Overview

This dataset (submitCode_(2).zip) is a complementary compilation designed to support research on maximum biomass (Mₘₐₓ), carbon fluxes (e.g., NPP, GPP), and their biogeographical patterns, drivers, and future changes across global woody plant communities (WPCs). They integrate field measurements, database extractions, digitized figure data, and processed global raster products, with consistent curation to ensure compatibility for modeling, validation, and biogeographical analysis.

The main folder submitCode.zip contains the following files:

keydata.csv

Integrates field-based measurements from multiple studies, with the core component being standardized data from Michaletz et al. (2014) — covering biomass, growth, mortality, and climate variables for 1,247 WPCs across broad climatic gradients. Supplementary data from Chen et al. (2016) and Sullivan et al. (2020) are included to enhance coverage of boreal and tropical forest dynamics. This dataset serves as the primary input for model fitting (e.g., IGMF parameterization) and initial validation.

ForC.csv

Contains harmonized records of ecosystem carbon stocks and fluxes extracted from the Global Forest Carbon (ForC) database (Anderson-Teixeira et al. 2018). Key variables include NPP, GPP, biomass, mortality, stand age, and mean annual temperature (MAT), with biomass-related values standardized to grams of biomass (assuming 50% carbon content). It is used for cross-validating model predictions of carbon fluxes.

Fig.csv

Comprises digitized data from published scientific figures, focusing on bioclimatic trend data such as biome-level tree lifespan–MAT relationships (Locosselli et al. 2020). These data are used to compare model-derived predictions of functional traits (e.g., lifespan) with observed biogeographical patterns.

xex.csv

Compiles field measurements of mature or old-growth forests extracted from the literature (e.g., Anderson-Teixeira et al. 2016; see Appendix S1: Table S2 for full citations). It includes variables related to aboveground biomass, growth, and mortality, and is used to validate model predictions of biomass dynamics in late-successional ecosystems.

global.csv

Contains site- and grid-level environmental variables extracted from global raster datasets via Google Earth Engine (GEE) and WorldClim. Variables include productivity (e.g., MODIS products), land cover, 19 bioclimatic variables, atmospheric CO₂ concentrations, soil properties (texture, moisture, organic matter), species richness, and future climate projections (2020–2100). All values are processed derivatives (spatial sampling, temporal averaging, or standardized products) rather than raw raster files, tailored for global-scale Mₘₐₓ estimation and future change projections.

Unified Data Processing: All datasets underwent systematic curation, standardization, and harmonization to ensure consistency.

Biomass conversion: Aboveground biomass was scaled to total biomass using taxon-specific ratios.

WPC classification: Pixels were identified as WPCs if vegetation cover exceeded 0.2 (based on land cover datasets).

NPP partitioning: NPP was split into structural (NPPₛ) and non-structural (NPPₙ) components to align with model requirements.

Unit harmonization: Carbon fluxes (NPP/GPP) and biomass variables were standardized to consistent units for cross-dataset comparison.

Note: Processed raster derivatives (e.g., WorldClim bioclimatic variables, MODIS productivity) comply with open-access license requirements and are ready for direct use in biogeographical analysis, ecosystem modeling, and climate change impact assessments.

Data Availability

Dataset

Corresponding Figures

keydata.csv: Fig.2A/B, Fig.3B/C/D

ForC.csv: Fig.2C/D, Fig.3D

Fig.csv: Fig.3A

xex.csv: Fig.3C

global.csv: Fig.4, Fig.5, Fig.6, Fig.7, Fig.S5, Fig.S6, Fig.S7, Fig.S10

Data Sources

All original data sources are publicly available with the following corrected citations and access links (addressed link confusion, invalid paths, and misaligned references):

1. Core Field & Database Data

Michaletz et al. (2014): Convergence of terrestrial plant production across global climate gradients (Nature).

https://doi.org/10.1038/nature13470

Chen et al. (2016): Climate change-associated trends in net biomass change are age dependent in western boreal forests of Canada (Ecology Letters).

https://doi.org/10.1111/ele.12649

Note: Separated from Luo et al. (2019) (previously conflated); this is the original research article for boreal forest dynamics, while Luo et al. (2019) is the associated dataset.

Luo et al. (2019): Data from: Divergent temporal trends of net biomass change in western Canadian boreal forests (Dryad Dataset).

https://doi.org/10.5281/dryad.8bg44b0

Anderson-Teixeira et al. (2018) (ForC database): ForC: A global database of forest carbon stocks and fluxes (Ecology).

Database access:

https://github.com/forc-db/ForC;

Original article: https://doi.org/10.1002/ecy.2408

Sullivan et al. (2020): Long-term thermal sensitivity of Earth’s tropical forests (Science) & associated data.

Research article: https://doi.org/10.1126/science.aaw7578; Data package: http://dx.doi.org/10.5521/forestplots.net/2020_2

Locosselli et al. (2020): Global tree-ring analysis reveals rapid decrease in tropical tree longevity with temperature (PNAS). 

https://www.pnas.org/doi/10.1073/pnas.2003873117

Anderson-Teixeira et al. (2016): Carbon dynamics of mature and regrowth tropical forests derived from a pantropical database (TropForC-db) (Global Change Biology). https://doi.org/10.1111/gcb.13226

2. Global Raster Data Sources (for global.csv)
The global raster data used to generate global.csv were extracted and processed as described in the Methods section of the manuscript, including MODIS products (productivity, land cover), WorldClim bioclimatic variables and future projections, Google Earth Engine processing, soil properties, biome classification, and species richness layers.

Variable Definitions

All variable names, descriptions, and units used in the datasets are provided in Variable_names.csv. This file contains the following columns:

Variable. Name — the name of the variable used in the datasets.

Units — the units of measurement for each variable.Description — a brief explanation of the variable, including ecological or methodological context where applicable.

Readers can refer to Variable_names.csv for complete and standardized information on all variables used in keydata.csv, ForC.csv, Fig.csv, xex.csv, and global.csv.

R Code Description

Figs 2_and_3_(including S5-S7).R: Plot Fig.2-3 & Fig.S5-S7 (s=supplementary): This code loads forest productivity datasets, fits two nonlinear biomass-temperature-age models, uses them to generate NPP predictions, and produces multiple comparative plots (A–D) evaluating fitted vs. observed NPP and related ecological parameters.

Figs_4_5_6_and_7_(including_s10).R: Plot Fig.4-7 & Fig.S10 (s=supplementary): This code rasterizes global point data to analyze and visualize the spatial and latitudinal patterns of Mₘₐₓ and its normalized ratio (M/Mₘₐₓ), comparing overall trends and differences between forest and shrub vegetation types using maps, latitudinal profiles, and boxplots.