today

1. Biodiversity databases: basic concepts

2. DarwinCore standard

3. Biodiversity data cleaning

4. A workflow in R

Biodiversity data

Museums, herbaria, collections

• A massive amount of data

• New research areas from the analysis and synthesis of these data: Biogeography, macroecology

• However: heterogeneous quality, inequality between countries and institutions

Biodiversity databases

• GBIF (Global Biodiversity Information Facility)

• Country-wise biodiversity information systems:

  • 🇧🇷 speciesLink, Sistema de Informações sobre Biodiversidade Brasileira SiBBr
  • 🇨🇴 Sistema de Información sobre Biodiversidad SibColombia
  • 🇲🇽 Sistema Nacional de Información sobre Biodiversidad SNIB
  • 🇦🇷 Sistema de Información de Biodiversidad SIB

• Countries that participate via ministries, agencies

• Other systems:

  • Natureserve/ BISON in the USA
  • iNaturalist

By taxonomic group

• World Flora Online

• Plants of the world

• International Plant Name Index (IPNI)

• Fishbase

• Mammal Diversity

Transforming science through open data, software & reproducibility

• R packages to make data available, data cleaning, APIs

• Package peer-review

• Packages related to biodiversity data retrieval, manipulation, and standardization: finch, rgbif, taxize, taxview ...

Looking at GBIF data

Let's download the data of a tree species from South America Myrsine coriacea from the Primulaceae family.

library(rgbif)
library(dplyr)
species <- "Myrsine coriacea"
occs <- occ_search(scientificName = species)
names(occs)
#glimpse(occs)
#names(occs$data)Copy code 

Column names returned from gbif follow the DarwinCore standard (https://dwc.tdwg.org).

Possible fields

• Species (taxonomic information)

• Locality, coordinates, geographic

• Collector notes and species attributes, IUCN status

• Collectors, determiners, and authors' names

DarwinCore

Multi-file DwC scheme

• a ZIP file with:

  • different data tables (relational databases)
  • description file (XML)
  • metadata files (EML Ecological Metadata Language)

• common columns for each table, ID

• IPT is used to publish your data, either in one single table or extended: species lists, field work samplings etc. Datasets are given a permanent identifier and citation

Brazil Flora G (2020): Brazilian Flora 2020 project - Projeto Flora do Brasil 2020. v393.274. Instituto de Pesquisas Jardim Botanico do Rio de Janeiro. Dataset/Checklist. doi:10.15468/1mtkaw URL: http://ipt.jbrj.gov.br/jbrj/

• 49.343 species

• 136.314 taxa

• Information about: distribution, habitat, endemism, references, synonyms

• IPT in R: package finch (Chamberlain 2020)

Primary data are not clean or updated

• Names and classifications change in time (synonyms):

  • International Commission on Zoological Nomenclature (ICZN), International Code of Botanical Nomenclature, International Association for Plant Taxonomy

  • The names of places: toponyms, we need gazetteers

• Taxonomic identification errors, georreferencing, typographic/orthographic

• Missing coordinates or localities

• Imprecise coordinates, badly attributed

Spatial error

Yesson et al 2007

Zero latitude and longitude

Yesson et al 2007

Taxonomic, nomenclatural erros

Data cleaning workflow

Check for

1. format
2. completeness
3. sense (dimensions, ranges)
4. outliers (geographic, temporary, environmental)

Good practices:

• Always flag 🚩 additions, corrections, never modify original data

• Document all steps -> reproducibility

• Cite packages and data sources

Know your shortfalls

informed ignorance is a powerful research tool

References

Chapman, A. D. (2005). Principles and methods of data cleaning. GBIF.

Hortal, J., de Bello, F., Diniz-Filho, J. A. F., Lewinsohn, T. M., Lobo, J. M., & Ladle, R. J. (2015). Seven shortfalls that beset large-scale knowledge of biodiversity. Annual Review of Ecology, Evolution, and Systematics, 46, 523-549.

Yesson, C., Brewer, P. W., Sutton, T., Caithness, N., Pahwa, J. S., Burgess, M., ... & Culham, A. (2007). How global is the global biodiversity information facility?. PloS one, 2(11), e1124.

Wieczorek, J., Bloom, D., Guralnick, R., Blum, S., Döring, M., Giovanni, R., ... & Vieglais, D. (2012). Darwin Core: an evolving community-developed biodiversity data standard. PloS one, 7(1), e29715.