| title | tags | authors | affiliations | date | bibliography | |||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
neotoma2: An R package to access data from the Neotoma Paleoecology Database |
|
|
|
6 November 2023 |
paper.bib |
The neotoma2 R package is a tool to access and manipulate data from the Neotoma Paleoecology Database (https://www.neotomadb.org) within the R environment. Neotoma is a community curated paleoecological data resource [@williams2018neotoma], containing nearly 9 million unique observations of paleoecological proxies with global coverage from 37 constituent databases. The package uses the Neotoma API v2.0 [@goring2023api] as a tool to import records from the Neotoma database, allowing researchers to examine taxonomic, spatial and temporal patterns across space and time over the last 5.4 million years. The R package allows researchers to download, and create new records using get and set functions (e.g., get_sites(), set_sites()) respectively. This provides researchers with the opportunity to develop dynamic workflows that include data generated locally, and not yet submitted to the Neotoma database. The set functions are intended as a precursor to utilities to upload data directly to Neotoma, although this functionality is not yet available.
The neotoma2 R package has been under dynamic development for over a year, but has been used for teaching and training [@Goring2023APD]. This release of the neotoma2 R package is a clean release of the package, with all of the core features provided and extensive test coverage implemented.
The neotoma R package [@goring2015neotoma] leveraged the Neotoma Paleoeocology Database v1.0 API and had been one of the primary tools for researchers working with data from Neotoma [@wang2023plants;@kujawa2016effects; @byun2021extensive]. Changes to the underlying database and a rebuilding of the API required new data objects within the R package to more closely align to the Neotoma data model [@grimm2008neotoma]. Additionally, the original v1.0 API that was accessed by the neotoma package was deprecated in 2020, meaning the neotoma package could no longer access data from Neotoma.
The broad user community for Neotoma [@williams2018neotoma; @goring2018nexus] requires a toolset that can access and manage data for each of the more than 40 dataset types within Neotoma and so extensive metadata must be accessed for each record. The neotoma2 package conforms to a tidyverse [@wickham2019tidyverse] approach for data management, with methods and data objects that are suited to piping using the %>% (or |>) pipe convention and with implementations for filtering and other common dplyr methods. The package also now uses "long" data.frames and tibbles by default, using the toWide() function to transform data into "wide" tables for use with common ecological data packages such as vegan [@oksanen2022vegan]. Data objects in the neotoma2 package now more closely resemble the underlying data model within Neotoma (https://open.neotomadb.org/dbschema) than in the previous package. Most importantly the neotoma2 package provides a toolset for paleoecologists, ecologists, conservation ecologists, archaeologists, and others, to access and examine the broad range of fossil data contained within the Neotoma Paleoecology Database.
Data from the Neotoma Database can be accessed through the public API (https://api.neotomadb.org) or a PostgreSQL database snapshot with a database client [@williams2018neotoma; https://www.neotomadb.org/data/db-snapshots] or through the EarthLife Consortium API [@uhen2021earthlife]. The R package will simplify many of the operations required to assemble and manipulate datasets, and provides functions that support researchers as part of their analytic workflows by linking Neotoma directly to packages within R used for ecological and earth science research, data visualization and statistical analysis.
We acknowledge contributions from the Neotoma Paleoecology Community, the participants of our workshops for the European Pollen Database, American Quaternary Association and the International Association of Limnologists/International Paleolimnology Association, and members of the EarthCube community. This work was funded through a grant to SJG from the National Science Foundation (NSF-1948926).