Both inter‐ and intraspecific trait variation are critical to species distribution along environmental gradients, but our understanding of these patterns predominantly relies upon species‐level trait means and variances. Trait integration, defined as how strongly multiple traits covary with one another, is a key indicator of the dimensionality of functional space for accommodating biodiversity in communities. As trait covariance can differ dramatically at the interspecific versus intraspecific levels, how intraspecific trait variability alters the strength of trait integration and eventually modulates biodiversity along environmental gradients has been rarely tested. Here, we measured nine functional traits (leaf area, specific leaf area, leaf and stem dry‐matter content, leaf nitrogen and phosphorus contents, specific stem length, Huber value and maximum height) paired with site‐specific soil fertility for 70 woody communities in subtropical Chinese forests. All species‐by‐site combinations were sampled to ensure a sufficient representation of intraspecific trait variation across sites. Community‐level trait integration was quantified from the variance of eigenvalues of the trait correlation matrix. The direct and/or indirect effects of soil fertility and trait integration on species richness and trait diversity were assessed through path analyses. Trait integration quantified from both inter‐ and intraspecific variances was on average 21.7% weaker than that from only interspecific variance, indicating a crucial role of intraspecific trait variability in promoting niche dimensionality. Whether accounting for intraspecific variation or not, less fertile sites had stronger trait integration, which in turn depressed both taxonomic and functional diversity, supporting the assumption that higher environmental stress demands stronger tradeoffs among multiple functions in viable strategies. Importantly, the negative association between trait integration and species richness became stronger when accounting for intraspecific variation, suggesting that species distribution and occurrence can be a consequence of intraspecific trait variability. This study highlights the importance of intraspecific trait variability in understanding functional tradeoffs underlying biodiversity patterns.