Geographic variation in Dendrocalamus latiflorus culm physical and mechanical traits is associated with source-climate gradients
Related Experiment Video
Updated: Oct 31, 2026

Relating Stomatal Conductance to Leaf Functional Traits
Published on: October 12, 2015
Ying Fang1, Hang Tao1, Jiannan Lin1
1College of Forestry, Fujian Agriculture and Forestry University, Fuzhou, China.
Introduction:
Geographic variation in bamboo culm traits can reveal associations between source environments and biomechanical performance and can help prioritize germplasm for further testing.
Methods:
We quantified 11 physical and mechanical traits in 31 geographic sources of Dendrocalamus latiflorus from nine provincial-level regions of China and related source-level means to 25 geographic and climatic variables. Analyses combined coefficients of variation, Pearson correlations with Benjamini-Hochberg false-discovery-rate correction, partial least squares regression (PLSR) with leave-one-out cross-validation and 1,000 bootstrap resamples, Moran's I tests of residual spatial autocorrelation, principal component analysis, objectively validated Ward clustering, and a seven-trait entropy-weighted TOPSIS evaluation with equal-weight sensitivity analysis.
Results:
Trait coefficients of variation ranged from 7.1% to 24.4%. After correction of 100 Pearson tests, 32 associations remained significant at q < 0.05; the strongest was between precipitation of the warmest quarter (BIO18) and compressive strength parallel to grain (CSPG; r = 0.658, q = 0.0058). PLSR showed the highest cross-validated performance for CSPG (Q2 = 0.599), moderate performance for air-dry and oven-dry density, and weak performance for modulus of rupture (Q2 = 0.076). Raw density and CSPG values were spatially autocorrelated, but PLSR residuals were not. Two validated trait classes showed no significant differences in six key source-climate variables. Entropy-weighted TOPSIS ranked P6, P4, and P19 highest; equal weighting produced strongly concordant ranks (Spearman rho = 0.911).
Discussion:
These results identify robust climate-associated trait patterns and candidate sources for further testing while avoiding causal inference from source-site sampling.
Related Concept Videos
Global Climate Change
Adaptations that Reduce Water Loss
Meristems and Plant Growth
What is Climate?
Responses to Drought and Flooding
Primary and Secondary Growth in Roots and Shoots
