Count unique species

[ecology-data-science]

Hello,

I have a quick question about rgbif - I thought I'd post here before potentially reinventing the wheel. The problem can be summarised as follows: I would like to define a bounding box around an area of interest, query GBIF and count the number of unique species that fall within that region, ideally across all taxa. In short, I would like to calculate a very rough and ready measure of species richness within an area.

Is there a reasonably straightforward way of doing this using rgbif (or perhaps spocc)?

Thanks in advance.

[dnoesgaard]

GBIF has recently introduced the ability to download a list of unique taxa in any occurrence query. Not sure this has been added to rgbif (needs "format": "SPECIES_LIST" in the query)—correct me if I'm wrong.

However, it seems rgbif is fine with importing these lists, e.g. did a simple bounding box of an island (http://api.gbif.org/v1/occurrence/download/0014128-190320150433242) and imported the download here:

> dl<-occ_download_get("0014128-190320150433242")
Download file size: 0.04 MB
On disk at ./0014128-190320150433242.zip
> occ_download_import(dl)
# A tibble: 801 x 21
   taxonKey scientificName acceptedTaxonKey acceptedScienti… numberOfOccurre… taxonRank taxonomicStatus kingdom kingdomKey phylum phylumKey class classKey order orderKey
      <int> <chr>                     <int> <chr>                       <int> <chr>     <chr>           <chr>        <int> <chr>      <int> <chr>    <int> <chr>    <int>
 1  7570897 Spergularia m…          7570897 Spergularia med…               25 SPECIES   ACCEPTED        Plantae          6 Trach…   7707728 Magn…      220 Cary…      422
 2  2481618 Stercorarius …          2481618 Stercorarius sk…                3 SPECIES   ACCEPTED        Animal…          1 Chord…        44 Aves       212 Char…  7192402
 3  3807718 Cerastium sub…          3807718 Cerastium subte…                4 SPECIES   ACCEPTED        Plantae          6 Trach…   7707728 Magn…      220 Cary…      422
 4  2252458 Terebratulina…          2252458 Terebratulina D…                1 GENUS     ACCEPTED        Animal…          1 Brach…       110 Rhyn…      210 Tere…      509
 5  4856677 Osculipora d'…          4856677 Osculipora d'Or…                1 GENUS     ACCEPTED        Animal…          1 Bryoz…        53 Sten…      226 Cycl…     1497
 6  7192432 Haematopus os…          7192432 Haematopus ostr…                2 SUBSPECI… ACCEPTED        Animal…          1 Chord…        44 Aves       212 Char…  7192402
 7  7473187 Draba verna L.          7473187 Draba verna L.                  4 SPECIES   ACCEPTED        Plantae          6 Trach…   7707728 Magn…      220 Bras…  7225535
 8  2480242 Vanellus vane…          2480242 Vanellus vanell…              225 SPECIES   ACCEPTED        Animal…          1 Chord…        44 Aves       212 Char…  7192402
 9  2480281 Charadrius mo…          2480281 Charadrius mori…               14 SPECIES   ACCEPTED        Animal…          1 Chord…        44 Aves       212 Char…  7192402
10  2480726 Pandion halia…          2480726 Pandion haliaet…               60 SPECIES   ACCEPTED        Animal…          1 Chord…        44 Aves       212 Acci…  7191147

Note, these lists also contain records of organisms identified to genus and higher. So you'll have to filter, if only interested in species.

[sckott]

thanks for your question @simontarr-aecom and thanks for your response @dnoesgaard

looks as though we don't support species list just yet, working on it now

[sckott]

reinstall remotes::install_github("ropensci/rgbif@download-format"), reload R, then try for example:

geometry <- "POLYGON((30.1 10.1,40 40,20 40,10 20,30.1 10.1))"
res <- occ_download(paste0("geometry within ", geometry), type = "within", format = "SPECIES_LIST")
res 
#> <<gbif download>>
#>   Username: sckott
#>   E-mail: myrmecocystus@gmail.com
#>   Format: SPECIES_LIST
#>   Download key: 0000182-190415153152247

occ_download_meta(res) # see when ready

x <- occ_download_get(res) # download
#> Download file size: 4.73 MB
#> On disk at ./0000182-190415153152247.zip

occ_download_import(x) # import into R

it's important to do geometry within <WKT> instead of geometry = <WKT>

[ecology-data-science]

Hi @dnoesgaard @sckott

Thank you both for your help with this and for applying a solution so quickly. This is really useful stuff and will save me a ton of time :)

[sckott]

@simontarr-aecom great, glad it works and that it will save you time

[simon-tarr]

Hi @sckott I know this issue is now closed but do you know of an easy way to filter out extinct species using the above method? A few Irish Elk have crept into my downloaded dataframe! I can't see an obvious way of filtering out extinct species.

[sckott]

having a look.

[sckott]

i think maybe occurrence API use to have an extinct parameter to toggle those records, but no longer exists. they do have an isExtinct parameter in the name lookup method, e.g., name_lookup(isExtinct=TRUE, limit=10)

@dnoesgaard is there any way to filter by exinction status with the occurrence download API? AFAIK there is not. there is isExtinct on the /species/search route, but i dont think there's a way to pass a taxon key to that route so e.g. a user could get a species list with the download API , then check extinction status with /species/search. And extinction status doesn't seem to be in the data for an extinct species e.g., http://api.gbif.org/v1/species/104070564