THE SPARROW LABORATORY INFORMATION MANAGEMENT SYSTEM — A TOOL FOR CONNECTING GEOCHEMICAL DATA TO CONTEXT AND COMMUNITY

Data sharing between geochemical laboratories is critically important for maintaining repeatable, comparable, and robust analytical workflows. Additionally, there is an increasing need across multiple geoscientific domains for large-scale access to standardized geochemical data. For instance, observation-driven digital Earth data resources such as Macrostrat and the Neotoma Paleoecology Database seek to calibrate ecological and stratigraphic proxies for environmental change against geochronology-driven assessments of geologic time; this process requires large-scale summaries of ages with basic geological context.

Archival facilities have been created to curate analytical data (e.g. Geochron.org) and samples (e.g. IGSN) at the community level. However, because most current geochronology laboratory information management systems are home-grown, labs often struggle with the manual work required to standardize their data products and contribute them to these centralized stores. Moreover, these workloads, already unsustainable, will further grow and evolve as more in-depth harmonization of data is sought between subfields of Earth science.

The Sparrow software project (https://sparrow-data.org), in development by a consortium of geochronology labs, is designed to improve laboratory data-management practices as an avenue towards supporting community data sharing. Sparrow is a simple data system, deployable at individual geochemistry labs, that augments current workflows for data collection and analysis. The system provides tools to manage metadata (e.g. projects, sample context, embargo status) and standard interfaces to automatically exchange data with outside facilities.

Sparrow is designed for flexibility, and we plan to support implementation by labs across the geochemical community. To date, Sparrow has been deployed at U-Pb, 40Ar/39Ar, SIMS, optically-stimulated luminescence, and cosmogenic nuclide facilities. Using examples from several early-stage implementations, we will show how the system eases management of data products, both for small labs and high-throughput facilities.