AusTraits v5.0.0

AusTraits v5.0.0

This is a major release of the AusTraits database.

• austraits-5.0.0.zip: contains the compiled dataset and details of
  structure
• austraits-5.0.0.rds: contains a version of the dataset for direct
  loading in R
• source code v5.0.0.zip: contains the source materials used to build
  the compiled dataset

For details on access, structure, and usage please visit
https://doi.org/10.5281/zenodo.3568417.

This release was generated from source materials available at
https://github.com/traitecoevo/austraits.build/releases/tag/v5.0.0 A
full set of changes in the source can be viewed at:
v4.2.0...v5.0.0

Compared to the last version, this release contains substantial
additions of new data and improvement of old data.

version	dataset_id	taxa	locations	traits	records
4.2.0	363	34262	5176	494	1848570
5.0.0	363	33402	4894	494	1813897

This release contains:

• 34673 records removed from the dataset: these datapoints were previously excluded (e.g. because they are exotics or poor quality), but that were inadvertently added into v 4.2.0. Now removed again. - restructruign of columns, as noted below.

This is the first AusTraits release to be compiled using the {traits.build} R package, available at https://github.com/traitecoevo/traits.build. The workflow is a refined version of the R-scripted pipeline previously used to compile AusTraits and the output structure has not changed, other than on-going minor fine-tuning, as detailed below.

Taxonomy: New AusTraits-specific functions relating to taxonomy have been written that utilise the package {APCalign}. One new function, build_align_taxon_names uses the function APCalign::align_taxa() to standardise syntax, correct typos, and ensure taxon names are aligned with some name on the APC or APNI lists. The second function, build_update_taxon_list uses the function APCalign:update_taxonomy to build a new taxon_list.csv file for the config/ folder. This file is then used by traits.build::dataset_update_taxonomy to update names to their currently accepted taxon name, when possible.

Edits were made to many dataset metadata files to align with these changes.

Changes to table structure:

• method_id was added, so that when the same trait was measured using multiple methods, these could be distinguished between
• the context identifiers were renamed to method_context_id, temporal_context_id, entity_context_id, plot_context_id, treatment_context_id to be more explicit
• austraits_curators became dataset_curators
• added repeat_measurements_id for trait measurements that are response curves, both to unify the repeated measurements that comprise a single "measurement" and also to capture the order of the measurements

No datasets have been added for this release. However, some metadata file changes exist, in particular to metadata[["taxonomic_updates"]], including removing duplicate taxonomic_updates or unneeded taxonomic updates and continued standardisation of taxonomic updates. All original_names in the taxonomic_updates tibble are now aligned to a specified taxon_name - either an informal name assigned through metadata[["taxonomic_updates"]] or a match to a name on the taxon_list.