The ATLAS project aims to build an extensive, harmonized database of plant species occurrences and functional traits within tropical homegardens worldwide. Given the complexity and diversity of these agroecosystems, assembling such a dataset requires a broad collaborative effort that goes beyond traditional literature reviews.
Homegarden data are scattered across multiple sources: published papers, grey literature, unpublished datasets, and ongoing research projects. Many relevant species inventories and trait datasets remain inaccessible without direct collaboration. Recognizing this, ATLAS adopts a multi-pronged data acquisition strategy focused on active engagement with the scientific community, research institutions, and practitioners globally.
This strategy includes:
By combining these approaches, ATLAS strives to create a comprehensive, high-quality dataset that reflects the rich ecological and cultural variation of tropical homegardens worldwide.
As part of our data acquisition, we performed a systematic literature search on June 2, 2025, in the Web of Science Core Collection, using the following query:
``` (“home garden” OR homegarden OR homestead* OR “household garden*“) AND (species OR”species richness” OR “plant diversity” OR biodiversity OR “floristic composition”) ```
This search yielded 2,437 records, supplemented by manual screening of references and citations to uncover additional relevant studies.
We included studies that:
We excluded reviews, conceptual articles without original data, and studies lacking direct species observations.
Using the semi-automated platform Abstrackr, which leverages active learning to efficiently prioritize relevant abstracts, we screened 1,540 abstracts. Screening was halted upon saturation—defined as 100 consecutive abstracts without further inclusion—yielding a final corpus of 300 studies.
From these studies, we manually extracted detailed metadata (e.g., study title, country, region, year), biological data (species counts, number of homegardens sampled), and ecological metrics (diversity indices, explanatory variables such as land use or management factors).
To ensure consistency, we standardized geographic names with the R package countrycode and applied uniform definitions for species richness and explanatory variables.
Importantly, because species-level data are often incomplete or embedded in supplementary materials, we prioritized contacting study authors to obtain raw or detailed datasets whenever possible. This direct collaboration ensures higher-quality data and fills critical gaps.
While literature and author outreach form a strong foundation, the full potential of ATLAS depends on active contributions from the global community.
If you are a researcher, practitioner, or institution holding data on plant species occurrences or functional traits in tropical homegardens, we encourage you to join ATLAS by sharing your datasets. Contributions of any size or format are valuable and help:
Data sharing within ATLAS respects contributors’ rights and includes proper attribution. We are committed to building a transparent, collaborative, and mutually beneficial network.
For details on how to contribute or collaborate, please visit our Contact page.
For any questions regarding this protocol or collaboration opportunities, please contact the ATLAS project team.
If you see mistakes or want to suggest changes, please create an issue on the source repository.
Text and figures are licensed under Creative Commons Attribution CC BY 4.0. Source code is available at https://github.com/dbeillouin/ATLAS_WEB, unless otherwise noted. The figures that have been reused from other sources don't fall under this license and can be recognized by a note in their caption: "Figure from ...".