Yoplac-Navarro et al. 2026 — Metals in Peruvian avocados: ecological and health risk across 8 regions

This study quantified As, Cd, Cr, Hg, Ni, and Pb in soils and avocado pulp across eight major Peruvian producing regions using microwave plasma atomic emission spectroscopy (MP-AES) with inductively coupled plasma sample introduction, applying probabilistic risk assessment (Monte Carlo simulation with Latin Hypercube sampling, 10 runs of 100,000 iterations) from a One Health perspective. The study is directly relevant to an active food-safety issue: in 2025 the EU RASFF system generated nine alerts for Peruvian avocados exceeding the EU cadmium maximum limit of 0.05 mg/kg. Results show substantial regional variation in soil metal concentrations, with Lima and La Libertad — the two highest-volume export regions — exhibiting the highest soil burdens. Bioaccumulation factors below 1 confirmed limited soil-to-fruit translocation, but fruit concentrations of Pb, Hg, Cd, As, and Ni varied widely enough to produce values approaching or exceeding regulatory benchmarks in several regions.

Key numbers

Soil concentrations (region-specific means ± SD, mg/kg dry weight; from Table 2). The highest-region values reported in the abstract are the Lima means.

(Tukey HSD letters in the source distinguish significant inter-region differences; not reproduced here.)

Soil regulatory benchmarks (Peruvian Environmental Quality Standards, ECA-Suelo):

• As 50 mg/kg — exceeded in Áncash (max 52.47), Ayacucho (max 71.63), Cusco (max 69.37), La Libertad (max 72.65), Lima (max 99.01)
• Cd 1.4 mg/kg — exceeded in La Libertad max 1.58 and Lima max 1.86
• Hg 6.6 mg/kg — not exceeded
• Pb 70 mg/kg — not exceeded
• Cr and Ni: no Peruvian ECA threshold cited

Avocado fruit (pulp) concentrations (dry-weight basis; samples dried at 70 °C / 24 h, ground, homogenized in metal-free mortar). Ranges across all 95 fruit samples (abstract):

• tAs: <0.003 to 0.192 mg/kg
• Cd: <0.005 to 0.130 mg/kg
• Cr: <0.003 mg/kg (below detection limit in every sample)
• tHg: <0.005 to 0.428 mg/kg
• Ni: <0.005 to 0.172 mg/kg
• Pb: <0.005 to 0.396 mg/kg

Per-region avocado fruit means (mg/kg dry weight; <LOD where the analyte was not detected):

(Tukey HSD letters in the source distinguish significant inter-region differences; not reproduced here.)

Detection-limit footprint: Cr below LOD in every avocado fruit sample. Most metals below LOD in Amazonas, Áncash, Cusco, and Huancavelica fruit samples; mean values reflect detected fraction only where stated.

Bioaccumulation factor (BAF = C_avocado / C_soil): all means <1 across all regions and all metals; 95th-percentile BAF exceeded 1 only for Hg in Huancavelica and Lima (Figure 5), suggesting limited but non-zero potential for Hg transfer under certain conditions.

Geo-accumulation index (Igeo): means <0 (uncontaminated) for Cd, Cr, Ni, Pb across all regions. As Igeo ranged from uncontaminated to moderately contaminated (0 ≤ Igeo ≤ 2) across all regions. Hg Igeo reached 0 ≤ Igeo ≤ 2 in La Libertad and Lima at the 95th percentile.

Ecological risk (ER): generally low (<40) for most metals; La Libertad and Lima reached considerable to high ER (80 ≤ ER < 320 at the 95th percentile) for Hg specifically.

Health risk (probabilistic):

• Hazard quotient (HQ) for Cd, Cr, Ni in avocado: all <1 in all regions.
• Hazard index (HI = ΣHQ): well below unity in all regions; highest in Ica, La Libertad, and Lima at 95th percentile (Figure 7; PDF p.14). Cd was the dominant contributor; Ni a minor contributor; Cr negligible.
• Cancer risk (CR):
  • Pb: 10⁻⁸ to 10⁻⁶ across regions (non-significant; <10⁻⁶ threshold).
  • Cd: 10⁻⁸ to 10⁻⁵ across regions (acceptable; below 10⁻⁴ threshold). Highest CR values for Cd at the 95th percentile in Ayacucho, Ica, La Libertad, and Lima.

Reference values used (mg/kgBw/day): Cd 0.001, Cr 1.5, Ni 0.02. Slope factors: Cd 0.38, Pb 0.0085. Ingestion rate modeled as lognormal (5th 3.54×10⁻³; 50th 2.38×10⁻²; 95th 7.02×10⁻²; kg/day). Body weight lognormal (5th 45.3; 50th 62.2; 95th 85.4 kgBw).

Soil order of accumulation: As > Cr > Pb > Ni > Cd > Hg.

Avocado fruit order of accumulation: Pb > Hg > As > Ni > Cd (Cr below LOD throughout).

Methods (brief)

Soil samples: top-layer (3–20 cm) composite samples (three subsamples per composite) collected with stainless-steel spoon, dried at 60 °C, ground, sieved through 0.15 mm nylon mesh. Avocado fruit samples: peel removed, pulp cut into 2–3 mm slices, dried at 70 °C for 24 h (Memmert SN260), ground in laboratory mill (Trittón Perú MP-100), homogenized in metal-free mortar (Eisco Scientific EIS-7115). Samples transported at 4 °C, stored at −20 °C.

Acid digestion: 0.5 g aliquot, 5 mL HNO₃ at 95 ± 5 °C / 15 min in HotBlock; additional 2.5 mL HNO₃, 30 min; H₂O₂ (1.5 mL plus up to 5 mL aliquots); finally 5 mL concentrated HCl; brought to 250 mL volume; filtered through Whatman No. 44.

Instrument: Agilent 4100 MP-AES with ICP sample introduction, standard torch, Inert One Neb nebulizer, double-pass glass cyclonic spray chamber. Wavelengths: As 228.802 nm, Cd 405.781 nm, Cr 193.695 nm, Hg 425.433 nm, Ni 214.160 nm, Pb 253.652 nm. Calibration via standards prepared from 1000 mg/kg stock. Analytical laboratory (LABISAG, Universidad Nacional Toribio Rodríguez de Mendoza de Amazonas) accredited to ISO/IEC 17025:2017 by Peru’s INACAL. LODs and LOQs defined by S/N 3:1 and 10:1, respectively; ten-replicate validation in Table S1 (not extracted).

Risk assessment: probabilistic Monte Carlo simulation with Latin Hypercube sampling, ten runs of 100,000 iterations, @Risk v8 (Palisade). Regional metal concentrations fitted to probability density functions before propagation through EDI, HQ, HI, and CR. BMDL01 Pb 1.5×10⁻³ mg/kgBw/day (cardiovascular); BMDL10 Pb 6.3×10⁻⁴ mg/kgBw/day (nephrotoxic).

No arsenic speciation (total As only). No mercury speciation (total Hg only). No chromium speciation (total Cr; reported below LOD throughout fruit samples).

Implications

Certification (HMTc context, not threshold-setting): This source contributes occurrence data for Peruvian avocado fruit (dry-weight basis) and matched soil across 8 producing regions, plus a 2025 RASFF Cd-alert history (nine alerts cited in the introduction; twelve batches reported in the discussion) the source itself flags as the policy backdrop. Dry-weight vs fresh-weight basis reconciliation against the EU 0.05 mg/kg Cd ML is deferred to the synthesis pass (see Verification notes). Avocado-inclusive HMTc product categories should account for origin variability; Peru-origin fruit in particular merits closer monitoring for Cd, and La Libertad / Lima for Hg.

Courses: Teaching case for the intersection of national mining legacy, agricultural soil quality, and food-safety compliance in an export crop. Peru’s position as a major mining nation and the world’s second-largest avocado exporter creates a structurally interesting supply-chain dynamic relevant to brand-supply-chain risk curriculum and to producer-side traceability.

App: The 8-region per-region breakdown provides the data structure for regional sub-profiles on the avocado ingredient page. La Libertad and Ica account for the bulk of Peruvian export volume; their regional distributions should anchor app default risk estimates for avocado in the absence of known origin. The wet/dry-weight basis caveat (see Verification notes) is load-bearing for any consumer-facing translation.

Verification notes

• 2026-05-25 merge-enhance (Claude Code, autonomous ingest skill v2.0). The pre-existing version of this page (updated 2026-05-14) had four schema-era defects: (1) raw_handle: manual-fetch-kimi (generic, not file-specific); (2) raw_path pointed to a non-existent stub P0311.pdf rather than the actual filename; (3) access_url, raw_sha256, no_doi_assigned absent; (4) products: [] despite this being a food paper (flagged as advisory in routing_malformed.csv). All four corrected. Also fixed: Key numbers section now distinguishes per-region soil/fruit means (presented as Table 2 in the source) from the “across-all-regions” range statements in the abstract; previously the page presented Lima-row means (76.17 ± 17.35 As, 0.55 ± 1.04 Cd, 25.35 ± 6.02 Pb) as if they were pooled across all eight regions.
• Basis caveat (dry weight). All fruit concentrations in this study are reported on dry-weight basis (samples dried at 70 °C / 24 h before grinding). The 2025 RASFF Peruvian-avocado Cd alerts cited in the introduction (≤0.10 mg/kg above the EU 0.05 mg/kg Cd ML for fruits) are on fresh weight per RASFF convention. The source itself does not state a moisture-correction factor and presents its dry-weight values directly against fresh-weight regulatory benchmarks. Resolution of this dry-weight / fresh-weight basis question belongs in the Part 9 synthesis pass on the avocado ingredient page or on the EU 2023/915 regulation page; this source page reports what the source reports.
• Speciation. Paper did not measure inorganic As, methylmercury, or Cr-VI. Frontmatter uses tAs, tHg, Cr accordingly.
• Brand firewall (Part 12): no brands named in the source; method-section vendor names (Agilent Technologies, Memmert, Eisco Scientific, Trittón Perú, Whatman/Cytiva, Palisade @Risk) are scientific-method references and retained per the 2026-05-17 audit exception.
• No HMTc threshold proposals or consumer translation in this page (Part 2 firewall). The Implications section flags relevance to HMTc and to the app but does not propose threshold values or write consumer-facing advisories.
• Audit subagent 2026-05-25 (REVISE): flagged that an earlier draft of the Basis caveat applied an external 70–75 % Hass moisture-content claim to compute fresh-weight equivalents (0.130 mg/kg dry-weight → 0.033–0.039 mg/kg fresh weight) and concluded the source’s maximum sample was “at or just below the EU ML… not above it.” Verified against source: the moisture-correction calculation was the wiki’s own cross-source synthesis, not in the paper. The conclusion also softened the source’s framing of the regulatory exceedance. Both removed; basis-reconciliation deferred to Part 9 synthesis. Also flagged: Ica omitted from the highest-HI-at-95th-percentile list (PDF p.14 names Ica, La Libertad, Lima). Verified and corrected.

Wiki pages this source may touch

• avocado
• fresh-fruit
• cadmium
• mercury
• arsenic
• lead
• chromium
• nickel

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.