Human exposure to soil contaminants in subarctic Ontario, Canada

Background Chemical contaminants in the Canadian subarctic present a health risk with exposures primarily occurring via the food consumption. Objective Characterization of soil contaminants is needed in northern Canada due to increased gardening and agricultural food security initiatives and the presence of known point sources of pollution. Design A field study was conducted in the western James Bay Region of Ontario, Canada, to examine the concentrations of polychlorinated biphenyls, dichlorodiphenyltrichloroethane and its metabolites (ΣDDT), other organochlorines, and metals/metalloids in potentially contaminated agriculture sites. Methods Exposure pathways were assessed by comparing the estimated daily intake to acceptable daily intake values. Ninety soil samples were collected at random (grid sampling) from 3 plots (A, B, and C) in Fort Albany (on the mainland), subarctic Ontario, Canada. The contaminated-soil samples were analysed by gas chromatography with an electron capture detector or inductively coupled plasma mass spectrometer. Results The range of ΣDDT in 90 soil samples was below the limit of detection to 4.19 mg/kg. From the 3 soil plots analysed, Plot A had the highest ΣDDT mean concentration of 1.12 mg/kg, followed by Plot B and Plot C which had 0.09 and 0.01 mg/kg, respectively. Concentrations of other organic contaminants and metals in the soil samples were below the limit of detection or found in low concentrations in all plots and did not present a human health risk. Conclusion Exposure analyses showed that the human risk was below regulatory thresholds. However, the ΣDDT concentration in Plot A exceeded soil guidelines set out by the Canadian Council of Ministers of the Environment of 0.7 mg/kg, and thus the land should not be used for agricultural or recreational purposes. Both Plots B and C were below threshold limits, and this land can be used for agricultural purposes.

ood security exists ''when all people, at all times, have physical and economic access to sufficient, safe and nutritious food to meet their dietary needs and food preferences for an active and healthy life'' (1). In Canada, Aboriginal people (First Nations, Inuit, and Metis) are disproportionately food insecure compared to the Canadian general population Á this is especially true for northern Aboriginal people Á where up to 70% of the households in some areas were found to be food insecure (2,3). Indeed, remote Aboriginal people face unique food security issues related to the high cost of market food, as this type of food is typically flown in (4,5), and the high cost associated with hunting and fishing, such as cost of fuel to travel to hunting sites and financial expenses of owning hunting equipment (6,7). Thus, it is not surprising that food localization projects have been planned and initiated in northern Canada with respect to gardening, at both the small (home gardens and small greenhouses) and medium (community gardens and large greenhouses) scales (3,8,9). However, the soils used in these gardening ae International Journal of Circumpolar Health 2015. # 2015 Ellen Stephanie Reyes et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), allowing third parties to copy and redistribute the material in any medium or format and to remix, transform, and build upon the material for any purpose, even commercially, provided the original work is properly cited and states its license.
initiatives have not been tested for contaminant levels, which is important because of known point sources in northern Canada for organochlorines from old radar lines (10Á13) and metals from mines (14,15), as well as persistent organic pollutants (POPs) travelling long distances via atmospheric transport from industrial countries and deposited into the soil (16).
Importantly, during the Cold War in the 1950s the Mid-Canada Radar Line (MCRL) in subarctic Canada was built by the Government of Canada during the 1950s, in response to the threat of a nuclear attack from the Soviet Union (13,17). The MCRL was deemed redundant in the mid-1960s by the Canadian military and was decommissioned (13,17). However, since the radar-line stations were typically not properly decommissioned at the time of closure, many MCRL sites have become point sources of environmental contamination (e.g. polychlorinated biphenyls, PCBs; dichlorodiphenyltrichloroethane, DDT) (12,13). Site 050 (located on Anderson Island in close proximity to the community of Fort Albany First Nation, Ontario, Canada) was the first MCRL site to be remediated (12). This abandoned radar-line site received remediation priority because of elevated levels of PCBs in soil (21,000 ppm) and vascular plants (up to 550 ppm), with !50 ppm considered hazardous waste in Canada (12,17). Potential sources of DDT exposure also include DDT-contaminated soil surrounding MCRL buildings and long-range atmospheric transport from industrial sites, to the extent where remediation was required (16,18). The people of Fort Albany historically worked, lived, and partook in traditional activities (e.g. harvesting plants, berries, fish, and small game) on and around Anderson Island (11). Prior to remediation, there was potential for human exposure and uptake of PCBs and DDT from Site 050 (11).
During operation and when MCRL site 050 was abandoned, materials and equipment were moved off site and buried around the community of Fort Albany (11). Another potential source of soil contaminants in Fort Albany may be associated with the historical (into the 1970s) agricultural use of lands on the mainland, by the Roman Catholic Mission as 2 community members recount how they threw some ''some powdery stuff'' over the fields to control the pests during their time in residential school (8, p.6). The old Roman Catholic Mission agricultural fields and surrounding area were being considered for a new agroforestry initiative; agroforestry is a more sustainable land-use system than conventional agriculture, as it uses woody perennials with crops to optimize beneficial biological interactions (8,9).
Since organochlorine contaminants are highly lipophilic and resist biodegradation in the environment, they tend to bioaccumulate in biota and biomagnify up the marine food chain (19). Consequently, Arctic populations are exposed to greater concentrations because they are at the highest trophic level of the food chain and have more of a reliance on a subsistence diet (19,20). Soil contamination was a relevant issue from both direct exposure (e.g. ingestion and dermal contact) and indirect exposure (e.g. water ingestion, ingestion of vegetation, and exposure from the food chain) perspectives. Three potential plots of land were considered for the agroforestry initiative (8,9). In the present study, soil contamination was assessed in the 3 plots with respect to soil contaminant concentrations to inform the siting of the agroforestry initiative.

Methods
Description of study site Fort Albany First Nation is located on the western shore of the James Bay region (52815?N, 81835?W) of Ontario, Canada (21). Approximately 850 people live in the community (21). Figure 1 shows that the community is situated on Sinclair Island, but First Nation members also live on the Mainland and Anderson Island (21). MCRL Site 050 was located on Anderson Island. As Fort Albany is a remote community, accessibility is limited with barges during late spring to early fall, ice/snow during the winter, and year-round access by aircraft (13).

Field sample collection
Ninety soil samples were collected at random (grid sampling) from 3 plots (A, B, and C) in Fort Albany (on the mainland), subarctic Ontario, Canada. The plot areas were 10 )10 feet and the soil samples were taken at root level due to concerns with potential plant uptake of contaminants. Six inch sample cores were collected and weighed before being refrigerated and shipped in Ziploc † bags for analysis. All soil samples were prepared and analysed for PCBs, DDT and its metabolites, other organochlorines, and metals/metalloids at the Analytical Services Unit, Queen's University, Kingston, Ontario.

Sample preparation
Sample preparation for organochlorines Samples were thoroughly homogenized before sampling for extraction and cleanup. Soil samples were subsampled for determination of wet/dry weight ratio. Accurately weighed 10 g of soil sample to which an aliquot of surrogate standard, dechlorobiphenyl, 40 g of sodium sulphate, and 20 g of Ottawa sand were added. Samples were extracted 3 times for 20 minutes with 50 mL of dichloromethane on an orbital shaker. The extract was then concentrated by rotoevaporation to approximately 1 mL, and 5 mL of hexane was added and again evaporated to 1 mL. This was repeated twice more, resulting in 1 mL of hexane solvent, which was then applied to a Florisil column for cleanup. The column was thoroughly rinsed with hexane and the eluent containing the organochlorines  Sample preparation for metals/metalloids (except for Hg) Samples were air-dried and ground to a fine powder with a mortar and pestles. Large stones were removed from the soil samples, as they would not be expected to contain any anthropogenic environmental contaminants. Accurately weighed 0.5 g of powdered soil sample was heated with 2 mL of nitric acid (HNO 3 ) and 6 mL of hydrochloric acid (HCl) and reduced the volume to 1Á2 mL. This solution was made up to 25 mL with deionized water and filtered through a Whatman No. 40 filter paper. Analysis and quality assurance for metals/metalloids (except for Hg) A 30-element suite of metals/metalloids was analysed: Ag, Al, As, B, Ba, Be, Ca, Cd, Co, Cr, Cu, Fe, K, Mg, Mn, Mo, Na, Ni, P, Pb, S, Sb, Se, Sn, Sr, Ti, Tl, U, V, and Zn. Samples were analysed in batches of up to 36, which comprised up to 28 samples, 2 blanks, 4 duplicates, and 2 reference materials, Mess-3 and SS-2, from the National Research Council of Canada. The control limits for Mess-3 and SS-2 are in Supplementary Tables I and II. 1 The typical percentages of each of the isomers is 77% p,p?-DDT, 15% o,p?-DDT, 4% p,p?-DDE, 0.1% o,p?-DDE, 0.3% p,p?-DDD, 0.1% o,p?-DDD, and a number of unidentified compounds (3.5%) (22). The typical percentages are added to highlight the importance of measuring each isomer separately. All samples were analysed by an inductively coupled plasma mass spectrometer.

Analysis and quality assurance
Analysis and quality assurance for mercury The Hg analysis was determined by cold vapour atomic absorption spectrophotometry. The instrument, a direct mercury analyser (DMA-80), allowed for measurement with little to no sample preparation as described in the U.S. Environmental Protection Agency (EPA) Method 7473 (23). Samples, which were weighed into quartz or nickel boats, enter the instrument's chamber where the sample is first dried and then thermally decomposed in a continuous flow of oxygen (O 2 ). The combustion products are carried off in the O 2 and are then further decomposed in a hot catalyst bed. The Hg vapours are trapped on a gold amalgamator tube and desorbed for spectrophotometric quantitation at 254 nm. Quality assurance/quality control procedures included method blanks and laboratory control samples throughout the entire sample preparation and analytical process (23).

Soil quality guidelines
The Canadian Council of Ministers of the Environment (CCME) developed soil quality guidelines for total DDT, total PCBs, and some metals/metalloids for the protection of the environment and human health (24). Table I presents the soil quality guidelines depending on land usage. The key total DDT metabolites that are found in the northern environment are p,p?-DDT, o,p?-DDT, and p,p?-DDE. Since DDT is a persistent chemical that tends to bioaccumulate and biomagnify throughout the food chain, as well as contribute with atmospheric transport to the Arctic from industrial areas, the concept of land usage has to be taken into consideration, particularly with agricultural and residential/parkland (22). For the purposes of this study, comparisons are made between the soil quality data from the study site in Fort Albany with the guideline for total DDT, total PCBs, and metals/ metalloids from the agricultural and residential/parkland usage.

Soil exposure factors and calculations
Ideally, all pathways should be considered when estimating the daily intake of a chemical contaminant. However,  only the exposure factors (EFs) for soil ingestion and soil dermal uptake were accounted for as these are the likely exposure routes as shown in Table II. The default factor values used in risk assessment analyses used to estimate exposure to pesticides frequently overestimate exposure but are the first preference for assessing the safety implications for the community (25). For instance, the exposure duration is 5 days a week for 52 weeks per year; however, the actual exposure duration is not as long. The default EFs were stratified according to age group to indicate how often the individual is exposed to the contaminant during a year and the number of years this pattern has been repeating (26). Health Canada's recommendations for maximum estimate of soil ingestion intake of 35 mg/day for the 5Á11 year age group, and 20 mg/day for the 12Á20 year and 21' year age groups were not used for this study (26). A soil ingestion study done by Harper et al. (27) recommended a maximum estimate of soil ingestion rate of approximately 400 mg/day because it is a conservative parameter used to evaluate health risks associated with the contaminated sites. This EF is based on Aboriginal practices that involve consuming traditional food sources that can become contaminated with soil particles, gardening, gathering, and preservation techniques that can increase the level of soil contact, and other additional environmental activities (e.g. outdoor recreation for children and cultural activities) (27,28). As a worst-case scenario, a bioavailability value of 1 (100%) was used in exposure estimation. The EF is calculated to estimate an average dose over the exposure period Health Canada (26)  The amount of total DDT absorbed into the body by soil ingestion (ED s ) is estimated with the following equation (26): where ED s 0is the estimated dose through soil ingestion expressed as milligrams of contaminant eaten per kilogram of body weight per day (mg/kg/day); C0the concentration of the contaminant in the soil in milligrams per kilogram of soil (mg/kg); IR=the soil ingestion rate, the amount of soil an individual eats in a day in milligrams (mg/day); EF0the exposure factor, which indicates how often the individual has been exposed to the contaminant over a lifetime; and BW 0the body weight, that is, the average body weight in kilograms based on an individual's age group (kg).
The amount of total DDT that is absorbed into the body through dermal contact with contaminated soil (ED ss ) can be estimated with the following equation (26): where ED ss 0is the estimated dose through dermal contact with soil expressed as milligrams of the contaminant absorbed through the skin per kilogram of body weight per day (mg/kg/day); C0the concentration of the contaminant in the soil in milligrams per kilogram of soil (mg/kg); A0the total soil adherence, amount of soil that sticks to an individual expressed in milligrams per day; BF0the bioavailability factor, the percentage of the contaminant in the soil that is actually free to move out of the soil and through the skin (unitless); EF0the exposure factor, indicates how often the individual has been exposed to the contaminant over a lifetime; and BW 0body weight, average body weight in kilograms based on an individual's age group (kg).

Recommended estimated maximum intake values
Regulatory agencies developed guidelines and advisories regarding the usage of DDT, DDE, and DDD in the environment. The details of the recommended intake values applicable to DDT are summarized in Table III.

Soil quality
The concentration ranges of total DDT found in the soil plots were distributed heterogeneously with values ranging from below the detection limit to 4.19 mg/kg. Table IV indicates that Plot A had the highest total mean DDT concentration of 1.12 mg/kg, followed by Plot B and Plot C, which were 0.09 and 0.01 mg/kg, respectively. The concentrations of the PCBs and other organochlorines were below the detection limit and hence we have not presented the data as these contaminants do not pose a health risk. The metal concentrations in the 3 soil plots are presented in Supplementary Table III. In comparison  to the available Canadian soil quality guidelines from  Table I, contamination in the plot sites by metal pollutants is not of concern. The concentration levels of the toxic metals are well below the soil quality guidelines, and thus meet the benchmark for safe usage of relevant land resources (24).

Soil exposure analysis for DDT
To assess the potential health risks due to the contamination of DDT compounds in the soil plots, exposure model calculations were applied. Figure 2 presents the exposure to total DDT by direct soil ingestion. The data clearly showed that Plot A had a much higher level of DDT compounds compared to B and C. In general, the level had the order: child !teen!adult. The mean exposure concentration9standard deviation of total DDT by soil ingestion was 4.33 )10 (5 95.04 )10 (5 mg/kg/day, with ranges not detectable to 9.68)10 (4 mg/kg/day. Table V shows that the estimated daily intake (EDI) of total DDT was averaged to be 4.35)10 (5 mg/kg/day over a lifetime of 70 years (hazard index00.00435). The EDI is tabulated by adding each possible combination of exposure pathway, and it is noted that the estimated dose is calculated separately for each age group. The results show that soil dermal uptake is the main exposure pathway to total DDT.

Discussion
Comparison of soil quality guidelines and recommended maximum intake values There was a high degree of variability for total DDT between each soil plot. The total DDT levels of both Plots B and C (0.09 and 0.01 mg/kg, respectively) were orders of magnitude below the maximum threshold limits developed by CCME for agricultural and parkland/usage of 0.7 mg/kg (22). Plot A had a total DDT level of 1.12 mg/kg and this result was higher than Canada's soil guidelines. It is worth mentioning that DDT was prohibited and removed from major use in Ontario since 1972 but still persists in the natural environment (13,37). Since Plot A exceeded government guideline threshold limits, this plot should not be used for agricultural or recreational purposes.
Interestingly, the results from the present study indicate that if there were potential health concerns due to DDT exposure, this would mainly occur through dermal contact. This finding contrasts with other studies that suggest that direct ingestion is the most common route of exposure to DDT (16,38,39). However, these studies noted that food consumption is the main source of intake. Bard's (16) study focused on POP contamination from consuming fish and other marine mammals, and Dougherty et al. (38) and MacIntosh et al.'s (39) studies assessed the potential hazards of consuming various food products, such as milk, beef, and fish, as sources of exposure to DDT. Lastly, the results from the estimated exposure assessment to Reference dose (RfD): an estimated daily oral exposure of a chemical to the human population (including sensitive groups) that is likely to be without an appreciable health risk over a lifetime (35). c Tolerable daily intake (TDI): an estimated amount of a substance in food, drinking water, or air that can be ingested over a lifetime without deleterious, non-carcinogenic effects (36). total DDT by direct soil ingestion were below regulatory guidelines set out by the U.S. EPA and Health Canada for all soil plots and age groups. Since the results from the soil exposure analysis were below the reference dose and tolerable daily intake, this indicates that the exposure level to DDT via the soil is not likely to pose any risk to human health, even using a bioavailability factor of 1 (100%).  (41,43).

Comparison of soil samples from different locations
In comparison to soils from southern Poland for the cities of Katowice (mean: 0.110 mg/kg) and Kraków (mean: 0.260 mg/kg), the contamination level of DDT found in Fort Albany is similar (52). The soil plots in India for the District Dibrugarh (mean: 0.757 mg/kg), District Nagaon (mean: 0.903 mg/kg), and Agra (mean: 1.01 mg/kg) have relatively elevated total DDT values compared to Fort Albany (50,51), which is not surprising, as these soil concentration values are for a country that continues to produce and use organochlorine pesticides such as DDT as a vector control agent (50).
Both China and East Antarctic have lower contamination levels of total DDT compared to Fort Albany with ranges from BLoD to 6.36)10 (2 mg/kg (45Á47, 49). However, the exception was an industrial soil site in Beijing, China, that had unevenly distributed total DDT concentrations of 3.02Á67.43 mg/kg in different soil layers (48). This is of particular concern since this contaminated-soil site is currently a paint factory, but plans are developed for future restoration and residential development (48).

Conclusion
The soil used in the First Nation community of Fort Albany, Ontario, Canada, is primarily contaminated by DDT, but also by low concentrations of some metals/ metalloids (e.g. As, Ba, Co, Cr, Cu, Ni, Pb, V, and Zn). The soil exposure and estimated analysis revealed no known health risks to humans, as the results were well below  government thresholds recommended by the U.S. EPA and Health Canada, even though the SDDT concentration in the Plot A soil was above Canada's soil quality guidelines. Nonetheless, it is prudent for any agricultural initiative in northern Canada to first test the soil for contamination, as there are many sources of contaminants in the north other than long-range transport that may impact the quality of the food produced. The methods in the Fort Albany, Ontario study can also be used to measure contaminants in soil of communities located near known or suspected point sources of pollution, such as military sites (e.g. White Alice sites in Alaska) (53) and near extraction industries (e.g. Russia's European High North) (54) to address environmental health concerns across the Circumpolar North.