Data from: Climate-driven changes in habitat suitability of two keystone species in dry Afromontane forests of Ethiopia

This dataset comprises raw and processed environmental and species-occurrence data used to assess current and future habitat suitability of Juniperus procera and Olea europaea subsp. cuspidata in Ethiopia. Current climatic conditions were represented by the 19 standard WorldClim bioclimatic variables, which were downloaded as global raster layers and cropped and masked to the geographical boundary of Ethiopia. Future bioclimatic projections for the SSP2-4.5 (SSP245) and SSP5-8.5 (SSP585) scenarios were obtained using the CMIP6_world() function in R for the selected Global Climate Model and future period and subsequently cropped and masked to Ethiopia using the terra package. Elevation data at approximately 40 m spatial resolution were obtained using the elevatr R package and processed to derive two additional topographic predictors: slope, expressed in degrees, and Topographic Position Index (TPI), calculated using a 3 × 3 focal neighbourhood. Areas corresponding to lakes were masked from the elevation-derived topographic layers. Species occurrence records for J. procera and O. europaea subsp. cuspidata were obtained from the Global Biodiversity Information Facility (GBIF) using the rgbif R package and Google Earth 3D view and geographically restricted to Ethiopia. The resulting environmental and occurrence datasets were prepared for species distribution modelling to evaluate changes in habitat suitability under current and projected future climate conditions. All data acquisition and processing procedures were implemented in R version 4.6.0, primarily using the terra, elevatr, and rgbif packages. The repository includes the processed environmental raster datasets, species occurrence data, and associated R scripts to facilitate transparency, reproducibility, and reuse of the study. Method 1. the global bioclimatic data were downloaded and subsequently cropped and masked to the geographical boundary of Ethiopia. Method for 2. The elevation data were downloaded using the elevatr R package and subsequently cropped to the geographical boundary of Ethiopia. Areas corresponding to Ethiopian lakes were masked using lake boundary polygons. Method for 3. The Future climate projections were obtained using the CMIP6_world() function in R for the selected Global Climate Model (GCM) and future time period. The resulting global raster layers were subsequently cropped and masked to the geographical boundary of Ethiopia using functions available in the terra package, producing Ethiopia-specific future bioclimatic datasets for subsequent species distribution modelling. Method for 4. The data were obtained using the CMIP6_world() function in R for the selected Global Climate Model and future time period, following the same procedures described for the SSP2-4.5 scenario. The resulting global raster layers were cropped and masked to the geographical boundary of Ethiopia using the terra package. Method for 5. Slope was derived from the elevation data described above using the terrain() function in the terra R package: library(terra)  slope <- terrain(Ethiopia_elev,                  v = "slope",                  unit = "degrees") The resulting slope raster was subsequently masked using the boundary polygons of Ethiopian lakes. Method for 6. TPI was derived from the elevation raster using a 3 × 3 focal neighbourhood. It was calculated as the difference between the elevation of each raster cell and the mean elevation of its surrounding cells using functions available in the terra package: library(terra) # Calculate mean elevation within a 3 × 3 neighbourhood mean_elev <- focal(   Ethiopia_elev,   w = matrix(1, 3, 3),   fun = mean,   na.rm = TRUE ) # Calculate Topographic Position Index tpi <- Ethiopia_elev - mean_elev The resulting TPI raster was subsequently masked using the boundary polygons of Ethiopian lakes. Positive TPI values indicate locations that are higher than their surrounding areas, whereas negative values indicate locations that are lower than their surroundings. Method for 7: Records were retrieved by specifying the scientific name of the species and were subsequently clipped to the geographical boundary of Ethiopia. Method for 8: Records were retrieved by specifying the scientific name of the species and were subsequently clipped to the geographical boundary of Ethiopia. # Data from: Climate-driven changes in habitat suitability of two keystone species in dry Afromontane forests of Ethiopia Dataset DOI: [10.5061/dryad.q573n5tz5](doi.org) This repository contains the raw and processed data used in the manuscript **“Climate-Driven Changes in Habitat Suitability of Two Keystone Species in Dry Afromontane Forests of Ethiopia,”** accepted for publication. The repository includes the following datasets and associated R scripts. **1. Ethiopia_current_clim.tif** This raster dataset contains the 19 standard WorldClim bioclimatic variables representing current climatic conditions. The global bioclimatic data were downloaded and subsequently cropped and masked to the geographical boundary of Ethiopia. **2. Ethiopia_elevation_no_lakes.tif** This raster dataset contains elevation data for Ethiopia at approximately 40 m spatial resolution. The elevation data were downloaded using the elevatr R package and subsequently cropped to the geographical boundary of Ethiopia. Areas corresponding to Ethiopian lakes were masked using lake boundary polygons. **3. Ethiopia_future_245.tif** This raster dataset contains future bioclimatic projections under the SSP2-4.5 (SSP245) climate scenario. Future climate projections were obtained using the CMIP6_world() function in R for the selected Global Climate Model (GCM) and future time period. The resulting global raster layers were subsequently cropped and masked to the geographical boundary of Ethiopia using functions available in the terra package, producing Ethiopia-specific future bioclimatic datasets for subsequent species distribution modelling. **4. Ethiopia_future_585.tif** This raster dataset contains future bioclimatic projections under the SSP5-8.5 (SSP585) climate scenario. The data were obtained using the CMIP6_world() function in R for the selected Global Climate Model and future time period, following the same procedures described for the SSP2-4.5 scenario. The resulting global raster layers were cropped and masked to the geographical boundary of Ethiopia using the terra package. **5. Ethiopia_slope_no_lakes.tif** This raster dataset contains slope values expressed in degrees. Slope was derived from the elevation data described above using the terrain() function in the terra R package: library(terra)  slope <- terrain(Ethiopia_elev,                  v = "slope",                  unit = "degrees") The resulting slope raster was subsequently masked using the boundary polygons of Ethiopian lakes. **6. Ethiopia_tpi_no_lakes.tif** This raster dataset contains the Topographic Position Index (TPI). TPI was derived from the elevation raster using a 3 × 3 focal neighbourhood. It was calculated as the difference between the elevation of each raster cell and the mean elevation of its surrounding cells using functions available in the terra package: library(terra) \# Calculate mean elevation within a 3 × 3 neighbourhood mean_elev <- focal(   Ethiopia_elev,   w = matrix(1, 3, 3),   fun = mean,   na.rm = TRUE ) \# Calculate Topographic Position Index tpi <- Ethiopia_elev - mean_elev The resulting TPI raster was subsequently masked using the boundary polygons of Ethiopian lakes. Positive TPI values indicate locations that are higher than their surrounding areas, whereas negative values indicate locations that are lower than their surroundings. **7. Occurrence data for** ***Juniperus procera*** (Occurence_data_of_Juniperus_procera.xls) Occurrence records for *Juniperus procera* were obtained from the Global Biodiversity Information Facility (GBIF) using the rgbif R package. Records were retrieved by specifying the scientific name of the species and were subsequently clipped to the geographical boundary of Ethiopia. Additional data were generated using Google Earth 3D and information in herbarium sheets of the collection at the National Herbarium of Addis Ababa University **8. Occurrence data for** ***Olea europaea*** **subsp.** ***cuspidata*** (Occurence_data_of_Olea_europaea.xls) Occurrence records for *Olea europaea* subsp. *cuspidata* were obtained from the Global Biodiversity Information Facility (GBIF) using the rgbif R package. Records were retrieved by specifying the scientific name of the species and were subsequently clipped to the geographical boundary of Ethiopia. Additional data were generated using Google Earth 3D and information in herbarium sheets of the collection at the National Herbarium of Addis Ababa University. Neither species is considered endangered in Ethiopia. However, both are important native tree species of the dry Afromontane forests of the Ethiopian Highlands, which are recognized as areas of high biodiversity and conservation importance. Although the species are not currently classified as endangered, their populations and habitats may be subject to increasing pressures from habitat degradation, land-use change, deforestation, and climate change, rather than from targeted hunting. Therefore, reporting the geographical coordinates to four decimal places does not, in this context, present a meaningful additional conservation risk or facilitate the targeted removal of individuals of either species. **Software Requirements** The data processing and analysis scripts were developed and executed in **R version 4.6.0**. The principal R packages used for data acquisition and processing include: * terra * elevatr * rgbif