,
continent = NULL,
geometry = NULL,
geom_big = "asis",
geom_size = 40,
geom_n = 10,
recordedBy = NULL,
recordedByID = NULL,
identifiedByID = NULL,
basisOfRecord = NULL,
datasetKey = NULL,
eventDate = NULL,
catalogNumber = NULL,
year = NULL,
month = NULL,
decimalLatitude = NULL,
decimalLongitude = NULL,
elevation = NULL,
depth = NULL,
institutionCode = NULL,
collectionCode = NULL,
hasGeospatialIssue = NULL,
issue = NULL,
search = NULL,
mediaType = NULL,
subgenusKey = NULL,
repatriated = NULL,
phylumKey = NULL,
kingdomKey = NULL,
classKey = NULL,
orderKey = NULL,
familyKey = NULL,
genusKey = NULL,
establishmentMeans = NULL,
protocol = NULL,
organismId = NULL,
publishingOrg = NULL,
stateProvince = NULL,
waterBody = NULL,
locality = NULL,
limit = 500,
start = 0,
fields = "all",
return = NULL,
facet = NULL,
facetMincount = NULL,
facetMultiselect = NULL,
skip_validate = TRUE,
curlopts = list(),
...
)
##### Arguments
taxonKey

(numeric) A taxon key from the GBIF backbone. All included and synonym taxa are included in the search, so a search for aves with taxononKey=212 (i.e. /occurrence/search?taxonKey=212) will match all birds, no matter which species. You can pass many keys by passing occ_search in a call to an lapply-family function (see last example below).

scientificName

A scientific name from the GBIF backbone. All included and synonym taxa are included in the search.

country

The 2-letter country code (as per ISO-3166-1) of the country in which the occurrence was recorded. See here https://en.wikipedia.org/wiki/ISO_3166-1_alpha-2

publishingCountry

The 2-letter country code (as per ISO-3166-1) of the country in which the occurrence was recorded.

hasCoordinate

(logical) Return only occurence records with lat/long data (TRUE) or all records (FALSE, default).

typeStatus

Type status of the specimen. One of many options. See ?typestatus

recordNumber

Number recorded by collector of the data, different from GBIF record number. See http://rs.tdwg.org/dwc/terms/#recordNumber for more info

lastInterpreted

Date the record was last modified in GBIF, in ISO 8601 format: yyyy, yyyy-MM, yyyy-MM-dd, or MM-dd. Supports range queries, smaller,larger (e.g., '1990,1991', whereas '1991,1990' wouldn't work)

continent

Continent. One of africa, antarctica, asia, europe, north_america (North America includes the Caribbean and reachies down and includes Panama), oceania, or south_america

geometry

Searches for occurrences inside a polygon described in Well Known Text (WKT) format. A WKT shape written as either POINT, LINESTRING, LINEARRING POLYGON, or MULTIPOLYGON. Example of a polygon: POLYGON((30.1 10.1, 20, 20 40, 40 40, 30.1 10.1)) would be queried as http://bit.ly/1BzNwDq See also the section WKT below.

geom_big

(character) One of "axe", "bbox", or "asis" (default). See Details.

geom_size

(integer) An integer indicating size of the cell. Default: 40. See Details.

geom_n

(integer) An integer indicating number of cells in each dimension. Default: 10. See Details.

recordedBy

The person who recorded the occurrence.

recordedByID

(character) Identifier (e.g. ORCID) for the person who recorded the occurrence

identifiedByID

(character) Identifier (e.g. ORCID) for the person who provided the taxonomic identification of the occurrence.

basisOfRecord

Basis of record, as defined in our BasisOfRecord enum here https://gbif.github.io/gbif-api/apidocs/org/gbif/api/vocabulary/BasisOfRecord.html Acceptable values are:

• FOSSIL_SPECIMEN An occurrence record describing a fossilized specimen.

• HUMAN_OBSERVATION An occurrence record describing an observation made by one or more people.

• LITERATURE An occurrence record based on literature alone.

• LIVING_SPECIMEN An occurrence record describing a living specimen, e.g.

• MACHINE_OBSERVATION An occurrence record describing an observation made by a machine.

• OBSERVATION An occurrence record describing an observation.

• PRESERVED_SPECIMEN An occurrence record describing a preserved specimen.

• UNKNOWN Unknown basis for the record.

datasetKey

The occurrence dataset key (a uuid)

eventDate

Occurrence date in ISO 8601 format: yyyy, yyyy-MM, yyyy-MM-dd, or MM-dd. Supports range queries, smaller,larger (e.g., '1990,1991', whereas '1991,1990' wouldn't work)

catalogNumber

An identifier of any form assigned by the source within a physical collection or digital dataset for the record which may not unique, but should be fairly unique in combination with the institution and collection code.

year

The 4 digit year. A year of 98 will be interpreted as AD 98. Supports range queries, smaller,larger (e.g., '1990,1991', whereas '1991,1990' wouldn't work)

month

The month of the year, starting with 1 for January. Supports range queries, smaller,larger (e.g., '1,2', whereas '2,1' wouldn't work)

decimalLatitude

Latitude in decimals between -90 and 90 based on WGS 84. Supports range queries, smaller,larger (e.g., '25,30', whereas '30,25' wouldn't work)

decimalLongitude

Longitude in decimals between -180 and 180 based on WGS 84. Supports range queries (e.g., '-0.4,-0.2', whereas '-0.2,-0.4' wouldn't work).

elevation

Elevation in meters above sea level. Supports range queries, smaller,larger (e.g., '5,30', whereas '30,5' wouldn't work)

depth

Depth in meters relative to elevation. For example 10 meters below a lake surface with given elevation. Supports range queries, smaller,larger (e.g., '5,30', whereas '30,5' wouldn't work)

institutionCode

An identifier of any form assigned by the source to identify the institution the record belongs to. Not guaranteed to be que.

collectionCode

An identifier of any form assigned by the source to identify the physical collection or digital dataset uniquely within the text of an institution.

hasGeospatialIssue

(logical) Includes/excludes occurrence records which contain spatial issues (as determined in our record interpretation), i.e. hasGeospatialIssue=TRUE returns only those records with spatial issues while hasGeospatialIssue=FALSE includes only records without spatial issues. The absence of this parameter returns any record with or without spatial issues.

issue

(character) One or more of many possible issues with each occurrence record. See Details. Issues passed to this parameter filter results by the issue.

search

Query terms. The value for this parameter can be a simple word or a phrase.

mediaType

Media type. Default is NULL, so no filtering on mediatype. Options: NULL, 'MovingImage', 'Sound', and 'StillImage'.

subgenusKey

(numeric) Subgenus classification key.

repatriated

(character) Searches for records whose publishing country is different to the country where the record was recorded in.

phylumKey

(numeric) Phylum classification key.

kingdomKey

(numeric) Kingdom classification key.

classKey

(numeric) Class classification key.

orderKey

(numeric) Order classification key.

familyKey

(numeric) Family classification key.

genusKey

(numeric) Genus classification key.

establishmentMeans

(character) EstablishmentMeans, possible values include: INTRODUCED, INVASIVE, MANAGED, NATIVE, NATURALISED, UNCERTAIN

protocol

(character) Protocol or mechanism used to provide the occurrence record. See Details for possible values

(character) The type license applied to the dataset or record. Possible values: CC0_1_0, CC_BY_4_0, CC_BY_NC_4_0, UNSPECIFIED, and UNSUPPORTED

organismId

(numeric) An identifier for the Organism instance (as opposed to a particular digital record of the Organism). May be a globally unique identifier or an identifier specific to the data set.

publishingOrg

(character) The publishing organization key (a UUID).

stateProvince

(character) The name of the next smaller administrative region than country (state, province, canton, department, region, etc.) in which the Location occurs.

waterBody

(character) The name of the water body in which the locations occur

locality

(character) The specific description of the place.

limit

Number of records to return. Default: 500. Note that the per request maximum is 300, but since we set it at 500 for the function, we do two requests to get you the 500 records (if there are that many). Note that there is a hard maximum of 100,000, which is calculated as the limit+start, so start=99,000 and limit=2000 won't work

start

Record number to start at. Use in combination with limit to page through results. Note that we do the paging internally for you, but you can manually set the start parameter

fields

(character) Default ('all') returns all fields. 'minimal' returns just taxon name, key, latitude, and longitute. Or specify each field you want returned by name, e.g. fields = c('name','latitude','elevation').

return

Defunct. All components (meta, hierarchy, data, media, facets) are returned now; index to the one(s) you want. See occ_data() if you just want the data component

facet

(character) a character vector of length 1 or greater. Required.

facetMincount

(numeric) minimum number of records to be included in the faceting results

facetMultiselect

(logical) Set to TRUE to still return counts for values that are not currently filtered. See examples. Default: FALSE

Faceting: All fields can be faceted on except for last "lastInterpreted", "eventDate", and "geometry"

You can do facet searches alongside searching occurrence data, and return both, or only return facets, or only occurrence data, etc.

skip_validate

(logical) whether to skip wicket::wkt_validate call or not. passed down to check_wkt(). Default: TRUE

curlopts

list of named curl options passed on to HttpClient. see curl::curl_options for curl options

...

##### Details

protocol parameter options:

• BIOCASE - A BioCASe protocl compliant service.

• DIGIR - A DiGIR service endpoint.

• DIGIR_MANIS - A DiGIR service slightly modified for the MANIS network.

• DWC_ARCHIVE - A Darwin Core Archive as defined by the Darwin Core Text Guidelines.

• EML - A single EML metadata document in any EML version.

• FEED - Syndication feeds like RSS or ATOM of various flavors.

• OAI_PMH - The Open Archives Initiative Protocol for Metadata Harvesting.

• OTHER - Any other service not covered by this enum so far.

• TAPIR - A TAPIR service.

• TCS_RDF - Taxon Concept data given as RDF based on the TDWG ontology.

• TCS_XML - A Taxon Concept Schema document.

• WFS - An OGC Web Feature Service.

• WMS - An OGC Web Map Service.

Multiple parmeters: Note that you can pass in a vector to one of taxonKey, scientificName, datasetKey, catalogNumber, recordedBy, geometry, country, publishingCountry, recordNumber, search, institutionCode, collectionCode, decimalLatitude, decimalLongitude, depth, year, typeStatus, lastInterpreted, continent, or mediatype parameters in a function call, but not a vector >1 of these parameters at the same time

Hierarchies: hierarchies are returned wih each occurrence object. There is no option no to return them from the API. However, within the occ_search function you can select whether to return just hierarchies, just data, all of data and hiearchies and metadata, or just metadata. If all hierarchies are the same we just return one for you.

Data: By default only three data fields are returned: name (the species name), decimallatitude, and decimallongitude. Set parameter minimal=FALSE if you want more data.

Nerds: You can pass parameters not defined in this function into the call to the GBIF API to control things about the call itself using curlopts. See an example below that passes in the verbose function to get details on the http call.

Scientific names vs. taxon keys: In the previous GBIF API and the version of rgbif that wrapped that API, you could search the equivalent of this function with a species name, which was convenient. However, names are messy right. So it sorta makes sense to sort out the species key numbers you want exactly, and then get your occurrence data with this function. GBIF has added a parameter scientificName to allow searches by scientific names in this function - which includes synonym taxa. Note: that if you do use the scientificName parameter, we will check internally that it's not a synonym of an accepted name, and if it is, we'll search on the accepted name. If you want to force searching by a synonym do so by finding the GBIF identifier first with any name_* functions, then pass that ID to the taxonKey parameter.

WKT: Examples of valid WKT objects:

• 'POLYGON((-19.5 34.1, 27.8 34.1, 35.9 68.1, -25.3 68.1, -19.5 34.1))'

• 'MULTIPOLYGON(((-123 38,-116 38,-116 43,-123 43,-123 38)),((-97 41,-93 41,-93 45,-97 45,-97 41)))'

• 'POINT(-120 40)'

• 'LINESTRING(3 4,10 50,20 25)'

• 'LINEARRING' ???' - Not sure how to specify this. Anyone?

Note that GBIF expects counter-clockwise winding order for WKT. You can supply clockwise WKT, but GBIF treats it as an exclusion, so you get all data not inside the WKT area. occ_download() behaves differently in that you should simply get no data back at all with clockwise WKT.

Long WKT: Options for handling long WKT strings: Note that long WKT strings are specially handled when using occ_search or occ_data. Here are the three options for long WKT strings (> 1500 characters), set one of these three via the parameter geom_big:

• asis - the default setting. This means we don't do anything internally. That is, we just pass on your WKT string just as we've done before in this package.

• axe - this option uses the geoaxe package to chop up your WKT string in to many polygons, which then leads to a separate data request for each polygon piece, then we combine all dat back together to give to you. Note that if your WKT string is not of type polygon, we drop back to asisas there's no way to chop up linestrings, etc. This option will in most cases be slower than the other two options. However, this polygon splitting approach won't have the problem of the disconnect between how many records you want and what you actually get back as with the bbox option.

This method uses chop, which uses GridTopologyfrom the sp package, which has two parameters cellsize and cells.dim that we use to chop up polygons. You can tweak those parameters here by tweaking geom_size and geom_n. geom_size seems to be more useful in toggling the number of WKT strings you get back.

See wkt_parse to manually break make WKT bounding box from a larger WKT string, or break a larger WKT string into many smaller ones.

• bbox - this option checks whether your WKT string is longer than 1500 characters, and if it is we create a bounding box from the WKT, do the GBIF search with that bounding box, then prune the resulting data to only those occurrences in your original WKT string. There is a big caveat however. Because we create a bounding box from the WKT, and the limit parameter determines some subset of records to get, then when we prune the resulting data to the WKT, the number of records you get could be less than what you set with your limit parameter. However, you could set the limit to be high enough so that you get all records back found in that bounding box, then you'll get all the records available within the WKT.

Range queries: A range query is as it sounds - you query on a range of values defined by a lower and upper limit. Do a range query by specifying the lower and upper limit in a vector like depth='50,100'. It would be more R like to specify the range in a vector like c(50,100), but that sort of syntax allows you to do many searches, one for each element in the vector - thus range queries have to differ. The following parameters support range queries.

• decimalLatitude

• decimalLongitude

• depth

• elevation

• eventDate

• lastInterpreted

• month

• year

Issue: The options for the issue parameter (from https://gbif.github.io/gbif-api/apidocs/org/gbif/api/vocabulary/OccurrenceIssue.html):

• BASIS_OF_RECORD_INVALID The given basis of record is impossible to interpret or seriously different from the recommended vocabulary.

• CONTINENT_COUNTRY_MISMATCH The interpreted continent and country do not match up.

• CONTINENT_DERIVED_FROM_COORDINATES The interpreted continent is based on the coordinates, not the verbatim string information.

• CONTINENT_INVALID Uninterpretable continent values found.

• COORDINATE_INVALID Coordinate value given in some form but GBIF is unable to interpret it.

• COORDINATE_OUT_OF_RANGE Coordinate has invalid lat/lon values out of their decimal max range.

• COORDINATE_REPROJECTED The original coordinate was successfully reprojected from a different geodetic datum to WGS84.

• COORDINATE_REPROJECTION_FAILED The given decimal latitude and longitude could not be reprojected to WGS84 based on the provided datum.

• COORDINATE_REPROJECTION_SUSPICIOUS Indicates successful coordinate reprojection according to provided datum, but which results in a datum shift larger than 0.1 decimal degrees.

• COORDINATE_ROUNDED Original coordinate modified by rounding to 5 decimals.

• COUNTRY_COORDINATE_MISMATCH The interpreted occurrence coordinates fall outside of the indicated country.

• COUNTRY_DERIVED_FROM_COORDINATES The interpreted country is based on the coordinates, not the verbatim string information.

• COUNTRY_INVALID Uninterpretable country values found.

• COUNTRY_MISMATCH Interpreted country for dwc:country and dwc:countryCode contradict each other.

• DEPTH_MIN_MAX_SWAPPED Set if supplied min>max

• DEPTH_NON_NUMERIC Set if depth is a non numeric value

• DEPTH_NOT_METRIC Set if supplied depth is not given in the metric system, for example using feet instead of meters

• DEPTH_UNLIKELY Set if depth is larger than 11.000m or negative.

• ELEVATION_MIN_MAX_SWAPPED Set if supplied min > max elevation

• ELEVATION_NON_NUMERIC Set if elevation is a non numeric value

• ELEVATION_NOT_METRIC Set if supplied elevation is not given in the metric system, for example using feet instead of meters

• ELEVATION_UNLIKELY Set if elevation is above the troposphere (17km) or below 11km (Mariana Trench).

• GEODETIC_DATUM_ASSUMED_WGS84 Indicating that the interpreted coordinates assume they are based on WGS84 datum as the datum was either not indicated or interpretable.

• GEODETIC_DATUM_INVALID The geodetic datum given could not be interpreted.

• IDENTIFIED_DATE_INVALID The date given for dwc:dateIdentified is invalid and cant be interpreted at all.

• IDENTIFIED_DATE_UNLIKELY The date given for dwc:dateIdentified is in the future or before Linnean times (1700).

• MODIFIED_DATE_INVALID A (partial) invalid date is given for dc:modified, such as a non existing date, invalid zero month, etc.

• MODIFIED_DATE_UNLIKELY The date given for dc:modified is in the future or predates unix time (1970).

• MULTIMEDIA_DATE_INVALID An invalid date is given for dc:created of a multimedia object.

• MULTIMEDIA_URI_INVALID An invalid uri is given for a multimedia object.

• PRESUMED_NEGATED_LATITUDE Latitude appears to be negated, e.g. 32.3 instead of -32.3

• PRESUMED_NEGATED_LONGITUDE Longitude appears to be negated, e.g. 32.3 instead of -32.3

• PRESUMED_SWAPPED_COORDINATE Latitude and longitude appear to be swapped.

• RECORDED_DATE_INVALID A (partial) invalid date is given, such as a non existing date, invalid zero month, etc.

• RECORDED_DATE_MISMATCH The recording date specified as the eventDate string and the individual year, month, day are contradicting.

• RECORDED_DATE_UNLIKELY The recording date is highly unlikely, falling either into the future or represents a very old date before 1600 that predates modern taxonomy.

• REFERENCES_URI_INVALID An invalid uri is given for dc:references.

• TAXON_MATCH_FUZZY Matching to the taxonomic backbone can only be done using a fuzzy, non exact match.

• TAXON_MATCH_HIGHERRANK Matching to the taxonomic backbone can only be done on a higher rank and not the scientific name.

• TAXON_MATCH_NONE Matching to the taxonomic backbone cannot be done cause there was no match at all or several matches with too little information to keep them apart (homonyms).

• TYPE_STATUS_INVALID The given type status is impossible to interpret or seriously different from the recommended vocabulary.

• ZERO_COORDINATE Coordinate is the exact 0/0 coordinate, often indicating a bad null coordinate.

Counts: There is a slight difference in the way records are counted here vs. results from occ_count. For equivalent outcomes, in this function use hasCoordinate=TRUE, and hasGeospatialIssue=FALSE to have the same outcome using occ_count with isGeoreferenced=TRUE

##### Value

An object of class gbif, which is a S3 class list, with slots for metadata (meta), the occurrence data itself (data), the taxonomic hierarchy data (hier), and media metadata (media). In addition, the object has attributes listing the user supplied arguments and whether it was a 'single' or 'many' search; that is, if you supply two values of the datasetKey parameter to searches are done, and it's a 'many'. meta is a list of length four with offset, limit, endOfRecords and count fields. data is a tibble (aka data.frame). hier is a list of data.frames of the unique set of taxa found, where each data.frame is its taxonomic classification. media is a list of media objects, where each element holds a set of metadata about the media object.

##### Note

Maximum number of records you can get with this function is 100,000. See https://www.gbif.org/developer/occurrence

##### References

https://www.gbif.org/developer/occurrence#search

downloads(), occ_data(), occ_facet()

• occ_search
##### Examples
# NOT RUN {
# Search by species name, using \code{\link{name_backbone}} first to get key
(key <- name_suggest(q='Helianthus annuus', rank='species')$data$key[1])
occ_search(taxonKey=key, limit=2)

# Return 20 results, this is the default by the way
occ_search(taxonKey=key, limit=20)

# Get just metadata
occ_search(taxonKey=key, limit=0)$meta # Instead of getting a taxon key first, you can search for a name directly ## However, note that using this approach (with \code{scientificName="..."}) ## you are getting synonyms too. The results for using \code{scientifcName} and ## \code{taxonKey} parameters are the same in this case, but I wouldn't be surprised if for some ## names they return different results occ_search(scientificName = 'Ursus americanus') key <- name_backbone(name = 'Ursus americanus', rank='species')$usageKey
occ_search(taxonKey = key)

# Search by dataset key
occ_search(datasetKey='7b5d6a48-f762-11e1-a439-00145eb45e9a', limit=20)$data # Search by catalog number occ_search(catalogNumber="49366", limit=20) ## separate requests: use a vector of strings occ_search(catalogNumber=c("49366","Bird.27847588"), limit=10) ## one request, many instances of same parameter: use semi-colon sep. string occ_search(catalogNumber="49366;Bird.27847588", limit=10) # Get all data, not just lat/long and name occ_search(taxonKey=key, fields='all', limit=20) # Or get specific fields. Note that this isn't done on GBIF's side of things. This # is done in R, but before you get the return object, so other fields are garbage # collected occ_search(taxonKey=key, fields=c('name','basisOfRecord','protocol'), limit=20) # Use paging parameters (limit and start) to page. Note the different results # for the two queries below. occ_search(datasetKey='7b5d6a48-f762-11e1-a439-00145eb45e9a',start=10,limit=5)$data
occ_search(datasetKey='7b5d6a48-f762-11e1-a439-00145eb45e9a',start=20,limit=5)$data # Many dataset keys ## separate requests: use a vector of strings occ_search(datasetKey=c("50c9509d-22c7-4a22-a47d-8c48425ef4a7", "7b5d6a48-f762-11e1-a439-00145eb45e9a"), limit=20) ## one request, many instances of same parameter: use semi-colon sep. string v="50c9509d-22c7-4a22-a47d-8c48425ef4a7;7b5d6a48-f762-11e1-a439-00145eb45e9a" occ_search(datasetKey = v, limit=20) # Occurrence data: lat/long data, and associated metadata with occurrences ## The data slot has a data.frame of all data together ## for easy manipulation occ_search(taxonKey=key, limit=20)$data

# Taxonomic hierarchy data
## In the hier slot
occ_search(taxonKey=key, limit=10)$hier # Search by recorder occ_search(recordedBy="smith", limit=20) # Many collector names occ_search(recordedBy=c("smith","BJ Stacey"), limit=20) # recordedByID occ_search(recordedByID="https://orcid.org/0000-0003-1691-239X", limit=20) # identifiedByID occ_search(identifiedByID="https://orcid.org/0000-0003-4710-2648", limit=20) # Pass in curl options for extra fun occ_search(taxonKey=2433407, limit=20, curlopts=list(verbose=TRUE))$hier
occ_search(taxonKey=2433407, limit=20,
curlopts = list(
noprogress = FALSE,
progressfunction = function(down, up) {
cat(sprintf("up: %d | down %d\n", up, down))
return(TRUE)
}
)
)$hier # occ_search(taxonKey=2433407, limit=20, # curlopts = list(timeout_ms = 1)) # Search for many species splist <- c('Cyanocitta stelleri', 'Junco hyemalis', 'Aix sponsa') keys <- sapply(splist, function(x) name_suggest(x)$data$key[1], USE.NAMES=FALSE) ## separate requests: use a vector of strings occ_search(taxonKey = keys, limit=5) ## one request, many instances of same parameter: use semi-colon sep. string occ_search(taxonKey = paste0(keys, collapse = ";"), limit=5) # Search using a synonym name # Note that you'll see a message printing out that the accepted name will be used occ_search(scientificName = 'Pulsatilla patens', fields = c('name','scientificName'), limit=5) # Search on latitidue and longitude occ_search(decimalLatitude=48, decimalLongitude=10) # Search on a bounding box ## in well known text format ### polygon occ_search(geometry='POLYGON((30.1 10.1,40 40,20 40,10 20,30.1 10.1))', limit=20) ### multipolygon wkt <- 'MULTIPOLYGON(((-123 38,-116 38,-116 43,-123 43,-123 38)), ((-97 41,-93 41,-93 45,-97 45,-97 41)))' occ_search(geometry = gsub("\n\\s+", "", wkt), limit = 20) ## taxonKey + WKT key <- name_suggest(q='Aesculus hippocastanum')$data$key[1] occ_search(taxonKey=key, geometry='POLYGON((30.1 10.1,40 40,20 40,10 20,30.1 10.1))', limit=20) ## or using bounding box, converted to WKT internally occ_search(geometry=c(-125.0,38.4,-121.8,40.9), limit=20) # Search on a long WKT string - too long for a GBIF search API request ## We internally convert your WKT string to a bounding box ## then do the query ## then clip the results down to just those in the original polygon ## - Alternatively, you can set the parameter geom_big="bbox" ## - An additional alternative is to use the GBIF download API, see ?downloads wkt <- "POLYGON((-9.178796777343678 53.22769021556159, -12.167078027343678 51.56540789297837, -12.958093652343678 49.78333685689162,-11.024499902343678 49.21251756301334, -12.079187402343678 46.68179685941719,-15.067468652343678 45.83103608186854, -15.770593652343678 43.58271629699817,-15.067468652343678 41.57676278827219, -11.815515527343678 40.44938999172728,-12.958093652343678 37.72112962230871, -11.639734277343678 36.52987439429357,-8.299890527343678 34.96062625095747, -8.739343652343678 32.62357394385735,-5.223718652343678 30.90497915232165, 1.1044063476563224 31.80562077746643,1.1044063476563224 30.754036557416256, 6.905187597656322 32.02942785462211,5.147375097656322 32.99292810780193, 9.629796972656322 34.164474406524725,10.860265722656322 32.91918014319603, 14.551671972656322 33.72700959356651,13.409093847656322 34.888564192275204, 16.748937597656322 35.104560368110114,19.561437597656322 34.81643887792552, 18.594640722656322 36.38849705969625,22.989171972656322 37.162874858929854, 19.825109472656322 39.50651757842751,13.760656347656322 38.89353140585116, 14.112218847656322 42.36091601976124,10.596593847656322 41.11488736647705, 9.366125097656322 43.70991402658437,5.059484472656322 42.62015372417812, 2.3348750976563224 45.21526500321446,-0.7412967773436776 46.80225692528942, 6.114171972656322 47.102229890207894,8.047765722656322 45.52399303437107, 12.881750097656322 48.22681126957933,9.190343847656322 48.693079457106684, 8.750890722656322 50.68283120621287,5.059484472656322 50.40356146487845, 4.268468847656322 52.377558897655156,1.4559688476563224 53.28027243658647, 0.8407344726563224 51.62000971578333,0.5770625976563224 49.32721423860726, -2.5869999023436776 49.49875947592088,-2.4991092773436776 51.18135535408638, -2.0596561523436776 52.53822562473851,-4.696374902343678 51.67454591918756, -5.311609277343678 50.009802108095776,-6.629968652343678 48.75106196817059, -7.684656152343678 50.12263634382465,-6.190515527343678 51.83776110910459, -5.047937402343678 54.267098895684235,-6.893640527343678 53.69860705549198, -8.915124902343678 54.77719740243195,-12.079187402343678 54.52294465763567, -13.573328027343678 53.437631551347174, -11.288171777343678 53.48995552517918, -9.178796777343678 53.22769021556159))" wkt <- gsub("\n", " ", wkt) #### Default option with large WKT string fails # res <- occ_search(geometry = wkt) #### if WKT too long, with 'geom_big=bbox': makes into bounding box res <- occ_search(geometry = wkt, geom_big = "bbox")$data
library("rgeos")
library("sp")
plot(wktsp)
coordinates(res) <- ~decimalLongitude+decimalLatitude
points(res)

#### Or, use 'geom_big=axe'
(res <- occ_search(geometry = wkt, geom_big = "axe"))
##### manipulate essentially number of polygons that result, so number of requests
###### default geom_size is 40
###### fewer calls
(res <- occ_search(geometry = wkt, geom_big = "axe", geom_size=50))
###### more calls
(res <- occ_search(geometry = wkt, geom_big = "axe", geom_size=30))

# Search on country
occ_search(country='US', fields=c('name','country'), limit=20)
isocodes[grep("France", isocodes$name),"code"] occ_search(country='FR', fields=c('name','country'), limit=20) occ_search(country='DE', fields=c('name','country'), limit=20) ### separate requests: use a vector of strings occ_search(country=c('US','DE'), limit=20) ### one request, many instances of same parameter: use semi-colon sep. string occ_search(country = 'US;DE', limit=20) # Get only occurrences with lat/long data occ_search(taxonKey=key, hasCoordinate=TRUE, limit=20) # Get only occurrences that were recorded as living specimens occ_search(taxonKey=key, basisOfRecord="LIVING_SPECIMEN", hasCoordinate=TRUE, limit=20) # Get occurrences for a particular eventDate occ_search(taxonKey=key, eventDate="2013", limit=20) occ_search(taxonKey=key, year="2013", limit=20) occ_search(taxonKey=key, month="6", limit=20) # Get occurrences based on depth key <- name_backbone(name='Salmo salar', kingdom='animals')$speciesKey
occ_search(taxonKey=key, depth="5", limit=20)

# Get occurrences based on elevation
key <- name_backbone(name='Puma concolor', kingdom='animals')$speciesKey occ_search(taxonKey=key, elevation=50, hasCoordinate=TRUE, limit=20) # Get occurrences based on institutionCode occ_search(institutionCode="TLMF", limit=20) ### separate requests: use a vector of strings occ_search(institutionCode=c("TLMF","ArtDatabanken"), limit=20) ### one request, many instances of same parameter: use semi-colon sep. string occ_search(institutionCode = "TLMF;ArtDatabanken", limit=20) # Get occurrences based on collectionCode occ_search(collectionCode="Floristic Databases MV - Higher Plants", limit=20) occ_search(collectionCode=c("Floristic Databases MV - Higher Plants","Artport")) # Get only those occurrences with spatial issues occ_search(taxonKey=key, hasGeospatialIssue=TRUE, limit=20) # Search using a query string occ_search(search = "kingfisher", limit=20) # search on repatriated - doesn't work right now # occ_search(repatriated = "") # search on phylumKey occ_search(phylumKey = 7707728, limit = 5) # search on kingdomKey occ_search(kingdomKey = 1, limit = 5) # search on classKey occ_search(classKey = 216, limit = 5) # search on orderKey occ_search(orderKey = 7192402, limit = 5) # search on familyKey occ_search(familyKey = 3925, limit = 5) # search on genusKey occ_search(genusKey = 1935496, limit = 5) # search on establishmentMeans occ_search(establishmentMeans = "INVASIVE", limit = 5) occ_search(establishmentMeans = "NATIVE", limit = 5) occ_search(establishmentMeans = "UNCERTAIN", limit = 5) # search on protocol occ_search(protocol = "DIGIR", limit = 5) # search on license occ_search(license = "CC_BY_4_0", limit = 5) # search on organismId occ_search(organismId = "100", limit = 5) # search on publishingOrg occ_search(publishingOrg = "28eb1a3f-1c15-4a95-931a-4af90ecb574d", limit = 5) # search on stateProvince occ_search(stateProvince = "California", limit = 5) # search on waterBody occ_search(waterBody = "AMAZONAS BASIN, RIO JURUA", limit = 5) # search on locality res <- occ_search(locality = c("Trondheim", "Hovekilen"), limit = 5) res$Trondheim$data res$Hovekilen$data # Range queries ## See Detail for parameters that support range queries occ_search(depth='50,100') # this is a range depth, with lower/upper limits in character string occ_search(depth=c(50,100)) # this is not a range search, but does two searches for each depth ## Range search with year occ_search(year='1999,2000', limit=20) ## Range search with latitude occ_search(decimalLatitude='29.59,29.6') # Search by specimen type status ## Look for possible values of the typeStatus parameter looking at the typestatus dataset occ_search(typeStatus = 'allotype', fields = c('name','typeStatus')) # Search by specimen record number ## This is the record number of the person/group that submitted the data, not GBIF's numbers ## You can see that many different groups have record number 1, so not super helpful occ_search(recordNumber = 1, fields = c('name','recordNumber','recordedBy')) # Search by last time interpreted: Date the record was last modified in GBIF ## The lastInterpreted parameter accepts ISO 8601 format dates, including ## yyyy, yyyy-MM, yyyy-MM-dd, or MM-dd. Range queries are accepted for lastInterpreted occ_search(lastInterpreted = '2014-04-02', fields = c('name','lastInterpreted')) # Search by continent ## One of africa, antarctica, asia, europe, north_america, oceania, or south_america occ_search(continent = 'south_america')$meta
occ_search(continent = 'africa')$meta occ_search(continent = 'oceania')$meta
occ_search(continent = 'antarctica')$meta # Search for occurrences with images occ_search(mediaType = 'StillImage')$media
occ_search(mediaType = 'MovingImage')$media occ_search(mediaType = 'Sound')$media

# Query based on issues - see Details for options
## one issue
occ_search(taxonKey=1, issue='DEPTH_UNLIKELY', fields =
c('name','key','decimalLatitude','decimalLongitude','depth'))
## two issues
occ_search(taxonKey=1, issue=c('DEPTH_UNLIKELY','COORDINATE_ROUNDED'))
# Show all records in the Arizona State Lichen Collection that cant be matched to the GBIF
# backbone properly:
occ_search(datasetKey='84c0e1a0-f762-11e1-a439-00145eb45e9a',
issue=c('TAXON_MATCH_NONE','TAXON_MATCH_HIGHERRANK'))

# Parsing output by issue
(res <- occ_search(geometry='POLYGON((30.1 10.1,40 40,20 40,10 20,30.1 10.1))', limit = 50))
## what do issues mean, can print whole table, or search for matches
gbif_issues()[ gbif_issues()$code %in% c('cdround','cudc','gass84','txmathi'), ] ## or parse issues in various ways ### remove data rows with certain issue classes library('magrittr') res %>% occ_issues(gass84) ### split issues into separate columns res %>% occ_issues(mutate = "split") ### expand issues to more descriptive names res %>% occ_issues(mutate = "expand") ### split and expand res %>% occ_issues(mutate = "split_expand") ### split, expand, and remove an issue class res %>% occ_issues(-cudc, mutate = "split_expand") # If you try multiple values for two different parameters you are wacked on the hand # occ_search(taxonKey=c(2482598,2492010), recordedBy=c("smith","BJ Stacey")) # Get a lot of data, here 1500 records for Helianthus annuus # out <- occ_search(taxonKey=key, limit=1500) # nrow(out$data)

# If you pass in an invalid polygon you get hopefully informative errors

### the WKT string is fine, but GBIF says bad polygon
wkt <- 'POLYGON((-178.59375 64.83258989321493,-165.9375 59.24622380205539,
-147.3046875 59.065977905449806,-130.78125 51.04484764446178,-125.859375 36.70806354647625,
-112.1484375 23.367471303759686,-105.1171875 16.093320185359257,-86.8359375 9.23767076398516,
-82.96875 2.9485268155066175,-82.6171875 -14.812060061226388,-74.8828125 -18.849111862023985,
-77.34375 -47.661687803329166,-84.375 -49.975955187343295,174.7265625 -50.649460483096114,
179.296875 -42.19189902447192,-176.8359375 -35.634976650677295,176.8359375 -31.835565983656227,
163.4765625 -6.528187613695323,152.578125 1.894796132058301,135.703125 4.702353722559447,
127.96875 15.077427674847987,127.96875 23.689804541429606,139.921875 32.06861069132688,
149.4140625 42.65416193033991,159.2578125 48.3160811030533,168.3984375 57.019804336633165,
178.2421875 59.95776046458139,-179.6484375 61.16708631440347,-178.59375 64.83258989321493))'

# occ_search(geometry = gsub("\n", '', wkt))

### unable to parse due to last number pair needing two numbers, not one
# wkt <- 'POLYGON((-178.5 64.8,-165.9 59.2,-147.3 59.0,-130.7 51.0,-125.8))'
# occ_search(geometry = wkt)

### unable to parse due to unclosed string
# wkt <- 'POLYGON((-178.5 64.8,-165.9 59.2,-147.3 59.0,-130.7 51.0))'
# occ_search(geometry = wkt)
### another of the same
# wkt <- 'POLYGON((-178.5 64.8,-165.9 59.2,-147.3 59.0,-130.7 51.0,-125.8 36.7))'
# occ_search(geometry = wkt)

# wkt <- 'LINESTRING(3 4,10 50,20 25)'
# occ_search(geometry = wkt)

### Apparently a point is allowed, but errors
# wkt <- 'POINT(45 -122)'
# occ_search(geometry = wkt)

## Faceting
x <- occ_search(facet = "country", limit = 0)
x$facets x <- occ_search(facet = "establishmentMeans", limit = 10) x$facets
x$data x <- occ_search(facet = c("country", "basisOfRecord"), limit = 10) x$data
x$facets x$facets$country x$facets$basisOfRecord x$facets$basisOfRecord$count
x <- occ_search(facet = "country", facetMincount = 30000000L, limit = 10)
x$facets x$data
# paging per each faceted variable
(x <- occ_search(
facet = c("country", "basisOfRecord", "hasCoordinate"),
country.facetLimit = 3,
basisOfRecord.facetLimit = 6,
limit = 0
))
x$facets # You can set limit=0 to get number of results found occ_search(datasetKey = '7b5d6a48-f762-11e1-a439-00145eb45e9a', limit = 0)$meta
occ_search(scientificName = 'Ursus americanus', limit = 0)$meta occ_search(scientificName = 'Ursus americanus', limit = 0)$meta
# }

Documentation reproduced from package rgbif, version 3.3.0, License: MIT + file LICENSE

### Community examples

Looks like there are no examples yet.