Search

Ecological Niche Modelling of Tropical Dry Forest Species Raw data and code base for the research approach on climate change-driven biotic changes in tropical dry forests. Briefly, the code reproduces the traditional Ecological Niche Modelling (ENM) and the Ensemble of Small Models (ESM) framework; perform assemblage-level analysis in the resulting outputs; and then illustrate the building of figures used in the main manuscript. The modelling framework was built using the directory structure informed in the README.pdf file. The R-code provided have steps designed to replicate the directory structure as informed, but understanding it is a good starting point to navigate the output produced. The files provided below were designed for illustrative purposes and to avoid the need to re-run all ENM and ESM models in order to replicate the major findings of the manuscript (the complete set of outputs surpass 600GB and 1 million of output files). There are five zip files whose content we described below: Occurrences Folder.zip: it represents the "Occurrence Folder" directory, as illustrated in the README.pdf. This zip includes three files representing the occurrence data for the plant species used in the modelling framework. Datasets.zip: it represents the "Datasets" folder directory, as illustrated in the README.pdf. This zip one the 'PlantLifeForms.txt' file, which informs plant growth form for 2656 tropical dry forest species. This is the only file that will not be produced by R-scripts provided. The other 94 csv files in this zip are provided for illustrative purposes to avoid the need of the user to re-run all ENM and ESM models. Among these csv files, we provide (i) species-level outputs (measures of species range shift), (ii) assemblage-level outputs (measures of richness, beta-diversity, ecological generalism, and woodiness of plant assemblages), (iii) summary statistics (ENM-ESM model performance metrics, kruskal wallis tests, spatial correlations), and (iv) processed tables used to build the main figures in the manuscript. Shapefiles.zip: it represents the "Shapefiles" folder directory, as illustrated in the README.pdf. There are five shapefiles in this zip. (i) Caatinga_wgs84, spatial polygon illustrating the study area. (ii) cea_grid_cells, 10x10 km equal-area grid cells created for the assemblage-level analysis. (iii) ne_110m_admin_0_countries, Natural Earth Vector draws boundaries of countries (see https://www.naturalearthdata.com/downloads). (iv) ne_110m_graticules_20, Natural Earth Vector draws graticules of the globe (see https://www.naturalearthdata.com/downloads). (v) wwf_realm, the WWF biogeographical realms, modified from Dinerstein et al. (2017, see https://ecoregions.appspot.com/). AggregatedUncertainty.zip: it represents the "AggregatedUncertainty" folder directory, as illustrated in the README.pdf. Aggregated uncertainty is a combined measure of variation in species habitat suitability across five generalized circulation models (GCM). This zip file contains 16 raster files in format tif, with each raster representing a combination between SSP scenario (ssp245 or ssp585) and extrapolation constraints (MOP00, MOP70, MOP80, or MOP90). RData.zip: it represents the "RData" folder directory, which is designed to store RData files produced during the running of the R-scripts. It currently includes two files to facilitate the plotting of main figures. Predictors Folder.zip: it includes the raster files produced for the folder "Predictors Folder". There are three subfolders: PCA, WorldClim, SoilGrids. All rasters were generated at the spatial resolution of 5 arc-minutes. Projection_PCA.zip: it includes the raster files produced for the folder "Projection_PCA". There are 20 subfolders, each named according to a combination of GCM (5 options) and SSP scenario (2 options). All rasters were generated at the spatial resolution of 5 arc-minutes. R-code: Moura_etal_JEcol2023_Script1_ENM_DataAnalysis.R: perform data cleanning, prepare predictor layers, run traditional ecological niche models, get ensemble models across GCMs, apply extrapolation constraints, and compute species geographical range shifts. Moura_etal_JEcol2023_Script2_ESM_DataAnalysis.R: run ensemble of small models for rare species, get ensemble models across GCMs, apply extrapolation constraints, and compute species geographical range shifts. Moura_etal_JEcol2023_Script3_AssemblageLevelAnalysis.R: build species-level tables to classify plants with respect to ecological generalism and woodiness, compute aggregated model uncertainty, extract presence-absence matrices, compute assemblage-level metrics, and assess changes in biodiversity patterns across regions subject to homogenization or heterogenization. Moura_etal_JEcol2023_Script4_Figures_MainText.R: build figures 1 to 6 reported in the main text. AdaptedRFunctions.R: auxiliary script containing functions modified from existing R-packages. References Moura, M.R., Nascimento, F.A.O., Paolucci, L.N., Silva, D.P., & Santos, B.A. (2023). Pervasive impacts of climate change on the woodiness and ecological generalism of dry forest plant assemblages. Journal of Ecology, doi: 10.1111/1365-2745.14139. Correspondence to: mariormoura@gmail.com, The RCode.1 (Moura_etal_JEcol2023_Script1_ENM_DataAnalysis.R) has been revised to eliminate invalid subdirectories, and compressed zip files now contain rasters of predictor layers for the current and future time.

Ecological Niche Modelling ofTropical Dry Forest Species Raw data and code base for the research approach on climate change-driven biotic changes in tropical dry forests. Briefly, the code reproduces the traditional Ecological Niche Modelling (ENM) andthe Ensemble of Small Models (ESM) framework; perform assemblage-level analysis in the resulting outputs; and then illustrate the building of figures used in the main manuscript. The modelling framework was built using the directory structure informed in the README.pdf file. The R-code provided have steps designed to replicate the directory structure as informed, but understanding it is a good starting point to navigate the output produced. The files provided below were designed for illustrative purposes and to avoid the need to re-run all ENM and ESM models in order to replicate the major findings of the manuscript (the complete set of outputs surpass 600GB and 1 million of output files).There are five zip files whose content we described below: Occurrences Folder.zip: it represents the "Occurrence Folder" directory, as illustrated in the README.pdf. This zip includes three files representing the occurrence data for the plant species used in the modelling framework. Datasets.zip: it represents the "Datasets" folder directory, as illustrated in the README.pdf. This zip one the 'PlantLifeForms.txt' file, which informs plant growth form for 2656 tropical dry forest species. This is the only file that will not be produced by R-scripts provided. The other 94 csv files in this zip are provided for illustrative purposes to avoid the need of the user to re-run all ENM and ESM models. Among these csv files, we provide (i) species-level outputs (measures of species range shift), (ii) assemblage-level outputs (measures of richness, beta-diversity, ecological generalism, and woodiness of plant assemblages), (iii) summary statistics (ENM-ESM model performance metrics, kruskal wallis tests, spatial correlations), and (iv) processed tables used to build the main figures in the manuscript. Shapefiles.zip: it represents the "Shapefiles" folder directory, as illustrated in the README.pdf. There are five shapefiles in this zip. (i) Caatinga_wgs84, spatial polygon illustrating the study area. (ii) cea_grid_cells, 10x10 km equal-area grid cells created for the assemblage-level analysis. (iii) ne_110m_admin_0_countries, Natural Earth Vector draws boundaries of countries (see https://www.naturalearthdata.com/downloads). (iv) ne_110m_graticules_20, Natural Earth Vector draws graticules of the globe (see https://www.naturalearthdata.com/downloads). (v) wwf_realm, the WWF biogeographical realms, modified from Dinerstein et al. (2017, see https://ecoregions.appspot.com/). AggregatedUncertainty.zip: it represents the "AggregatedUncertainty" folder directory, as illustrated in the README.pdf. Aggregated uncertainty is a combined measure of variation in species habitat suitability across five generalized circulation models (GCM). This zip file contains 16 raster files in format tif, with each raster representing a combination between SSP scenario (ssp245 or ssp585) and extrapolation constraints (MOP00, MOP70, MOP80, or MOP90). RData.zip: it represents the "RData" folder directory, which is designed to store RData files produced during the running of the R-scripts. It currently includes two files to facilitate the plotting of main figures. R-code: Moura_etal_JEcol2023_Script1_ENM_DataAnalysis.R: perform data cleanning, prepare predictor layers, run traditional ecological niche models, get ensemble models across GCMs, apply extrapolation constraints, and compute species geographical range shifts. Moura_etal_JEcol2023_Script2_ESM_DataAnalysis.R: run ensemble of small models for rare species, get ensemble models across GCMs, apply extrapolation constraints, and compute species geographical range shifts. Moura_etal_JEcol2023_Script3_AssemblageLevelAnalysis.R: build species-level tables to classify plants with respect to ecological generalism and woodiness, compute aggregated model uncertainty, extract presence-absence matrices, compute assemblage-level metrics, and assess changes in biodiversity patterns across regions subject to homogenization or heterogenization. Moura_etal_JEcol2023_Script4_Figures_MainText.R: build figures 1 to 6 reported in the main text. AdaptedRFunctions.R: auxiliary script containing functions modified from existing R-packages. References Moura, M.R., Nascimento, F.A.O., Paolucci, L.N., Silva, D.P., & Santos, B.A. (2023). Pervasive impacts of climate change on the woodiness and ecological generalism of dry forest plant assemblages. Journal of Ecology, in press. Correspondence to: mariormoura@gmail.com