Vetiver Grass Strips Cadmium From a Toxic Vietnam Airbase, and a Fire May Have Helped It

Bien Hoa Airbase in southern Vietnam is one of the most notorious legacy-contamination sites on Earth, a place where decades of Agent Orange storage and use left soils laced not only with dioxins but with a cocktail of toxic trace metals. Now, a team of Vietnamese and Belgian researchers has delivered something rare in the phytoremediation world: nearly five years of continuous field evidence showing exactly how the metal contamination behaves in the soil profile as vetiver grass works its roots through it. The study, published in Environmental Geochemistry and Health, tracked depth-dependent metal speciation, bioavailability and soil-to-plant transfer across 55 months of vetiver treatment, and its findings challenge some of the most common assumptions about how green cleanup should be judged.

The research focused on vetiver grass, Chrysopogon zizanioides, a dense, deep-rooted tropical grass that has become a global workhorse of phytoremediation. What makes this study unusual is its longevity and its resolution. Rather than a single snapshot of topsoil chemistry, the team combined a spatial survey of 53 surface soil samples with depth-resolved profiling at 13 locations per horizon, sampling three distinct layers: 0 to 10 centimeters, 10 to 30 centimeters, and 30 to 50 centimeters. This vertical structure matters enormously, because the metals that threaten human health are not uniformly distributed through the ground, and the mechanisms that plants use to intercept them operate differently at different depths.

To understand what the plants were actually dealing with, the researchers applied a modified BCR sequential extraction procedure, a laboratory technique that chemically partitions each metal into fractions of increasing binding strength. The first two fractions, known as F1 and F2, represent the exchangeable and reducible pools, the most labile and bioavailable forms of a metal. Metals locked in these fractions can dissolve into soil pore water, be taken up by roots, and move through the food chain. Metals trapped in residual fractions, by contrast, are essentially geochemically inert, bound within mineral lattices and largely irrelevant to immediate exposure risk. The distinction between these pools is the difference between contamination that matters and contamination that merely exists.

The results revealed a striking element-specific pattern. Cadmium emerged as the dominant contaminant and the clear star of the show, occurring predominantly in labile fractions, with the sum of F1 and F2 exceeding 68 percent of total cadmium in the soil. In other words, more than two-thirds of the cadmium at Bien Hoa was sitting in forms that plants can readily access. Chromium and nickel told a completely different story: both remained largely residual, locked away in forms that vetiver could barely touch. This divergence is not a technical footnote. It means that a single total-metal measurement would have painted a misleading picture of the site, lumping together a metal that is actively cycling through the rhizosphere with metals that are essentially frozen in place.

The cadmium story grew more dramatic over time. The bioaccumulation factor, or BAF, which expresses the ratio of metal concentration in plant tissue to its concentration in soil, was the only metric that consistently exceeded 1.0 for any element, meaning vetiver was actively concentrating cadmium in its tissues above ambient soil levels. Over the 40-month assessment period, this BAF increased 2.3-fold, a signal that the plants were becoming progressively more efficient at pulling cadmium from the ground as the vetiver stand matured and its root systems expanded deeper into the profile.

One of the most intriguing findings concerns the vertical distribution of bioavailable metal. Although the relative proportion of labile fractions was actually higher in vetiver-treated plots, the absolute concentrations of bioavailable metals declined more rapidly with depth in those same plots. The researchers interpret this as consistent with rhizosphere-mediated mobilisation coupled with plant interception: the roots and their associated microbial communities chemically mobilise metals from the solid phase, and the plant then intercepts the mobilised pool before it can migrate. The authors are careful to note that the observational field design does not allow this mechanism to be established directly, a candid acknowledgment that distinguishes correlation from causation in a system as complex as a living soil profile.

Then nature intervened. At month 52, an accidental fire swept through the site, removing the vetiver canopy but sparing the root system beneath the soil surface. This unplanned event, unwelcome as it was, created an accidental experiment. When the researchers returned at month 55, they found enhanced bioaccumulation factors and enhanced translocation factors, the latter measuring how efficiently metals move from roots to shoots. The team suggests this may reflect post-fire rejuvenation, a burst of new growth drawing on an intact root system and possibly on nutrient pulses from ash, though they caution that confirmation requires further monitoring. The episode nonetheless hints at a practical management insight: controlled disturbance that stimulates regrowth might actually sustain phytoextraction rather than interrupt it.

The ecological risk indices told a nuanced story. The ECI, or ecological contamination index, classified 75 percent of locations as non- to moderately contaminated, with values below 4. On its face, that sounds reassuring. But the toxicity-weighted mRAC, a modified risk assessment code that weights labile fractions by each metal’s toxicity, remained high across the site. This discrepancy is the study’s central message: total concentrations and even broad contamination indices can mask a persistent, mobile, plant-accessible pool of the most dangerous metals. The toxicity-weighted picture remained consistent with continued rhizosphere-mediated mobilisation of labile cadmium, meaning the cleanup was working but the hazard had not been neutralised.

What emerges from 55 months of field data is a set of practical recommendations that could reshape how legacy-contaminated sites are managed. The findings support managed harvest-regrowth cycling, in which vetiver biomass is periodically cut and removed, taking its accumulated cadmium with it, while the root system persists and continues extracting from depth. This is not a passive green cover but an active industrial process, one that requires deliberate intervention to sustain. The fire at month 52, whatever its immediate costs, inadvertently demonstrated that the root system is the true engine of the operation, and that stimulating its regrowth may be the key to long-term extraction efficiency.

Perhaps the most consequential implication is methodological. The authors argue that bioavailability-based indicators, rather than total concentrations, should guide decisions about when residual contamination still warrants exposure control. This is a deceptively simple statement with profound regulatory consequences. Sites are routinely classified and released based on total metal loads, a metric that can overstate risk for geochemically bound metals and understate it for labile ones. At Bien Hoa, a site judged moderately contaminated by broad indices harbored a cadmium pool that was overwhelmingly bioavailable and actively moving into plant tissue. The lesson extends far beyond Vietnam: for any legacy-contaminated soil, the question that matters is not how much metal is present, but how much of it is reachable, and by what.

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

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

Article References: Depth-dependent trace metal speciation, bioavailability and soil–plant transfer across 55 months of vetiver phytoremediation at Bien Hoa Airbase, Vietnam. (n.d.). https://doi.org/10.1007/s10653-026-03499-9

Image Credits: AI Generated

DOI: 10.1007/s10653-026-03499-9

Keywords: Depth-dependent, trace, metal, speciation, bioavailability, soil, plant, transfer, across, months, vetiver, phytoremediation