Heavy Metal Accumulation in Rice and Aquatic Plants Used as Human Food: A General Review

This 2021 narrative review in Toxics synthesizes the global literature on heavy metal accumulation in rice (Oryza sativa) and edible aquatic plants — water spinach (Ipomoea aquatica), Indian lotus (Nelumbo nucifera), watercress (Nasturtium officinale), Chinese water chestnut (Eleocharis dulcis), water mimosa (Neptunia oleracea), taro (Colocasia esculenta), and others. The authors compile occurrence values from sites across Bangladesh, Sri Lanka, Thailand, China, Australia, and Nigeria, showing that Cd, As, Cr, Pb, and Hg repeatedly appear in these food plants at concentrations exceeding WHO/FAO permissible limits. The review identifies rice as the dominant Hg exposure pathway for inland Asian populations through paddy methylation, in contrast with the fish-mediated pathway elsewhere.

Key numbers

Permissible limits compiled from WHO/FAO references [63,64] in Table 2 (mg/kg, cereals and vegetables):

• As 0.1–0.2; Cd 0.05–0.4; Co 0.01; Cr 1.3; Cu 20.0; Hg 0.03; Ni 0.1; Pb 0.05–0.3; Zn 60.0.

Observed concentrations compiled across cited studies (Table 2, mg/kg wet produce; asterisk = exceeds the permissible limit in at least one cited study):

• Rice (Oryza sativa): As 0.09–0.13, Cd 0.003–0.06, Cr 0.12–0.37, Cu 2.6–5.3, Hg 0.002–0.034*, Ni 0.25*, Pb 0.01–0.53*, Zn 16–36.
• Water spinach (Ipomoea aquatica): Cd 0.06–1.10*, Co 0.04–0.09, Hg 1.44*, Pb 0.28.
• Indian lotus (Nelumbo nucifera): As 0.1–1.3*, Cd 0.04–0.09, Cr 1.6–2.2*, Cu 4.4–7.4, Pb 0.3–0.8*, Zn 9.8–15.4.
• Watercress (Nasturtium officinale): As 2.0*, Cd 0.10, Co 0.30*, Cr 0.34, Ni 0.34*, Pb 0.86*.

Site-level findings cited in the body:

• Greater Bangkok region, Thailand: nine-site survey of I. aquatica reported Hg “up to 1.440 µg kg⁻¹ dry weight” in the body text (p.4, citing Göthberg et al. 2002 [38]); the same reference is summarized in Table 2 as Hg 1.44 mg/kg wet produce. The two values cannot both be correct; the paper does not reconcile them (see verification notes).
• Yichang City, China (unmanaged pond): Nelumbo nucifera tubers exceeded the national food standard for Cd and As; edible-tuber peel concentrations were 1.3–9.0× higher than inner flesh (p.4, citing Luo et al. 2017 [39]).
• Rice–fish co-culture: Cd in rice grain reached 5.86 mg/kg when Cd was added to the water, well above the cereal threshold (p.4, citing Luo et al. 2020 [42]).

Bioconcentration factors (BCF) compiled for brown rice (p.5, citing Lee et al. 2017 [47]): As 0.001–0.224, Cd 0.001–2.434, Pb 0.001–0.048.

Exposure/health-effect anchors:

• Daily Intake Rate equation (p.12, Eq. 1): DIR = C × IR / BW, where C is the heavy metal concentration in the food (mg/kg fresh weight), IR is plant-food ingestion rate (g fresh weight per person per day), and BW is mean adult body weight (kg).
• Nigeria rice study (p.10, citing Otitoju et al. 2014 [71]): mean blood Pb in children 1–6 years old was 10.6 µg/dL, exceeding the CDC reference value of 10 µg/dL; Pb concentrations in studied rice varieties exceeded the recommended tolerable weekly intake.

Methods (brief)

Narrative review; no primary measurements. The authors aggregate values from approximately 140 cited references (1949–2021), with most occurrence data drawn from regional surveys using ICP-MS, ICP-OES, AAS, and HG-AAS reported in the original studies; analytical detail varies by source and is not standardized in the review. Table 2 compiles permissible-limit and occurrence data without normalizing wet-weight vs dry-weight basis. Table 3 catalogues main oxidation states and natural/anthropogenic sources for As, Cd, Cr, Pb, and Hg. Tables 4 and 5 catalogue health endpoints and mitigation techniques. The review does not separate inorganic from total arsenic, nor methylmercury from total mercury, in its compiled occurrence values — those are reported in the source page as tAs and tHg. The authors explicitly note as limitations: (i) absence of focus on plant adaptation/coping mechanisms; (ii) absence of international policy and regulatory framework analysis.

Implications

Certification: Contributes occurrence data for rice (As, Cd, Pb, Hg, Ni) and four edible aquatic-plant categories (water spinach, Indian lotus, watercress, taro). The Pb-in-rice range cited (0.01–0.53 mg/kg, with upper end above WHO/FAO cereal limits) is a rice-grain occurrence range, not specific to any processed rice product. The peel-vs-flesh contrast in lotus tuber (1.3–9.0× higher in peel) is a single-study finding cited via Luo et al. 2017 that contributes occurrence data on within-tuber distribution.

Courses: Useful overview-level source for modules on aquatic food-plant contamination, paddy methylation of Hg, bioconcentration factor (BCF), and the distinction between essential micronutrients (Fe, Mn, Cu, Mo, Ni, Zn) and non-essential toxic metals (As, Cd, Cr, Pb, Hg). The DIR equation and worked Nigeria blood-Pb anchor support exposure-translation modules.

App: Reinforces rice as a high-priority ingredient for As and Cd profiling. Supports flagging water spinach, lotus root, and watercress as higher-risk aquatic vegetables when sourced from regions with documented industrial or mining contamination of irrigation water.

Verification notes

• Enhanced 2026-05-18 from prior 2026-05-14 revision. Cite_key, DOI, raw_path, license preserved.
• Numerical corrections to Table 2 transcription. The prior revision had column-shifted values for Rice, Water Spinach, and Watercress rows. Corrected against the source PDF Table 2 (p.8): Rice Pb is 0.01–0.53* (not 0.12–0.37), Rice Cr is 0.12–0.37 (not 2.6–5.3), Rice Cu is 2.6–5.3 (not previously reported), Rice Ni is 0.25* (not 0.01–0.53). Water spinach Cd is 0.06–1.10* (not Pb), water spinach Hg is 1.44* (not Cd); Zn 16–36 belongs to the Rice row, not Water Spinach. Watercress Cr is 0.34 (not 0.30), watercress Co is 0.30* (newly included), watercress Ni 0.34* added. All values cross-verified to Table 2 columns.
• Paper-internal Hg inconsistency for water spinach. The body (p.4) cites Göthberg et al. 2002 as “Hg concentrations of up to 1.440 µg kg⁻¹ dry weight”; Table 2 summarizes the same reference as 1.44 (mg/kg wet produce). The difference is three orders of magnitude. Reported here as the paper presents both, without choosing one over the other. Downstream synthesis should consult Göthberg et al. 2002 directly before using either value.
• Speciation discipline. The review does not separate iAs/tAs or MeHg/tHg in its occurrence tables, so frontmatter uses tAs and tHg. The metals field expanded from the prior revision to include Cu, Zn, and Co because Table 2 reports occurrence values for those three across the cited species.
• Raw handle correction. Changed from non-standard papers-cube to PCMF_article-1-copy-5 per the manual-fetch handle convention.
• Matrices. Updated from [rice, aquatic-vegetables, irrigation-water] to [rice, leafy-vegetable, root-vegetable, irrigation-water] to align with the controlled vocabulary in docs/gpt-collaboration/system-prompt.md; aquatic-vegetables is not in the vocabulary, and leafy-vegetable + root-vegetable better describe what the review actually compiles (water spinach/watercress leaves; lotus tuber, taro corm).
• Routing advisory. products: [] is correct for this review — it is a synthesis across food-source categories, not a product-category study; the advisory entry in routing_malformed.csv is expected and non-blocking for a review of this kind.
• Missing-ingredient-slug proposals. Four declared ingredient slugs do not currently exist as wiki pages (verified 2026-05-18 against wiki/ingredients/ directory): water-spinach, lotus-root, watercress, taro. Each is contributed to by this A-tier review; the freq-2+ auto-stub script will pick them up if a second contributing source appears, per Part 10. Documenting here so the audit’s slug-vocabulary finding is explicit rather than implicit.
• Implications-section tightening 2026-05-18 (subagent audit). Audit subagent (2026-05-18) flagged two Implications phrasings as boundary-extrapolating beyond what this single review supports (“rice-based infant cereals and rice flours” — verified not in source; “root-vegetable certification” scope decision — verified not in source). Both verified against source and softened to stay strictly within “contributes occurrence data” framing.

Wiki pages updated on ingest

• arsenic
• cadmium
• lead
• chromium
• mercury
• nickel
• copper
• zinc
• cobalt
• rice
• water-spinach
• lotus-root
• watercress
• taro

Page history

The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.