PUBLIC HEALTH ASSESSMENT
KENNECOTT (SOUTH ZONE)
COPPERTON, SALT LAKE COUNTY, UTAH
Demographics, Land Use and Natural Resource Use
C. Demographics, Land Use, and Natural Resource Use
Bingham Creek Operable Unit
- General Population Information
Downstream from 4800 West downstream to nearly the Jordan River, the areas adjacent and nearthe creek are largely residential. ATSDR staff's review of census block data for that area showsthere are about 13,000 people residing within 1500 feet of the creek; of these, about 5,000 areage 5 or less (19). The census indicates the population is about 94 percent white; a few percent are Hispanic.
- Bingham Flats, Brick Plant, Gravel Plant
ATSDR staff saw no residences close to this segment of the creek. In the part of Bingham Flatsbeyond Kennecott's property, staff saw a large electrical substation and a large water tank; theremainder of the area appeared to have been used for agriculture at one time. Downstream, thecreek crosses a remote southern corner of the brick plant property at a point far from theproduction area and then passes through the county's gravel quarry property.
- 4800 West to 3200 West
ATSDR staff saw that residential development near the creek begins at 4800 West. The first2,700 feet of creek downstream is bounded on the north by residences and on the south by a golfcourse. The next 4,100 feet is primarily residential, with some vacant property to the north. Along those two segments, the channel bottom is 25 to 40 feet lower than the surroundingtopography. Lot boundaries do not extend into the channel (17).
For the next 2,000 feet observed downstream, residences border the south side of the creek, andthe land immediately north is predominately vacant. Most of the next 1,100 feet (up to 9000South) has homes on the north side and predominately vacant land on the south. Along thesesegments, the channel bottom is 10 to 15 feet below surrounding topography, and some lotboundaries extend into the channel (17).
From 9000 South to the Utah Lake Distribution Canal (1,400 feet) (Figure 2B), the north side ofthe creek is dominated by Jordan Valley Hospital parking lots and then by vacant land. The southside of the creek within that segment is bordered by residential property. For the next 1,600 feet(to 3200 West), the land immediately north and south of the creek is vacant (17).
- 3200 West to 2700 West
The first 2,100 feet of channel (from 3200 West to 8600 South) lies within Jordan View Estates,is concrete lined (since 1986) and fenced on both sides, and has residences on both sides (43). The next 1,100 feet of channel (from 8600 South to 2700 West) is enclosed in a pipe and crosseslarge tracts of undeveloped property. ATSDR staff saw that those properties are fenced along the street edge.
- 2700 West to Brookside Trailer Park
ATSDR staff saw that, from 2700 West to Sugar Factory Road (Figure 2B), the first 1,700 feet ofcreek has residences on one side of the channel and a golf course the other. The creek then iscarried within a pipe for about 1,600 feet (18). The channel then reappears and crosses about1,700 feet of undeveloped land that is bounded on the southwest by residential properties. ATSDR observed that a high fence is present between the channel and the residences.
From Sugar Factory Road up to Brookside Trailer Park (1,500 feet), the channel crossesproperties that are predominantly undeveloped. Within the trailer park, ATSDR staff saw that the channel segment (about 1,500 feet) is very close to many homes. Within that reach the channel is shallow and narrow.
- Other Land Use and Natural Resource Use
The channel is essentially a dry creek bed throughout most of the year, and is accessible at manypoints throughout the residential areas. Evidence such as bicycle tracks, caves in the channelbanks, and slide marks on the banks shows children use the channel as a recreational area. Children also have been observed in the channel (3). The ball park at Skye Drive and playgroundsat Brookside Trailer Park and the Meadow Greens subdivision also are locations for children'sactivities.
The creek, being essentially dry, is not likely to be used for fishing or swimming. Intermittent,localized wading is plausible following runoff episodes. Fishing does occur in the Jordan River;swimming also is plausible there.
ATSDR staff observed several schools in the creek vicinity, but none lie within the affected areas. The nearest school is approximately 2,000 feet from the channel.
The Jordan Valley Hospital property on 9000 South lies along the creek. ATSDR staff saw thatthe South Valley Care Center, a nursing facility on 9000 South adjacent to the hospital property,is several hundred feet west of the creek channel, on land that has not been shown to have beenaffected by creek flooding.
Copperton Soils Operable Unit
ATSDR staff's review of census block data for the Copperton vicinity indicates that the town hasa 1990 total population of about 550 and about 40 children age 5 or less (19). The population is about 97 percent white.
The former Bingham High School on the north edge of town is now a middle school. A largecommunity park with children's play areas is on the south side of town. The area on the east sideof town on which homes were relocated from the Lark community was formerly a play area forchildren (2). No hospitals or nursing care facilities are in the community. Copperton obtains itspublic water supply from two deep wells (44). ATSDR staff learned from the water improvementdistrict chairman that the town has been in existence since 1928 and every residence is connectedto the system, which was installed in 1932.
Butterfield Creek Operable Unit
ATSDR staff's review of census block data for the Herriman vicinity indicates that the area has a1990 total population of about 916 and about 110 children age 5 or less (19). The population isestimated to be 98 percent white, 1 percent Hispanic, and 1 percent other racial designations.
Southwest Salt Lake County Groundwater Operable Unit
Southwest Salt Lake County encompasses 3 incorporated cities (West Jordan, South Jordan andRiverton) and 3 unincorporated communities (Copperton, Herriman, and High Country Estates). The following table summarizes 1992 population estimates for the cities and communities.
|High Country Estates||795|
The 1990 predominant land use categories were mining, industrial, non-irrigated agriculture,residential, irrigated agricultural and commercial. The majority of the residential areas, whichincludes typical suburban shopping areas and schools, is in the eastern half of the area closer tothe Jordan River but spreading westward toward the Oquirrh Mountains. In 1995, residential andcommercial are rapidly replacing agricultural lands.
Natural Resource Use
Groundwater is the source of domestic water for the majority of the people living in SouthwestSalt Lake County. Table B2 summarizes the groundwater usage by communities in southwestSalt Lake County.
|Community||1992 Annualwater use inmillions ofgallons (mg)||Number ofMunicipal Wells||Per cent ofGroundwaterpumped fromstudy area|
|West Jordan||3, 373||1,676||6 with 4 in use||50%|
|South Jordan*||1, 174||N/A||none||N/A|
|Copperton||90||90||3 with 2 in use||100%|
Source: Modified from Table 6-2 in Sverdrup, 1993, "Preliminary Risk Evaluation of Sulfate,Report for Bingham Creek/Ground Water".
Kennecott Corporation completed a detailed well inventory of southwest Salt Lake County in1995. Wells within a 120 square mile area were surveyed for location and use. This informationis summarized in Table 8
|Drinking water wells||347|
|Non-drinking water wells||219|
|Kennecott Utah Copper (KUC)monitoring wells||427|
|Non-KUC monitoring wells||90|
|Not in use water wells||601|
- Source: March 7, 1996, letter from Elaine Dorward-King, Kennecott Utah Copper Corporation, to Max Howie, ATSDR
Selection of Key Contaminants for Public Health Evaluation
ATSDR reviews contaminant data and selects those that warrant subsequent further evaluationfor public health implications. Identification of these contaminants does not imply that humanexposure does occur or that exposure would actually result in adverse health effects.
Contaminant selection considers the following factors:
- Concentrations of contaminants in media.
- Sample locations, field data quality, and laboratory data quality.
- Relationship of concentrations to ATSDR's public health assessment comparison values; also, the unavailability of suitable comparison values.
- Community health concerns.
The soil contaminants selected for further evaluation and the media in which sampling shows theyhave occurred are summarized in the data tables in Appendix D. The data tables also identifyspecific public health assessment comparison values ATSDR considered in the selection process. An Environmental Media Evaluation Guide (EMEG) is an estimated comparison concentrationthat is based on information determined by ATSDR from its toxicological profiles for a specificchemical. Reference Dose Media Evaluation Guide (RMEG) comparison concentrations arebased on EPA's estimates of the daily exposure to a contaminant that is unlikely to cause adversehealth effects. An Action Level (AL) is an EPA regulatory concentration that, if exceeded in apublic water system, requires the system operators to initiate specified response actions. CancerRisk Evaluation Guide (CREG) is a comparison concentration that is based on an excess cancerrate of one in a million persons and is calculated using EPA's cancer slope factors. TheOccupational Safety and Health Administration (OSHA) and National Institute for OccupationalSafety and Health (NIOSH) standards for air in the workplace have been divided by 400 todevelop a conservative comparison value for community exposure. Ambient Air Quality Standard(NAAQS) was also selected as a comparison value. An estimated (Est.) comparison value isbased on ATSDR staff review of toxicologic data for a contaminant.
ATSDR also reviewed the EPA Toxic Chemical Release Inventory (TRI) for 1992 to learnwhether that database would disclose any supplemental information about contaminant releases inthe area.
Identification of Exposure Pathways
ATSDR identifies human exposure pathways by examining environmental and human componentsthat might lead to contact with contaminants of concern. A pathway analysis considers fiveelements: a source of contamination, transport through an environmental medium, a point ofexposure, a route of human exposure, and an exposed population. Completed exposure pathwaysare those for which the five elements are evident, indicating that exposure to a contaminant hasoccurred in the past, is currently occurring, or will occur in the future. ATSDR regards peoplewho come in contact with contamination as exposed; for example, people reside in an area withcontaminants in air, or who drink water known to be contaminated, or who work or play incontaminated soil are considered exposed. Potential exposure pathways are those for whichexposure seems possible, but one or more of the elements is not clearly defined. Potentialpathways indicate that exposure to a contaminant could have occurred in the past, could beoccurring now, or could occur in the future.
Only exposure situations associated with completed pathways are discussed in this assessment;evaluations did not disclose any potential pathways likely to be of public health significance.
Determination of Public Health Implications
Determining the public health implications of a site is a two-track process: a toxicologicalevaluation, and, where appropriate, health outcome data evaluation.
- Toxicological Evaluation
Typically, the toxicological evaluation in a public health assessment is a comparison of theexposure dose (i.e., the amount of a substance individuals in an exposure pathway are exposed todaily) to an appropriate health guideline. If the contaminant being evaluated is a carcinogen, thenthe risk from exposure to that carcinogen is determined. The methodology for calculatingexposure doses and cancer risk is described in Appendix E.
- Health Outcome Data Evaluation
Health outcome data is information on the occurrence of cancer, birth defects, or other diseasesor conditions; or the results of testing for the contaminants of concerns in humans. Healthoutcome data are evaluated if it is biologically plausible for a health outcome to occur or if thecommunity is concerned about specific health outcomes; and if the appropriate data can beidentified to evaluate a health outcome. For biological plausibility, the decision to evaluate healthoutcome data depends on whether a completed exposure pathway exists for a chemical suspectedof causing the health outcome of concern (?). The selection of a noncarcinogenic health outcomeis based on a review of the toxicologic literature for that contaminant of concern.
|AL||USEPA Action Level|
|CREG||Cancer Risk Evaluation Guide|
|EMEG||Environmental Media Evaluation Guide|
|Est.||estimated comparison value (by ATSDR)|
|J||value estimated (by laboratory)|
|NAAQS||National Ambient Air Quality Standards|
|NIOSH||National Institute for Occupational Safety and Health|
|OSHA||Occupational Safety and Health Administration|
|PM10||airborne particles 10 microns and smaller in diameter (relevant to inhalationexposure)|
|ppb||parts per billion|
|ppm||parts per million|
|RMEG||Reference Dose Media Evaluation Guide|
|µg/m3||micrograms per cubic meter (ambient air)|
|<||less than the stated value|
|Contaminant||Maximum Concentration, Surface Samples (ppm)||Comparison Value|
|Locations for some samples are not clearly described in reference texts and figures. Data noted as being "Within channel orImmediately Adjacent" or "Off-channel" are based in part on ATSDR staff interpretations of available location information.|
* Values shown are for materials visually identified as tailings. Most samples from channel had less than 5,000 ppm lead. Onadjacent properties, lead and arsenic in essentially all samples were close to, or equal to, expected background levels. About 150 samples taken in 1990 and 1991 reviewed (3,?,5,6,7,8)
** In 1990, on Tarbert Circle, on a nearby property that backs up to a ditch that drains into the creek, 1,260 ppm lead wasfrom a backyard sample identified as "imported barrier"--seemingly implying the sample is of nonnative materials. Anearby sample on the property showed lead to be 480 ppm. (9)
*** By Skye Drive. Seventeen samples taken in 1990 and 1991 (6,10).
**** By Judd Drive. The 1,700 ppm lead and 70 ppm arsenic were the highest reported for two samples taken from the creekchannel. The 620 lead and 40 ppm arsenic were for samples taken elsewhere in park. Ten samples taken in 1990 and 1991(6,10).
***** Bike path by creek between 4800 West and Skye Drive. Nine samples taken, date not specified (11).
****** Few samples were analyzed for cadmium
|Contaminant||Maximum Concentration, Surface Samples (ppm)||Comparison|
|Locations for some samples are not clearly described in reference texts and figures. Data noted asbeing "Within channel or Immediately Adjacent" or "Off-channel" or "Residential Property" are basedin part on ATSDR staff interpretations of available location information.|
Approximately 140 samples taken in 1990, 1991 and 1992 (3,5,6,7,8). Many samples within thechannel, immediately adjacent, and from off-channel plumes substantively exceed expected leadbackground levels. Less than half of the samples were analyzed for arsenic. Of these, many exceedexpected background by a substantive margin. On residential property, it appears that lead and arseniclevels were close to or equal to expected background levels.
|Contaminant||Maximum Concentration, Surface Samples (ppm)||Comparison Value|
|Residential Subdivisions||IRECO/Rigbye||Jordan Viewa||Fahnian|
|Samples taken in 1990, 1991 and 1992 (12,6,9,13).|
a 487 samples. About 80 per cent of residential lots sampled exceeded expected background lead levels; most by a substantive margin. Manyarsenic levels exceeded expected background by a substantive margin.
b 83 samples. All residential lots sampled exceeded expected background lead levels; most by a substantive margin. A few arsenic levels alsoexceeded expected background by a substantive margin.
c 139 samples. Lead substantively exceeded background levels for about 1/3 of residential lots sampled (maximum 3,472 ppm) and some ofthe playground samples (maximum 3,794). Some arsenic levels exceeded expected background, but only a few did so by a substantivemargin. Questions about playground values resulted in a second and third resampling there that showed a reduced lead level for generalplay areas (maximum 1,900 ppm)--sandbox samples showed lead to be a maximum of 40 ppm.
d 26 samples. Lead at essentially all lots sampled was close to or equaled expected background levels. Arsenic did not exceed expectedbackground by a substantive margin.
e 10 samples. Lead in all samples substantively exceeded background lead levels. Two of the five samples analyzed for lead substantivelyexceeded background levels.
f Relatively few samples were analyzed for cadmium.
|Contaminant||Maximum Concentration (ppm)||Comparison Value|
|2700 West to|
|Sugar Factory |
Road to Redwood
|Locations for some samples are not clearly described in reference texts and figures. Data noted as being"Within Channel" or "Off-Channel Properties" are based in part on ATSDR staff interpretations of availablelocation information. No cadmium data. Sampling conducted in 1990, 1991 (6).|
a 5 samples. Lead exceeded expected background level for all samples.
b 27 samples. Lead on residential properties exceeded expected background levels for about 50 percent of samples.
c 37 samples. Lead exceeded expected background level for about 85 percent of samples.
d 17 samples. Lead on residential and undeveloped properties exceeded expected background level for all samples.
e 21 samples. Lead exceeded expected background level for all samples
f 13 samples. Lead on residential properties exceeded expected background for about 50 per cent of samples.
g Playground data. Lead does not exceed expected background levels.
Surface Samples (ppm)
|a 2020 ppm is the highest for nine samples taken on a residentialproperty closest to the conveyor. The highest copper was from asample taken at the downslope edge of the back yard by the Kennecottproperty fence; copper was less than 500 ppm for the other eightsamples taken on the property.|
b 597 ppm is the highest found on other properties
* Thirteen samples in 1994 and eight samples in 1993 (14,15,16).
** Three samples in 1994. (14)
Comparison of Exposure Dose to Health Guidelines
The exposure doses for soil ingestion were calculated in the following manner. The minimum and maximum concentration for a contaminantwere multiplied by the soil ingestion rates for adults, 0.0001 Kg/day; children, 0.0002 Kg/day, or pica children, 0.005 Kg/day. (The habit ofingesting large amounts of soil is called pica.) This product was divided by the average weight for an adult, 70 Kg (154 pounds) or for a child,10 Kg (22 pounds). Those calculations assume that there is frequent daily exposure to soil contaminated at the specified level. The results ofthe actual calculations are recorded in Tables E1-E3 which follow.
Calculation of Risk of Carcinogenic Effects
Carcinogenic risks from the ingestion of soil were calculated through the following. The adult exposure doses for ingestion of soil calculated asdescribed previously, were multiplied by the EPA's Cancer Slope Factor for the contaminants of concern. The result represents the maximumrisk for cancer after 70 years of exposure to the maximum concentration of the contaminant. A cancer slope factor was available only forarsenic. The results of the calculation of carcinogenic risk from exposure to arsenic can be found on Tables E1-E3 which follows. The resultsare discussed in the Public Health Implications portion of the Soils Exposure Situation Section.
Uncertainties in Calculating Cancer Risk
The actual risk of cancer is probably lower than the calculated number. The method used to calculate EPA's Cancer Slope Factor assumes thatthere is no safe level for exposure (31). There is little experimental evidence to confirm or refute this assumption. Lastly, the method computesthe 95% upper bound for the risk, rather the average risk, which results in there being a very good chance that the risk is actually lower, perhaps several orders of magnitude (31).
|TABLE E3 - ESTIMATED EXPOSURE DOSES AND CANCER RISK FOR CONTAMINANTS IN LOWER CHANNEL SOIL EXPOSURE SITUATION COMPARED TO HEALTH GUIDELINES FOR INGESTION1|
|Range of Estimated Exposure Doses in mg/kg/day||Health|
|ND - 80||0 - 0.0001||0 - 0.002||0 - 0.04||0.0003||MRL2||0 to|
2 in 10,0003
|Range of Lead|
|ND - 7,129||0 - 0.02||0 - 0.3||0 - 9||none||none||no cancer slopefactor isavailable|
|1 - An explanation of how these exposure doses and cancer risk was calculated can be found earlier in this appendix.|
2 - MRL = ATSDR's minimal risk level. For more information on the MRL for arsenic, see the ATSDR Toxicological Profiles for thIs chemical.
3 - Maximum additional lifetime risk of cancer
EXPLANATION ON REPRODUCTION AND USE OF PUBLIC COMMENTS
As mentioned in the introduction to this document, ATSDR received comments on the public commentdraft of the public health assessment from very few organizations and none from individual citizens. Comments from two of the organizations, Atlantic Richfield Company (via their legal counsel, Arnoldand Porter) and the Kennecott Corporation, are reproduced in this appendix. The reproduced commentsare incorporated into this document so that interested citizens may evaluate their effects upon the final document.
The Kennecott Corporation provided significant scientific references and information to support theirconcerns about the lack of health effects of sulfate below 1500 parts per million. ATSDR reviewed theinformation, including later reports from EPA Region VIII. Based upon evaluation of the informationprovided, ATSDR modified the draft document regarding ingestion of sulfate in drinking water at 500 ppm.
Comments from the University of Cincinnati, Utah Department of Environmental Quality (UDEQ), andthe Utah Department of Health (UDC) are not reproduced. The University of Cincinnati comments werenot reproduced for two reasons, the size of the comments and the concern about publishing informationderived directly from University's health study without written permission of the authors. However, theUniversity comments were carefully considered and the document modified as a result of the evaluation.
The comments from UDEQ and UDH were very helpful to the authors of PHA. As a result, someeditorial changes were made in the document but no significant changes in conclusions orrecommendations.
The comments presented by legal counsel, Arnold and Porter, on behalf of Atlantic Richfield Companywere also carefully evaluated. Some editorial changes were made to clarify issues but no significantchanges in conclusions or recommendations were deemed necessary.
ARNOLD & PORTER
August 20, 1996
Re: Public Health Assessment for Kennecott - Bingham (South Zone)
Dear Messrs. Howie and Mann:
The attached comments on the draft Public Health Assessment for Kennecott-Bingham (South Zone) aresubmitted on behalf of the Atlantic Richfield Company ("ARCO"). ARCO appreciates the opportunity tocomment once again on this document.
Please call me if you have questions concerning our comments.
COMMENTS OF ARCO ON ATSDR PUBLIC HEALTH ASSESSMENT FOR KENNECOTT-BINGHAM (SOUTH ZONE)
Page 1, paragraph 3: We fully agree that "present and future conditions of soils in the Bingham Creekarea" present "no apparent public health concern." However, we do not believe that ATSDR hassubstantiated its statement that "some areas of Bingham Creek . . . did pose a public health hazard" in thepast. At a minimum, this assertion is difficult to evaluate because the assessment does not define "publichealth hazard." The assessment certainly should make clear that a "public health hazard" is not equivalentto a risk to individuals.
To ARCO's knowledge, there is no instance of any human illness, disease or injury caused by the soils inthe Bingham Creek area. Moreover, in light of the (i) findings of the environmental health studyconducted in the area, (ii) studies conducted -- by ATSDR as well as by Potentially Responsible Parties --at other mining sites, and (iii) associated animal bioavailability studies on lead and arsenic in soils, all ofwhich demonstrate that soils mixed with mill tailings do not pose public health risks, the absence of anysuch illness, disease or injury is not surprising. We believe these studies constitute a strong, consistentbody of evidence demonstrating that residential exposures to mill tailings do no result in public healthhazards. In short, ATSDR has not documented or supported its assertion regarding the past existence of a"public health hazard" in "some" areas.
We also note that on page 7 the assessment asserts merely that "(t)he Bingham Creek situations may havebeen public health hazards in the past" (emphasis added), which is a considerably less emphatic positionthan that taken on page 1 of the assessment.
Page 8, first paragraph: It is misleading to suggest that tailings transport is a recent phenomena. Tailingsimpoundments diverted some or all of the tailings discharged into Bingham Creek beginning in 1906 or1907. Tailings production itself in Bingham. Canyon ended in about 1930, and Bingham Creek stoppedbeing a free-flowing drainage for the Canyon decades ago. In these circumstances -- where the asserted"public health hazard" has been present for virtually the entire century -- the absence of any human healthproblem caused by it should have received greater weight than it did.
Page 13, last paragraph: We agree with your conclusion that young children (who as a group haverelatively higher rates of soil ingestion) are not likely to have played in the "Bingham Creek UpperChannel." ATSDR presents no reasons why this would not also be true at other locations in BinghamCreek.
Page 14, fourth paragraph: This paragraph suggests that Bingham Creek flooded until Kennecott builtthe reservoir system. In fact, the "delta" area appears on the first available aerial photograph -- from 1937(Exhibit A) -- of what are now neighborhood areas. Moreover, historical evidence from the early years ofthe century -- regarding complaints due to tailings interference with agricultural activities -- stronglysuggests that the "delta" area was formed even before World War I.
Page 15, last paragraph to Page 16, fourth full paragraph: We believe that the most serious deficiency inthe draft assessment is the short shrift it gives to the data from the 1993 environmental health study. Inproviding our comments in march 1996, we noted that Dr. Bornschein had provided the results of thestudy to ATSDR in late 1994. We also offered to supply the data to ATSDR once again in computerizedformat for ease of analysis, and offered to respond to any questions or comments. Until ATSDR fullyevaluates all the data from this study -- which was the most comprehensive of its kind ever conducted at amining site in this nation -- its "public health assessment" cannot be scientifically valid.
ATSDR states that the "lack of systematic sampling . . . appears to be a problem" with the study(emphasis added). In fact, a comprehensive and systematic attempt was made in the Bingham Creekneighborhood areas to identify all young children and to sample their blood and urine. The fact that 75properties in the pertinent area were not included is no proof of the absence of systematic sampling, butrather indicates strict compliance with study protocols due to the absence of young children at thoseproperties (or children whose parents were not willing to fully participate in the study). In any case, anyrisks from exposures on these properties can be assessed from the slopes derived from the study. In thisregard, the regression analyses show a near non-existent relationship between lead and arsenic in soils andbiomarkers of exposure.
ATSDR next asserts that it received only a "partial" data set for the study and that certain "critical" datawere not available. As indicated above, ARCO has offered before to assist ATSDR in identifying the"missing" data, to the extent it is unable to do so on its own, and we renew that offer, even to the point offacilitating a direct exchange between ATSDR and Dr. Bornschein, so that any questions regarding thedata can be clarified.
In short, ATSDR has erred in ignoring the environmental health study, which demonstrated that the bloodlevels of children and participating adults who reside in the vicinity of the Bingham Creek channel arebelow the national average.
Furthermore, despite any limitations of the 1990 Utah Survey, we believe it should be given significantweight in evaluating risks, given the consistency of its results with many other mining site studies as wellas the 1993 study.
Page 17, second full paragraph: We agree that arsenic in soil is significantly less bioavailable thanarsenic in water. Bioavailability studies in a number of animal species consistently demonstrate thatbioavailability of arsenic in soils is in the 15-20% range. These studies are referenced in our comments ofMarch 1, 1996.
Page 18, second paragraph: We agree with your conclusions about the unlikelihood of cancer risks fromlead exposure.
Page 18, final paragraph: Aerial photographs (which we have enclosed at Exhibit B) show that theBingham Creek "floodplain" area was populated with only a few scattered houses during the 1962s withfarmland present until that time. Thus, ATSDR's reference to the "long-term residency in the delta area"is without historical foundation, and its assumption that the 70-year, lifetime risk of cancer "may" haveincreased in this area is wholly unsupported. Given this and other evidence, including the data on thereduced bioavailability of arsenic in soils and the low urine arsenic levels demonstrated in the BinghamCreek study, conclusions about elevated risks from arsenic exposures (see also the fourth paragraph on p.18 and p. 22) are not warranted.
Page 19, third paragraph: This paragraph fails to clearly distinguish the very low blood lead: soil leadslopes found at mining sites from other exposures, such as from operating smelters, which result in muchsteeper slopes. The notion that the blood lead: soil lead slope could be as high as 65 ug/dl per 1000 ppmincrease in lead in soils, with an average of 4-5 ug/dl per 1000 ppm is wholly without evidentiary supportat this site, or at any other mining site. In fact, the slope evidenced in the Bingham Creek study was lessthan 0.3 ug/dl per 1000 ppm. Such a minimal response makes clear that Bingham Creek children have notsuffered ill effects from lead exposures at the site.
You also may be interested in the attached slide (Exhibit C), which shows there was no relationshipwhatsoever between blood lead levels and proximity of the children's homes to Bingham Creek.
Page 19, first paragraph and Page 22, fourth paragraph: The correct LOAEL for cadmium noted in thedocument is 0.0075 mg/kg-day, not 0.075 as stated. Further, and more important, the assertion that achild with pica for soil might be at risk for proteinuria from cadmium exposures is completely withoutscientific support. Proteinuria develops only when cadmium concentrations in the kidney accumulate to acritical level, typically after several decades of chronic exposures. Transient exposures like those resultingfrom childhood pica must be averaged over a lifetime before being compared to a LOAEL, NOAEL orMRL. When this is done, it is clear that there is no risk from the Bingham Creek exposures. Thisconclusion is supported by ATSDR's studies in Palmerton, Pennsylvania, among others.
We also note that short term, high level exposures which might conceivably produce kidney damage orother toxic effects are not likely in the exposure conditions at Bingham Creek.
Dear Mr. Howie:
Kennecott Utah Copper Corporation (KUC) has reviewed the Public Health Assessment conducted by theAgency for Toxic Substances and Disease Registry (ATSDR) for the Kennecott (South Zone), Copperton,Salt Lake County, Utah (Bingham Creek area) and is providing both general and specific comments onthe June 28, 1996 draft released For Public Comment. KUC submitted comments (letters from ElaineDorward-King to Max Howie dated February 19, 1996 and March 7, 1996) on a preliminary draft of thisAssessment. ATSDR incorporated some of KUC's previous comments into the document, however, thisdraft still contains a number of inaccurate statements that will be misleading to the public if included inthe Final Assessment. We therefore have included some of our previous comments in addition toproviding comments on new sections of the document.
The amount of discussion spent on pre-removal exposure seems unnecessary given that a public healthassessment should focus on the current risks. The general statements that residents may be at risk frompast exposures are not sufficiently substantiated.
Overall, all report describes risk and hazard in vague terms with little discussion/explanation of theuncertainty in the literature and toxicological assessments of the chemicals of concern. For example, thepossibility of transient diarrhea is grouped under the same label (health hazard) as more serious effectssuch as cancer. More explanation of the process of the health evaluation should be made in the body ofthe text, because most members of the public are unlikely to read and understand the appendices. Thecalculations are more appropriate for a conservative screening for possible risk than for accuratecalculation of actual risks for residents. Because of the lack of explanation, however, the public maybelieve the calculations estimate their actual risks.
The Public Health Assessment could be improved with a discussion of the forms of arsenic, cadmium,copper, and lead in the soil. The form of these constituents determines their bioavailability and toxicity.Specifically, arsenic and lead species in soil as a result of mining are typically less bioavailable than thosederived from other sources. The U.S. Environmental Protection Agency's (EPA's) swine study ofBingham Creek soil reports 19% rather than the default 30% bioavailability for lead in soil (U.S. EPA,1995c).
The lead and arsenic exposure analysis of the University of Cincinnati also provides valuable informationfor the relative impact of these constituents in soil on residents' exposure. The results indicate that soil isa relatively minor source for exposure, contrary to the implications in the risk calculations included in theAssessment.
The public health assessment concludes that Southwest Salt Lake County groundwater is a public healthhazard because of sulfate levels in excess of EPA's proposed Maximum Contaminant Level Goal of 500mg/L. This conclusion is inaccurate and misleading to the public and contrary to the statements of EPA(1994) on the lack of health affects associated with chronic exposure to sulfate. The public healthassessment states that although residents may become accustomed to sulfate, they may be at risk forhaving kidney stones. This statement also implies more certainty than actually supported by the scientificand epidemiological data.
The assessment should state that EPA does not have a primary drinking water standard for sulfate becausehealth effects are poorly understood. It should be stated that an EPA proposed sulfate standard of 500mg/L was recently deferred for three years until additional study of sulfate health effects could bedetermined. The federal drinking water standards include only a secondary standard for sulfate which isbased on aesthetic qualities and not health effects. The report should also include that the State of Utahdoes have a primary drinking water standard of 1,000 mg/L and that no drinking water wells in the studyarea exceed that value.
Page 1, second paragraph: see comments above and below on the lack of "hazard" associated with sulfatebelow 1,200 ppm.
Page 1, third paragraph, page 27, second paragraph: Pre removal "hazards" are unlikely. At least someexplanation of the likelihood and nature of this calculated "hazard" should be provided. There is noevidence to support the statement. "Before actions, some areas of Bingham Creek contained soils that didpose a health hazard."
Page 1, last paragraph, page 27 third paragraph: The last sentence seems to imply that the blood leadmeasurements are not representative and that exposures were previously higher before the family wasnotified about the contamination. Nevertheless, the University of Cincinnati found that notification ofcontamination had little to no measurable effect on blood lead levels in Bingham Creek (University ofCincinnati, 1996). See also comments on page 5.
Page 2, Paragraph 1 The following addition is suggested: EPA has deferred listing on the NPL theproposed Kennecott South Zone per a Memorandum of Understanding between Kennecott Utah CopperCorporation, Utah Department of Environmental Quality, and the U.S. Environmental Protection Agencydated September 27, 1995.
Page 2, Paragraph 2 The following addition is suggested: EPA has deferred listing on the NPL theproposed Kennecott South Zone per a Memorandum of Understanding between Kennecott Utah CopperCorporation, Utah Department of Environmental Quality, and the U.S. Environmental Protection Agencydated September 27, 1995.
Page 2, Paragraph 2 The Kennecott South Zone is not an NPL Site. It should be referred to as aproposed site.
Page 5, Table 2, Bingham Creek Floodplain Soils and Lower Bingham Creek Soils There is no evidenceto support the statement made twice "Site did pose a public health hazard prior to removal and cleanup."A more accurate statement (used by ATSDR later in the text) would be "Site may have posed a publichealth hazard prior to removal and cleanup."
Page 7, Table 3 Frequency of Exposure; Page 13, Paragraph 4; Page 4, 5, 8; Page 15, Paragraph 2, andPage 17, Paragraph 2 and throughout document: Exposure estimates should include weather conditions.Exposure to soil contaminants would be unlikely on a daily basis due to typical weather conditions in thisarea. Exposure would be limited by snow and ice during portions of the fall, winter, and spring. Digging,tunneling, and bicycling would be next to impossible in these conditions.
Page 8, Paragraph 1: Should read "More recently, Kennecott Utah Copper Corporation (KUC) has beencontrolling mine and processing waters, and..."
Page 9-11, 13; Tables 4-7: The presentation of environmental data and exposure concentrations wouldbenefit from more statistical description of the data, rather than just presenting ranges. For example,means, upper and lower confidence limits would indicate the distribution of the data and the more likelyexposure for the community than the maximum and minimum. In addition, the casual reader would beconfused by the comparison of the "Maximum post-removal levels in mg/kg" to the "Range of pre-removal levels in mg/kg."
Page 15, first sentence under "Public Health Implications": The first sentence describes how the publichealth assessment compared health guidelines with carcinogenic and non-carcinogenic effects. It appearsTables D1 through D5 should be referenced here, because Tables E1 through E4 (referenced later in theparagraph) do not examine non-carcinogenic guidelines.
When comparing chemical concentrations to an ATSDR Cancer Risk Evaluation Guide (CREG) orEnvironmental Media Evaluation Guide (EMEG), some mention should be made that exceeding theselevels does not necessarily constitute a health threat because these levels are very conservative.Presentation of site concentrations in excess of these levels (e.g., Tables D1-D5) may seem to contradictthe conclusion that a health hazard is not apparent. Soil concentrations for post-removal soils do not appear on these screening tables.
Page 15, second paragraph under "Public Health Implications" and thereafter: The University ofCincinnati Lead and Arsenic Exposure study included children 72 months (6 years) of age and younger, not 7 years of age and younger.
Page 16, first two full paragraphs: The public health assessment argues that sampling the blood lead orurinary arsenic levels of 1,300 children under the age of seven in the Bingham Creek area with blood leadlevels from 284 children from the "contaminated" areas is not enough to reflect a true distribution becauseno children were recruited from 75 of the "contaminated" properties. Nevertheless, a statisticallyrepresentative population that accurately predicts the mean and distribution of blood lead levels iscertainly achievable without sampling every single child in a community. The University of Cincinnaticalculated that only 100 participants were required to predict the geometric mean within 0.3 ug/dl with a95% confidence. With a larger number of participants, the study is sufficient to represent an accurateestimate of the mean and distribution of blood lead levels. We therefore do not agree that these data areunusable for evaluating risk in the community. Actual, statistically valid biomonitoring andenvironmental data are far more accurate for assessing exposure than hypothetical risk calculations.
The geometric mean blood lead level is reported to be of 2.56 ug/dl with 0.8 percent (8 children) of bloodlead levels over 10 ug/dl in the 971 children tested (University of Cincinnati, 1996). This is lower thanother blood lead studies of mining or smelting areas in Utah (University of Cincinnati, 1990, 1996), Utahstate or county blood lead studies (Schlenker, 1995), and the NHANES III national blood lead studyeither for all children or all middle-income suburban children (Pirkle, 1995). Therefore, pre-removal soilconcentrations in Bingham Creek in 1993 did not result in blood lead levels elevated over state or national levels.
The University of Cincinnati study also shows through statistical analysis of their blood andenvironmental lead data that the influence of soil on blood lead is small, with only 3.2% of the variationin blood lead explained by soil concentration. In addition, of the eight children with blood lead levelsover 10 ug/dl, five lived on properties already remediated or with soil lead levels under 400 ppm, andthree were exposed to lead from family hobbies. Several of these children also were from the same familyand therefore should not be counted as independent observations.
Page 16, second full paragraph: This paragraph indicates that ATSDR could not obtain the informationon the number of children identified by census and tested in the University of Cincinnati's1993 study.This information is now available. Ninety-one percent of the occupants completed the census. Thesummary indicated a total of 1,157 families with children less than 72 months old and 1,706 children lessthan 72 months old. Of these, 971 children under 7 years of age in 728 families participated in the bloodlead screening.
Page 17, first paragraph under "Cadmium": The text notes that for pica children, the dose for the pre-removal Bingham Creek Flood Plain exceeded the health guideline (Explained in Appendix D as theEnvironmental Media Evaluation Guideline (EMEG) and that this exposure might result in kidney effects(proteinuria). The next should also state that the EMEG for cadmium is very conservative because isbased on a conservative Minimum Risk Level (MRL) that is even lower than the EPA reference dose.The MRL is inappropriate for assessing short-term exposure in a 10 kg child because it is based onaverage lifetime dose in adults that would result in proteinuria after cumulative lifetime exposure (asalluded to on page 18 but not clearly explained). This comment also applies to the fourth paragraph onpage 18 and first paragraph on page 19. The exposure dose for the pica child was also compared to theLOAEL from drinking water. The exposure dose should be compated to a LOAEL from food or soilingestion, because EPA has shown the LOAEL to be greater for cadmium in food than for cadmium in water.
Page 18, second paragraph under "Bingham Creek Floodplain Soil Exposure Situation" and Page 22,third paragraph of Section E: The phrase "resulted in a moderate increase in an individual's lifetime ofcancer" should be quantified. The word "moderate" is vague and could be construed to mean a higher thanacceptable increased risk.
Page 18, "Post-removal exposures". This paragraph indicates that "there is some chance of healtheffects...." "Some chance" is too vague and should also be quantified.
Page 18, last sentence under "Post-removal exposures". The statement that some chance exists for healtheffects for young children exposed where "contaminated" soil was not removed needs some explanation.Specifically, the likelihood of "some chance" for health effects and "contaminated soil" are not definedsufficiently to understand the magnitude of this risk and which residents might be affected. It is unclearwhether the unremoved contaminated soil is below the 1,100 ppm removal target, properties yet to becleaned up, or those that were missed for some reason. The University of Cincinnati study indicates thatany such chance is likely negligible. Permanent health effects also do not necessarily occur at blood leadlevels that exceed 10 ug/gl. The chance of such health effects depends on how high the blood lead is overan extended period of time.
Page 18, last paragraph: "increased lifetime risk of cancer" and "long-term residency" should be betterexplained as noted above. Given the numerous conservative assumptions of the risk calculations (e.g.,daily exposure to the maximum and 100% bioavailability) the "increased lifetime risk" is likelynegligible.
Page 19, first paragraph under "Lead": second sentence: This sentence states that lead levels in soil of1,000 ppm could increase blood levels from 0.6 to 65 ug/dl with an average increase of 4 to k ug/dl. Thisstatement lacks a summary of the calculations or supporting studies for both the baseline blood lead levelof 0.6 ug/dl and for the increase to 65 ug/dl. Even if the conservative EPA IntegratedExposure/Biokinetic Model (1994) is used with EPA's calculated upper possible bioavailability factor(19%), dust concentration (0.43 x Soil Pb +90 ppm), and GSD (1.43) (U.S. EPA, 1995a,b) for BinghamCreek; the geometric mean blood lead level would be 6.3 ug/dl with an approximate maximum blood leadlevel of about 20 ug/dl. Statistical analysis of actual blood lead and environmental lead data fromBingham Creek (University of Cincinnati, 1996) shows children have only a very small direct exposure tolead in soil from playing in either their yards or along the channel (3.2% of the variation in blood lead isexplained by soil concentration). Indirect exposure to soil through house dust is associated with anincrease in blood lead levels of 0.6 ug/dl as soil increases from 100 to 1,100 ppm (University ofCincinnati, 1996).
Page 19, first paragraph under "Lead," last sentence, and page 20, second paragraph under "Lead," lastsentence: Please reference the statement regarding decreased IQ and impaired hearing and growthassociated with blood lead levels in children at or below ug/dl. This statement requires more explanationof uncertainties. Although some studies have shown negative correlations between blood lead levels andthese health effects, other studies have not shown a significant effect particularly when confoundingvariables were accounted for (Schroeder el al., 1985; Schmitt and Anderson, 1992; Pocock et al., 1994).In addition, effects are generally extrapolated from large population studies in which such effects aremeasured over a large range in blood lead levels considerably in excess of 10 ug/dl. In these studiesshowing a correlation, the actual impacts of blood lead on IQ, for example, is very small compated to thelarge range in IQ among children in general (e.g., see summary graphs in U.S. EPA, 1990), and theimpacts on cognitive development may not persist with age, perhaps due to the greater impact of socialeconomic status, nutrition and other factors on cognitive development (Schroeder et al., 1985; Bellingeret al., 1990). Thus, any effects below 10 ug/dl are at best subtle and likely not of significant healthconsequence relative to the greater impacts of other influences on cognitive development such as theeducational environment, socioeconomic status, and diet.
Page 22, second, third, and fourth paragraphs of Section E: These paragraphs indicate that arsenic,cadmium and lead may have been high enough to result in health effects if young children were exposeddaily to lead and arsenic, and pica children were exposed to the maximum levels of cadmium. These risksare highly speculative, requiring many maximum exposure factors for the same individual (e.g., frequentexposure, living in the area for long periods of time, pica behavior, exposure to maximum soil levels,biological variability, near life time exposure to arsenic and cadmium) to produce a risk. Thishypothetical situation is unlikely to have occurred. Moreover, actual biomonitoring and environmentaldata confirm that hypothetical exposures and risks are overstated. As EPA's response to Concern 2indicates, children living in areas with elevated levels of lead tested very low for blood lead levels.
ATSDR acknowledges the unlikeliness of this hypothetical situation in Appendix E, page 55 of theAssessment. The discussion of uncertainties in calculating cancer risk states "The actual risk of cancer isprobably lower than the calculated number." It goes on to say, referencing Paustenbach, 1989. "Themethod computes the 95% upper bound for the risk, rather than the average risk, which results in therebeing a very good chance that the risk is actually lower, perhaps several orders of magnitude (31)."
Pate 22, third paragraph of Section E: This paragraph states that a moderate increase in cancer risk mayhave resulted in the Bingham Creek flood plain area due to arsenic prior to removal of soil. Thisstatement should be qualified by the numerous uncertainties in the arsenic slope factor (e.g., extrapolationof high dose data from Taiwan to the U.S.), and considerably lower bioavailability of arsenic in soil ascompared to drinking water which is the basis of the slope factor. Considerable evidence indicates thatarsenic is rapidly detoxified by the body at lower doses such as those associated with the estimatedexposure doses for children and adults in Appendix E. Feeding studies in monkeys using Anacondasmelter soil have shown relative bioavailabilities of around 20 to 28% for soil and dust (Freeman et al.,1995). By contrast, the calculations in the Public Health Assessment do not appear to be corrected for thedifference in bioavailability of arsenic in soil relative to water. In addition, the health assessment shouldnote that residential exposure to arsenic in soil in the U.S. has never been shown to be associated withcancer. In fact, an ecological study of skin cancer in the areas around the former Anaconda smelter andSilver Bow mine (with 6,500 acres of soil with arsenic above 90 ppm compared to Bingham Creek'saverage concentration of 27 ppm; University of Cincinnati, 1996) found lower skin cancer rates than twocontrol counties (Wong et al., 1992).
The biomonitoring and environmental data for the site also indicate that any arsenic exposure andassociated cancer risk prior to soil removal is likely negligible. Soil arsenic does not appear to be thesource of urinary arsenic levels measured in the Bingham Creek population. The University of Cincinnatifound that soil arsenic was negatively correlated with urinary arsenic )r=-0.04). Moreover, because leadand arsenic concentrations in residential soils were strongly correlated (r=0.96), these arsenic resultssuggest that house dust lead has a different source than soil (e.g., lead-based paint).
The assessment of pica children based on an adult lifetime MRL is inaccurate (see above comment on page 17).
Pate 22, fourth paragraph of Section E: This paragraph claims lead levels in two segments of the lowerchannel may have been high enough to result in health effects. More discussion is needed to explain thebasis of this statement. For instance, the exposure assumptions assumed should be stated.
Appendix D: The maximum concentrations of the metals were used to evaluate risk at the site. Althoughthis may be appropriate for a screening study, evaluation of exposure should use an average concentrationor an upper bound confidence limit of the mean. Children and adults are unlikely to be exposedconstantly to the maximum concentration.
Page 22 through 25, "Evaluation of Groundwater Contaminants". In comparison to other chemicals,sulfate does not pose a toxic health hazard. As summarized by U.S. EPA (1994a), sulfates have more of aphysical effect on increasing the retention of water in the gastrointestinal tract rather than a chemicallytoxic effect. Sulfates are naturally occurring substances and are an ingredient in over-the-counterlaxatives. The human body is able to acclimate to the effects of sulfates in a relatively short amount oftime. Accordingly, the expert panel convened by the U.S. concluded that "acute short-term effects are theappropriate focus for risk assessment." The benign nature of sulfate levels is demonstrated by numerouscommunities in the U.S. and Canada drinking water with sulfate levels in excess of 500 mg/L without a problem.
Visitors and infants have been the concern of EPA's proposed sulfate regulations. Nevertheless, theoccurrence of laxative effects due to sulfate is more complicated than represented and depends on anumber of factors, as noted in part by EPA's expert panel, such as osmolarity, presence of cations such ascalcium and magnesium, and physiological and dietary conditions. In addition, infants and visitors havebeen assumed to be sensitive to sulfate levels above 500 mg/L, but such an assumption is not supportedby the recent scientific or epidemiological data. In fact, if acclimation occurs by turn over of intestinalcell mucosa (U.S. EPA, 1997a), infants may be able to acclimate faster to the effects of sulfate because oftheir higher cell turn-over rates.
Levels of sulfate in wells currently at the Bingham Creek Site are reported in the Public HealthAssessment to be relatively low: maximum levels of 500 to 1,000 ppm in 6 out of 35 private wellscurrently in use. At these sulfate levels, whether laxative effects would occur even in visitors of infants isquestionable, and according to recent studies, doubtful (McGeehin, 1995; Heizer et al, 1994).
The data base considered by the U.S. EPA in proposing the 500 mg/L level is extremely weak. The fewstudies considered are generally old with small samples sizes and lack of controls for other confoundingfactors (U.S. EPA, 1994a). The results of recent studies in humans and piglets, which were not part of thedata base considered by the EPA in proposing the 500 mg/L level, indicate that sulfate levelsconsiderably in excess of 500 mg/L are required to cause laxative effects. A study by the Centers forDisease Control and Prevention, the South Dakota Department of Environment and Natural Resources,and the National Center for Environmental Health tracked the health of 276 newly born infants in homeswith high sulfate levels throughout South Dakota to assess whether the infants were experiencingincreased incidence of negative health effects (McGeehin, 1995). No increase in negative health effectswas found with consumption of water with sulfate levels in excess of 500 mg/L (maximum level = 2,787mg/L) as compared to those with lower sulfate levels. Sulfate intake was also not associated with asignificant increase in risk for diarrhea. Heizer et al. (1994) investigated the effects of high sulfate indrinking water in human volunteers and in piglets; the piglets were used as a model for human infantswhich are assumed to be a sensitive subpopulation. Diarrhea, the critical toxicological endpoint, was notobserved in the piglet study at concentrations less than 1,600 mg/L nor in the adult human volunteers atconcentrations up to 1,200 mg/L.
The South Dakota report also summarizes results from other recent studies. The North Carolina humanstudies, which were in part funded by the U.S. EPA, involved four adult volunteers ingesting increasinglevels of sulfate from 0 to 1,200 mg/L (increases occurred in 48 hours increments and a follow up studyof six volunteers ingesting sulfate levels of 1,200 mg/L for six days. Neither study reported diarrhea inthe subjects, although sample sizes of these studies are small. The South Dakota Department of Healthalso conducted its own study in 1992 at an institute for the mentally retarded and reported no healtheffects associated with sulfate unless levels above 1,200 mg/L were ingested over a considerable periodof time (Bureau of National Affairs, 1996). Two piglet studies in North Carolina were also funded by theEPA. These studies showed no affect on normal weight gain. Sulfate levels of 1,600 mg/L caused slightlymore soft and liquid stools, whereas sulfate levels at 1,800 mg/L and above were associated withsignificant diarrhea.
Collectively, these recent studies indicate that the onset of laxative effects occurs at sulfate levels on theorder of 1,600 to 1,800 mg/L rather than 500 mg/L.
Page 23, third paragraph: This paragraph states that the highest concentration of sulfate reported in aprivate well was 2,270 mg/L (well W337 in 1985). It should be noted that this well was abandoned in1986 and was replaced by paired wells P247A and P847B. Well P247A is screened in the upper 30 feet ofthe aquifer and sulfate concentrations have decreased from 2,200 mg/L in 1986 to 1,000 mg/L in 1995.Well P247B is screened approximately 400 feet below the water table and sulfate concentrations haveremained less than 500 mg/L since 1986. Therefore, the report should state that sulfate concentrations inthis area are decreasing and that long term exposure to sulfate concentrations exceeding 1,000 mg/L donot exist.
It is unclear what is meant by the statement that some wells have had persistent high levels of sulfate formore than one year. Does this refer to drinking water wells or monitoring wells and how is high definedin terms of sulfate concentrations? The report concludes that exposures to sulfate in drinking water are orwere greater than 500 mg/L but less than 1,000 mg/L.
Page 24, Table 9 Exposure to Contaminants in Groundwater: This table contains inaccurate information.Under the heading of key contaminant, sulfate concentrations from 505 - 2,270 mg/L are listed. Thereport has stated previously that exposure to sulfate was greater than 500 mg/L but less than 1,00 mg/L.In reality, sulfate concentrations in drinking water wells do not exceed 600 mg/L (see comment #5).Therefore, the 505 - 2,270 mg/L range should be replaced with the actual range of 536 - 558 mg/L (seecomment below, Page 24).
Page 24, first paragraph: This paragraph states that ATSDR has estimated that 35 existing privatedrinking water wells lie within or adjacent to a sulfate ground water plume. It is not possible to evaluatethe correctness of this number because a map or a list of the well locations has not been included with thedocument.
Page 24: Ground water quality data used by ATSDR for this Public Health Assessment was beingcollected by Kennecott Utah Copper (KUC) as part of the Southwest Jordan Valley Well Inventory. Thewell inventory was ongoing while ATSDR was performing the health assessment. KUC supplied ATSDRwith ground water quality data in April, 1995 and with an updated database in January, 1996. Thedatabase supplied to ATSDR by KUC in April, 1995 included the most recent analyses as of December,1994. Seven drinking water wells from the April, 1995 database (W41A, W322, W347, W396, W408,W415, and W417) were identified as having sulfate concentrations greater than 500 mg/L. As part of theongoing well inventory it was determined that two of the wells (W41A and W417) are being used forpurposes other than drinking water. An additional well (W347) is located outside of the well inventorystudy area and is outside of the influence of the sulfate plume. Therefore, only four drinking water wells(W322, W396, W408, and W415) in the well inventory study area exceeded sulfate concentrationsgreater than 500 mg/L.
The database supplied to ATSDR by KUC in January. 1996 included the most recent analyses as ofDecember, 1995. Five drinking water wells (W322, W415, EVG2233, W396, and W347) were identifiedas having sulfate concentrations over 500 mg/L. As stated previously well W347 is outside of theinfluence of the sulfate plume. Well EVG2233 was inventoried after the April, 1995 database wassubmitted to ATSDR. Well EVG2233 is located east of the Jordan River, however, and is most likelyaffected by other sources. Therefore, only three drinking water wells (W322, W415, and W396) arelocated within the influence of any KUC's sources and exceed 500 mg/L of sulfate. The followingsummarizes the latest measured sulfate concentrations in wells W415, W396, and W322:
|Well #||Last Date Sampled||Sulfate Concentration|
|March 8, 1996|
February 22, 1996
September 26, 1995
Well W322 is located in the Herriman area and wells W415 and W396 are located downgradient of the South Jordan Evaporation ponds.
Page 24, first paragraph: This paragraph states that some of the wells listed as currently not in use mayhave been used as drinking water supplies before contaminants were measured. This assumption isspeculative and no justifiable conclusion can be made from this assumption.
Page 25: The third paragraph mentions that although residents consuming high sulfate water may beacclimated to the laxative effects, sulfate increases the excretion of calcium in the urine and decreases thepH of the urine, both of which "some researchers have hypothesized" increase the risk of kidney stones.These statements are without documentation or reference and imply more risk than may actually exist.
Although there is published evidence that an increase in sulfate ingestion is associated with an increase inurinary calcium excretion and a decrease in urinary pH in humans and other animals (Tschope & Ritz,1985; Singh et al., 1993; Oetzel el al., 1994; Greger et al., 1991; Whiting et al., 1986 and 1991, cited inGreger et al., 1991) the physiological control mechanisms regulating calcium and sulfate levels are highlycomplex and affected by electrolyte and hormone levels and diet. The formation of kidney stones is also amultifactorial process that may occur, depending on the type of stone formed, with high or low urinarypH (Robertson, 1986). Summarizing the literature, Robertson (1986) stated that "the current consensus isthat the formation of abnormally sized particles of calcium-containing salts is due to the combination of anumber of chemical fores which between them control the relative rates of nucleation, growth andagglomeration of crystals of calcium oxalate and calcium phosphate in urine." Furthermore, we could findno evidence in human subjects of a link between sulfate ingestion alone and the development of kidneystones. In fact, sodium thiosulfate has been successfully used as a treatment for recurrent kidney stones(Yatzidis, 1985).
Even if sulfate ingestion were a definite risk factor for kidney stone formation, it is unclear whether themagnitude of the effects of sulfate in the 500 to 1,000 mg/L range would be sufficient to significantlyaffect the formation of kidney stones. Any effects of sulfate on the formation of kidney stones are likelysubtle as no increases in the incidence of kidney stones have been reported for populations ingestingnaturally elevated sulfate levels. EPA (1994a) has also stated that "there was no evidence of adversehealth effects in animals or humans from chronic exposure to sulfate in drinking water. The availablehealth data indicate that chronic exposure to sulfate is not harmful to health."
Page 26, last paragraph: There may be a typographical error in the first sentence. The two blood leadsamples described as ">5 ug/dl" are likely less than 5 ug/dl based on the other blood lead levels stated.The last sentence states that it is not possible to determine whether the results reflect typical exposureconditions. Nevertheless, the representativeness of the samples can be estimated based on several factorswhich unfortunately are not provided in this paragraph. These factors include the ages of the familymembers, their length of residency, whether they were mostly home prior to testing, and the time of theyear blood lead levels were tested. If the family members tested included young children, the testconducted in summer or fall, and the family had sufficient residence time, the results should be fairlyrepresentative of cumulative exposure. Although blood lead results are generally consideredrepresentative of the past months' exposure, researchers at the University of Cincinnati have found bloodlead levels of individual children to be relatively stable over time with respect to other children.
KUC appreciates the opportunity to provide these comments and requests that they be strongly consideredfor inclusion in the Final Assessment. KUC agrees with ATSDR's conclusion that the present and futureconditions of soils in the Bingham Creek area are no public health concern. KUC disagrees that a publichealth hazard was present in some areas of the Bingham Creek area prior to removal of contaminatedsoils. There is no toxicological or epidemiological evidence of adverse health effects due to arsenic,cadmium, or lead in soils. Actual biomonitoring results, in fact, indicate no risk to public health.ATSDR's references to potential risk from past exposures to contaminants in soil are not substantiated inthis Assessment, and appear to be qualitatively drawn using maximum soil concentrations andconservative exposure assumptions, even when actual exposure data are available.
Risk to public health from ingestion of sulfate in drinking water is over-stated and contradictory in theAssessment. ATSDR concludes that "one exposure situation is still considered a public health hazardbecause a few people still using private drinking water wells within the path of the ground watercontaminant plumes may be ingesting high sulfate concentrations present in the drinking water aquifer.""High sulfate concentrations" are not defined, and ATSDR goes on to say that "there is no informationthat indicates anyone is currently ingesting ground water contaminated with sulfate greater than 1000ppm." Based on the research results currently available, sulfate levels between 500 - 1000 ppm should notbe considered "high". It is doubtful whether even infants and transients experience laxative effects at theselevels. As detailed in our comments above, EPA has deferred establishing a drinking water standard forsulfate until additional research regarding sulfate health effects is completed. Also note that the State ofUtah does have a primary drinking water standard of 1000 mg/L, and no drinking water wells exceed this value.
In conclusion, while KUC largely agrees with the overall conclusions presented in the Assessment, it istroubling that much of the language in the body of the document alludes to potential exposures and risksthat are not realistic, cannot be substantiated, and have the potential to unnecessarily alarm the public.KUC considers this unfortunate and inappropriate, and hopes that these deficiencies are remedied in the final draft.
Please do not hesitate to call me if you have any questions regarding KUC's comments or suggestions.
Elaine J. Dorward-King, Ph.D.
Director, Environmental Affairs
Bellinger D, Leviton A, and Sloman J. 1990. Antecedents and correlates of improved cognitiveperformance in children exposed in utero to low levels of lead. Environmental Health Perspectives 89:5-11.
Bureau of National Affairs. 1996. Environment Reporter. Page 1614. January 12.
Freeman GB, Schoof RA, Ruby MV, Davis AO, Dill JA, Liao SC, Lapin CA, and Bergstrom PD. 1995.Bioavailability of arsenic in soil and house dust impacted by smelter activities following oraladministration in Cynomolgus Monkeys. Fundamental and Applied Toxicology 28:215-222.
Greger JL, Kaup SM, Behling AR. 1991. Calcium, magnesium, and phosphrous utilization by rats fedsodium and potassium slats of various inorganic anions. J Nutr 121:1382-1388.
Heizer WD, Sandler RS, Seal Jr. E, Murray SC, Gomez GG, Busby MG, Scliebe BG, Pusek SN. 1995.The effect of sulfate in drinking watr on intestinal function: studies in normal adult human volunteers andnewborn, artificially-reared piglets. In press.
Marcus AH and Elias RW. 1994. Estimating the contribution of lead-based paint to soil lead, dust lead,and childhood blood lead. In: ME Beard and SD Allen Iske (eds), Lead in Paint, Soil, and Dust: HealthRisks, Exposure Studies, Control Measures, Measurement Methods and Quality Assurance, ASTM STP1226. American Society for testing and Materials, Philadelphia, Pennsylvania.
McGeehin MA. 1995. Evaluation of Human Health Effects Associated with Drinking Water ContainingElevated Levels of Sulfate: A Cohort Investigation in South Dakota. Cover letter and attached reportfrom MA McGeehin, National Center for Environmental Health to S. Lance, South Dakota Departmentof Health. December 1.
Oetzel Gr, Fettman Mj, Hamar DW, Olson Jd. 1991. Screening of anionic salts for palatability, effects onacid-base status, and urinary calcium excretion in dairy cows. J Dairy Sci 74:965-971.
Pirkle JL, Brody DJ, Gunter EW, Kramer RA, Paschal DC, Flegal KM, Matte TD. 1994. The decline inblood lead levels in the United States: the National Health and Nutrition Examination Surveys(NHANES). Journal of the American Medical Association 272(4):284-291.
Pocock SJ, Smith M, and Baghurst P. 1994. Environmental lead and children's intelligence: a systematicreview of the epidemiological evidence. BMJ309:1189-97.
Robertson WG. 1986. Pathophysiology of stone formation. Urol Int 41:329-333.
Schlenker T. 1995. Letter to Mayor Tom Dolan. Director, Salt Lake City-County Health Department.December 18.
Schmitt N and Anderson TW. 1992. Blood lead in children. Letters to Editor. The Lancet, February 8.
Schroeder SR, Hawk B, Otto DA, Mushak P, and Hicks RE. 1985. Separating the effects of lead andsocial factors on IQ. Environmental Research 38:144-154.
Singh PP, Hussain F, Gupta RC, Pendse AK, Kiran R, Ghosh R. 1993. Effect of dietary methionine andinorganic sulfate with and without calcium supplementation, on urinary calcium excretion of guinea pigs(cavia porcellus). Ind J Exp Biol 31:96-97.
U.S. Environmental Protection Agency (U.S. EPA). 1990. Technical Support Document on Lead. ECAO-CIN-757. First Draft. September.
U.S. Environmental Protection Agency (U.S EPA). 1994a. Drinking Water; National Primary DrinkingWater Regulations--Sulfate; National Primary Drinking Water Regulation Implementation. 40 CRF Parts141, 142, and 143. 59 Federal Register 65578, December 20.
U.S. Environmental Protection Agency (U.S. EPA). 1994b. EPA/540/R-93/081. Guidance Manual forthe integrated Exposure Uptake Biokinetic Model for Lead in Children. Office of Emergency andRemedial Response.
U.S. Environmental Protection Agency (U.S. EPA). 1995a. Kennecott Site-Specific IEUBK ModelingApproach. Unpublished. Presented by Life Systems under contract to U.S. EPA Region VIII.
U.S. Environmental Protection Agency (U.S. EPA). 1995b. Site-specific IEUBK modeling riskassessment task force meeting. Present by Life Systems under contract to EPA Region VIII, BinghamCreek Utah. May 23.
U.S. Environmental Protection Agency (U.S. EPA). 1995c. Unpublished results of a bioavailability studyof Bingham Creek area soils in swine. Gerry Henningsen. May.
University of Cincinnati. 1990. Midvale Community Lead Study, Final Report. Department ofEnvironmental Health.
University of Cincinnati. 1996. Bingham Creek Environmental and Human Health Lead and ArsenicExposure Study. Final Report. Department of Environmental Health.
Wong O, Whorton MD, Foliart DE, Lowengart R. 1992. An ecological study of skin cancer andenvironmental arsenic exposure. International Archives of Environmental Health 64:235-241.
Yatzidis H. 1985. Successful sodium thiosulphate treatment for recurrent calcium urolithiasis. ClinNephrol 23:63-67.
1. ATSDR Staff (John Crellin, Don Gibeaut, John Mann, andGlenn Tucker) visited the site vicinity during the weeks ofSeptember 4 and October 23, 1994. Pertinent information obtainedduring those visits is described in appropriate sections of thisdocument.