W07 - Improving research linking and visibility of specimens and physical samples through standardised identifiers and metadata

Session Type: Workshop
Full Title: W07 - Improving research linking and visibility of specimens and physical samples through standardised identifiers and metadata
Short Title: Standardising specimen identification
Organizer(s): Donald Hobern, GBIF Secretariat
Contributors: Rob Guralnick, Simon Cox, Kerstin Lehnert, Ramona Walls

Unsolicited contributions considered? Yes


Specimens are a key foundation to our knowledge of biodiversity. They are physical samples that represent and enable us to characterise biological entities. It is critically important to record the relationships of these specimens to occurrences of taxa and ecosystems, to observational systems used in scientific investigations (e.g. sites and facilities, site visits, campaigns, observation activities), and to the data and information products derived from these (images, datasets, reports, publications, assessments and recommendations). However, our ability to do this reliably is constrained by the lack of a generally adopted approach to specimen identification, which is compatible both with scholarly usage and with relevant IT systems. TDWG previously adopted Life Science Identifiers (LSIDs) as the recommended system for this purpose, but these have recently been deprecated. Successful universal identifier systems used in the scholarly community in general, or in some nearby disciplines include DOI (Digital Object Identifier), ORCID (Open Researcher and Contributor ID), BOLD (Barcode of Life Data Systems), each of which have specific strengths and limitations. While Darwin Core provides standardized metadata to describe biodiversity samples and standards such as DataCite can support citation and basic linking among identified objects, no existing infrastructure fully implements the ability to identify and link biodiversity specimens to the natural entities from which they were derived, or the physical and digital entities derived from them. The International GeoSample Number (IGSN) has a similar scope to the requirement of the biodiversity community, but until now has largely been confined to the earth sciences. An identifier system for biodiversity samples should be compatible with these to allow interoperability across discipline boundaries. This workshop is a follow-up to a symposium on ‘Linking environmental data and samples’ held in Canberra, May 2017 (Cox et al. 2018). The goal of this workshop is to discuss and converge on the requirements for long-term stable identification of specimens in biodiversity, evaluate existing solutions, and assess any progress that may have been made since a workshop held in conjunction with TDWG 2014 (Guralnick et al. 2015).