Data and code from: Biodiversity–carbon relationships diverge in forests dominated by different species within the same plant functional type
Data files
Apr 22, 2026 version files 228.12 KB
-
Linear_mixed-effects_models.R
88.34 KB
-
Piecewise_structural_equation_modeling.R
22.38 KB
-
Plot_Cunninghamia_lanceolata.csv
57.66 KB
-
Plot_Pinus_massoniana.csv
45.18 KB
-
Quantile_age-sequence_analysis.R
10.20 KB
-
README.md
4.37 KB
Abstract
Theoretical and experimental evidence suggests that tree biodiversity enhances forest ecosystem functioning. However, whether these relationships hold within forest types dominated by different species of the same plant functional type remains unclear. We analysed 772 plots in subtropical China, contrasting high- and low-carbon-sequestration plantations of Cunninghamia lanceolata and Pinus massoniana along a chronosequence, to explore causal links among aboveground carbon storage (AGC), biodiversity, stand structure, and environmental factors. We found that the effects of biodiversity on AGC varied across forest types. Specifically, species diversity and functional diversity enhanced AGC in high-carbon-sequestration C. lanceolata plantations, whereas species and functional diversity of co-occurring broadleaf species had negative effects. In high-carbon-sequestration P. massoniana plantations, phylogenetic diversity and diameter at breast height (DBH) variation positively influenced AGC, outweighing the negative effects of species diversity. Low-carbon-sequestration plantations exhibited weak or non-significant biodiversity–AGC relationships. Additionally, soil attributes directly reduced AGC in low-carbon-sequestration C. lanceolata plantations, indicating nutrient limitations under intensive management. We demonstrate that biodiversity–ecosystem functioning relationships diverge among forest types dominated by different species within the same plant functional type, and that biodiversity effects on AGC vary across these plantations. In soil-limited conifer plantations (C. lanceolata), increasing species diversity enhances AGC, whereas in light-demanding conifer plantations (P. massoniana), phylogenetic diversity and DBH variation, capturing species similarity and niche occupancy, better explain AGC than species or functional diversity. We advocate for tailored management strategies that consider dominant tree species identity to optimise biodiversity and carbon sequestration.
Dataset DOI: 10.5061/dryad.np5hqc08x
Description of the data and file structure
The dataset supports analyses of biodiversity–ecosystem functioning relationships and aboveground carbon storage in managed plantation forests, including the generation of figures presented in the associated manuscript.
Files and variables
File: Linear_mixed-effects_models.R
Description: R script used to fit linear mixed-effects models (LMMs) to evaluate the effects of biodiversity, stand structure, and environmental variables on aboveground carbon storage (AGC).
File: Piecewise_structural_equation_modeling.R
Description: R script used to perform piecewise structural equation modelling (PSEM) to quantify direct and indirect relationships among biodiversity, stand structure, environmental variables, and AGC.
File: Quantile_age-sequence_analysis.R
Description: R script used to estimate aboveground carbon sequestration rates along stand age using quantile regression (10th, 50th, and 90th percentiles).
File: Plot_Cunninghamia_lanceolata.csv
Description: Plot-level dataset for Cunninghamia lanceolata plantations.
Variables
- No: Row number
- Plot: Unique identifier for each sampling plot
- RegionName: Administrative region (county-level) where the plot is located
- AverageDBH: Mean diameter at breast height of trees in the plot (cm)
- Density: Stand density (number of trees per hectare, stems ha -1)
- DBH_CV: Coefficient of variation of DBH
- Shannon: Shannon diversity index of all tree species
- shan_Broadleaf: Shannon diversity index of broadleaf species only
- PD: Phylogenetic diversity (Faith’s PD) of all species
- PD_Broadleaf: Phylogenetic diversity of broadleaf species
- FDis: Functional dispersion of all species
- FD_Broadleaf: Functional dispersion of broadleaf species
- MAT: Mean annual temperature (°C)
- MAP: Mean annual precipitation (mm)
- SoilThickness: Soil depth (cm)
- HumusLayerThickness: Thickness of the humus layer (cm)
- LitterThickness: Thickness of the litter layer (cm)
- Altitude: Elevation above sea level (m)
- AGC: Aboveground carbon storage (Mg ha -1)
- Age: Stand age (years)
- Group: Classification of plots into low-carbon-sequestration (LCS = 1) and high-carbon-sequestration (HCS = 2) based on AGC quantiles
- MaxDBH: Maximum DBH (cm)
File: Plot_Pinus_massoniana.csv
Description: Plot-level dataset for Pinus massoniana plantations.
Variables
- No: Row number
- Plot: Unique identifier for each sampling plot
- RegionName: Administrative region (county-level) where the plot is located
- AverageDBH: Mean diameter at breast height of trees in the plot (cm)
- Density: Stand density (number of trees per hectare, stems ha -1)
- DBH_CV: Coefficient of variation of DBH
- Shannon: Shannon diversity index of all tree species
- shan_Broadleaf: Shannon diversity index of broadleaf species only
- PD: Phylogenetic diversity (Faith’s PD) of all species
- PD_Broadleaf: Phylogenetic diversity of broadleaf species
- FDis: Functional dispersion of all species
- FD_Broadleaf: Functional dispersion of broadleaf species
- MAT: Mean annual temperature (°C)
- MAP: Mean annual precipitation (mm)
- SoilThickness: Soil depth (cm)
- HumusLayerThickness: Thickness of the humus layer (cm)
- LitterThickness: Thickness of the litter layer (cm)
- Altitude: Elevation above sea level (m)
- AGC: Aboveground carbon storage (Mg ha -1)
- Age: Stand age (years)
- Group: Classification of plots into low-carbon-sequestration (LCS = 1) and high-carbon-sequestration (HCS = 2) based on AGC quantiles
- MaxDBH: Maximum DBH (cm)
Code/software
All analyses were conducted in R version 4.3.2.
Access information
Other publicly accessible locations of the data:
- None.
Data was derived from the following sources:
- Forest inventory data were collected from permanent plots in Hunan Province, China. Functional trait data were compiled from the Flora of China, the Global Wood Density Database, the China Plant Trait Database, and the TRY database. Climatic data (mean annual temperature and precipitation) were obtained from the National Tibetan Plateau Data Center.
