Dataset Desmodes overview

[DimEvil]

• [ ] InstitutionID was not used correctly. Now it forms the basis of occurrenceID as the identifier of the unique records DwC info: DwC:institutionID --> new unique occurrenceID's generated [rowNumber; institutionID, institutionCode -> hash values
• [ ] scientificName --> Desmodus rotundus | Desmodus desmodus rotundus | Desmodus d. rotundus |Desmodus rotundus rotundus | Desmodus rotundus - exposição keep only Desmosus rotundus & Desmodes rotundus rotundus ?
• [ ] occurrenceRemarks --> republished from https://doi.org/10.6084/m9.figshare.18745316
• [ ] Data coming from Lee dataset (Lee DN, Papes M, Van Den Bussche RA. 2012. Present and potential future distribution of common vampire bats in the Americas and the associated risk to cattle. PLoS One 7:e42466.) also mentioned data downloaded from GBIF. Any means to identify those ones to avoid duplicates?
• [ ] in dwc:family also 'SUBSPECIES' as a value? (with institutionID 860.015.542-6)
• [ ] coordinateUncertaintyInMeters ranges from 0,04 to 262883
• [ ] Tried to establish relation with already published GBIF records by resourceRelationShip
• [ ] I completed empty values in basisOfRecord with MaterialCitation & occurrence

occurrence %<>% mutate(dwc_basisOfRecord = case_when ( datasetName == 'Lee' ~ "MaterialCitation" ,datasetName == 'Zarza' ~ "MaterialCitation" ,basisOfRecord == 'UNKNOWN' ~ "occurrence" ,datasetName == 'literature' ~ "MaterialCitation" ,datasetName == 'Piaggio' ~ "MaterialCitation" ,datasetName == 'Juan Luis_Personal_Database' ~ "MaterialCitation" ,datasetName == 'Literature' ~ "MaterialCitation" ,datasetName == 'Bio_Diversi_Data_UY' ~ "MaterialCitation" ,datasetName == 'PCMS' ~ "MaterialCitation" ,datasetName == 'Literature' ~ "MaterialCitation" ,datasetName == 'Streicker_Peru' ~ "MaterialCitation"
,datasetName == 'SAG' ~ "MaterialCitation" ,TRUE ~ basisOfRecord ))

• [ ] added IUCN as nomenclaturalCode
• [ ] added occurrenceStatus --> present
• [ ] changed locality_proper --> locality

[pvandevuurst]

• First task looks good, as it looks like you are assigning each occurrence a GBIF ID
• Scientific Name should be Desmodus rotundus
• Occurrence remark looks good, as that would direct to the original figshare database
• Lee paper does use occurrences from GBIF, but those should have already been removed (i.e., no duplicates). If you would like to exclude all occurrence points from the Lee paper, I believe that would also be acceptable. Let me know if that makes sense.
• If basis of record is not listed then it is unknown. I am unsure of what MaterialCitation signifies within the GBIF system, but I believe that would the most accurate description.
• Locality_proper to locality is fine, they are the same thing. Locality_proper is simply corrected for capitalization
• I am unsure of what you mean by dwc:family also ‘SUBSPECIES’? Family should be Phyllostomidae. However, if there is an error, please let me know.

[DimEvil]

• First task looks good, as it looks like you are assigning each occurrence a GBIF ID --> I'm assigning each occurrence a new unique ID, this is not the GBIF ID. In the original dataset I could not find GUID (Global Unique ID's, which we need for publication on GBIF. What came close was the institutionID (I guess this is the unique record number in the original database). DwC:institutionID is actually an ID for the institute, so I guess the term was not used correctly. If I download the data from https://doi.org/10.15468/dl.my64ap, I get the whole DwC (also all the empty columns) and the uniqueID can be found back in occurrenceID

• Scientific Name should be Desmodus rotundus

Ok, fixed

• Occurrence remark looks good, as that would direct to the original figshare dataset

ok

• Lee paper does use occurrences from GBIF, but those should have already been removed (i.e., no duplicates). If you would like to exclude all occurrence points from the Lee paper, I believe that would also be acceptable. Let me know if that makes sense.

If the Lee paper data here (Desmodus_dataset_Dec_2021.csv) is not already on GBIF , we will keep them in the dataset. :)

• If basis of record is not listed then it is unknown. I am unsure of what MaterialCitation signifies within the GBIF system, but I believe that would the most accurate description.

More info on the term can be found here: https://github.com/tdwg/dwc/issues/329 IT's referring to evidence of an occurrence from literature.

• Locality_proper to locality is fine, they are the same thing. Locality_proper is simply corrected for capitalization

ok

• I am unsure of what you mean by dwc:family also ‘SUBSPECIES’? Family should be Phyllostomidae. However, if there is an error, please let me know.

I think there is an error somewhere. Where family and taxonRank changed places.

Also here is something not correct, as we have 100% Desmodus rotundus, family should be the same in the whole dataset.

fixed

I also changed M & F in Dwc:sex to male and female as these are the controlled vocabulary terms. :)

[pvandevuurst]

Sounds good. Looks like two of the columns might have been switched at some point (family and taxonRank). Perhaps try downloading the most recent version from Figshare (https://figshare.com/articles/dataset/Desmodus_rotundus_Occurrence_Record_Database/15025296). (i.e., Desmodus_dataset_Apr_2022).

[DimEvil]

Some more questions, remarks in April 22 version

I have these values in basisOfRecord MachineHUMAN_OBSERVATIONbPRESERVED_SPECIMENervatiHUMAN_OBSERVATIONn HumanHUMAN_OBSERVATIONbPRESERVED_SPECIMENervatiHUMAN_OBSERVATIONn

Still some errors in family <-> taxonRank (I created fixed terms here now)

All taxonrank is now species (as all scientificName = Desmodus rotundis (no subspecies)

[pvandevuurst]

Okay, gotcha. Just went through and fixed those columns, so they should be okay now. For taxonRank species is fine, just make sure that Desmodus rotundus is spelled correctly and that should be good to go!

[pvandevuurst]

For the basis of record it looks like some things have been collated. Should be HUMAN_OBSERVATION, PRESERVED_SPECIMEN, or LIVE_SPECIMEN. That specific record is unknown, so it should be left black or listed as UNKNOWN. Does that make sense?