Elbagory et al. 2025 — Potentially toxic elements in four melon varieties, Ganges-Yamuna River Basin

This study assessed eight potentially toxic elements (PTEs: Cd, Cr, Cu, Pb, As, Fe, Mn, Zn) in four melon varieties — two watermelons (Citrullus lanatus var. Arka Shyama and Crimson Sweet) and two muskmelons (Cucumis melo var. Cantaloupe and Kajri) — grown at 10 riverbank sites in the Ganges-Yamuna Basin of Northern India in summer 2024. Concentrations were measured in dry-weight edible pulp by ICP-OES after diacid digestion. Cd, Pb, and As levels in several samples exceeded WHO/FAO permissible limits on a dry-weight basis. Health risk indices (HRI, DIM, THQ) remained below 1 across all four varieties, with HRI for Cd, Cr, Pb, and Fe approaching 0.5 in watermelon and muskmelon varieties.

Key numbers

All concentrations in mg/kg dry weight (DW). Means ± SD across 10 sampling sites; each composite sample (5 fruits per site) analyzed three times.

Watermelon — Arka Shyama (Table 1):

Watermelon — Crimson Sweet (Table 1):

Muskmelon — Cantaloupe (Table 2):

Muskmelon — Kajri (Table 2):

Regulatory comparison (paper-internal, Section 3.1, p. 7): authors note that Cd in certain samples surpassed the 0.05 mg/kg threshold (WHO/FAO MPL for fruit; the paper cites the FAO/WHO Codex Committee on Contaminants in Foods, reference [40]) and that Pb concentrations approached 0.11 mg/kg at site L10. Several As values also exceed the 0.05 mg/kg level. Comparisons in the paper are stated on a dry-weight basis; the WHO/FAO MPL is normally applied to fresh weight, so direct comparability is limited (see Verification notes).

Health risk indices (Table 3):

DIM (mg/kg/day):

HRI (dimensionless):

THQ values were ≤0.0007 across all PTEs and varieties (Table 3). PCA reduced the dataset to two principal components capturing 95.65% (Arka Shyama: PC1 77.25 + PC2 18.40), 96.22% (Crimson Sweet: 87.53 + 8.69), 98.73% (Cantaloupe: 87.30 + 11.43), and 98.02% (Kajri: 93.74 + 4.28) of variance. Fe and Zn showed the most dominant distributions on L3 and L6 sites per the Arka Shyama PCA biplot (Figure 3b,d). HCA cluster-similarity pairs reported by the paper (Section 3.2): L2–L3 and L5–L7 for Arka Shyama; L3–L8 and L4–L5 for Crimson Sweet; L2–L9 and L7–L10 for Cantaloupe; L5–L9 and L4–L7 for Kajri. Per Section 3.1, Pb peaks at 0.11 mg/kg at site L10 (Crimson Sweet) and several samples exceed the WHO/FAO Cd MPL of 0.05 mg/kg DW.

Methods (brief)

ICP-OES (Perkin Elmer 7300 DV, Boston, MA, USA), p. 4 (Section 2.2). Edible pulp separated from ripened fruit, oven-dried at 60 °C to constant weight, and ground to fine powder. Digestion: 2 g dry powder + 20 mL HNO3/HClO4 (3:1, diacid mixture per USEPA Method 3050B), self-digested overnight (12 h), then heated on a hot plate at 150 °C for 1 h in a closed system with reflux condenser. Final volume adjusted to 50 mL with 3% HNO3 and filtered through Whatman No. 41 filter paper. Validation against certified reference material BCR679 (Merck, Mumbai, India) and standard reference materials (1000 mg/L stock, Merck, Mumbai, India). Statistics: PCA and HCA performed in OriginPro 2023 (OriginLab, Northampton, MA, USA); risk indices computed in Microsoft Excel 365.

Risk-index formulas (Section 2.3, p. 5): HRI = DIM/RfD; DIM = SS × HMc/Bw, with SS = 300 g dried weight/day, HMc = PTE concentration (mg/kg), Bw = 70 kg adult; THQ = 10⁻³ × (Ef × Ed × Ir × HMc/Bw × Cp × RfD), with Ef = 365 days, Ed = 365 days, Ir = 0.30 kg/day, Cp = 25,550 days, and RfDs in mg/kg/day (Cd 1.0×10⁻³, Cr 5.0×10⁻³, Cu 4.0×10⁻², Pb 4.0×10⁻⁴, As 3.0×10⁻⁴, Fe 7.0×10⁻¹, Mn 1.4×10⁻², Zn 3.0×10⁻¹).

Note: As reported as total arsenic (tAs); the paper does not speciate inorganic vs organic arsenic. Cr reported as total Cr; no Cr-VI speciation. LOD/LOQ values are not reported. Composite sampling structure (5 fruits pooled per site, 10 sites, 4 varieties) means per-fruit variability is not separable from per-site variability.

Implications

Certification: Contributes occurrence data for fresh melon (watermelon and muskmelon) grown under riverbank cultivation in Northern India; Cd, Pb, and As exceedances are reported on a dry-weight basis, with Pb peaking at 0.11 mg/kg DW at site L10. Wet-weight equivalents (assuming ~88% moisture in melon pulp) are roughly an order of magnitude lower than DW figures, which constrains direct comparison to fresh-fruit regulatory limits typically expressed on a wet-weight basis.

Courses: Strong case study for riverbank cultivation as a contamination pathway — sewage discharge, irrigation with contaminated surface water, and localized soil pollution all drive PTE accumulation in summer fruits. PCA and HCA methodology is well documented and reproducible for site-clustering exercises.

App: Origin-specific signal for elevated Cd, Pb, and As in melon fruit (watermelon and muskmelon) when declared origin is the Ganges-Yamuna agricultural basin of Northern India (Saharanpur/Haridwar districts).

Verification notes

Page enhanced 2026-05-19 from a prior revision dated 2026-05-14. Corrections applied against the source PDF (DOI 10.3390/horticulturae11020216):

• metals: previously listed Al (aluminum); the paper does not measure aluminum (PTE set is Cd, Cr, Cu, Pb, As, Fe, Mn, Zn per Section 2.2). Corrected to [Cd, Cr, Cu, Pb, tAs, Fe, Mn, Zn].
• ingredients: previously declared [[ingredients/muskmelon]], which is not a slug in the current taxonomy snapshot. Replaced with the umbrella [[ingredients/melon]] plus the existing [[ingredients/watermelon]] and [[ingredients/cantaloupe]] (cantaloupe is the named muskmelon variety in the study; Kajri is also a muskmelon cultivar but has no dedicated slug).
• raw_handle: previously the generic literal manual-fetch-kimi; corrected to the per-paper handle MFK_assessment-of-potentially-toxic-elements-in-four-m.
• raw_path: filename was truncated to ...Four Melon Fruit Vari.pdf; corrected to the actual filename on disk.
• matrices: previously included watermelon, muskmelon, and dry-weight, which are commodity names and a reporting basis respectively rather than matrix descriptors. Reduced to [fresh-fruit, melon-pulp, fruit].
• license: upgraded from CC BY to CC BY 4.0 per the paper’s copyright page.
• Tables for Crimson Sweet, Cantaloupe, and Kajri were previously abbreviated to selected metals; expanded to all eight PTEs to match Table 1 and Table 2 in the source.
• DIM and HRI matrices added from Table 3 (Section 3.3) — previously only narrative summaries were present.
• sample_population expanded to capture the L1–L4 (Yamuna), L5–L8 (Hindon/Solani tributaries), L9–L10 (Ganges) site partitioning (Section 2.1).
• Regulatory comparison clarified: the paper compares dry-weight concentrations to a WHO/FAO MPL normally applied on fresh weight, which limits direct interpretability; this caveat was added rather than removed because the paper itself makes the comparison.
• Legacy ## Wiki pages updated on ingest section removed per the current Part 6 source-page template.

Audit subagent (2026-05-19) findings — verified independently against the PDF and applied:

• ❌ Kajri As CV%: subagent flagged wiki value 11.84 as a column-shift duplicate of the Fe CV; Table 2 Kajri CV row reads 41.47/21.18/7.04/38.84/53.84/11.84/16.03/21.46 (Cd/Cr/Cu/Pb/As/Fe/Mn/Zn). Verified; corrected As CV to 53.84.
• ❌ Kajri As range: subagent flagged wiki value 0.01–0.07 as understating the minimum; Table 2 Kajri As column L1=0.02, no value below 0.02. Verified; corrected to 0.02–0.07.
• ⚠️ Kajri PCA variance sum: subagent flagged wiki “97.99%” as arithmetic drift; PC1=93.74 + PC2=4.28 = 98.02. Verified; corrected to 98.02 and added per-component variances for all four varieties.
• ⚠️ HCA hotspot attribution: subagent flagged that the wiki sentence “L3 and L6 … and L4 … as hotspots for elevated Cd, Pb, and As” is not in the source. Section 3.2 actually names Fe/Zn (not Cd/Pb/As) as dominant on L3/L6 in the Arka Shyama biplot, and identifies cluster-similarity pairs not “hotspots.” Verified; rewrote the sentence to track what Section 3.2 actually reports (similarity pairs L2–L3/L5–L7 for Arka Shyama, L3–L8/L4–L5 for Crimson Sweet, L2–L9/L7–L10 for Cantaloupe, L5–L9/L4–L7 for Kajri) and moved the regulatory-exceedance summary up from the prior hotspot sentence to a faithful restatement of Section 3.1.
• ⚠️ Wet-weight conversion in Implications/Certification: subagent flagged as borderline Part 2 (wiki-side arithmetic on top of a source-internal regulatory claim). Rejected as a correction: the conversion is framed as a comparability caveat (“constrains direct comparison”), not as a threshold proposal, and is necessary to honestly report the paper’s own DW-vs-WW MPL comparison. Leaving in place.
• ⚠️ matrices: melon-pulp not flagged as a new-vocabulary proposal: noted here. Matrices vocabulary per docs/gpt-collaboration/system-prompt.md is open (“Add a new term only if none of these fits”); melon-pulp is unambiguous and consistent with naming conventions like fish-muscle, cocoa-powder, egg-yolk already in use across the corpus. No correction required.
• Checks 2, 3, 4 returned ✅. Verdict was REVISE on the basis of the four findings above; after applying the three real corrections and rewriting the hotspot sentence, the page should clear a re-audit on Checks 1 and 5.

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.