Supporting data and code for A global map of root biomass across the world's forests

This dataset contians supporting data and code for the manuscript: A global map of root biomass across the world's forests. 

Folder data_orin include raw data and references. Python script data_merge_fill.py merges different raw databases, removes repeated entries and fills data gaps in environmental factors. Final output of data_merge_fill.py is site_filled_1kmv1.csv. Python scripts in fig folder include scripts for calculation and producing figures in the manuscript. pred_1km_15cover_1000_99.py is an example code to train a random forest model and to make a global forest root biomass map with a unit of weight biomass per tree. pred_1km_15cover_allometry_1000_99.py generates data used in Supplementary Figure 7 which estimates global forest root biomass through allometric equations. Folder budget_uncertainty contains code and calculations of the uncertainty of global and biome scale root biomass budget. 

The  pergridarea_agb.nc contains global forest aboveground biomass,  pergridarea_bgb.nc is the belowground biomass,  err_bgb_15_99_ncomm_revise.nc is uncertainty of belowground biomass maps with a resolution of 1km and the unit of Mg/ha (gridcell area).