As data repositories strive to meet the FAIR data principles and to solve the reproducibility crisis in science, it is essential to implement a comprehensive solution for data citation with persistent identifiers. Despite this imperative need, data citation is a fairly new effort in the research community that requires technical and cultural change. Fortunately, funders and journal publishers are now recognizing datasets as first class research outputs, accelerating the demand for data citations.

To address this challenge, Ocean Networks Canada (ONC) has undertaken an enormous effort to allocate Digital Object Identifiers (DOIs) to our dynamic datasets. Often DOIs are assigned to research objects after work is complete, when the object is static and unlikely to change. However, ONC provides access to near real-time data, such that citation capabilities are needed before the end of an instrument deployment and data is still accumulating. Moreover, datasets are versioned as needed for corrections and enhancements. Assigning DOIs for dynamic data has several challenges.

This session will describe the technical implementation of DOIs for the substantial dataset collection in ONC’s Oceans 3.0 digital infrastructure, as well as how we are approaching educational and cultural changes. While we leveraged the Research Data Alliance Data Citation Working Group recommendations to guide our efforts, there were few examples to emulate since we were early adopters; thus, we aim to share the insights we gained through our decision-making and implementation.

We will introduce the concept of data citation, including persistent identifiers (PIDs) and the benefits to all stakeholders. We will then provide a detailed overview of the Oceans 3.0 implementation, including approaches for dataset metadata, granularity, and versioning, and access to citation information within tools and data products.

Speakers