Plant Functional Traits Are Associated with Canopy-to-Soil Hydrological Regulation in Tropical Forests: Evidence from Smallholder Plantations and Secondary Forests

Tropical forest conversion to smallholder plantations alters plant community structure, but the coupled eco-hydrological consequences of trait simplification remain poorly characterised. We compared paired smallholder plantation and secondary forests in Hainan, China (n = 3 plots per type, one-year campaign) by jointly measuring rainfall partitioning, soil physico-chemical properties, steady-state infiltration, fine-root biomass, understory diversity (H′), bark roughness, DBH, LAI and canopy density. Two robust descriptive patterns emerged: smallholder plantations showed approximately 3-fold higher stand-scale stemflow fraction (≈21% vs. 7%) and markedly lower steady-state infiltration than secondary forests, while the DBH–stemflow slope was about 6-fold steeper in the smallholder plantation. Secondary forests concentrated more fine-root biomass in the upper 0–20 cm layer (68% vs. 42% of total) and exhibited higher root–water matching (0.89 vs. 0.41). The four variables implicated in the proposed aboveground route co-varied along the expected sequence (r = 0.78, 0.82, 0.69 and 0.74), but no association reached significance; with only six plot-level observations we report them descriptively rather than as a fitted path model. We propose a coupled management–trait syndrome framework to organise trait-based restoration thinking beyond canopy cover, and frame candidate—illustrative, not statistically derived—indicators (understory H′ ≥ 1.5; BRI ≥ 2.0; MC ≥ 0.80) as candidate indicators for future validation in factorial experiments that cross weeding frequency with species mixtures. Given the paired two-site design (n = 3 plots per type), a ~500 mm site-rainfall difference, together with four non-significant co-variation coefficients, the present contribution is explicitly a hypothesis-generating paired baseline rather than a confirmatory test—providing (i) a multi-indicator reference dataset, (ii) a reusable root–water matching metric, and (iii) candidate restoration indicators for future validation within a tropical Plant–Water–Nutrient programme.