Diet choices determine mercury exposure risks for people living in gold mining regions of Peru

Artisanal and small-scale gold mining (ASGM) is the largest global anthropogenic mercury (Hg) source and is widespread in the Peruvian Amazon. Consuming Hg-laden foods exposes people to this potent neurotoxin. While numerous studies have examined fish Hg content near ASGM, Hg accumulation in other commonly consumed animal-and plant-based foods from terrestrial environments is often overlooked. In this study, we aim to address understudied dietary Hg exposures. To understand Hg exposure from food staples in the Peruvian Amazon, we measured total and methyl Hg in local crops, fish, chicken meat, chicken feathers, and eggs from ASGM-impacted and upstream (reference) communities. Diet surveys were used to estimate probable weekly Hg intake from each food. Fish and chicken stable carbon and nitrogen isotope signatures were analyzed to evaluate trophic magnification. Though few crops exceeded food safety recommendations, rice methyl Hg proportions were high (84%). Trophic level was an expected key predictor of fish Hg content. 81% (17 of 21) of local carnivorous fish exceeded WHO and EPA recommendations. Compared to upstream communities, mining-impacted communities demonstrated elevated total Hg in crops (1.55 (interquartile ranges (IQR): 0.60–3.03) μg kg−1 upstream versus 3.38 (IQR: 1.62–11.58) in mining areas), chicken meats (2.69 (IQR: BDL–9.96) μg kg−1 versus 19.68 (IQR: 6.33–48.1)), and feathers (91.20 (IQR: 39.19–216.13) μg kg−1 versus 329.99 (IQR: 173.22–464.99)). Chicken meats from mining areas exhibited over double the methyl Hg concentrations of those upstream. Methyl Hg fractions in chicken muscle tissue averaged 93%. Egg whites and livers exceeded Hg recommendations most frequently. Proximity to mining, but not trophic position, was a predictor of chicken Hg content. Our results demonstrate that terrestrial and aquatic foods can accumulate Hg from mining activity, introducing additional human Hg exposure routes. However, locally sourced carnivorous fish was the largest contributor to an estimated three-fold exceedance of the provisional tolerable weekly Hg intake.


Introduction
Mercury (Hg) is a pervasive, bioaccumulative, and neurotoxic trace metal [1][2][3].Diet is considered the primary human Hg exposure route, especially in communities where fish is a protein staple, as it contains large proportions of methyl Hg (MeHg)-the organic form that carries the greatest health risks due to its high bioavailability [4,5].However, dietary Hg exposure is not limited to fish.For regions impacted by geogenic and anthropogenic Hg sources, terrestrial food sources such as rice can accumulate substantial total and MeHg quantities [6][7][8][9][10][11]. Rice and other crops receive atmospheric Hg inputs through their leaves by intercepting particulate Hg, taking up gaseous elemental Hg through their stomata, or absorbing Hg through their roots [12,13].Moreover, poultry and livestock raised in Hg-contaminated environments can accumulate Hg in their muscles, organs, and eggs [14][15][16].Diets rich in these Hg-laden foods have the potential to cause deleterious human health consequences.
Hg accumulation in chickens raised near pollution sources is not well-studied, but chicken remains an essential protein source in Madre de Dios, Peru.In a 307-person Madre de Dios survey, chicken was the most frequently consumed protein among both indigenous and non-indigenous populations, with 85% of participants consuming chicken at least weekly [17].Although chicken consumption surpassed fish, the survey did not specifically evaluate locally raised chickens or measure Hg in chicken meat [17].A critical knowledge deficit exists for chicken and crop Hg accumulation in mining-impacted areas of South America.
Areas near Hg contamination are most likely to produce foods with elevated Hg concentrations.Hg pollution is a major issue in Madre de Dios due to artisanal and small-scale gold mining (ASGM) activity [18][19][20][21].In ASGM, miners take advantage of the affinity of Hg for gold to efficiently extract gold from sediment [5].After Hg is added to sediment and binds gold particles, gold is isolated by burning the amalgam [1].The process releases Hg vapor into the community and Hg mine tailings into the environment [1].In recent decades, ASGM has become the largest global anthropogenic Hg emission source [21].In Madre de Dios, an estimated 180 tons of Hg are used in ASGM annually [22], leading to terrestrial and aquatic ecosystem contamination [18,23].For children in Madre de Dios, prior investigations revealed associations between elevated hair Hg concentrations and neurodevelopmental deficits, such as decreased IQ [24], along with attenuated vaccine immune responses [25] and elevated anemia risk [26].
A previous Madre de Dios study found that despite a significant correlation between local fish consumption and Hg in human hair, the fish oral Hg reference dose underestimated observed hair Hg concentrations [27].The discrepancy in hair Hg concentrations was thereby attributed to dietary and environmental sources besides fish.Still, elevated hair Hg could not be explained through diet surveys on grains (quinoa, kiwicha), fruits (tomato, banana), or organ meat [27].Another regional study found positive associations between human Hg burden and daily fruit consumption in areas upstream of mining [28].The same study observed positive correlations between hair Hg concentrations and fish consumption only in communities with mining activity [28].By directly measuring Hg content in food, we seek to address fish and non-fish dietary factors contributing to elevated human exposures.
In this study, we aim to determine the contribution of different local foods to Hg exposure for people living in ASGM-impacted areas.We analyzed local crop, fish, and chicken samples from Madre de Dios for Hg.Through a comparative analysis of foods from mining-impacted and upstream areas, we asked: (1) is ASGM activity associated with elevated Hg concentrations in locally available foods? (2) Which foods are potential Hg exposure sources for people living in ASGM-affected areas?We hypothesized that foods sourced from ASGM-impacted locations would have higher Hg content than those from upstream.For meats, we anticipated that animals consuming dietary items higher on the food chain or occupying high trophic levels would have increased Hg concentrations in their internal tissues (muscles, organs).

Study area
Madre de Dios in the southern Peruvian Amazon is a hotspot for biodiversity and ASGM.The region houses approximately 70% of Peru's mining activity [21].Considering growing knowledge surrounding the human health impacts of Hg exposure, regional and federal governments have attempted to regulate ASGM and reduce Hg use.Since 2010, the Peruvian government has ratified the Minamata Convention, used the army to disrupt informal mining sites, declared a state of emergency in Madre de Dios, and attempted to formalize permissible mining concessions [29].However, informal mining persists because of economic opportunities afforded by increasing gold prices, persistent global gold demand, and the lack of alternative livelihoods for some families [30].In this study, we compared the Hg content in crop, fish, and chicken samples from areas heavily and minimally impacted by mining (reference sites).The study sites (figure 1, table 1) were classified by degree of mining based on maps from Diringer et al [20], satellite imaging data from Caballero Espejo et al [31], and observations of mining activity during sample collections.

Sample collection & preparation
Fifty crop samples were collected in 17 Madre de Dios communities from June to August 2018.We purchased crops from every community we passed while traveling 200 km by boat along the Madre de Dios River from Shipetiari to Laberinto.Crops were purchased from markets after vendors confirmed they were grown locally.Crop samples were washed with ultrapure water, placed in individual plastic bags, and stored in a freezer.They were transported frozen and stored at −20 • C upon arrival to Duke University (Durham, NC, USA) until lyophilization and grinding.
Fish samples represented a variety of species, trophic levels, and origins.Fifteen locally accessible and commonly consumed species were selected.From June to August 2018, we purchased 98 fish samples from aquaculture farms, markets, and fishermen in eight Madre de Dios communities.River-caught, locally farmed, and imported ocean fish were collected.
Sourcing fish from sellers rather than the river directly ensures the samples appropriately represent available foods and differentiates these data from prior regional studies [19,20,32].Fish species and their origins were identified by vendors and confirmed using the Instituto de Investigaciones de la Amazonía Peruana (IIAP, Research Institute of the Peruvian Amazon) guide to regional fish [33].Since the whole fish was available for purchase in most cases, total length (mouth to tail) and standard length (mouth to last vertebra, excluding tail) were measured.Next, small muscle fillets (<5 g fresh-weight) were placed in plastic bags, stored in a cooler with ice packs, and transferred to a Credo Cube Series 20 M during the remaining fieldwork.Samples were transported frozen to Duke University, skin was removed, and samples were stored at −20 • C until lyophilization and grinding.Fish samples were weighed before and after drying.Moisture content was calculated as follows: %moisture = fresh weight − dry weight fresh weight * 100.
From July to August 2019, we sampled feathers (n = 38), eggs (n = 16), and meat (n = 8) from backyard chickens raised in a mining-impacted community (Boca Colorado) and an upstream community (Boca Manu).Fifteen commonly consumed chicken cuts were collected (liver, skin, leg, breast, intestine, gizzard, tail, back, thigh, neck, wing, heart, spleen, feet, fat).Multiple feathers were collected from the breast and back of each chicken.All internal tissue and feather samples were rinsed with ultrapure water, placed in individual plastic bags, and stored on dry ice inside a Credo Cube Series 20 M until lyophilization at the IIAP Laboratorio de Mercurio y Química Ambiental (Mercury and Environmental Chemistry Laboratory, Madre de Dios).Lyophilized samples were transported to Duke University for analysis.

Diet survey & participants
Food frequency questionnaires were administered during a larger study of anemia in Madre de Dios.Families across eight communities were offered free anemia screenings and invited to participate in an interview.In Shintuya, Boca Manu, Diamante, Mazuko, and Huepetuhe, schoolchildren were screened for anemia and households of children with anemia were invited to complete the questionnaire.Additional free door-to-door anemia screenings were performed in Boca Manu, Diamante, and Mazuko.Households with children or mothers with anemia were invited to enroll in the study.Health posts in Mazuko, Huepetuhe, Laberinto, Puerto Maldonado, and Tres Islas offered free anemia screenings and households with a child or mother with anemia were invited to complete the questionnaire.Trained interviewers administered questionnaires after informed consent was obtained from the heads of household or their spouses.Food items included in the survey were fish (by species), rice, potatoes, yuca, corn, plantains, tomatoes, beans, quinoa, kiwicha, Brazil nuts, chicken, bush meat, and pork.Adults from 86 households provided meat, grain, fruit, and vegetable consumption data.Foods analyzed for total Hg were used in the exposure calculations (section 2.5).Participants described their daily, weekly, or monthly consumption of each food and the quantity they typically consume per sitting to approximate an average long-term diet (table S4) [34].To maintain cultural sensitivity, participants were asked to describe how much food they eat in handfuls rather than kilograms.Handful measurements were converted to kilograms for analysis (table S1).
Additional surveys were performed during the Boca Manu and Boca Colorado chicken sample collections (table S3).Consenting participants were asked about their primary protein sources and how frequently they consume different chicken cuts.Participants who raised chickens were asked about their chickens' living conditions and diets.

Laboratory analyses & quality control
Total Hg content of dried, homogenized crop and fish muscle tissue was measured at Duke University via thermal decomposition, amalgamation, and atomic absorption spectrophotometry (Environmental Protection Agency (EPA) method 7473) on a Milestone direct mercury analyzer (DMA-80) per methodology in Gerson et al [35].Samples were run in duplicate, and the average result was used for data analysis.If the relative percent difference (RPD) exceeded 10% between the two sample runs, samples were rerun in duplicate.Instrument calibration was performed using the Brooks Rand Instruments Total Mercury Standard (1.0 ng L −1 ).Continuous calibration verification (CCV), quality control standard, matrix spikes (MS), and blank measurements were performed after every tenth sample.National Institute of Standards and Technology (NIST) standard reference materials for crops, 1633c (coal fly ash) and 2709a (San Joaquin Soil), had percent recoveries of 95% ± 3% SD (n = 39) and 94% ± 3% SD (n = 8), respectively.Blanks contained <0.0037 mg kg −1 total Hg.Standard reference materials for fish tissue, NIST 1633c (coal fly ash) and DORM-4 (fish protein), had percent recoveries of 96% ± 3% SD (n = 52) and 91% ± 7% SD (n = 27), respectively.All blanks contained <0.0023 mg kg −1 [total Hg].The instrument detection limit for fish and crop total Hg analysis was 0.5 ng Hg.For fish, dry-weight (dw) total Hg was converted to fresh-weight (fw) total Hg as follows: .
Chicken tissues and feathers were analyzed at the Biodiversity Research Institute (Portland, ME, USA) for total Hg following EPA method 7473 on a Milestone DMA-80.Samples were run in duplicate and rerun if the RPD exceeded 15%.For quality control, empty sample boats (blank boats, n = 7) and blanks (no boat or sample, n = 19) were run to test for residual Hg in the analyzer and the sample boats, respectively.All blanks and blank boats returned <0.002 mg kg −1 [total Hg].Standard reference materials used for feathers and tissues were DOLT-5 (dogfish liver; 97% recovery ± 0.7% SD, n = 25) and CE464 (tuna fish; 99% recovery ± 2% SD, n = 25).The instrument detection limit for internal chicken tissue, egg, and feather total Hg analysis was 0.003 ng Hg.
MeHg is reported as absolute values and percent MeHg.For samples with detectable MeHg, percent MeHg is defined as follows: Different instrument detection limits led to MeHg values higher than total Hg in some samples.To avoid skewed representations of MeHg proportions, all MeHg percentages exceeding 100% were rounded down to 100%.
Stable carbon and nitrogen isotope analyses were performed on ground feather samples and dried fish muscle tissue at the University of California-Davis Stable Isotope Facility (SIF; Davis, CA, USA) and the Duke Environmental Stable Isotope (DEVIL) Laboratory (Durham, NC, USA), respectively.δ 13 C and δ 15 N signatures were determined via mass spectrometry using an Elementar Micro Cube elemental analyzer interfaced to a PDZ Europa 20-20 isotope ratio mass spectrometer for feather samples and two Thermo-Finnigan Delta Plus XL continuous flow mass spectrometer systems for fish muscle tissue.For quality control, samples were interspersed with several replicates of internal laboratory standards, which were calibrated against standard reference materials.δ 13 C and δ 15 N values were normalized to correct for changes in instrument environment, parameters, and gas drift.Stable isotope ratios are expressed in δ notation in parts per 1000 (‰) relative to Vienna PeeDee Belemnite standards for 13 C and atmospheric N 2 for 15 N, as follows: where X represents 13 C or 15 N and R is 13 C/ 12 C or 15 N/ 14 N.

Human exposure estimations
We combined food total Hg measurements with diet survey data to assess the probable weekly intake (PWI) of Hg.The PWI was estimated as described by Li et al [40] using the following equation: .
The average adult weight in Peru, 68 kg, was used [41].Median dw total Hg was converted to fw Hg content for each food to facilitate comparisons with international fw Hg guidelines.We used literature values for percent moisture because crop, chicken meat, and egg moisture content were not obtained (table S1).Typical consumption quantities and frequencies were standardized on a per-week basis for analysis (tables S4 and S11).

Statistical analysis
All variables except chicken feather δ 15 N and δ 13 C signatures did not follow normal distributions before or after log transformation according to the Shapiro-Wilk test.Accordingly, non-parametric Wilcoxon Rank Sum, Kruskal-Wallis, and Dunnett's tests compared Hg content across locations, species, and origins.Ordinary Least Squares linear regressions determined trophic magnification using stable isotope signatures and log-transformed Hg data.Non-parametric Kendall's tau correlation coefficients (τ) determined the relationship between fish Hg content, length, and moisture, which could not be normalized.Statistical outcomes were considered significant if p <0.05.Summary statistics are reported as medians with interquartile ranges (IQR).We acknowledge the small sample sizes and insufficient power due to availability constraints of different locally sourced foods in each community.Thus, p-value inflation is possible.Figure creation and statistical analyses were performed in R version 4.0.2.

Ethical considerations
Institutional Review Boards at Duke University (#2019-0530) and the University of Peru Cayetano Heredia (chicken: #104284; total diet: #102134) provided ethical clearance for this project.All participants provided verbal and written consent in the chicken collection survey and the total diet survey, respectively.Survey data were deidentified for analysis.The Duke University Animal Care and Use Committee approved field methods involving live animals (#A161-19-07).

Estimated dietary Hg exposure risk from crops
Because neither the US Environmental Protection Agency (EPA) nor the World Health Organization (WHO) has released recommendations for crop total Hg content, total Hg concentrations are compared to China's Maximum Contaminant Level (MCL; 20 µg kg −1 fw for grains, 10 µg kg −1 fw for vegetables) [42].Crop MeHg advisories are not available.10% (5 of 50) of crop samples contained concerning fw total Hg concentrations by the MCL standard (table 2).Four of the five crop samples inadvisable for consumption were from mining areas.Maximum total Hg concentrations for 43% (3 of 7) of the crop varieties collected exceeded the recommendation.The highest measured total Hg concentrations exceeded the MCL by 1.5-fold in vegetables and 2.3-fold in rice. 1 of 7 rice samples (30.52 µg kg −1 fw) and 0 of 2 corn samples exceeded the grain MCL.12% (2 of 17) of plantain samples (13.77 and 16.17 µg kg −1 fw) and 1 of 2 potato samples (22.91 µg kg −1 fw) exceeded the vegetable MCL.

Fish muscle mercury content
All 98 fish samples contained detectable total Hg concentrations (figure 3).Fish total Hg content varied among the eight communities for all fish (representing spatial trends) and for non-imported fish only (representing fish protein available to community members) (p <0.041).Across all purchased fish, total Hg concentrations were lower on average in Boca Manu-an upstream community with minimal mining activity-than in two mining and amalgam-burning towns (Laberinto, p = 0.006 and Puerto Maldonado, p = 0.012).

Table 2.
Total mercury (Hg) and methyl Hg (MeHg) concentrations in crops from 17 Madre de Dios communities.Median dry-weight (dw) total Hg concentrations were converted to fresh-weight (fw) concentrations based on the literature moisture content of each plant (table S1) and compared to the maximum contaminant level (MCL) b values for Hg set by the Chinese Ministry of Health.Samples are categorized by crop type, perennial versus annual life cycle, and origin (crops from mining and amalgam-burning communities versus upstream minimal mining communities).All MeHg percentages greater than 100% were set to 100%.Percent MeHg was not calculated for samples where total Hg was below the detection limit because it was not possible to infer the amount of total Hg.Among non-imported fish caught in Madre de Dios, carnivore Hg concentrations varied by community as well (p = 0.013), with higher total Hg content in mining communities (1.34 (IQR: 0.67-1.84)mg kg −1 fw) compared to upstream communities (0.92 (IQR: 0.73-1.08mg kg −1 fw)).The same pattern was observed among non-imported herbivores (p = 0.007; mining: 0.14 (IQR: 0.07-0.15)mg kg −1 fw; upstream: 0.12 (IQR: 0.07-0.12)mg kg −1 fw).Omnivore Hg content did not significantly differ by community or mining presence (p = 0.265; mining: 0.01 (0.01-0.03) mg kg −1 fw; upstream: 0.02 (0.02-0.03) mg kg −1 fw).Note that sample size was insufficient to compare Hg content on a species-by-species or community-by-community basis for non-imported fish.
Within the communities, the purchase site (farm, household, market, restaurant, fisherman) did not strongly influence Hg content (p = 0.341).Total Hg content varied by fish origin (lake, river, farm, ocean; p <0.001; table 4).Paco was the most common farmed fish, with 21 of 23 samples originating from aquaculture.Median total Hg in river-caught fish was 15 times higher than in farmed fish (p <0.001; table 4).
Fish total Hg data were characterized by comparison to WHO and EPA Hg guidelines.Total Hg concentrations in 20% of fish (19 of 98) exceeded the EPA Fish Tissue Residue Criterion (FTRC; 0.3 mg kg −1 fw) [44], and total Hg concentrations in 19% (18 of 98) of fish exceeded the WHO Guideline Level (GL; 0.5 mg kg −1 fw) [45] (table 4).Excluding imported fish, 25% of samples exceeded both recommendations.All fish exceeding the guidelines were obligate carnivores and piscivores.81% (17 of 21) of carnivorous fish sourced from Madre de Dios exceeded WHO and EPA Hg guidance (table 4; figure 3).On average, samples exceeding total Hg advisories were 2.7 times higher than the WHO GL and 5.4 times higher than the EPA FTRC.The highest total Hg concentration in a single fish was 4.76 mg kg −1 fw from a river-caught bagre, 9.5 times the WHO GL and 15.9 times the EPA FTRC.

Fish muscle stable C and N isotope signatures
δ 13 C signatures ranged from −15.1‰ (paco) to −41.2‰ (boquichico; table 3).δ 15 N signatures ranged from 4.2‰ (boquichico) to 17.9‰ (jurel).δ 13 C and δ 15 N differed by species (p <0.001; p <0.001; figure S1) and assigned trophic levels (p <0.001; p <0.001).Due to the lack of local isotopic baselines for the large fish diversity of the Amazon basin, δ 15 N was used as a proxy for trophic position as performed in Teffer et al [46].Accordingly, a clear δ 15 N separation is apparent for detritivores compared to piscivores and omnivores, indicating distinct food sources (figure 4).A linear regression across species demonstrated a positive correlation between δ 15 N and log-adjusted total Hg (R 2 = 0.32; p <0.001; figure 4).The trophic magnification factor, which measures exponential contaminant magnification in food webs, was calculated as 1.48 following methodology in Coelho et al [47].When stratifying by fish origin, non-imported carnivores from mining areas demonstrated higher trophic magnification factors than fish from upstream sites (2.73 mining, 1.92 upstream).The opposite pattern was observed in omnivores and herbivores/detritivores, with lower trophic magnification factors found in mining areas compared to upstream sites (omnivores: 0.79 mining, 1.52 upstream; herbivores/detritivores: 0.84 mining, 1.49 upstream).

Chicken feather, internal tissue, and egg mercury content
Total Hg and MeHg were measured in eggs and internal tissues (muscles and organs) of backyard chickens from Boca Colorado (high mining activity) and Boca Manu (upstream reference, minimal mining activity).Total Hg was measured in chicken feathers.Detectable total Hg quantities were present in all feather samples (n = 38) and 89% (127 of 143) of internal tissue and egg samples.Total Hg content proved higher in feathers and internal tissues from chickens raised near mining than those raised upstream (p <0.001; tables 5 and 6).For chickens raised near mining activity, median total Hg concentrations were 7.3-fold and 3.6-fold higher in internal tissues and feathers, respectively, compared to samples from the upstream reference site.Median internal tissue and feather total Hg content from the mining-impacted area was 19.68 (IQR: 6.33-48.1)µg kg −1 dw and 329.99 (IQR: 173.22-464.99)µg kg −1 dw, respectively, compared to 2.69 (IQR: Total mercury (Hg) concentrations in fish from Madre de Dios, Peru, percentage of samples exceeding EPA and WHO Hg recommendations, isotopic ratios, length, and moisture content.Estimated trophic positions d are assigned to each species.Fresh-weight total Hg concentrations (mg kg −1 ) are compared to the WHO Guideline Level (GL) for total Hg in non-predatory muscle tissue (0.5 mg kg −1 ) and the EPA Fish Tissue Residue Criterion (FTRC; 0.3 mg kg −1 ).Average total length (measured from mouth to tail) and standard length (measured from mouth to last vertebra, excluding tail) are reported for each species (cm).Samples are categorized alphabetically by fish species.a Excludes canned and ocean fish from outside the region (jurel, trucha, canned atún, canned caballa).Average total Hg content differed significantly among internal chicken tissues (p <0.001).The highest recorded total Hg value was in a chicken liver from the mining community (600.23 µg kg −1 dw).Chicken livers from the mining community had the highest average total Hg content of all internal tissue samples (222.81 (IQR: 49.76-437.65)µg kg −1 dw).Median liver total Hg content was significantly higher than eight (gizzard, intestine, leg, heart, neck, skin, thigh, wing) of the other 14 internal tissues (p <0.027; table 5).Total Hg content differences among the remaining internal tissues were insignificant.Egg whites had higher average Hg than all sampled internal tissues (175.15(IQR: 113.71-337.22)µg kg −1 dw; table 5).

Chicken feather stable isotope signatures
δ 13 C signatures ranged from −12.2‰ to −21.8‰ (table 6; figure S2).δ 15 N signatures ranged from 6.4‰ to 10.1‰.The chicken feather isotopic profile closely resembled omnivorous fish (figure S3).δ 15 N data did not vary by proximity to mining (p = 0.286).δ 13 C signatures, however, were higher in chickens raised in the upstream community (p = 0.012).No correlation between log-transformed total Hg concentrations and δ 15 N was observed in chicken feathers (p = 0.103), so trophic magnification calculations cannot be applied.

Estimated dietary Hg exposure risk from chicken and egg consumption
Fw total Hg content of meat and eggs was compared to the maximum contaminant level from the Chinese Ministry of Health (MCL: 50 µg kg −1 fw) [42].No standards currently exist for MeHg in poultry meat.Among all edible samples, egg whites had the highest average fw total Hg concentrations (21.02 (IQR: 13.65-40.47)µg kg −1 fw) and a maximum value of 184.81 µg kg −1 fw, 3.7 times higher than the accepted MCL.Egg yolks had a lower total Hg burden (13.98 (IQR: 11.50-27.81)µg kg −1 fw) and maximum concentration (461.90 µg kg −1 fw).Notably, most Hg in eggs was in the MeHg form (87% for whites, 67% for yolks; table 5).The majority (>50%) of Hg in all internal tissues studied was present as MeHg.19% (3 of 16) of egg whites and 13% (2 of 16) of egg yolks exceeded the MCL.

Chicken husbandry and dietary survey
Select survey results are presented in table 7 from participants who contributed chicken eggs, meat, and/or feathers to the study.All survey responses are summarized in table S3.In the mining-impacted community of Boca Colorado, 95.8% (23 of 24) of participants raised backyard chickens.Upstream in Boca Manu, 71.4% (11 of 15) of participants raised chickens.Because most households did not meet their demand for meat with backyard chickens, they purchased supplementary meat from local markets, restaurants, neighbors, and vendors outside the area.Most participants did not consider local chicken as their primary protein source Table 5. Mercury (Hg) concentrations in chicken eggs and internal muscle and organ tissues from Madre de Dios, Peru.Dry-weight (dw) Hg concentrations were converted to a fresh-weight basis (fw) based on literature moisture content values (table S1).Median total Hg and methyl Hg (MeHg) concentrations are compared to the maximum contaminant level (MCL) for Hg set by the Chinese Ministry of Health (50 µg kg −1 fw) [42].All MeHg percentages greater than 100% were set to 100%.Percent MeHg was not calculated for samples where total Hg was below the detection limit because it was not possible to infer the amount of total Hg.

Total
Hg   either seasonally or throughout the year (37.5% in Boca Colorado, 36.1% in Boca Manu).In both communities, most full-grown chickens were free-range for at least part of the day.In Boca Colorado, most owners fed their chickens corn (87.5%) or food scraps (25.0%).In Boca Manu, most chickens ate corn (90.0%) or prepackaged food (10.0%).Of note, corn sampled in the crop Hg analysis portion was purchased from markets and was not designated by sellers as chicken feed.The market corn is likely most reflective of human Hg intake as we do not know whether some participants fed this corn to their chickens.Accordingly, chicken Hg intake from corn cannot be inferred from our market samples.

An overview of dietary Hg exposure in Madre de Dios, Peru
Our results demonstrate that crops, fish, chicken muscle and organ tissues, and chicken eggs sourced from mining-impacted areas of Madre de Dios, Peru had greater Hg concentrations than the same foods originating upstream in minimally-impacted reference areas.The estimated average dietary Hg intake from fish in this study exceeded WHO safety guidelines by over two-fold.Although Hg intake from crops, chicken meat, and eggs did not exceed WHO safety guidelines, they do compound dietary Hg exposure risk.
Virtually all crop samples contained detectable total Hg concentrations, but few exceeded established exposure risk guidelines.All imported and local fish muscle samples contained detectable total Hg, but only carnivorous fish samples exceeded the guidelines.Carnivorous species caught in Madre de Dios had the highest average total Hg concentrations and are not recommended for frequent consumption.Total Hg content in internal chicken tissues and feathers differed considerably by mining presence in the community, presumably due to environmental Hg contamination, but not isotopic profile.Hg content of several chicken eggs and livers from the mining-impacted area exceeded dietary safety recommendations, but all other muscle and organ tissues contained permissible levels.Chickens in this study demonstrated omnivorous diets with markedly similar isotopic profiles and total Hg accumulation to omnivorous fish.
Most crop samples did not contain detectable MeHg quantities.Aside from rice and cocona, MeHg concentrations were consistently low, and all other crop MeHg proportions fell below 14%.Average rice MeHg proportions were comparable to fish, giving rise to potentially concerning exposures over time.Across all chicken meats, MeHg accounted for the majority of total Hg (54% in liver to 100% in spleen and back meat).MeHg concentrations were highest in egg whites and lowest in skin.Internal chicken tissues from the mining-impacted area contained over two-fold higher MeHg than those from upstream.These measurements demonstrate Hg enhancement of local staple foods, underscoring a need for informed dietary choices to minimize potential associated human health threats and steps to mitigate terrestrial and aquatic anthropogenic Hg pollution.

Exposure characterization
To contextualize dietary exposure, total Hg concentrations were coupled with food frequency survey data to generate the PWI of Hg from each food.The estimated PWIs were then compared to the WHO provisional tolerable weekly intake (PTWI; 1.6 µg total Hg per kg of body weight (bw) per week) [48].Hg intakes above this value pose toxicity risks for sensitive populations.Particularly vulnerable groups include pregnant women, children under six years old, people with impaired renal function, and those with heightened immune responses to trace metals [48][49][50].The MeHg PTWI remains under review and subject to debate.The Faroe Islands established a stricter MeHg standard, 1.3 µg kg −1 bw, following adverse effects in child cohort studies [51].Consumption of fish, rather than vegetables or other meats, and the beneficial modifying effects of omega-3s were used to develop this standard [51].

Dietary Hg exposure risk from crop consumption
Although nearly all crop samples had detectable total Hg quantities, most would be considered fit for consumption by international food safety standards.The total PWI from crops (0.87 µg total Hg kg −1 bw per week; table S11) reached 54% of the PTWI safety guideline.Rice and potatoes contributed the most to the crop PTWI.The contribution of MeHg to the PWI was low and entirely from rice (0.14 µg kg −1 bw per week; 9% of the PTWI).Neurotoxic effects in the general population are not typically observed until a two-fold PTWI exceedance (3.2 µg kg −1 bw per week), so fruit and vegetable consumption carry small expected health risks [52,53].Further, most Hg in all crops besides rice was inorganic (5% MeHg in mandarins to 13% MeHg in plantains) and not readily absorbed through digestion.Rice was an exception because although only 1 of 7 samples exceeded the recommended Hg level, MeHg proportions were high (84%) and comparable to chicken and fish muscle tissue.
The total Hg concentrations in most of our samples are considered acceptable for consumption.However, concern is warranted long-term, given the high bioavailability of MeHg and the common practice of daily rice consumption in Madre de Dios.It should also be recognized that we exclusively analyzed locally grown crops in this study, and individuals likely supplement their diet with imported crops of unknown Hg content.Consequently, complete dietary Hg exposure to individuals through crops has yet to be elucidated.

Crop Hg accumulation patterns
These seven terrestrial foods are accumulating Hg from their growing environments, posing health concerns for those reliant on local produce.Hg uptake and associated dietary risks varied by plant species, likely due to distinct absorption and accumulation mechanisms [40,54].Average total and MeHg concentrations in fruits (cocona, plantain, mandarin) were lower than those in grains (rice, corn) and root vegetables (potatoes, yuca).As Madre de Dios soils accumulate Hg from mining [18], root vegetables can directly absorb soil Hg and could pose dietary risks if consumed in large quantities.Ssenku et al [55] previously reported total Hg values up to 50 330 ± 5290 µg kg −1 dw in sweet potatoes and 4 ± 0 µg kg −1 dw in yuca from ASGMimpacted areas [55].In an ASGM-impacted area of Brazil, Egler et al [56] measured 320 µg kg −1 dw total Hg in sweet potatoes and up to 2700 µg kg −1 dw in yuca.Further extensive crop studies including root vegetables will provide a more generalizable understanding of Hg content and associated dietary risks.
Previous research has observed low Hg concentrations in fruit and noted that its vitamins, minerals, and fiber may modulate Hg absorption [57].An improved understanding of Hg absorption in fruiting plants is needed, but our results suggest fruit consumption confers minimal immediate human health concerns.
In contrast to fruit, rice contained high MeHg proportions.The tendency of plants to accumulate metals will depend on bioavailability, which is impacted by soil pH, organic matter, cation exchange capacity, moisture, clay content, and redox potential [58].When rice is grown under flooded, anoxic conditions that promote microbial methylation, most total Hg is present as MeHg [59,60].Field experiments propose that rice assimilates evaded gaseous Hg and Hg from surrounding soil and water into its tissues [61].However, rice in Madre de Dios is often planted using dry methods, which do not provide the same anoxic methylation environment.Cultivation information for each rice sample was not available.Different farming methods could explain observed MeHg variability, as dry methods have previously demonstrated reduced rice Hg uptake [62].Soil and growth conditions undoubtedly contribute to crop Hg accumulation trends and deserve additional investigation.
The continuous environmental Hg input from expanding mining activity across Madre de Dios suggests that soil and atmospheric Hg loads will increase in the region over time [23,31,63].Long-term exposure concerns are especially relevant for people in Hg-contaminated areas worldwide who depend on small-scale gardening, local foods, and rice for food security [64].
Annual crops (including rice) contained three-fold higher median total Hg concentrations than perennials.While some authors have found higher total Hg concentrations in perennial plants due to increasing Hg accumulation over time [65,66], other authors assert that young and annual plants are more susceptible to metal pollution due to their rapid growth and aging [67,68].For example, small annual plants near a cinnabar mine demonstrated the highest total Hg content, explained by an inverse relationship between plant height and Hg uptake [69].Similarly, in a phytoremediation pesticide study, annual plants demonstrated greater and more efficient pollution uptake than perennials, largely because quick-growing plants are more likely to be contaminant hyperaccumulators [70].The same study noted that agri-food crops assimilated greater quantities of contaminants than grasses [70].Annual plants have also demonstrated higher nitrogen uptake from soil [71].There may thus be a shared mechanism between increased nutrient and Hg absorption in annual plants.
Crop total Hg content reported here is slightly lower than literature values from Hg-polluted areas globally [6], including E-waste recycling [7], cinnabar mining [8], industrial emissions [9], and coal-energy facilities [40].However, outside of rice, references are lacking for crops grown in ASGM-impacted regions.Rice Hg burden from this study comprised a smaller proportion of the estimated daily Hg intake than in China, but MeHg proportions were similarly high [7,11].Aside from a handful of reports on rice contamination in ASGM-affected areas of Asia, there are knowledge gaps surrounding ASGM-attributable Hg accumulation in other crops [72][73][74].Our study represents a critical first step in determining dietary Hg exposure risk for common South American crops and necessitates further investigation at ASGM sites globally.

Dietary Hg exposure risk from fish consumption
Detectable Hg contamination was found in every fish sample, emphasizing fish consumption as the predominant human Hg exposure pathway.As described above, the PWI was compared to the WHO PTWI (1.6 µg kg −1 bw), which accounts for the benefits derived from nutrients like omega-3 fatty acids while considering Hg toxicity [75].The average PWI from typical fish consumption was 4.17 µg kg −1 bw per week (table 8), exceeding safe recommendations by 2.6-fold and posing neurotoxicity risks in adults [52,53].Among all food groups, Hg intake from fish poses the greatest and most immediate human health risks.
Carnivorous fish consumption accounted for the majority of estimated Hg exposure.Exposure from the reported dorado consumption alone (1.78 µg kg −1 bw per week, 111.4% PTWI) exceeded the PWTI.Doncella accounted for the second-highest percentage of the PTWI (67.6% PTWI, 1.08 µg kg −1 bw per week).Fish with higher Hg concentrations-bagre and zungaro-were consumed less frequently and accounted for a minor fraction of the PTWI (table 8).Other studies also show fish to comprise a large proportion of dietary Hg intake and recommend that they are consumed cautiously to avoid health risks [2,5].Limiting carnivorous fish from ASGM areas and increasing herbivorous fish consumption is critical to reduce human Hg exposure.
Each fish species did not contribute equally to the PTWI; not every fish presented high exposure risks.Non-carnivorous species contributed minimally to the total fish PWI.Paco, an omnivorous fish farmed and caught in Madre de Dios, had the lowest average total Hg concentrations (0.01 mg kg −1 fw).Langeland et al [77] found similarly low total Hg concentrations in farmed paco due to insufficient time for methylation during the few years fish are raised in ponds.In aquaculture, paco typically eat outside feed instead of foraging [78], so there is little dietary Hg input from pond food webs, which represent methylation hotspots Table 8.Probable weekly intake (PWI) of total mercury (Hg) and methyl Hg (MeHg) by food group.(a) PWI components for crop consumption, (b) PWI components for fish consumption, (c) PWI components for chicken consumption.The PWI values (µg of total Hg per kg of body weight (bw) per week) are based on weekly consumption of each food (kg per week) according to diet surveys and total Hg concentrations on a fresh-weight (fw) basis.Dry-weight (dw) total Hg measurements were converted to fw based on moisture content presented in the literature (table S1 S2. [23].We found that fish with low trophic positions and farmed fish contribute relatively little to human Hg exposure, making them viable nutritional substitutes for carnivorous fish.

Fish Hg accumulation patterns
We found that trophic level, but not length, effectively predicted fish Hg accumulation.Length was not associated with total Hg content for most (13 of 15) species.Paco did demonstrate a significant size-Hg relationship, likely due to its aquaculture origins, which give rise to similar environmental conditions and longevity across samples from that species [79].As previously reported in the Amazon basin, feeding behavior and trophic level typically predict Hg content better than fish size [19,20,80].Total Hg and nitrogen signatures (‰ δ 15 N) of fish samples determined the trophic magnification factor, which indicates bioaccumulation at values over 1 [47].The trophic magnification factor across all studied Madre de Dios fish species was 1.48, demonstrating Hg bioaccumulation (figure 4).
Trophic magnification was more pronounced in carnivores compared to lower trophic levels.Non-imported carnivorous fish samples showed markedly increased trophic magnification in mining areas compared to upstream areas (2.73 versus 1.92), whereas lower trophic levels demonstrated higher trophic magnification in upstream areas.This finding is supported by the 'trophic guild hypothesis' in Barocas et al [81], which proposes that biomagnification will be higher in fish from higher trophic levels.In addition to diet, larger body sizes, increased lifespan, and ecological drivers make higher trophic guilds more susceptible to Hg pollution from mining [81][82][83].
Contrary to previous work, herbivorous and detritivorous fish contained higher Hg in their muscle tissues than omnivorous fish [20,84].Kasper et al [85] propose that this pattern may occur because detritivores' feeding habits expose them to large amounts of sediment Hg.Although we did not measure fish MeHg, we assume MeHg percentages do not drastically vary by species.Prior research suggests MeHg percentages are comparable across trophic levels for fish of similar sizes and ages, though carnivorous fish may assimilate MeHg slightly more efficiently [86,87].Average fish total Hg concentrations were similar to, or higher than, other regional studies [19,20,53].Bagre, noted by other authors to present low Hg exposure risks by EPA standards [19,20], had the second-highest total Hg concentrations in our study.Continued biomonitoring efforts to better understand the factors driving fish Hg uptake near ASGM are necessary due to Hg's continued use and potential for legacy contamination.
Prior Madre de Dios studies hypothesize a spatial relationship between ASGM activity and fish Hg content, which is supported by our finding of elevated Hg in carnivorous and herbivorous fish from mining areas compared to those upstream [20,81,83].Additionally, studies in this region suggest that ASGM is a significant but not exclusive contributor to environmental Hg contamination in the Amazon.Other point-and non-point pollution sources include biomass burning, soil erosion, hydroelectric dams and, overarchingly, deforestation [88][89][90], leading to elevated concentrations of Hg in non-carnivorous fish outside ASGM areas in this study and others [83,91].

Dietary Hg exposure risk from chicken consumption
To our knowledge, Hg concentrations in free-range chicken eggs and select meat samples reported here are among the highest in the literature (table 9).Several previous studies have demonstrated a clear link between proximity to gold mining and Hg accumulation in poultry blood [92,93] and tissues [15].Robust data on Hg accumulation in poultry meat and eggs are lacking, however, despite their global importance as protein sources.Nevertheless, it is well-established that ASGM is a major driver of Hg accumulation in numerous avian species.In an analysis of mercury concentrations in neotropical birds across Central and South America, Sayers et al [94] noted that birds within 7 km of mining activity exhibited four times higher Hg concentrations than other sites.In an analysis from Madre de Dios that included birds of all trophic guilds, feather samples from mining-impacted sites contained 6.7 times higher Hg [95].The samples from ASGM-impacted areas were also the highest in the published literature to date [94,95].
Detectable total Hg was present in all chicken feathers, all eggs, and most internal tissues.However, it should be emphasized that Hg content measured in fish samples often exceeded that of chicken tissues by an order of magnitude or more.Although the average total Hg content of chicken eggs and internal tissues was low relative to crops and fish, organic Hg proportions were high.Thus, consuming chicken meat could represent an additional path of long-term Hg exposure, especially in ASGM-impacted sites.And, similar to fish, chicken muscle tissues contained 93% MeHg.Egg whites contained 87% MeHg, and 67% yolks contained MeHg.Organ tissues, including liver, gizzard, and spleen, had 54%, 94%, and 100% MeHg, respectively.MeHg percentages were higher in egg and internal tissues from upstream sites, possibly due to more sensitive MeHg instrument detection or ecological attributes such as habitat type, endogenous/ exogenous egg formation reserves, and embryo age.Similar to previous investigations, a considerable portion of samples reached 100% MeHg, and percent MeHg variability was highest at low Hg exposure levels, resulting in large percent differences in the proportion of total Hg as MeHg.Dietary Hg intake from chicken meat and eggs relative to the WHO-recommended PTWI (1.6 µg kg −1 bw) is low (2.6% total Hg and MeHg) compared to the contribution from fish (260.5%) and crops (54.3% total Hg, 9.0% MeHg) (figure 6).Egg total Hg concentrations were the highest across all edible tissues (21.02 (IQR: 13.65-40.47)µg kg −1 fw in egg whites; 13.98 (IQR: 11.50-27.81)µg kg −1 fw in egg yolks), and they were consumed more frequently than backyard chicken meat.Unsurprisingly, eggs contributed the most to the estimated dietary total Hg intake (2.0% and 2.2% of the PTWI from total and MeHg, respectively).Livers made the second-highest individual contribution to the PTWI (0.2% and 0.1% for total and MeHg; Median total Hg = 13.64 (IQR: 5.16-47.23)µg kg −1 fw).All other internal chicken tissues contributed 0.6% total Hg and 0.4% MeHg to the PTWI (table 8).Most survey participants were inclined to raise their chickens for daily egg production rather than slaughtering them for meat.Backyard chickens are reliable food sources, capital assets, and a form of savings.Chickens can be slaughtered or sold during food shortages, special occasions, or a need for extra cash [110].Besides offering future food and financial security, chickens provide eggs as a constantly renewing protein source or commodity.Most chicken meats contained modest total Hg concentrations.However, people relying on backyard chickens for food in mining-impacted areas should be cognizant of liver and egg consumption.Slightly elevated Hg concentrations across all locally available foods will compound dietary Hg exposure risk and should be considered in dietary recommendations.

Chicken Hg accumulation patterns
Observing backyard chickens' feeding habits in communities along the Madre de Dios River motivated our study of their isotopic profiles.We saw diverse diets ranging from scavenged invertebrates to food leftovers, and seldom observed commercial grain diets.The isotopic profiles and total Hg concentrations of chickens mirrored omnivorous fish species, indicating similar Hg accumulation despite terrestrial rather than aquatic origins (figure S3).Isotope profiles, reflecting chicken diet, were comparable in mining-impacted and upstream communities (figure S2), suggesting a ubiquitous omnivorous diet in the region.
Total Hg concentrations in feathers and internal tissues from the mining-impacted area were 7.3 and 3.6 times higher than those from upstream (tables 5 and 6).Since chicken trophic position did not entirely explain observed spatial patterns in total Hg content, environmental Hg exposure from ASGM appears to predict Hg accumulation better than dietary differences.Higher total Hg concentrations in the environment and food sources likely explain elevated total Hg burdens in chickens from the mining-impacted region.
Total Hg distribution patterns were distinct across different tissues, which will influence human Hg exposure.Eggs contained substantially higher total Hg concentration than internal tissues, explained by the fact that laying eggs is an avian Hg excretion mechanism [111].Egg whites demonstrated elevated total Hg and MeHg concentrations compared to yolks.Egg white and yolk compositions may underlie different molecular mimicry opportunities and observed Hg content.Egg whites contain virtually no lipids and predominantly contain water, structural proteins, glycoproteins (ovalbumin and protease inhibitors), lysosomes, and peptides [112].Like human serum albumin, ovalbumin has multiple thiol groups [113].MeHg binds albumin with high affinity, forming MeHg-sulfhydryl conjugates which readily undergo membrane transport and are strongly implicated in toxicity [114,115].An extracellular matrix physically separates egg whites and yolks.Egg yolks almost exclusively contain lipids (84%), with less than 10% soluble proteins [112], allowing for aggregation of lipophilic compounds rather than Hg [116].
Like eggs, feathers are an avian metal excretion mechanism.Elevated feather total Hg concentrations are frequently noted in monitoring efforts [117,118].The feather:liver:muscle Hg ratio is approximately 7:3:1, although this exact conversion should be used cautiously [117].In adult birds, feather total Hg can represent Hg exposure, circulating Hg, and the tissue Hg pool mobilized during feather growth [118,119].However, molt patterns, foraging behavior, body condition, and age are among the many variables influencing feather and internal tissue Hg concentrations [118,119].
Given the role of the liver in avian trace metal detoxification, specifically Hg demethylation [120], the greatest total Hg accumulation in internal tissues was observed in chicken livers, and 54% of the total Hg in liver was present as MeHg.Exercising caution when consuming eggs and liver from backyard chickens in mining communities is advisable to limit human Hg exposure.

Proximity to mining influences Hg intake
We found that mining presence in the area where a crop or meat is harvested strongly influences Hg accumulation.The dietary Hg risk of consuming foods from mining-impacted locations exceeds those of upstream regions.Our analysis assumed similar diets between communities because the survey/community populations were not large enough to reliably differentiate diets in sites with high versus minimal mining activity.
If it is not possible to source high-risk foods from outside the community, which is likely the case for many crops and fish, alternating fish species or turning to chicken meat could substantially reduce dietary Hg exposure.The PWIs calculated for fish and chicken were 4.17 µg kg −1 bw and 0.04 µg kg −1 bw, respectively, relative to the PTWI recommendation (1.6 µg kg −1 bw).If fish protein were swapped for chicken protein by increasing chicken intake four-fold and decreasing fish intake four-fold, the PWIs would drop to 1.04 µg kg −1 bw for fish and 0.17 µg kg −1 bw for chicken.This substitution is just one potential change communities could make while consuming local protein and maintaining the proportions of each fish species consumed, which may be necessary given seasonal accessibility constraints.Typical meals with different protein sources are depicted in figure 7 to illustrate this concept.

Conclusions
We sought to provide a comprehensive examination of total Hg and MeHg content in various common foods and locally available fish from ASGM-impacted areas of Madre de Dios, Peru, to assist affected individuals in balancing dietary risks and benefits.Foods from terrestrial and aquatic ecosystems are accumulating Hg in ASGM-impacted areas.The estimated weekly Hg intake from local crop, fish, chicken, and egg consumption suggests that intentional diet choices are needed to avoid potential adverse health consequences.In an average diet shown through our survey results, the combined PWI from crops, fish, and chicken products exceeds WHO recommendations by more than 3-fold, suggesting a generally elevated baseline dietary Hg exposure for people living in Madre de Dios.It must be noted that weekly exposure calculations used adult body weight, and children likely face comparatively higher exposures given their lower body weight.These findings are valuable to community health-particularly for indigenous populations with a firm reliance on local fish-and governments seeking to further conservation efforts.
The widespread Hg pollution from ASGM in the Peruvian Amazon confers human health risks, which can be combated in part by an improved understanding of dietary exposure pathways.However, forming the most effective and beneficial public health recommendations presents challenges.Fish are important sources of protein, fat-soluble vitamins, and essential fats like omega-3 polyunsaturated fatty acids [4].Diversifying fish species consumption is also necessary to reap all macro and micronutrient benefits [121].Restricting or eliminating fish intake to reduce Hg intake would introduce new health risks by limiting valuable nutrients and could threaten cultural food traditions [4].Similarly, chicken eggs are a regional protein staple and should not be avoided entirely.Rather, understanding how Hg accumulates in fish and chicken eggs can promote exposure reduction.Finally, we acknowledge that while current safety standards are based on the best available data, no absolute threshold for health effects exists because there is no safe Hg exposure level [122].Pollution control efforts must be taken wherever possible to protect population health.

Figure 1 .
Figure 1.Study chicken, crop, and fish sample collection sites in Madre de Dios, Peru.Known mining-impacted areas based on geospatial imaging in Caballero Espejo et al [31] are denoted in red.Adapted with permission from Biorender.com.Created with BioRender.com.CC BY-NC-ND 4.0.

[
total Hg fw] = [total Hg dw] sample dw sample fw

Figure 2 .
Figure 2. Mercury (Hg) concentrations in crops collected from 17 Madre de Dios communities arranged by median total Hg concentrations.(a) Total Hg concentrations on a dry-weight basis (dw), (b) Methyl Hg (MeHg) concentrations (dw), and (c) percent of total Hg present as MeHg.Adapted with permission from Biorender.com.Created with BioRender.com.CC BY-NC-ND 4.0.

Figure 3 .
Figure 3.Total mercury (Hg) content in fish samples collected in Madre de Dios, Peru by trophic level.Hg measurements are compared to guidelines set by the EPA of 0.3 mg kg −1 total Hg (horizontal line) and the WHO guideline of 0.5 mg kg −1 total Hg.Fish species are presented from highest to lowest average total Hg content.Vertical dotted lines divide species by trophic level.Adapted with permission from Biorender.com.Created with BioRender.com.CC BY-NC-ND 4.0.

Figure 4 .
Figure 4. Total mercury (Hg) and δ 15 N (‰) relationship in fish collected in Madre de Dios, Peru.Fish were assigned a trophic level based on a literature review.From the slope of the regression line (y = 0.17x-2.73),the trophic magnification factor (TMF = 10 β ) was 1.48.

Figure 5 .
Figure 5.Total mercury (Hg) and methyl Hg (MeHg) concentrations in chicken eggs and internal tissues from Madre de Dios, Peru by proximity to mining.(a) Total Hg content by tissue type on a dry-weight basis (dw), (b) MeHg content, and (c) percent of total Hg present as MeHg in chicken eggs and meat.

Figure 6 .
Figure 6.Probable weekly intake (PWI) of a) total mercury (Hg) and b) methyl Hg (meHg) by food group (µg of Hg per kg of body weight per week).Fresh-weight (fw) total Hg content was used in the calculations.The horizontal line represents the provisional tolerable weekly intake (1.6 µg kg −1 body weight per week) from the World Health Organization.The MeHg PWI was calculated for fish assuming that MeHg comprises 95% of total Hg content, although natural variation exists among different species [109].

Figure 7 .
Figure 7. Average total mercury (Hg) intake estimated for three common meals in Madre de Dios.Meal 1 represents a typical breakfast, meals 2a and 2b depict a typical lunch, and meal 3 represents a typical dinner.Each meal is color-coded by Hg exposure risk (red = high exposure risk, orange = moderate exposure risk, green = low exposure risk).

Table 1 .
Locations of crops, fish, and chicken samples collected from 19 Madre de Dios communities.Areas are characterized by the presence of amalgam burning and mining, indigenous communities, and urbanization.

Table 4 .
Comparison of mercury (Hg) concentrations in fish by trophic level and origin in Madre de Dios, Peru.Fresh-weight (fw) total Hg concentrations (mg kg −1 ) are compared to the WHO Guideline Level (GL) for total Hg in fish muscle tissue (0.5 mg kg −1 ) and the EPA Fish Tissue Residue Criterion (FTRC; 0.3 mg kg −1 ).

Table 6 .
Isotopic profiles and total mercury (Hg) content of chicken feathers collected in Madre de Dios, Peru.Samples are categorized by the presence of mining in the community of purchase.Total Hg data are presented as medians with interquartile ranges (IQR) on a dry-weight basis (dw), and isotopic data are presented as means ± standard error (SE) because the data are normally distributed.

Table 7 .
Diet survey results from participants who provided chicken meat, eggs, and feathers to the study.
[76]WI exceeds the Provisional Tolerable Weekly Intake (PTWI) for Hg established by the WHO of 1.6 µg per kg body weight per week.Note that PTWIs are only established for total Hg intake, not MeHg intake.bTheHgintakefrom one egg was estimated by combining [total Hg] from the yolk and the white under the assumption that chicken eggs are approximately 67% white and 33% yolk by mass[76].
a c Total Hg & MeHg fw concentrations, weekly consumption, and PWI data for individual internal chicken tissues are presented in table

Table 9 .
A comparison of total mercury (Hg) and methyl Hg (MeHg) content of chicken internal tissues and eggs in this study versus previous studies.Results are presented on a dry weight (dw) or fresh-weight (fw) basis.