| layout | title |
|---|---|
page |
iNaturalist Case Study |
iNaturalist is a social network and online platform for sharing, mapping and identifying biodiversity observations around the world. While collaboratively building a biodiversity database, they frame their primary aim as “connect[ing] people to nature”. iNaturalist began in 2008 as a Master’s final project at the UC Berkeley’s School of Information and is now a joint initiative between the California Academy of Sciences and the National Geographic Society, with almost 5.7 million registered users and over 130 million uploaded observations as of September 2022.
The common research object of iNaturalist is a biodiversity database comprising recorded sightings of species, complemented with metadata such as location, time, and photographs. The database is intended for non-commercial reuse, but contributors have the option to apply licenses for their own photos. While most database entries are visible to the public, the precise location might be obscured in some cases to protect endangered species or for privacy concerns.
In iNaturalist, participants mainly contribute either by uploading new observations, or by adding species annotations to existing observations. Additional tasks include writing journal posts, discussing in a forum, and creating data collection projects that encourage users to provide observations specific to certain areas or species. Users in the curator role can edit the taxonomy, create and edit wiki pages, resolve flags on taxa and user content, hide comments and suspend users for violating community guidelines.
In iNaturalist the main tasks are of small size and well-defined. The smallest task would be to annotate existing data points, followed by uploading new observations. Integration of the main modules is automatic: observations are automatically added to the database and provide the data pool for future annotations. Once the species has been collectively identified (see quality control), the observation is automatically added to the Global Biodiversity Information Facility (GBIF) database (GBIF: The Global Biodiversity Information Facility, 2022).
Contributors can select from the range of tasks mentioned above. They are free to decide the type and quantity of observations they would like to annotate or add to the database, without automatic assignment, and anyone can start a data-collection project on the platform. There is a curator role that can be granted by iNaturalist staff for users who have engaged continuously over several months: curators have additional rights, e.g. to edit iNaturalist’s taxonomy or to hide comments that violate community guidelines.
There is a communal validation process for species identification, using the following mechanism (Boone & Basille, 2019): New observations are automatically tagged with “needs ID”. As the observations and annotations are generally publicly accessible, users can see other users' annotations, and can agree, disagree, or make a guess more specific. For example, if the person uploading the observation specified it as “ladybird”, other users can either confirm this identification, specify what species of ladybird it is or disagree by identifying it as a different insect. The resulting annotation is called “community taxon” (see Figure A2.1. in Appendix A.2.). To achieve research grade, the ID needs to be supported by at least two users, and either have a community taxon at species level or lower. This research grade database is publicly accessible under an open license but some metadata, like specific location, can be obscured, to protect the privacy of underage users, or to protect endangered species. In addition to voting on the species, users can also add comments.
To learn how to contribute to the database, there is a “Getting started” guide with tutorials on uploading observations and species identification. Users are free to address bigger tasks, like creating their own projects, or offline data-collection events, called “bioblitz”, for which there are also help pages, or become curators after having collected experience through regular engagement.
There are features to coordinate work, like “needs ID” tag in the communal validation process, projects that call for specific observations, or filters to help users identify observations in areas of their expertise. For direct communication, there is a forum, as well as comment and blog functionality.