Location: Ministry Home > Publications > Technical Studies and Reports > 4022App_X.pdf > HTML version

This is a HTML version of the original PDF document. The HTML version is being provided for reading purposes only and is not the official version of the document.

APPENDIX X

UPDATED PQRA:

ASSESSMENT OF RISKS TO HUMAN
HEALTH FROM EXPOSURE TO
SEDIMENTS IN STUDY AREA A
(DELORO MINE SITE TO THE INLET
OF MOIRA LAKE)

Distribution:

2 Copies- Ministry of Environment, Kingston, Ontario
1 Copy - GlobalTox International consultants Inc, Guelph, Ontario
1 Copy - Golder Associates ltd, Calgary, Ontario
1 Copy - Golder Associates Ltd, Mississauga, Ontario

Golder Associates

TABLE OF CONTENTS

SECTION

• 1.0 INTRODUCTION
  • 1.1 Scope of work
  • 1.2 Study Approach
• 2.0 SEDIMENT SAMPLING AND ANALYSIS
  • 2.1 Methods
  • 2.2 Analytical Results
• 3.0 PRELIMINARY QUANTITATIVE RISK ASSESSMENT
  • 3.1 Exposure Assessment
  • 3.2 Receptor Selection
    • 3.2.1 Non-cancer Endpoints
    • 3.2.2 Cancer Endopoints
  • 3.3 Toxicity Assessment
  • 3.4 Risk Characterization
    • 3.4.1 Acute Effects
    • 3.4.2 Chronic Non-Cancer Endpoints
    • 3.4.3 Cancer
  • 3.5 Uncertainity Analysis
• 4.0 CONCLUSIONS AND RECOMMENDATIONS
  • 4.1 Conclusions
  • 4.2 Recommendation
• 5.0 REFERENCES

• LIST OF TABLES

• Table 1 Arsenic Concentrations Detected at Study Area A Sampling Sites
• Table 2 Exposure Pathways
• Table 3 Exposure Pathway Equations and Input Variables for Accute Effects
• Table 4 Exposure Pathway Equations and Input Variables for Chronic and Sub-Chronic Non-Cancer Effects
• Table 5 Exposure Pathway Equations and Input Variables for Cancer Assessment
• Table 6 Receptor Characteristics - Scenario 1
• Table 7 Receptor Characteristics - Scenario 2
• Table 8 Receptor Characteristics (Cancer) - Scenario 1
• Table 9 Receptor Characteristics (Cancer) - Scenario 2
• Table 10 Toxicological Criteria and Sediment Bioavailability Factors for Arsenic
• Table 11 Estimated Arsenic Exposure Levels and Exposure Ratios for Acute Effects (Average)
• Table 12 Estimated Arsenic Exposure Levels and Exposure Ratios for Acute Effects (Maximum)
• Table 13 Arsenic Exposure Levels and Exposure Ratios for Chronic Non-Cancer Health Effects (Average)
• Table 14 Arsenic Exposure Levels and Exposure Ratios for Chronic Non-Cancer Health (Maximum)
• Table 15 Estimated Arsenic Exposure Levels and Associated Cancer Risk levels (Average)
• Table 16 Estimated Arsenic Exposure Levels and Associated Cancer Risk Levels (Maximum)

• LIST OF FIGURES

• Figure 1 - sediment Sampling Locations in Study

1.0 INTRODUCTION

This Updated Preliminary Quantitative Risk Assessment (PQRA) was produced in response to the findings of the PQRA presented in the Technical Report of the Phase II Moria River Study (Golder, 2000) The original PQRA was a conservative, screening level assessment It was used to determine whether are potential current and future human health risks associated with substances released to the Moria River system from the Delore Mine Site The results of the PQRA were used to determine whether a more detailed risk assessment was needed

The PQRA considered how users of the rive system might come into contact with metals present in water, sediments and fish and then compared estimated exposures to benchmark values that represent safe or acceptable exposure levels The PQRA used conservative assumptions to estimate exposure therefore, the exposure estimates were "worst case", and are likely to overestimate true exposure levels and associated risks If the "worst Case" exposure estimates were below the benchmark values, then there was a high level of confidence that the exposure pathways and metal concentrations being considered did not require further assessment If the "worst case: exposure estimates were above the benchmark values, than further assessment of those particular pathways and metals was indicated

The PQRA showed that all estimated exposures to the metals of concern were below benchmark values except for arsenic in Study Area a (Deloro Mine Site to the Inlet of Moria Lake) and Study Area B (Moira Lake) the PQRA indicated that residents of Study Areas A and B would have greater estimated levels of exposure to arsenic than the Typical Ontario Resident (TOR) under the conditions assumed in the assessment Use of water from the river system for drinking water was the most significant pathway of exposure to arsenic The drinking water reminder issued by the MOE and the Hastings and Prince Edward Counties Health Unit (Appendix IX) was and continues to be appropriate.

Exposure to sediments in Study Area A was another exposure pathway that produced estimated exposures above benchmark values This section of the river is not well suited for in-water recreational activities and has only a limited number of residences However, recreational use of this section of the river system may be an important exposure pathway, at least for some individuals. The amount of exposure through this pathway, both in absoluteterms and relative to total arsenic exposure, required further study.

The overall goal of this Updated PQRA was to determine whether exposure to sediments in Study Area A contributes significantly to potential risk to persons who come into contact with these sediments the results would indicate whether specific risk management options for sediments are required.

1.1 Scope of Work

The scope of work for this Updated PQRA is to assess the potential risk to recreatiional users of the Moria River in Study Area A Associated with exposure to arsenic in sediment.

1.2 Study Approach

The Updated PQRA was completed using similar methodology and assumptions adopted for the PQRA, as discussed in Section 3 of the Technical Report However, this Updated PQRA was focused on locations where children acutally used the river for recreational activities such as swimming and wading, whereas the initial PQRA addressed sediment quality at representative downstream locations To evaluate potential health risks, this assessment considers the fowllowing three exposure and toxicological effects endpoints.

• Acute Effects: (effects other than cancer that occur after a short-term exposure duration, usually defined in hours or days)
• Chronic and Subchronic Non-Cancer Effects: (effects other than cancer that occur after a medium to long-term exposure duration, usually defined in months or years)
• Chronic Cancer Effects: (cancer effects are associated with a long-term exposure duration, usually defined in years or lifetime)

It was necessary to evaluate all three of these endpoints to ensure that the potential risks to individuals exposed to arsenic in sediment in Study Area A were considered A difference between the PQRA (as contained in the Moria River Study report) and the Updated PQRA, is that the PQRA focused on chronic and subchronic non-cancer effects and chronic cancer effects, whereas the Updated PQRA also evaluated acute effects from sediment exposure. A discussion of the specific approach adopted for this updated assessment is described in the following sections.

2.0 SEDIMENT SAMPLING AND ANALYSIS

A supplemental sediment sampling survey was conducted by Golder Associates during the period July 31 to August 3, 2000, to identify access points and potential locations for swimming/bathing/wading along the Moria river within Study Area. A The results of this survey have been complied and included with the results from the initial sediment sampling program, as discussed in Section 212 (Results). A complete discussion of the previous sediment sampling program is presented in Section 22 of the Technical Report (Golder,2000).

It should be noted that only "near-shore" sample locations, defined as locations with 1 m or less water depth, were included in this assessment. The selection of shallow water depths is related to the choice of a child aged 5-11 as the receptor most likely to have the highest exposures. The retionale for the choice of a child receptor is presented in Section 32 (below).

2.1 Methods

Supplemental sediment sampling locations within Study Area A were selected based on a combination of factors that influence the potential for resident and cottager exposure to sediments these include:

• Public or cottage access;
• water depth;
• substrate type; and
• evidence of recreational use

All near-shore recreational activities and evidence of recreational use (docks, boats, beaches, swimming platforms, rope swings, etc) were noted and photographed. Use of an area by children was of particular importance. This use was evidenced by swing sets or playgrounds on shore and toys or children heard or seen playing either on land or in the water. Most of these indications of use were observed during the collection of sediment samples.

Sediment samples were collected from water depths less than 1 meter using a Petite Ponar grab sampler deployed from a small boat. Three grab samples were taken at each site and combined in roughly equal proportions in large stones were removed. A sub-sample was then placed into glass jars for submission under chain of custody to Philip Analytical Services for analysis of total arsenic concentrations Depth, substrate characteristics and current velocities were also recorded at each sampling location

Sediment samples were collected from nine locations (designated as locations 1 through 9) between Bronson's Rapids (Location 1) and the inlet to Moira Lake (Location 9) (see Figure 1). The nine sample locations for the supplemental sampling program were sampled in areas that consisted exclusively of private access areas and varied from sand beaches to marshy shorelines. The area upstream of Bronson's Rapids was not sampled becauses to marshy shorelines. The area upstream of Bronson's Rapids was not sampled because there were very few access points and there was an abundance of poison ivy in the area. Thus, it is unlikely that this area of the river would be used for swimming.

2.2 Analytical Results

Total arsenic concentrations in the supplemental sediment samples ranged from 10.2 mg/kg to 238.0 mg/kg (Table 1) Four samples contained greater than 50% fine organic sediments, with total arsenic concentrations raging from 141.0 mg/kg to 232 mg/kg Two sites had between 39 and 50% fine sediments Arsenic concentrations at these sites were 95.0 mg/kg and 200.0 mg/kg Three sites (RAMR-A-5, RAMR-A-7 AND RAMR-A-9) contained less than 20% fine sediments and the total arsenic concentrations at these stations ranged from 10.2 mg/kg to 25.4 mg/kg.

The sediment samples collectted at water depths less than 1 metre during the first sampling program in 1999 were not used in this assessment these sites are in the section of the river that has very few access points and abundant poison ivy along the banks use of the river for recreational purposes is highly unlikely in this portion of the river

Figure 1 Sediment Sampling Locations in Study Area A

TABLE 1

Arsenic Concentrations Detected at Study Area A Sampling Sites

¹ equals percent silt and percent clay

² probably private access point, but no cottage observed

The average arsenic concentration of 129 mg/kg in sediment samples from the supplemental sampling program was used in the exposure assessment in the Updated PQRA Use of this value is appropriate for the exposure assessment because it is representative of the sediment quality in Study Area A at areas where known human access to the river occurs In addition, modelling using the maximum arsenic concentration of 238 mg/kg was also employed to represent a "worst case" exposure scenario

3.0 PRELIMINARY QUANTITATIVE RISK ASSESSMENT

The Updated PQRA is a semi-quantitative evaluation of the potential risk to human health resulting from exposure to arsenic present in sediments from Study Area A. The methodology and assumptions adopted for this assessment are similar to those used in the PQRA, as described in Section 3 of the Technical Report.

A deterministic modelling approach has been adopted for this updated PQRA. Health-protective point estimates for all parameters used to describe exposure and toxicity have been incorporated into this assessment. for example, maximum plausible estimates of receptor characteristics including body weight, body surface area and water consumption have been used. This approach is expected to result in exposure estimates that are greater than what actually occurs. In addition, both maximum and average arsenic concentrations were considered.

3.1 Exposure Assessment

The exposure assessment describes and quantifies how humans may come into contact with arsenic in sediments from Study Area A. Considering that people may reside on the river system all year and may come into contact with arsenic in sediments, various potential pathways of exposure were examined.

As described above, the average concentration of arsenic in sediments from Study Area A collected at depths of I metre or less was used for the exposure modelling (Table I). Use of these shallow water sediment samples is appropriate since it is considered unlikely that a child aged 5 - 11 (i.e., the receptor) will come into contact with sediments at depths much greater than I metre. Exposure was estimated using sediment concentrations, along with receptor body weight, ingestion rate, dermal contact rate, time activity patterns (i.e., time spent swimming) and chemical bioavailability.

Table 2 is a summary of the exposure pathways that have been considered to be important for this assessment.

TABLE 2

Exposure Pathways

Equations and a detailed description of the input variables that were used for the exposure modelling for these two pathways are given in Table 3 (acute effects), Table 4 (chronic non-cancer health effects) and table 5 (cancer effects).

TABLE 3

Exposure Pathway Equations and input Variables for Acute Effects

Exposure Pathway	Description of Exposure	Exposure Pathway Calculations	Value
1	Accidental Ingestion of sediment where	AIS = [SI x CS x OB]/BW AIS = Accidental ingestion of sediment (µg/kg/day) SI = Sediment ingestion (J/day) CS = Arsenic concentration in sediment (µg/g) OB = Oral bioavailability (unitless) BW = Body weight (kg)	Calculated see Tables 6 and 7 129 or 238 µg/g See Table 10 32.9 kg
2	Dermal contact with sediment where	DSE = (C x A x BF x 10-3) / BW DSE = Dermal sediment expsoure (µg/kg/day) C = Arsenic concentration in sediment A = Soil adherence BF = Bioavailability factor BW = Body weight (kg)	Calculated 129 or 238 µg/g see Tables 6 and 7 See Table 10 32.9 kg

The dose estimates calculated using these equations identify the daily dose of arsenic that the child receptor receives. Comparison of this dose estimate to the acute toxicity reference value for arsenic will determine whether the receptor may be expected to experience acute arsenic toxicity effects.

Arsenic can also cause chronic and sub-chronic non-cancer health effects in humans. To determine the potential for these types of adverse effects to arise in exposed individuals, it is necessary to calculate the average daily dose, occurring in any given year. The equations and a detailed description of the input variables that were used for the exposure modelling for the two exposure pathways are given in table 4 (chronic and sub-chronic non-cancer endpoints).

TABLE 4

Exposure Pathway Equations and input Variables for Chronic and Sub-Chronic Non-Cancer Effects

Exposure Pathway	Description of Exposure	Exposure Pathway Calculations	Value
1	Accidental ingestion of sediment where:	AIS = [SI x CS x OB] x EF x ED/BW x AT AIS = Accidental ingestion of sediment (µg/kg/day) SI = Sediment ingestion (g/day) CS = Arsenic concentration in Sediment (µg/g) OB = Oral bioavailability (unitless) EF = Exposure Frequency ED = Exposure Duration BW = Body weight (kg) AT = Averaging Time	Calculated see Tables 6 and 7 129 or 238 µg/g See Table 10 60 or 14 days 7 years 32.9 kg 7 years
2	Dermal contact with where:	DSE = [ C x A x BF x 10-3] x EF x ED / BW x AT DSE = Dermal sediment exposure (µg/kg/day) C = Arsenic concentration in sediment A = Soil adherence BF = Bioavailability factor EF = Exposure Frequency ED = Exposure Duration BW = Body weight (kg) AT = Averaging Time	Calculated 129 or 238 µg/g see Tables 6 and 7 See Table 10 60 or 14 days 7 years 32.9 kg 7 years

Since arsenic can cause cancer, it is necessary to estimate the average daily dose of arsenic to which a person is exposed over an entire lifetime. This dose, termed the Lifetime Average Daily Dose (LADD), is multiplied by the cancer potency factor for arsenic to estimate the associated increased cancer risk. Equations and a detailed description of the input variables that were used for the exposure modelling for these two pathways are given in Table 5 (cancer endpoints).

Table 5

Exposure Pathway Equations and Input Variables for Cancer Assessment

Exposure Pathway	Description of Exposure	Exposure Pathway Calculations	Value
1	Accidental ingestion of sediment Where:	AIS =([SI x CS x OB x F] / (BW x AT) AIS = Accidental ingestion of sediment (µg/kg/day) SI = Sediment ingestion (g/day) CS = Arsenic concentration in sediment OB = Oral bioavailability (unitless) F =Frequency of events AT = Averaging time BW = Body weight (kg)	Calculated See Tables 6 and 7 129 or 238 µg/g See table 10 See Tables 6 and 7 365 days See tables 8 and 9
2	Dermal contact with sediment Where:	DSE = ([C x A x BF x 10-3 x F] / (BW x AT) DSE = Dermal sediment exposure (µg/kg/day) C = Arsenic concentration in sediment A = Soil adherence BF = bioavailability factor F = frequency of evemts AT = averaging time BW = body weight (kg)	Calculated 129 or 238 µg/g See Tables 6 and 7 See Table 10 See Tables 6 and 7 365 days See Tables 8 and 9

3.2 Receptor Selection

3.2.1 Non-Cancer Endpoints

For the assessment of non-cancer endpoints, a child aged 5 to 11 years old was indentified as the most appropriate human receptor to evaluate potential non-cancer effects. This age group was selected for the receptor because the physical characteristics and activity patterns for children of this age make them the most likely to have the greatest exposure on a body weight basis. For example. they generally spend more time outdoors swimming and playing along the river and lakes than either adults or younger children, and are expected to have higher levels of exposure on a body weight basis due to their physical characteristics. Therefore, use of this receptor represents a health-protective approach to the Updated PQRA.

This assessment considered two different receptor scenarios, as summarized below:

• Receptor scenario 1: This scenario represents a resident child receptor that swims in the Moira River in study Area a for 60 days each year. The child receptor characteristics for this scenario are indentical to those indentified in the pQRA. as discussed in the Technical Report.
• Receptor Scenario 2: This scenario represents a child receptor that vacations for a two week period in Study Area A and uses the river for swimming over this period. For this scenario, it is assumed that the child receptor uses the river more intensively over the two-week period. Therefore, this receptor is assumed to have a 50 percent higher sediment ingestion rate and a 25 percent higher soil adherence factor than Receptor Scenario i because of more time spent in the water per day. Receptor Scenario 2 was selected to reflect the differences in activity and exposure patterns associated with a short, but intense, use of the river.

Parameters describing the physiological and behavioral characteristics of the child receptor for each of the two receptor scenarios are given in Table 6 (scenario 1) and Table 7 (Scenario 2). Maximum plausible receptor characteristics such as ingestion rate and body surface area were employed and are expected to overestimate actual exposure in most cases.

TABLE 6

Receptor Characteristics - Scenario 1

TABLE 7

Receptor Characteristies-Scenario 2

The exposure rates in Tables 6 and 7 were selected to represent a reasonable "worst case" exposure scenario. The number of swimming events for Scenario 1 (60 days) was an assumed value selected to represent the maximum number of days that a child is expected to swim in the river during the summer. For Scenario 2, the number of days was selected to represent a tow-week period for a family that uses the river while on summer vacation.

3.2.2 Cancer Endpoints

To estimate the risk of cancer, arsenic exposures were estimated based on a lifetime of exposure over four discrete life stages (i.e., preschool child, child, adolescent and adult). The physiological and behavioral characteristics of each life stage used for the exposure modelling are given in Tables 8 and 9.

TABLE 8

Receptor Characteristics (Cancer) - Scenario 1

TABLE 9

Receptor Characteristics (Cancer) - Scenario 2

3.3 Toxicity Assessment

The toxicological criteria and bioavailability factors used in this assessment were adopted from the PQRA (Golder 2000), and were derived from the environmental health risk assessment conducted for the village of deloro (CanTox 199). Table 10 is a summary of the toxicological criteria and the bioavailability factors for arsenic. Regulatory agencies such as Health Canada and the US EPA have developed these values. A description of the toxicological properties of arsenic is given elsewhere (CanTox 1999 and Golder, 2000).

TABLE 10

Toxicological Criteria and Sediment Bioavailability Factors for Arsenic

Endpoint	Route	Criterion/Value	Effect	Study	Agency
Acute effects	Oral	PMTDI 2 µg/kg/day	acute arsenic intoxication such as abdominal pain vomiting, muscular pain and weakness	Joint FAO/WHO Expert Committe on food Additives	JECFA 1989
Non-cancer	Oral	RfD 0.3 µg/kg/day	hyperpigmentation, keratosis, possible vascular complications (human)	Tseng et al., 1968 Tseng, 1977	US EPA 1998
Cancer	Oral	q1* 0.0015 (µg/kg/day)-1	skin cancer, basal and squamous cell carcinoma (human)	Tseng et al., 1968 Tseng, 1977	US EPA 1998
Bioavailability	Dermal	0.8 - 1.9 percent	NA	Webster et al., 1993	NA
Bioavailability	Oral (sediment)	14 percent (soil)	NA	Freeman et al., 1995	NA

NA: not applicable

Information adopted from CanTox (1999) and Golder (2000).

3.4 Risk Characterization

Risk characterization is the final step in the Updated PQRA and involves comparing the estimated exposures to exposure values that are safe. Comparison of the estimated exposures to arsenic to those typically encountered by the general population is also an important aspect of risk characterization, since it provides a frame of reference to evaluate local conditions. Finally, uncertainties in the assessment, and their impact on its conclusions, are indentified and discussed.

3.4.1 Acute effects

Acute exposure to elevated levels of arsenic is associated with various effects including abdominal pain, vomiting, muscular pain and weakness. To characterize the potential risks associated with acute arsenic exposure it is necessary to compare the estimated daily dose (i.e., exposure) to a health-protective toxicity reference value. For this assessment, the Provisional Maximum Tolerable Daily intake Value (PMTDI) of 2 µg/kg/day, which was originally derived by the FAO/WHO Joint Expert Committee on Food Additives, was selected as the appropriate toxicity reference value (JEFCA, 1989). This is the same value that was adopted as an acute toxicity reference value for the environmental health risk assessment conducted for the village of Deloro (Cantox, 1999). To determine whether there is a potential for acute effects, an Exposure Ratio (ER) is calculated by dividing the estimated exposure by the PMTDI, as summarized below.

Exposure Ratio = Estimated Exposure
PMTDI

ER values less than one indicate that no acute adverse effects are expected. Conversely, ER values greater than one may indicate a health risk.

The estimated acute exposure to arsenic present in sediments from Study Area a and the calculated ERs for each of the two receptor scenarios is given in Table 11 (calculated based on the average arsenic concentration) and Table 12 (calculated based on the maximum detected arsenic concentration).

TABLE 11

Estimated arsenic Exposure Levels and Exposure Ratios for Acute Effects (Average)

The calculated ER values for acute effects are approximately 0.2 for both receptor scenarios, which are well below the level at which health effects may be observed (i.e., ER value equal to 1).

The ER values for Receptor Scenarios 1 and 2 using the maximum observed arsenic concentration of 238 mg/kg are 0.3 and 0.4 respectively (Table 12). These ER values are less than half the value at which there is a potential risk of acute health effects.

TABLE 12

Estimated Arsenic Exposure Levels and Exposure Ratios Acute Effects (Maximum)

The ER values for Receptor Scenarios 1 and 2 in Tables 11 and 12 do not consider exposure to arsenic from other pathways. The Typical Ontario Resident (TOR)is exposed to arsenic via food and the sources. Therefore, any arsenic exposures received by residents in Study Area A because of contact with sediments would be incremental (in addition) to exposures received as a TOR.

Exposure of the residents of Study Area A via other pathways may be evaluated based on the exposures received by the TOR. The ER values calculated for the TOR are representative of the health risks typically encountered by the average Ontario resident and were calculated using standard risk assessment methodology. As outlined in the environmental health risk assessment for the Village of Deloro, the estimated daily Exposure to arsenic from all sources is approximately 0.6 µg/kg/day for the 50^th percentile TOR and approximately 1.4 µg/kg/day for the 95^th percentile TOR (CanTox, 1999). The estimated total exposure from sediments and other exposure pathways for the 50^th and 95^th percentile TOR are discussed below.

Total Exposures Using The 50^th Percentile TOR:

Residents or users of the river in Study Area A would have a total estimated daily exposure to arsenic of 1 and 1.1 µg/kg/day for Receptor Scenarios 1 and 2, respectively. This total exposure estimate is based upon the average arsenic sediment concentration in Study Area A and the 50^th percentile TOR exposure value. The resulting acute health risk ER value for both scenatrios is approximately 0.5. The ER value assuming the maximum arsenic sediment concentration of 1.23 µg/kg/day is 0.6 for receptor Scenario 1 and 0.7 for Receptor Scenario 2. All ER values for the 50^th percentile TOR plus exposure to sediments in Study Area A are less than 1. Thus, the incremental exposure to arsenic in sediments from Study Area a does not represent an acute health risk.

Total Exposures Using the 95^th Percentile TOR:

The total estimated daily arsenic intake is approximately 1.7 and 1.9 µg/kg/day for Receptor Scenarios 1 and 2, respectively if the 95^th percentile TOR exposure value and average arsenic sediment concentration are assumed. The corresponding ER values for both scenarios are 0.9 if the maximum arsenic sediment concentration is used, than the cumulative estimated exposure for Receptor Scenario 1 is 2 µg/kg/day and 2.3 µg/kg/day for receptor scenario 2. The corresponding ER values for both scenarios are 1. The cumulative exposure values calculated for the 95^th percentile TOR, which may be considered a "worst case" scenario, are approximately 1. In light of the conservative nature of the assessment, residents and users of the Moira River in Study Area A are unlikely to experience acute health effects due to exposure to arsenic.

In summary, the above findings indicate that the concentrations of arsenic detected in sediment in Study Area A are not expected to represent an acute health risk to area residents and other users of the river, even under worst casa exposure conditions and considering exposure to arsenic from other sources. However the margin of safety between the predicted exposure values and the acute toxicity reference value may be small.

3.4.2. Chronic Non-Cancer Endpoints

Non cancer effects do not occur after chronic or sub-chronic exposure unless the dose is above a certain threshold. Risk characterization for these chemicals involves comparing the total estimated daily dose (i.e., exposure) to the Reference Dose (RfD). The RfD defines the amount of chemical that humans, including sensitive sub-populations, may be exposed to on a daily basis for an entire lifetime without an adverse effect. To determine the potential for chronic and sub-chronic non-cancer health effects, the ER was calculated by dividing the estimated exposure by the RfD, as summarized below.

Exposure Ratio = Estimated Exposure
RfD

As described above, ER values less than one indicate that no adverse effects are expected whereas ER values greater than one may indicate a health risk. However, because of the conservative approach used in this assessment, ER values greater than one may not represent a significant risk. In these instances, additional interpretation of the results is needed.

The estimated exposure to arsenic present in sediments from Study Area and the calculated ERs based on these exposure estimates for each of the two receptor scenarios is given in Table 13 (calculated based on average arsenic concentrations) and Table 14 (calculated based on maximum detected arsenic concentrations).

TABLE 13

Arsenic Exposure Levels and Exposure Ratios for Chronic Non-Cancer Health Effects
(Average)

As shown in Table 13, the calculated ER values based on average arsenic sediments concentrations are significantly less than one for both exposure scenarios. This result indicates that adverse chronic non-cancer health effects in residents and other users of the river system exposed to arsenic in sediment in Study Area A are not expected.

ER values calculated based on the maximum concentrations of arsenic detected in sediment are 0.3 and 0.1 for Receptor Scenarios 1 and 2, respectively (see Table 14). These ER values are both considerably less than 1 and suggests that adverse chronic non-cancer health effects in individuals exposed to the sediments are unlikely.

TABLE 14

Arsenic Exposure Levels and Exposure Ratios for Chronic Non-Cancer Health Effects
(Maximum)

As discussed previously, the ER values for Receptor Scenarios 1 and 2 do not consider exposure to arsenic from other pathways. To determine the significance of the above findings when other sources of exposure to arsenic are considered, the estimated ER values were compared to ER values for arsenic calculated for the 50^th percentile TOR and 95^th percentile TOR as described elsewhere (Cantox, 1999).

Total Exposures Using the 50^th Percentile TOR:

The Total estimated arsenic exposures for a 50^thpercentile TOR using the average arsenic sediment concentration in Study Area A are 0.7 and 0.6 µg/kg/day for Receptor Scenario 1 and 2, respectively. The corresponding ER values are 2 for Receptor Scenarios 1 and 2. These values are not significantly different than the ER value of 2 calculated for the 50^th percentile TOR. Therefore, exposure to sediments in Study Area A does not represent a significant exposure pathway, relative to other sources of exposure experienced by the TOR. Residents of Study Area A who use the river for swimming or bathing would not receive a significant incremental exposure to arsenic compared to the TOR.

Total Exposures Using the 95^th percentile TOR:

Residents of Study Area A who use the river for swimming or bathing would not be exposed to more arsenic than the 95^th percentile TOR. The total estimated cumulative arsenic exposures using the 95^th percentile TOR and the average arsenic concentration are 1.5 µg/kg/day and 1.4 µg/kg/day for Receptor Scenario 1 and 2, respectively. The corresponding exposure values for Receptor Scenarios 1 and 2 are 5. Since the ER value for the 95^th percentile TOR is 5, this result indicates that exposure to maximum arsenic concentrations in sediments from Study Area A does not represent a significant exposure pathway relative to other sources of exposure experienced by the TOR.

The ER values for the TOR are greater than one because arsenic occurs naturally in air, water, soil and food. For most people, the diet is the largest source of exposure, with concentrations in foods usually ranging from 20 µg/kg to 140 µg/kg. Although concentrations may be substantially higher in certain seafoods, much of this is in a nontoxic from.

Use of the RfD to predict the potencial for adverse non-cancer health effects is a conservative approach. The RfD adopted for this study was derived to protect individuals from developing skin lesions (e.g., hyperpigmentation and keratosis) following continuous long-term (i.e., lifetime) exposures to arsenic. The induction of these skin lesions was apparent only in humans continuously exposed to much higher levels of arsenic than has been estimated in this study, rather than the short intermittent duration of exposure considered in this assesment (60 days/year for Scenario 1 and 14 days/year for Scenario 2). Also, skin lesions associated with arsenic exposures have not been observed in children less than 11 years of age, even when exposed to extremely high concentrations of arsenic in drinking water (US EPA, 1998), which is the highest risk age group identified in this assessment.

In summary, exposure of residents to river sediments in Study Area A results in ER values that are only marginally greater than those received by the 50^th percentile TOR and are less than those received by the 95^th percentile TOR. Therefore, exposure to arsenic in sediments from Study Area A is not expected to result in a significant incremental risk of non-cancer health effects from chronic exposure.

3.4.3 Cancer

In conducting risk assessments, carcinogens are usually assumed to have a non-threshold type dose response. That is, it is assumed that there is some risk associated with any exposure, no matter how low. Risk characterization of carcinogens involves calculating a Cancer Risk Level (CRL). The CRL is calculated by multiplying the estimated Lifetime Average Daily Dose (LADD) of a chemical by the cancer potency factor (i.e., slope factor, or q^*) of that chemical, as outlined below.

CRL=LADDxq^*

The CRL defines the predicted excess risk that an individual has of developing cancer in their lifetime as a result of a particular exposure to a substance.

For carcinogens, the toxicological endpoint of interest is the incremental lifetime risk of development cancer. Since a lifetime is commonly defined as 70 years, the LADD is calculated by amortizing the total lifetime dose over 70 years, even if the exposure is less than a lifetime. The CRL is calculated by multiplying the LADD by the chemical-specific cancer potency factor (q*). The q* is used to define the ability of a chemical to cause cancer in humans. Comparison of the calculated CRL values to cancer risk levels that are de Minimus (i.e., negligible) is done to determine if the calculated cancer risks are acceptable. Regulatory agencies generally consider CRLs ranging from 1 x 10^-6 to 1 x 10^-4 as de minimus risk levels.

The estimated exposure to arsenic present in sediments from Study Area a and the calculated CRL based on these exposure estimates for each of the two child receptor scenarios and assuming an average arsenic sediment concentration is given in Table 15.

TABLE 15

Estimated Arsenic Exposure Levels and Associated Cancer Risk Levels (Average)

The lifetime cancer risk for Receptor Scenario 1 is 1 x 10^-5 and 4 x 10 ^-6 for Receptor Scenario 2. These risk estimates are within the range that regulatory authorities generally consider negligible (i.e., 1 x 10^-4 to 1 x 10^-6).

The corresponding lifetime cancer risk estimates based on the maximum detected arsenic concentration in sediments are given in Table 16.

TABLE 16

Estimated Arsenic Exposure Levels and Associated Cancer Risk Levels (Maximum)

The lifetime cancer risks for Receptor Scenario 1 is 2 x 10^-5 and for Receptor Scenario 2 the risk is 8 x 10^-6. These risk estimates are approximately twice those estimated based on the average arsenic concentration, but are well within the range that regulatory authorities consider negligible(i.e., 1 x 10^-4 to 1 x 10^-6).

These risk estimates were compared to the CRLs derived for the 50^thpercentile TOR (CRL = 5.4 x 10^-4) and 95^th percentile TOR (CRL - 8.2 x 10^-4), as discussed elsewhere (Can Tox, 1999).

Total Exposures Using the 50^th Percentile TOR:

The Increase in the CRL for the 50^th percentile TOR because of exposure to arsenic in sediments from Study Area A is very small and does not result in risks exceeding negligible levels. The CRL for the 50^th percentile TOR is 5.4 x 10^-4. The estimated CRLs based on the average sediment concentration are 1 x 10^-5 and 4 x 10^-6 for Receptor Scenarios 1 and 2, respectively. When these risks are added to the CRL calculated for the 50^th percentile TOR, the resulting CRL values are not significantly different than the result for the 50^th percentile TOR. Based on maximum sediment concentrations, the estimated CRL values are 2 x 10^-5 and 7 x 10^-6 for Receptor Scenarios 1 and 2, respectively. The corresponding cumulative CRLs are only marginally greater (approximately four percent) than the CRL for the 50^th percentile TOR.

Total Exposures Using the 95^th Percentile TOR:

Exposure to arsenic in sediments from Study Area A is expected to increase the CRL for the 95^th percentile TOR only marginally. The CRL for the 95^thpercentile TOR and for the cumulative CRL based on the average sediment concentration are 8 x 10^-4. The cumulative CRLs based on the maximum arsenic sediment concentration are 8 x 10^-4 for Receptor Scenarios 1 and 2. The corresponding cumulative CRLs are similar to the CRL for the 95^th percentile TOR.

In summary, residents for the Moira River system exposed to sediments in Study Area A are predicted to have CRL values only slightly greater than those received by the 50^th or 95^th percentile TOR. Therefore, exposure to arsenic in sediments from Study Area A does not significantly increase cancer risks due to arsenic exposure.

3.5 Uncertainty Analysis

In a risk assessment, the does and benchmarks are estimated from the available information about how exposure might take place and the doses that cause effects. There is always uncertainty associated with these estimations, depending on the quality, quantity and variability associated with the available information. When information is uncertain, it is standard practice in a risk assessment to make assumptions that are biased towards safety, so as to be protective of human health. Every effort is made to ensure that assumptions are specific to the site being evaluated.

There are several layers of safety applied in this Updated PQRA. For instance, the exact amount of time a child spends outside in contact with sediments in Study Area a is uncertain because it varies with the weather, other activities or the child's family, and the child's own preferences. The risk assessment assumed, as a worst case, that a child would spend time in contact with sediments in Study Area A every day for 60 days (Scenario 1) or every day with even more direct contact with the skin for 14 days (Scenario 2). These are conservative scenarios since children would be spending time in other activities as well. Therefore, scenarios 1 and 2 add layers of safety. In this way, if the risk assessment indicates that estimated doses are less than benchmark doses for this "maximally exposed" child, then we can conclude that all children will be protected.

As another example, the quantity of arsenic in sediment in specific areas of Study Area A may be uncertain because the information is based on a limited number of samples at one time period and/or there is considerable variability in concentrations of arsenic between samples in the area. In this case, a layer of safety is added by calculating a dose based on the maximum observed arsenic concentration, even if this was only in one very small area. If the dose based on the maximum arsenic concentration is less than the benchmark dose, then we can be confident that there is no problem across the whole Study Area A.

Use of maximum sediment arsenic concentrations in combination with using the 95^th percentile TOR to estimate total exposures adds another layer of safety. It is highly unlikely that residents or users of the river in Study Area A will be exposed to the maximum arsenic sediment concentration and also be exposed to the 95^th percentile arsenic does as a TOR.

There is also uncertainty associated with estimating benchmark doses. Benchmark doses are based on toxicity information available from government databases and published scientific literature. The majority of toxicity information comes from the results of experiments with laboratory animals. Some additional information on human health effects is also available for some substances where cases of workplace exposures and associated health effects have been documented. There is some uncertainty in extrapolating from animal studies and workplace case studies to the possible effects that may result from exposure to arsenic in sediments in Study Area A. It is standard practice in a risk assessment to set the benchmark dose at a much lower level than the animal benchmark (typically 100 to 1000 times lower). This is done to ensure that minor exceedances of these benchmark doses will not cause adverse health effects.

This assessment also assumed that individuals would be exposed to inorganic arsenic, which is more toxic than organic forms. Inorganic arsenic in groundwater and surface water can be reduced (methylated) by micro-organisms, albeit to a limited extent. Therefore, it is likely that a portion of the arsenic measured in sediment and water samples along the river system is present in the less toxic organic from. To the extent that some arsenic is present in organic forms, this approach may lead to overestimates of the toxic potency of the arsenic present in the Moira River sediments, and risk estimates may therefore be overstated. The extent of this overestimation cannot be determined based upon the available data.

Because of all of the layers of safety incorporated into the calculation of an ER, values that are at or marginally greater than one do not indicate that adverse effects are occurring or are likely to occur. As explained above, minor exceedances of the benchmark (resulting in ER values slightly above one) are within the layers of safety incorporated into the risk assessment.

4.0 CONCLUSIONS AND RECOMMENDATIONS

This Updated PQRA identifies the specific risks of exposure to arsenic in sediments in Study Area A. It was performed in order to identify whether specific risk management measures were required with respect to exposure to sediments in Study Area A.

No findings in this report should be taken to indicate that adverse health impacts are occurring among river system users.

4.1 Conclusions

Acute Effects:

• Exposure to the range of concentrations of arsenic detected in sediment in Study Area A will not result in adverse health effects after short-term exposure of area residents and other users of the river.
• Exposure to sediments in Study Area A combined with the exposure to arsenic as experienced by the TOR will not result in adverse health effects after short-term exposure of area residents and other users of the river.

Chronic Non-Cancer Effects:

• Exposure to the range of arsenic concentrations in sediments from Study Area A will not result in adverse health effects after long-term exposure of area residents and other users of the river.
• Exposure to sediments in Study Area A combined with the exposure to arsenic as experienced by the TOR will not result in adverse health effects after long-term exposure of area residents and other users of the river.

Cancer Effects:

• The risk of developing cancer due to exposure to the range of arsenic concentrations in sediments from Study Area A is within the range considered negligible or acceptable by regulatory authorities.
• Exposure to sediments in Study Area A combined with the exposure to arsenic as experienced by the TOR will not increase cancer risk significantly; the incremental risk is extremely small.

Based on the results of this investigation, it is concluded that further assessment of the potential health risks associated with arsenic in sediments in Study Area A is not warranted.

4.2 Recommendation

• Based on the results of this investigation, further assessment of the existing arsenic concentrations in sediment is not warranted. Should conditions change due to a major flood event, or in the event of a significant accidental release of sediments from Young's Creek, follow up sediment sampling would be recommended to confirm that exposure to sediments in study area A continues to represent no significant incremental health risk. Following completion of the final cleanup program for the Deloro Mine Site, including Young's Creek, would also be an appropriate time for follow up sediment sampling.

The concentration of arsenic in sediment in the follow-up sampling event(s)should be compared with the arsenic concentrations presented in the Updated PQRA and the need for further assessment determined at that time. This will help to ensure that exposure to sediments in Study Area A continues to represent no significant incremental health risk to users of the Moira River.

5.0 REFERENCES

Cantox. 1999. Deloro Village exposure assessment and health risk characterization for arsenic and other metals. Prepared by Cantox Environmental Inc. December, 1999.

Environment Canada. 2000. Annual Meteorological Summary. October 2000.

Golder Associates. 2000. Phase II Moira River Study Impacts of the Former Deloro Mine Site on the Moira River System.

JECFA. 1989. Joint FAO/WHO Expert Committee on Food Additives. Toxicological evaluation of certain food additives and contaminants. Cambridge University Press, 1989; 155-162.

Tseng, W.P. 1977. Effects and dose-response relationships of skin cancer and blackfoot disease with arsenic. Environ. Health Prospect. 19;109-119.

Tseng, W.P., Ghu, H.M., How, S.W., Fong J.M., Lin, C.S., and Yeh, S. 1968. Prevalence of skin cancer in an endemic area of chronic arsenicism in Taiwan. J. Nat. Cancer Instit. 40(3):453-463.

US EPA. 1998. Integrated Risk Information System (IRIS). United States Environmental Protection Agency.