Vulnerability to climate changes of tropical forests across Africa
Data files
Feb 27, 2025 version files 13.41 MB
-
data_VULNERABILITY_TFA_DivDis_24_02_25.zip
13.41 MB
-
README.md
4.75 KB
Sep 30, 2025 version files 159.06 MB
-
R_code_final.zip
159.06 MB
-
README.md
3.12 KB
Abstract
Aim: Global climate projections identify tropical regions as hotspots of climate change during the 21st century. The few ground data in tropical Africa confirm significant warming and drying over the last decades, but how plant communities will tolerate these new climate conditions remain vastly uncertain. In this study, we assess the climatic vulnerability of tropical moist forests across Africa.
Location: Tropical Africa.
Methods: We mapped climate change exposure across the tropical moist forest biome, focusing on mean annual temperature (MAT), mean annual precipitation (MAP), and climatological water deficit (CWD) using climate projections for 2085 from five regional models under RCP4.5 and RCP8.5. Using occurrence records for 3,536 tree and shrub species, we estimated species’ climatic limits and safety margins, then averaged these margins at the community level. Finally, we combined exposure and safety margins to assess species- and community-level risk by 2085.
Results: Under RCP4.5, corresponding to an average warming of 2.4°C by 2080, 92% of species (3,256) could be at risk in at least one community where they occur. This rate increases to 96% (3,405 species) under RCP8.5, with an average warming of 4.3°C. In all scenarios, the most at-risk communities are concentrated in low-elevation regions, where species have few opportunities to migrate if their climatic limits are exceeded. The high risk across the forest biome results from the combination of significant and widespread temperature increases and the relatively narrow safety margins of the species. Specifically, 50% of species have an average safety margin below 1.6°C above baseline temperatures, suggesting they are already near their tolerance limits.
Main conclusions: Beyond refining our understanding of the vulnerability of tropical moist forests across Africa, our results have far-reaching implications for conservation, allowing to target species and communities of interest for further monitoring and conservation efforts.
https://doi.org/10.5061/dryad.7h44j105d
R_code_final.zip
This repository contains species occurrence data, spatial layers, and R scripts used to reproduce the main results of the article: Vulnerability to climate change of tropical forests across Africa.
climatic_data/
Climate layers (baseline and projections under RCP4.5 / RCP8.5):
- CWD (Climatic Water Deficit) – in mm
- MAP (Mean Annual Precipitation) – in mm
- MAT (Mean Annual Temperature) – in °C
Also includes exposure layers for each variable (files prefixed withexpo_, e.g.expo_mat_rcp45.tiffor MAT under RCP4.5).
communities_grid/
Polygons/pixels representing forest communities:
- communities: community boundaries (polygons/pixels)
- forest_communities: communities including species occurrences
floristic_data/
Tree and shrub occurrences extracted from RAINBIO (https://gdauby.github.io/rainbio/download_page.html).
The file rainbio_selected.csv contains information on species occurrences. The key variables included are:
- tax_sp_level: binomial scientific name of the species
- ddlat: latitude
- ddlon: longitude
- growth_form_level_1_1_3: species growth form (tree, shrub, etc.)
- final_accuracy: spatial accuracy of the occurrence coordinates
output_data/
Results generated by R scripts (CSV), e.g.:
table_safety_margin_sp.csv: climate safety margins for species.
output_maps/
Maps generated by the analyses. The naming convention of the output maps is as follows:
- First, the vulnerability index (e.g. Exposure),
- then the climate variable (e.g. CWD),
- followed by the climate scenario (e.g. RCP4.5).
For example: Exposure_CWD_RCP45.
shp_africa/
Base shapefiles (African boundaries and reference layers), e.g.:
africa_cropped.*: cropped African boundaries used to delimit analysis extentATMF_sf.*: African Tropical Moist Forest (ATMF) grid cells used in spatial analyses (in main folder)
R scripts
- 1_climate_safety_margin_analysis.R — computes climate safety margins for tree and shrub species.
- 2_climate_exposure_analysis.R — analyzes spatial exposure of communities to projected climate changes.
- 3_climate_risk_by_community.R — assesses climate risk scores for each community.
Software requirements
- R (version ≥ 4.0)
- Packages:
sf,terra,dplyr,ggplot2
Data sources
The floristic data were derived from RAINBIO, available at:
https://gdauby.github.io/rainbio/download_page.html
The climate date were derived from AFRICLIM 3.0, available at:
https://webfiles.york.ac.uk/KITE/AfriClim/
Change Log
- September 2025 : The
R_code_finalfolder has replaceddata_VULNERABILITY_TFA_DivDis_24_02_25.zipand README updated
