Data from: Phylogenetic diversity outperforms functional diversity in predicting aboveground biomass across tree strata in a subtropical forest
Data files
Jun 09, 2026 version files 203.62 KB
-
env(topographicy_edaphic_factors).xlsx
15.06 KB
-
overstorey_trait.xlsx
13.91 KB
-
plot_inventory.xlsx
155.67 KB
-
README.md
5.65 KB
-
understory_trait.xlsx
13.33 KB
Abstract
Conducted in a subtropical forest in southwestern China, this study evaluates how various dimensions of plant biodiversity regulate aboveground biomass (AGB) across forest strata under consistent edaphic and topographic conditions, addressing the insufficiently elucidated question of which biodiversity dimension most prominently impacts AGB productivity. It reveals that tree species with acquisitive strategies in the overstory and those with conservative strategies in the understory respectively promote AGB accumulation, a key finding linking tree functional strategies to AGB dynamics across different forest layers. Importantly, phylogenetic diversity outperforms functional diversity in explaining AGB accumulation across tree strata, which provides evidence opposite to most existing empirical studies and serves as a valuable supplement to research on how biodiversity affects AGB in subtropical forests. The submitted dataset comprises four integrated components: (1) plot-level tree inventory data, including species identity, diameter at breast height (DBH), and height for all individual trees; (2) environmental covariates, comprising topographic variables (e.g., elevation, slope, aspect) and edaphic properties (e.g., soil pH, organic carbon content, nutrient availability); (3) functional trait values for tree species sampled from the overstorey layer; and (4) functional trait values for tree species sampled from the understory layer. All data were collected under standardized protocols in 30 evergreen broad- leaved forest plots in subtropical southwestern China.
Dataset DOI: 10.5061/dryad.v15dv42c4
Description of the data and file structure
The data we have submitted contain forest plot inventory, environmental factors including topographic and edaphic factors, as well as functional traits of tree species from the overstory and understory layers.
Files and variables
File: plot_inventory.xlsx
Description: Thirty 20 m × 20 m plots were randomly set along an altitudinal gradient from 2450 to 2650 m within the reserve, with the average spacing between adjacent plots exceeding 200 m. We identified species and measured DBH and height for all healthy trees (DBH ≥ 1 cm) in the plots. Tree species were identified following the taxonomic nomenclature of Flora of China and Flora of Yunnan.
Variables
- quadrat (quadrat number)
- LN (abbreviated forms of plant scientific names of the tree species )
- CC (tree trunk circumference measured at breast height, cm)
- DBH ( diameter at breast height (DBH) calculated based on CC, cm)
- H (Tree height, m).
File: env(topographicy_edaphic_factors).xlsx
Description: A total station instrument was used to measure the relative heights of the corners and the center of 30 plots to facilitate subsequent calculations of topographic features. Elevation was determined by calculating the average value of the elevations at each four corners of the plot. Convexity was defined as the altitude of the plot center minus the average altitudes of the four corners. Slope was determined as the mean angle between the plane formed by any three corners and the horizontal plane. Slope aspect was calculated as the angle between the projection of the slope normal line in the horizontal plane and due north.
Each plot was divided into four quadrants, the surface litter layer from the center of the quadrants was removed, then soil samples were gathered at a depth ranging from 0 to 20 cm and homogenized thoroughly. Edaphic factors were determined as following: Walkley-Black analysis for organic matter (SOM), potentiometry for pH, Kjeldahl method for total nitrogen concentration (TN), MoeSb colorimetry for total phosphorus concentration (TP), flame photometry for total potassium concentration (TK), diffusion method for available nitrogen concentration (AN), Bray I for available phosphorus concentration (AP), ammonium acetate method for available potassium concentration (AK) and azomethine-H method for available boron concentration (AB).
Variables
- Elevation (m)
- Convex (convexity,no unit)
- Aspect (slope aspect, no unit)
- Slope (no unit)
- SW (soil water content, %)
- pH (soil pH value, no unit)
- SOM (soil organic matter content, g/kg)
- TN (total nitrogen content, %)
- TP (total phosphorus content, %)
- TK (total potassium content, %)
- AN (available nitrogen content, mg/kg)
- AP (available phosphorus content, mg/kg)
- AK (available potassium content, mg/kg)
- AB (available boron content, mg/kg).
File: understory_trait.xlsx
Description: Specific leaf area (SLA) was determined as one-side area of a fresh leaf divided by its oven-dry mass. Kjeldahl method, MoeSb colorimetry, flame photometry and potassium dichromate method were applied to determine leaf nitrogen (LN), phosphorus (LP), potassium (LK) and carbon (LC). Woody density (WD) was measured as dry mass (via constant-weight drying) divided by volume ( via water displacement).
Variables
- SLA (cm2/g)
- N (leaf nitrogen content, g/kg)
- P (leaf phosphorus content, g/kg)
- K (leaf potassium content, g/kg)
- C (leaf carbon content, %)
- WD (woody density, g/cm3).
File: overstorey_trait.xlsx
Description: Specific leaf area (SLA) was determined as one-side area of a fresh leaf divided by its oven-dry mass. Kjeldahl method, MoeSb colorimetry, flame photometry and potassium dichromate method were applied to determine leaf nitrogen (LN), phosphorus (LP), po-tassium (LK) and carbon (LC). Woody density (WD) was measured as dry mass (via constant-weight drying) divided by volume ( via water displacement).
Variables
- include SLA (cm2/g)
- N (leaf nitrogen content, g/kg)
- P (leaf phosphorus content, g/kg)
- K (leaf potassium content, g/kg)
- C (leaf carbon content, %)
- WD (woody density, g/cm3).
Code/software
Data analysis was conducted in R (version 4.3.1; R Core Team, 2023) using the following workflow and packages: Principal component analysis (PCA) was performed on the selected topographic and edaphic variables in order to reduce the dataset dimensionality. Functional diversity metrics were calculated based on above measured functional traits using R package "FD". Based on the constructed phylogeny, the phylogenetic diversity indices were calculated using the ‘picante’ package in R. The relationship between biodiversity indices and AGB was evaluated by applying partial least squares (PLS) regression.The “shaving” function was utilized to remove the less informative predictors of AGB. The variable importance for projection (VIP) of the retained predictors was computed via the “plsreg2” function.Linear regression models were fitted to depict the relationships between AGB and the informative predictors for both the overstory and understory layers, using the R package “ggplot2”.
Access information
Other publicly accessible locations of the data:
- n/a
Data was derived from the following sources:
- field inventory and lab analysis