Importance of Considering theFramework Principles in RiskAssessment for Metals1

C H A R L E S A . M E N Z I E *

Exponent, Alexandria, Virginia

L I N D A M . Z I C C A R D IY V E T T E W . L O W N E Y

Exponent, Boulder, Colorado

A N N E F A I R B R O T H E RS C O T T S . S H O C KJ O Y C E S . T S U J I

Exponent, Bellevue, Washington

D I E M H A M A ID E B O R A H P R O C T O R

Exponent, Irvine, California

E L I Z A B E T H H E N R YS T E A V E H . S U

Exponent, New York, New York

M I C H A E L W . K I E R S K IM A R G A R E T E . M C A R D L E

Exponent, Maynard, Massachusetts

L I S A J . Y O S T

Exponent, St. Paul, Minnesota

The recent EPA Framework for Metals Risk Assessmentprovides the opportunity for contextual risk assessment forsites impacted by metals (such as the depicted Dauntless Minein Colorado).In 2007, three documents were published that emphasizedimportant considerations when evaluating exposure to andpotential health and environmental effects of metals in theenvironment. The Metals Environmental Risk AssessmentGuidance (MERAG) project was launched by the InternationalCouncil of Mining and Metals and Eurometraux, withregulatory endorsement provided by the UK Department forEnvironmental Food and Rural Affairs (DEFRA). The guidanceand associated fact sheets were developed to providescientific and regulatory guidance on the most advancedscientific concepts for metals (1). The guidance document

highlighted the importance of considering background,essentiality, speciation, mobility, and bioavailability whenconsidering the potential adverse health effects of metals. Italso addressed issues related to bioaccumulation and bio-magnification. A related document focused on how theseissues should be considered specifically for human healthrisk characterization (2).

Also in 2007, the U.S. Environmental Protection Agency(EPA) published the Framework for Metals Risk Assessment(3). The Framework is based on a set of principles that areconsistent with the scientific considerations set forth in theMERAG guidance and are special attributes of metals. Thefive Framework principles are

1. Metals are naturally occurring constituents in theenvironment and vary in concentrations across geo-graphic regions.

2. All environmental media have naturally occurringmixtures of metals, and metals often are introducedinto the environment as mixtures.

3. Some metals are essential for maintaining proper healthof humans, animals, plants, and microorganisms.

4. The environmental chemistry of metals strongly influ-ences their fate and effects on human and ecologicalreceptors.

5. The toxicokinetics and toxicodynamics of metals de-pend on the metal, the form of the metal or metalcompound, and the organism’s ability to regulate and/or store the metal.

These principles have been increasingly recognized asimportant for addressing exposure to and risk from metalsin the environment. The publications cited above provide abroad technical and regulatory foundation for incorporatingthe principles into metals risk assessments. Our experiencewith metals risk assessments indicates that some principlesare especially important for some metals and less importantfor others, and thus, case-specific information is important.This paper provides our view of the relative importance ofthe Framework principles for exposure and risk assessmentof particular metals and metalloids. The ratings of relevant

1 Editor’s Note: To our delight at ES&T, we have started to receiveFeatures and Viewpoints by independent author(s) coincidentallyoverlapping both in topic and review schedule. Within days of thispaper’s acceptance another specifically concerning selenium in theenvironment was accepted. The choice was thus made to presentboth manuscripts in the same issue (November 15, 2009; 43[22]).Readers of this piece by Menzie et al. are therefore encouraged toread that by Luoma and Presser (DOI 10.1021/es900828h).

LIN

DAZI

CCAR

DI

Environ. Sci. Technol. 2009, 43, 8478–8482

8478 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 22, 2009 10.1021/es9006405 2009 American Chemical SocietyPublished on Web 08/31/2009

importance presented in Table 1 reflect the authors’ experi-ence with both human health and ecological risk assessmentsfor metals associated with mining and manufacturingoperations, products, and hazardous waste sites, particularlyin North America, but also in other countries. We havedeveloped a thorough understanding of the many factorsthat influence metal species (e.g., valence states), metal forms(e.g., inorganic and organic), and metal states (phases) (e.g.,in solution, or bound to dissolved organic matter orparticulates), and the resulting bioavailability of such chemi-cal species. Table 1 incorporates our knowledge that certainforms of metals are more toxic than others, and that regulatorytoxicity values are developed for forms of metals that mayor may not represent the forms present in the environment.

Five Framework Principles. The relative importance ofthe individual principles for metals most commonly en-countered in site-specific risk assessments is summarized inTable 1. Although the Framework focused on inorganic metalsand metalloids such as Sb and As that share properties ofboth metals and nonmetals, we also consider application ofthe principles to organometallics such as methyl mercury,organoarsenic, and organoselenium where relevant, becausethese chemicals can be important in a risk assessmentcontext. The information summarized for the metals andmetalloids (hereafter collectively referred to as “metals”) listedin Table 1 reflects the cumulative knowledge and riskassessment experience of the authors. We considered the 23“metals and metalloids of interest” as identified in theFramework, and then selected the chemical species includedin Table 1 based on our collective expertise and metals thatare known “risk drivers”. In what follows, our intent as metalsassessment practitioners was not to provide a comprehensive

review on how each principle influences each metal, butrather, to focus on metals of greatest interest to the riskassessment community and for which our experience hasshown that one or more principles are especially important.

Principle 1. It is well established (e.g., 4-6) that con-centrations of metals in native rocks and soils vary consider-ably across geographic areas. Table 1 indicates that naturaloccurrence becomes an increasingly important considerationwhen risk-based limits approach or fall below the naturalbackground. This can occur because exposure assumptionsused to derive the risk-based values, and the actual factorsthat determine exposure, are inconsistent (e.g., see discussionof bioavailability under Principle 4), or because the risk-based concentrations are based on estimates of theoreticalrisk extrapolated from toxicity observed at high doses, whichmay not apply at lower exposure levels. Further, it is importantto consider regional variations and that plant and animalpopulations are shaped, at least in part, by local conditions,including soil and sediment mineralogy (7).

We identified the consideration of natural occurrence ashighly important for human health and/or ecological riskassessments for Al, As, Ba, Be, Cr(III), Fe, Pb, and Se. Al andAs are discussed further, as examples.

EPA’s soil screening guidance (8) recognized that Al, thethird-most abundant element in the Earth’s crust, is oftenidentified as a chemical of concern in ecological riskassessments because of the conservative nature of thescreening benchmarks and toxicity reference values (TRVs)that are based on toxicity tests using Al in solution. Relyingon TRVs derived from toxicity studies using soluble formscan result in hazard quotients (HQs) from food-web modelingthat indicate risks (i.e., HQs >1) for Al concentrations that

TABLE 1. Relative Importance of Metals Principles in Human Health and Ecological Risk Assessmenta

a Notes: L, low, never or rarely a consideration; M, moderate, sometimes a consideration; H, high, often a consideration;RfD, reference dose; hh, human health; eco, ecological.

VOL. 43, NO. 22, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 8479

are below natural background. While Al toxicity is associatedwith soluble Al, soil and sediment analytical chemistrymethods measure total Al, including natural Al, most of whichis bound in insoluble mineral complexes such as alumino-silicates which are not bioavailable (8). Based on theseconsiderations, EPA (8)now recommends that Al be identifiedas a chemical of concern only for sites with soil pH < 5.5,because as pH decreases below this level, Al becomesincreasingly soluble, bioavailable, and toxic.

An example of a health-based screening level that fallsbelow natural background is provided by As. EPA screeninglevels for As in soil, based on cancer risk, are 0.39 and 1.6mg/kg (note that all soil concentrations are in dry weight)for residential and industrial exposure, respectively. Naturalbackground As concentrations in U.S. soils range from 1 to>20 mg/kg, and even higher in naturally mineralized areas(4). Setting health-based criteria below the natural back-ground is a dilemma for As and has resulted in target risksat the upper end of the risk management range, and in theuse of state, regional, or local background concentrations inlieu of more stringent health-based criteria. Exposures viaenvironmental media may also be small relative to normalbackground exposures, because of the ubiquitous presenceof inorganic As in the diet and in drinking water. As a result,exposure to inorganic As from incidental soil ingestion isgenerally lower than intake from background dietary sourcesuntil soil concentrations exceed 100 mg/kg (9). Thus, althoughcancer risk assessments may predict elevated excess risksassociated with As in soil, comparison to background dietaryexposures helps place such risks in perspective with ordinaryexposures.

Principle 2. The risks associated with mixtures of metalsin the environment depend on the degree to which one metalinfluences either exposure to or the effects of another. Metalsin mixtures may interact to produce antagonistic, additive,or synergistic effects. Table 1 indicates that mixture-relatedinfluences on exposure and health effects are moderately tohighly important for many of the metals. Additional mixture-related influences important to metals risk assessments willlikely be identified in the future. As indicated in Table 1,consideration of mixtures is of high importance for Cd, Cu,Fe, Pb, Mn, Mo, Ni, Se, Ag, and Zn.

Mixture effects have been shown or considered to beimportant in many human health and ecological riskassessments. For example, populations exposed to elevatedlevels of As and Cd in China show interactive effects thatincrease renal toxicity (10). On the other hand, at low doses,exposure and toxicity for metal mixtures may be reduced byinteractions in which these constituents form less solublemineral complexes in the gastrointestinal tract, or competefor metal carrier or binding proteins in the body (e.g.,metallothionein). Many interactions among metals at lowerdoses are thus antagonistic (11-14), with examples includingZn/Cu, Zn/Cd, Fe/Cd, Fe/Co, Mn/Fe, Mn/Al, Mn/Cd, As/Se,and Hg/Sesmixtures of these metals are less toxic than thesum of each of the component metals, individually. For thesemetals, more accurate estimates of risk are achieved if mixtureeffects are taken into account.

A common example of accounting for mixture effectsamong metals is the additivity of effects from cationic metalsduring acute exposures of aquatic organisms. In the case ofuptake from solution, the relative potency of cations isdetermined by differential binding efficiencies at the gillsurface (15).

Another example of interactions in metals mixtures isthat between Se and As, which have been reported to haveadditive or even synergistic effects in vivo and in vitro (14),as well as some reported antagonistic effects (16). Arsenitemay potentiate or attenuate the protective effect of Se againstcarcinogenicity, depending on the methylation state of the

Se compound (14). Se is also reported to increase biliary andurinary excretion of As in animal studies, reduce cellulartoxicity of As in biological systems, and protect againstchromosomal damage by As in cultured human lymphocytes(summarized in (17)). An intervention trial in an As-exposedpopulation in Inner Mongolia reported that 14 months ofsupplementation with Se reduced hair and blood levels ofAs and greatly improved various arsenical skin lesions (16).Se’s action is likely related to its role in antioxidant enzymes(17, 18).

Principle 3. For risk assessments, the essentiality of ametal becomes an important consideration when risk-basedlimits fall below concentrations that are essential formaintaining proper health, and also when natural levels ofmetals fall below those needed for maintaining proper health(deficiency). Finally, essentiality is important when thresholddoses for nutrition and toxicity are close to one another. Thedose-response curve for essential metals should include lowlevels at which nutritional deficits occur, as well as the higherlevels at which toxic effects occur (resulting in the classic“U”-shaped curve for essential metals; Figure 1).

We caution that, because the nutritional and toxic levelsof essential metals are specific to organism, sex, and life stage,care should be taken when evaluating dose-response curvesof essential metals and when extrapolating from one organ-ism, sex, or life stage to others. Table 1 reflects which essentialmetals are most likely to cause problems in metals riskassessments due to close threshold doses or other consid-erations of excess exposure vs deficiency.

The importance of considering essentiality when evaluat-ing certain metals has led to proposed new terminology todescribe their dose-response features and avoid confusionwith the concept of “no-effect levels” used for toxicity athigh doses (2). These new terms include “sufficient dietaryintake” and “deficiency effect levels.” Because risk assess-ments typically assess exposures in addition to dietary intake,another boundary term describing effect levels for riskassessment of essential metals is the range of “supplementa-tion levels,” such as those developed by Health Canada (19)for multivitamin/mineral supplements. Similar terms foressential nutrients, including trace metals, in the U.S. includedietary reference daily intakes (RDIs). The parallel develop-ment of dose-response relationships for nutrition andtoxicity is emerging as a key feature of assessing risksassociated with essential metals.

Se is an example of a potentially toxic element that is alsoan essential nutrient in the diet of many wildlife species. Insome parts of the world where Se levels are low in soils, Sedeficiencies can occur. Conversely, irrigation of soils withhigher Se concentrations can mobilize Se due to its relativelyhigh solubility, and produce toxic concentrations in low-lying areas. One of the key challenges to evaluating the risksassociated with Se is that there is a fairly narrow rangebetween therapeutic and toxic intake.

FIGURE 1. A simple representation of a “U-shaped” dose-response curve for an essential metal (EAR ) estimatedaverage requirement; RDA ) recommended dietary allowance;UL ) tolerable upper intake level).

8480 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 22, 2009

Principle 4. Risks of metals and other chemicals dependon exposure concentrations at the site of toxic action, andexposure can be strongly influenced by environmentalchemistry. The chemical form of a metal often dictates itsenvironmental fate and transport and its potential for uptake,metabolism, or toxic response. Important and variableenvironmental factors, such as pH, dissolved oxygen, organiccarbon, the presence of sulfides, and cation-exchangecapacity, can influence the bioavailability of metals. Thus,relative bioavailability of a metal in the environmentcompared to the form in toxicity tests is an importantconsideration in assessing risk.

Site-specific bioavailability is important for evaluatingexposure to metals such as Pb and Ba. In 2007, EPA releasedtargeted guidance for assessing site-specific oral bioavailabilityof metals (20). The agency also released a validation of an invitro method for evaluating the relative bioavailability of leadfrom soils or soil-like materials. With these new methods andguidance, it is now possible to adjust risk assessments to addressPb bioavailability on a site-specific basis.

Ba has received much attention, because it is commonlypresent in elevated levels in soils, but primarily in forms withlow bioavailability. For example, when Ba was assumed tobe 100% bioavailable in dust at an Alaskan mining site, food-web modeling indicated a potential for adverse effects tosmall mammals (21). However, when site-specific geologyand mineral characteristics were reviewed, it was clear thatthis assumption was likely to be very conservative. An invitro study was conducted using simulated gastric fluids toevaluate site-specific bioaccessibility of Ba from a variety ofsoil and tundra materials (22). The relatively low bioacces-sibility values confirmed the expectation that the Ba exposurewas less significant than exposure to other metals of concern(23), presumably because the environmental form differedfrom the forms used to establish the benchmark values.Menzie et al. (24) confirmed that the ecological benchmarksfor Ba in soil derived from soluble Ba compounds did notreflect exposure to BaSO4, the relatively insoluble form of Bathat is most common in surface soils.

Principle 5. Many risk assessments for metals historicallyhave been based on simplified estimates of exposure andeffects that too often neglect toxicokinetics (the passage ofa toxic agent or its metabolites through an organism) andtoxicodynamics (the physiological mechanisms by whichtoxins are taken up [absorbed], distributed, metabolized, andexcreted [ADME]). However, these factors have been criticalfor some metals and, as new data and assessment methodsemerge, will likely increase in importance.

The route of exposure and form of metal are importantto consider when predicting toxicity, including carcinoge-nicity. In addition, the sensitivity to and risks from metalsvary with age, sex, pregnancy, nutritional status, genetics,and life stage. Toxicokinetics, particularly elimination rates,are important when addressing the potential for metalbioconcentration and bioaccumulation. While certain metalsand forms of metals are known to bioaccumulate and transferto other trophic levels via the food chain, inorganic forms ofmetals generally do not biomagnify. Many organisms canregulate metal uptake and/or accumulation, or store metalsas nonbioavailable granules (25). Our experience suggeststhat consideration of toxicokinetics and toxicodynamics isimportant for most metals risk assessments.

Cr is an example of a metal for which toxicokinetics areespecially important when assessing risks. Trivalent Cr(Cr[III]) and insoluble forms of hexavalent Cr (Cr[VI]) do notreadily pass through cellular membranes; therefore, Crabsorption via all pathways of exposure is dependent onmetal species and solubility. For example, Cr(VI) generallyis not stable in the acidic and reducing conditions of thestomach and is detoxified through reduction to Cr(III) prior

to absorption. Most forms of Cr(III) are not readily bioavail-able and are of relatively low toxicity. Only at exposures thatoverwhelm the reducing capacity of the stomach willsignificant systemic absorption occur, precipitating anincreased risk of adverse health effects.

Mn is another metal for which toxicokinetics and toxico-dynamics are particularly important to consider. Mn exposurevaries significantly by the route of exposure, chemical form,and the nutritional status and developmental stage of theindividual. Because it is an essential nutrient, dietary intake ofMn is critical to the maintenance of numerous enzymaticfunctions and developmental processes. The bioavailability ofdietary Mn is tightly controlled by homeostatic mechanismsthat maintain relatively steady levels of Mn in the body. Thesecontrols can be disrupted by micronutrient imbalances in theindividual, resulting in increased absorption of Mn. In contrastto ingested Mn, inhalation delivers a larger bioavailable fractionby facilitating transport to the systemic circulation for distribu-tion to the brain, the target site of Mn-induced neurotoxicity.Therefore, the route of exposure to Mn is especially importantin affecting toxicity.

Discussion and ConclusionsThere are challenges in conducting metals risk assessmentsgiven the complexity of factors that influence metals’chemistry, behavior in the environment, bioavailability, andpharmacokinetics. Emerging metals risk assessment guidanceincorporating concepts such as the Framework principlessignificantly expands on the methods and considerationsthat have traditionally been applied in the evaluation of risksfrom exposure to metals.

Regulatory agencies within the U.S. and internationallyincreasingly recognize that it is important to incorporatethese principles as the risk assessment process becomes moresophisticated. For example, EPA has issued general guidancethat indicates the need to consider such issues as essentiality(26), and very specific guidance regarding evaluation ofbioavailability of metals from soils (20). In another example,several states are now recognizing the importance ofconsidering the natural occurrence of metals. For instance,although health-based screening levels for As in soil are basedon strict toxicity considerations, many state agencies haverecognized that risk-based screening levels for As fall wellbelow natural background, and so have recommendedpreliminary remedial goals or screening levels targeted atconcentrations representative of state-wide background, andseveral states have implemented guidance that specificallyallows for comparison against background on a site-specificbasis (e.g., New Jersey (27), New York (28), Georgia (29), andFlorida (30), among others).

In another regulatory application of the Frameworkprinciples, EPA in 2007 revised the aquatic life Water QualityCriteria for Cu to consider site-specific speciation andbioavailability (31), and has a draft strategy for revising othercriteria that consider advancements in ecological risk as-sessment methodology (32). EPA has also developed anapproach for estimating the aquatic toxicity of divalent metalcations that considers the bioavailable metal fraction thatcan be measured in pore water and/or predicted based onthe relative sediment concentrations of acid volatile sulfide(AVS), simultaneously extracted metals (SEM), and the totalorganic carbon (TOC) content of the sediment (33).

This paper provides the authors’ view on the relativeimportance of the Framework principles, based on ourcollective experience as risk assessors and scientists. Otherrisk assessors will undoubtedly have additional insights thatcan contribute to understanding where and how the prin-ciples can be effectively applied. Our article is intended tostimulate further consideration and application of the

VOL. 43, NO. 22, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 8481

principles within regulatory and research arenas. Recognizingwhere such information has the greatest utility is a startingpoint that will help focus future work.

The authors are from Exponent’s Environmental Sciences and HealthSciences practice groups, which focus on understanding the humanhealth and environmental effects of chemical exposure. The authors’collective experience in the fate, transport, toxicology, and riskassessment of metals and other chemicals in the environment, products,and food provides broad insight on the potential for and consequencesof chemical exposure. Dr. Charles A. Menzie is a Principal Scientistand Director of Exponent’s Ecological Sciences practice. His primaryarea of expertise is the environmental fate and effects of physical,biological, and chemical stressors on terrestrial and aquatic systems.Dr. Menzie is recognized as one of the leaders in the field of riskassessment and was awarded the Risk Practitioner Award by the Societyfor Risk Analysis. He chaired the external EPA working group thatassisted EPA in the development of the principles discussed in thispaper. Please address correspondence regarding this paper tocamenzie@exponent.com.

AcknowledgmentsWe acknowledge the extensive work put forth by theInternational Council of Mining and Metals and Eurometrauxin developing the Metals Environmental Risk AssessmentGuidance (MERAG), and the U.S. Environmental ProtectionAgency in the development of the Framework for MetalsRisk Assessment.

Supporting Information AvailableExpanded version of Table 1 that summarizes the relativeimportance of the metals principles in human health andecological risk assessment, which includes key consider-ations. This information is available free of charge via theInternet at http://pubs.acs.org.

Literature Cited(1) MERAG: Metals environmental risk assessment guidance; 978-

0-9553591-2-5; International Council of Mining and Met-als (ICMM), 2007; http://www.euras.be/assets/files/MERAG/MERAG.pdf.

(2) HERAG: Health risk assessment guidance for metals; 978-0-9553591-4-9; International Council of Mining and Metals(ICMM), 2007; http://www.ebrc.de/ebrc/downloads/HERAG.pdf.

(3) Framework for metals risk assessment; EPA 120/R-07/001; U.S.Environmental Protection Agency, Office of the Science Advisor,Risk Assessment Forum, 2007; http://www.epa.gov/raf/metalsframework/pdfs/metals-risk-assessment-final.pdf.

(4) Major- and Trace-Element Concentrations in Soils from TwoContinental-Scale Transects of the United States and Canada;Open-File Report 2005-1253; U.S. Geological Survey, 2005.

(5) Dragun, J.; Chiasson, A. Elements in North American Soils;Hazardous Materials Control Resources Institutes: Greenbelt,MD, 1991.

(6) Determination of Background Concentrations of Inorganics inSoils and Sediments at Hazardous Waste Sites; EPA/540/S-96/500; U.S. Environmental Protection Agency, Office of Emergencyand Remedial Response: Washington, DC, 1995.

(7) McLaughlin, M. J.; Smolders, E. Background zinc concentrationsin soil affect the zinc sensitivity of soil microbial processes - Arationale for a metalloregion approach to risk assessments.Environ. Toxicol. Chem. 2001, 2639–2643.

(8) Ecological soil screening level for aluminum. Interim Final;OSWER Directive 9285.7-60; U.S. Environmental ProtectionAgency, Office of Solid Waste and Emergency Response:Washington, DC, 2003.

(9) Tsuji, J. S.; Yost, L. J.; Barraj, L. M.; Scrafford, C. G.; Mink, P. J.Use of background inorganic arsenic exposures to provideperspective on risk assessment results. Regul. Toxicol. Phar-macol. 2007, 48, 59–68.

(10) Norberg, G. F.; Jin, T.; Hong, F.; Zhang, A.; Buchet, J. P.; Bernard,A. Biomarkers of cadmium and arsenic interactions. Toxicol.Appl. Pharmacol. 2005, 206, 191–197.

(11) Greger, J. L.; Zaikis, S. C.; Abernathy, R. P.; Bennett, O. A.;Huffman, J. Zinc, nitrogen, copper, iron, and manganese balancein adolescent females fed two levels of zinc. J. Nutr. 1978, 108(9), 1449–1456.

(12) O’Dell, B. L. Mineral interactions relevant to nutrient require-ments. J. Nutr. 1989, 112 (ISS 12 Suppl.), 1832–1838.

(13) Golub, M. S.; Han, B.; Keen, C. L.; Gershwin, M. E. Developmentalpatterns of aluminum in mouse brain and effects of dietaryaluminum excess on manganese deficiency. Toxicology 1993,81, 33–47.

(14) Norberg, G. F.; Gerhardsson, L.; Brogerg, K.; Mumtaz, M.; Ruiz,P.; Fowler, B. A. Interactions in metal toxicology. In Handbookon the Toxicology of Metals; Nordberg, G. F., Fowler, B. A.,Nordberg, M., Friberg, L. T., Eds.; Elsevier: New York, 2007; pp117-146.

(15) Di Toro, D. M.; Allen, H. E.; Bergman, H. L.; Meyer, J. S.; Paquin,P. R.; Santore, R. C. Biotic ligand model of the acute toxicity ofmetals. 1. Technical basis. Environ. Toxicol. Chem. 2001, 20,2383–2396.

(16) Levander, O. A. Metabolic interrelationships between arsenicand selenium. Environ. Health Perspect. 1977, 19, 159–164.

(17) Yang, L.; Wang, W.; Hou, S.; Peterson, P. J.; Williams, W. P.Effects of selenium supplementation on arsenism: An interven-tion trial in Inner Mongolia. Environ. Geochem. Health 2002,24, 359–374.

(18) Kenyon, E. M.; Hughes, M. E.; Levander, O. A. Influence of dietaryselenium on the disposition of arsenate in the female B6C3F1mouse. J. Toxicol. Environ. Health 1997, 51, 279–299.

(19) Multi-Vitamin/Mineral Supplement Monograph; InformationUpdate 2007-156; Health Canada, October 22, 2007.

(20) Guidance for Evaluating the Oral Bioavailability of Metals inSoils for Use in Human Health Risk Assessment; OSWER Directive9285; U.S. Environmental Protection Agency: Washington, DC,pp 7-80.

(21) Exponent. Draft DMTS Fugitive Dust Risk Assessment; Preparedfor Teck Cominco Alaska Incorporated, Anchorage, AK; Expo-nent: Bellevue, WA, 2005; http://www.dec.state.ak.us/spar/csp/docs/reddog/01dmts_ra_text_4_05.pdf.

(22) Shock, S. S.; Bessinger, B. A.; Lowney, Y. W.; Clark, J. L.Assessment of the solubility and bioaccessibility of barium andaluminum in soils affected by mine dust deposition. Environ.Sci. Technol. 2007, 41 (13), 4813–4820.

(23) Exponent. DMTS fugitive dust risk assessment; Prepared for TeckCominco Alaska Incorporated, Anchorage, AK; Exponent: Belle-vue, WA, 2007; http://www.dec.state.ak.us/spar/csp/docs/reddog/rafinal/1_dmts_ra_text.pdf.

(24) Menzie, C. A.; Southworth, B.; Stephenson, G.; Feisthauer, N.The importance of understanding the chemical form of a metalin the environment: The case of barium sulfate (Barite). HumanEcol. Risk Assess. 2008, 14 (5), 974–991.

(25) Luoma, S. N.; Rainbow, P. S. Why is metal bioaccumulation sovariable? Biodynamics as a unifying concept. Environ. Sci.Technol. 2001, 39, 1921–1931.

(26) Risk Assessment Guidance for Superfund, Volume I, HumanHealth Evaluation Manual (Part A), Interim Final; EPA/540/1-89/002; U.S. Environmental Protection Agency, Office ofEmergency and Remedial Response: Washington, DC, 1989.

(27) Guidance Documents, Soil Cleanup Criteria; New Jersey De-partment of Environmental Protection, 2008; http://www.nj.gov/dep/srp/guidance/scc; last updated, June 4, 2008.

(28) Determination of Soil Cleanup Objectives and Cleanup Levels;TAGM 4046; New York State Department of Conservation,Division of Technical and Administrative Guidance.

(29) Georgia Environmental Protection Division Guidance for Select-ing Media Remediation Levels at RCRA Solid Waste ManagementUnits; Georgia Environmental Protection Division, 1996.

(30) Development of Cleanup Target Levels (CTLs) For Chapter 62-777, F.A.C.; Technical Report prepared by the University ofFlorida; Florida Department of Environmental Protection, 2005;www.dep.state.fl.us/water/drinkingwater/standard.htm.

(31) Aquatic Life Ambient Freshwater Quality Criteria - Copper, 2007Revision; U.S. Environmental Protection Agency, 2007; http://www.epa.gov/waterscience/criteria/copper/2007/criteria-full.pdf.

(32) Draft strategy: Proposed revisions to the “Guidelines for DerivingNumerical National Water Quality Criteria for the Protection ofAquatic Organisms and Their Uses”; U.S. Environmental Pro-tection Agency, 2003; http://www.epa.gov/waterscience/criteria/library/alcg_sab_draft.pdf.

(33) Procedures for the Derivation of Equilibrium PartitioningSediment Benchmarks (ESBs) for the Protection of BenthicOrganisms: Metal Mixtures (Cadmium, Copper, Lead, Nickel,Silver and Zinc); EPA-600-R-02-011; U.S. Environmental Pro-tection Agency; Office of Research and Development: Wash-ington, DC, 2005.

ES9006405

8482 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 22, 2009