Depth-dependent trace metal speciation, bioavailability and soil-plant transfer across 55 months of vetiver phytoremediation at Bien Hoa Airbase, Vietnam.

Long-term field evidence on how trace metal speciation and bioavailability evolve during phytoremediation of legacy-contaminated soils remains scarce, limiting exposure-relevant appraisal of residual contamination. Over 55 months, this study evaluated the performance of vetiver grass (Chrysopogon zizanioides) across depth-stratified soil profiles (0-10, 10-30, and 30-50 cm) at the contaminated Bien Hoa Airbase, Vietnam, combining a spatial survey of surface soils (n = 53) with depth-resolved profiling (n = 13 per horizon). Modified BCR sequential extraction, plant tissue analyses and ecological risk indices (mRAC, mHQ, ECI) were applied. Cd was the dominant contaminant and the only element with bioaccumulation factors (BAF) consistently exceeding 1.0, with its BAF increasing 2.3-fold over the 40-month assessment period. Speciation was element-specific: Cd occurred predominantly in labile fractions (F1 + F2 > 68%), whereas Cr and Ni remained largely residual. Although higher relative labile fractions were observed in vetiver-treated plots, absolute bioavailable concentrations declined more rapidly with depth, a pattern consistent with rhizosphere-mediated mobilisation coupled with plant interception, although the observational field design does not allow this mechanism to be established directly. An accidental fire at month 52 removed the canopy but spared the root system; enhanced BAF and translocation factors at 55 months may reflect post-fire rejuvenation, although confirmation requires further monitoring. Ecological risk indices classified 75% of locations as non- to moderately contaminated (ECI < 4), whereas the toxicity-weighted mRAC remained high, consistent with continued rhizosphere-mediated mobilisation of the labile metal pool. These findings support managed harvest-regrowth cycling to sustain phytoextraction and show that bioavailability-based indicators, rather than total concentrations, should guide decisions on when residual contamination still warrants exposure control.