Six months after slash-and-burn, regenerating individuals maintain NSC pools comparable to undisturbed trees while exhibiting belowground allocation constraints and acquisitive leaf functional shifts during structural recovery. Non-structural carbohydrates (NSC) are critical for tree persistence following disturbances, yet the physiological mechanisms governing biomass recovery in tropical dry forests remain underexplored. Here, we investigated how carbon reserves and leaf functional traits are coordinated during early post-disturbance regeneration. We conducted an in situ experiment in a Caatinga dry forest (six species; 75 undisturbed, 75 disturbed trees). Trees were cut at ground level and biomass burned at the end of a dry season; 23 burned individuals regenerated within the subsequent six-month rainy season. We quantified soluble sugars, starch, and total NSC (soluble sugars + starch) in leaves, stems, and roots, measured structural characteristics, and assessed six leaf functional traits, namely leaf area, specific leaf area, thickness, dry matter content, succulence, and construction cost. Regenerating individuals maintained soluble sugar gradients similar to undisturbed trees, with concentrations in leaves and stems four- and two-fold higher than those in roots, respectively. They exhibited NSC concentrations comparable to undisturbed trees but developed thinner leaves with lower dry matter content. Structural recovery was negatively correlated with root NSC, a pattern consistent with short-term allocation constraints during early recovery. Trait–NSC relationships revealed coordinated variation among carbon pools and leaf functional traits. In summary, regenerating individuals display coordinated adjustments in reserve concentrations and leaf traits consistent with resource-acquisitive strategies during early recovery. However, reserves availability alone does not guarantee persistence. Under climatic change or recurrent disturbance, such recovery-stage allocation dynamics may become increasingly constrained, potentially increasing forest vulnerability.
Vanderlei et al. (Sat,) studied this question.