Data from: Bioclimatic envelope models predict a decrease in tropical forest carbon stocks with climate change in Madagascar

1. Recent studies have underlined the importance of climatic variables in determining tree height and biomass in tropical forests. Nonetheless, the effects of climate on tropical forest carbon stocks remain uncertain. In particular, the application of process-based dynamic global vegetation models have led to contrasting conclusions regarding the potential impact of climate change on tropical forest carbon storage. 2. Using a correlative approach based on a bioclimatic envelope model and data from 1771 forest plots inventoried during the period 1996-2013 in Madagascar over a large climatic gradient, we show that temperature seasonality, annual precipitation and mean annual temperature are key variables in determining forest aboveground carbon density. 3. Taking into account the explicative climate variables, we obtained an accurate (R2 = 70% and RMSE = 40 Mg.ha-1) forest carbon map for Madagascar at 250 m resolution for the year 2010. This national map was more accurate than previously published global carbon maps (R 2 ≤ 26% and RMSE ≥ 63 Mg.ha −1 ). 4. Combining our model with the climatic projections for Madagascar from seven IPCC CMIP5 global climate models following the RCP 8.5, we forecast an average forest carbon stock loss of 17% (range: 7-24%) by the year 2080. For comparison, a spatially homogeneous deforestation of 0.5% per year on the same period would lead to a loss of 30% of the forest carbon stock. 5. Synthesis: Our study shows that climate change is likely to induce a decrease in tropical forest carbon stocks. This loss could be due to a decrease in the average tree size and to shifts in tree species distribution, with the selection of small-statured species. In Madagascar, climate-induced carbon emissions might be, at least, of the same order of magnitude as emissions associated to anthropogenic deforestation. Above-ground carbon density (ACD) dataThis file includes the geographical coordinates and above-ground carbon density (ACD in Mg.ha-1) of 1771 forest plots in Madagascar.ACD.csvNational forest carbon map for MadagascarNational forest carbon map at 250m for Madagascar for the year 2010. GeoTIFF raster file, 4.2 Mo, EPSG: 32738, 250m resolution, Int16.ACD_for10.tifNational carbon map for MadagascarNational carbon map for Madagascar for the year 2010. GeoTIFF raster file, 12.4 Mo, EPSG: 32738, 250m resolution, Int16. This map doesn't take into account the forest mask for the year 2010.ACD_RF_2010.tifRandom Forests modelRandom Forests model used to derive the national forest carbon map. R data file (.rda), 4.7 Mo.RFmodel.rdaClimatic projections for MadagascarClimatic projections from seven IPCC CMIP5 global climate models (GCMs) for years 2050 and 2080 following two representative concentration pathways (RCPs): RCP 4.5 and RCP 8.5. The seven GCMs are: ACCESS 1.0 (ac), CCSM4 (cc), GISS-E2-R (gs), HadGEM2-ES (he), IPSL-CM5A-LR (ip), MIROC5 (mc) and NorESM1-M (no). The projections are for the following bioclimatic variables: bio1=mean annual temperature, bio4=temperature seasonality, bio12=annual precipitation.climproj.zipShapefile of Madagascar ecoregionsShapefile of Madagascar ecoregions. Code: 1=spiny, 2=mangrove, 3=moist, 4=dry.ecoregions_shp.zipMultiband raster of explicative variables.Multiband raster of explicative variables for the ACD model. The raster includes the following bands: elevation, EVI, VCF, mean annual temperature, temperature seasonality and annual precipitation. EVI is scaled between 0 and 10000 so that the raster type can be "INT2S". EVI must be rescaled between 0 and 1 to be used with the Random Forests model.Xvar.tifMultiband raster for computing ACD by forest typeMultiband raster for computing ACD by forest type. The raster includes four bands for: Madagascar forest extent in 2010, ecoregions, Saatchi's carbon map and Baccini's biomass map.Zvar.tifR script for reproducibilityThe R script can be used to reproduce the results of the study. Associated data are available on the Dryad repository. Load the data in your working directory and create a "./results" directory to save the outputs.Carbonmap_Madagascar_Vieilledent2016_JoE.R