What type of repository is this and who is it for?

The main purpose of this repo is to build AusTraits, a curated database of traits for the Australian flora. It contains code and data to build the harmonised AusTraits database outputs. So, the intended audiences are those who are interested in building AusTraits from scratch, or contributing data to AusTraits.

Those interested in simply using data from AusTraits, should visit and download the compiled resource from the versioned releases archived on Zenodo at doi: 10.5281/zenodo.3568417.

The repo is partway between a compendium and an R package. It is structured like an R package, and contains code that can be installed as a package. This allows us to use some of R’s package management tools:

  • use a DESCRIPTION file to document our dependencies
  • create our documentation, using the package pkgdown
  • run tests on our datasets (using testthat)
  • manage and build vignettes
  • use the devtools function to load functions.

It also contains data for rebuilding AusTraits. A key goal for us was to make the process for harmonising different datasets as transparent as possible. Our workflow is therefore fully-reproducible and open, meaning it exposes the decisions made in the processing of data into a harmonised and curated dataset (Figure 1); and can also be rerun by others.

Ways to contribute

We envision AusTraits as an ongoing collaborative community resource that:

  1. Increases our collective understanding of the Australian flora; and
  2. Facilitates the accumulation and sharing of trait data;
  3. Builds a sense of community among contributors and users; and
  4. Aspires to fully transparent and reproducible research of the highest standard.

Below are some of the ways you can contribute.

*Please note that the AusTraits project is released with a Contributor Code of Conduct. By contributing to this project you agree to abide by its terms.

Contributing new data

We gladly accept new data contributions to AusTraits. If you would like to contribute data, the requirements are:

  • Data were collected for Australian plant species growing in Australia
  • You collected data on one of the traits listed in the trait definitions table
  • You are willing to release the data under an open license for reuse by the scientific community
  • You make it is as easy as possible for us to incorporate your data by following the instructions.

If you want to contribute data, please review the instructions here on how to contribute data.

Reporting errors and improving documentation

Data contributors and data users who are less familiar with the AusTraits format and code than the custodians may determine that important descriptions or steps are omitted from this documentation file. We welcome additions and edits that make using the existing data or adding new data easier for the community.

If you notice a possible error in AusTraits, please post an issue on GitHub. If you can, please provide code illustrating the problem.

If you would like to value-add to AusTraits in some other way, please get in contact with an idea or offer of time.

Improving data quality

A core initiative of AusTraits from 2021-2023 is to refine and better document the trait names, definitions, and values that are the direct link from each contributor’s dataset to the harmonised database. This effort is funded by an Australian Research Data Commons (ARDC) grant through their Australian Data Partnerships program. It includes both a review of definitions by the core AusTraits team and a series of workshops to discuss clusters of related trait definitions.

The goal is to link as many trait names as possible to established, published definitions (e.g. in traits handbook, review paper on a method, manuscripts regularly cited as the standard for a specific trait). In addition, the list of allowable values for each categorical trait will be reviewed and revised.

If you are interested in contributing expertise to the revision of a given trait (or cluster of related traits), please contact us.

Compiling AusTraits

In this section we describe how to build the harmonised dataset. By “compiling” we mean transforming data from all the different studies into a harmonised common format. As described above and depicted in Figure 1, AusTraits is built so that you can rebuild the database from its parts at any time. This means that decisions made along the way, in how data are transformed or encoded, can be inspected and modified. And new data is easily incorporated.

The first step to compile AusTraits is to download a copy of the austraits.build repository from Github. Then open up the Rstudio project, or open R into the right repo directory.

Dependencies

To check you have the right packages installed, you can use the devtools package to run:

# install.packages("devtools)  # install devtools if needed
devtools::install_deps()

This command checks that the required packages, listed in the file DESCRIPTION, are available in your local machine.

Compile via remake

One of the packages that will be installed with the above is remake. This package manages the compiling, and also helps streamline the amount of recompiling needed when new sources are added.

Running the following command will rebuild AusTraits and save the assembled database into an RDS file located in export/data/curr/austraits.rds.

remake::make()

Remake can also load the compiled dataset directly into R by calling:

austraits <- remake::make("austraits")

Developing AusTraits

For those working with AusTraits code base or data, you may want to read about