Broad comparisons of structural or functional traits increase our understanding of organisms’ ecological strategies and evolutionary trajectories. Such analyses are increasingly dependent on large, well curated databases. One such database is AusTraits, an open-source, harmonized database of plant trait data with 1.25 million data records that include some data on nearly all of Australia’s 30,000 plant taxa for approximately 500 traits. Released in 2021, the R-code that assembles the database has now been transformed into a standalone R package, traits.build, allowing the reuse of the AusTraits workflow to establish databases for other taxonomic groups, global regions, or different traits. The pipeline aggregates individual datasets, aligning trait names, allowing substitutions for categorical trait values and transforming numeric trait values into common units. Metadata fields allow documentation of study locations and contextual details. The output is a collection of relational tables: a core traits table, supported by tables that document measurement locations, observation contexts, taxonomy, and measurement methods. The tables and columns adhere to, but extend, those described by DarwinCore and the Ecological Trait Data Standard (ETS), two ontologies familiar to the ecological research community. Reuse of this workflow to compile additional databases offers several advantages: 1) the database will adhere to an established format and be fully interoperable with AusTraits and other databases already using the traits.build package; 2) less effort is required to compile a new database, as a workflow exists and training materials are available; and 3) for plant databases, the AusTraits Plant Dictionary definitions can be reused for further cross-database integration. The global trait-based ecology community is increasingly committed to collaboration and workflows that allow the integration of data between distinct datasets, regions, and taxa; the release of traits.build will accelerate these efforts.
