Towards the conservation of Brazilian legumes: Summary, land cover and use analytics, and a comparative analysis of locations count methods (curated vs. automated [buffer-dissolution])

This dataset presents key findings from the study "Automating and Enhancing Species Extinction Risk Assessments with Historical Land Use and Land Cover Data."

The file `summary-threatened-legume-species.ods` provides a summary of all threatened species of the Leguminosae family native to Brazil, including IUCN category and criteria, location counts, and the year of the latest assessment, sourced from official records. Additionally, it includes calculated values for Area of Occupancy (AOO) and Extent of Occurrence (EOO), along with trend data indicating natural area change rates (decline [positive number, red] or growth [negative number, green]) within both AOO and EOO. Location counts are further detailed across various buffer radii (1-5 km), utilizing a buffer-dissolution method for automated location counting. Each buffer radius is analyzed to assess AOO and EOO decline, where '1' indicates a species is threatened and '0' indicates it is not. This data provides an efficient method for screening threatened species under criterion B of the IUCN Red List guidelines.

The file `overlay-analysis.ods` contains overlay analysis results for AOO and EOO of each species using MapBiomas land use and land cover (LULC) data (specifically MapBiomas Brazil, collection 7.1) from 1985 to 2021, covering all threatened legume species. This file provides both absolute area in square kilometers and percentages for each LULC class. The overlay analysis results support estimates of growth and decline trends for each LULC class.

The file `trend-analysis.ods` presents results of annual rate estimates from trend analysis across LULC classes, and including both natural and anthropic groupings. A complete JSON database with p-values and R² values is provided in `trend-analysis.json`.

This approach, combining all results for each species in a comprehensive, merged dataset, allows for effective filtering and ranking of the most threatened species as well as identification of their primary threats.

We recommend opening the ODS files with LibreOffice, as Microsoft Excel may experience issues parsing decimal formats accurately.

More detailed maps and graphs are available at https://github.com/lsbjordao/LULC-MapBiomas-Leguminosae.