ATSDR Attachment D Oak Ridge Y-12_PHA

Evaluation of Y-12 Mercury Releases

U.S. Department of Energy, Oak Ridge Reservation

Oak Ridge, Anderson County, Tennessee

EPA FACILITY ID: TN1890090003

March 30, 2012
�

THE ATSDR PUBLIC HEALTH ASSESSMENT: A NOTE OF EXPLANATION

This Public Health Assessment was prepared by ATSDR pursuant to the Comprehensive Environmental Response, Compensation, and
Liability Act (CERCLA or Superfund) section 104 (i)(6) (42 U.S.C. 9604 (i)(6)), and in accordance with our implementing regulations
(42 C.F.R. Part 90). In preparing this document, ATSDR has collected relevant health data, environmental data, and community health
concerns from the Environmental Protection Agency (EPA), state and local health and environmental agencies, the community, and
potentially responsible parties, where appropriate.
In addition, this document has previously been provided to EPA and the affected states in an initial release, as required by CERCLA
section 104 (i)(6)(H) for their information and review. The revised document was released for a 30-day public comment period.
Subsequent to the public comment period, ATSDR addressed all public comments and revised or appended the document as appropriate.
The public health assessment has now been reissued. This concludes the public health assessment process for this site, unless additional
information is obtained by ATSDR which, in the agency’s opinion, indicates a need to revise or append the conclusions previously
issued.

Agency for Toxic Substances & Disease Registry.....................................................Thomas R. Frieden, M.D., M.P.H., Administrator
Christopher J. Portier, Ph.D., Director
Division of Community Health Investigations (Proposed)…. ..................................................................................... (Vacant) Director
Tina Forrester, Ph.D., Deputy Director
Central Branch (Proposed).……………………………………………………………………………Richard E. Gillig, M.C.P., Chief

Eastern Branch (Proposed)…………………………………………………………………..Sharon Williams-Fleetwood, Ph.D. Chief

Western Branch (Proposed) ............................................................................................................. Cassandra Smith, B.S., M.S., Chief

Science Support Branch (Proposed) ................................................................................................................ Susan Moore, M.S., Chief

Use of trade names is for identification only and does not constitute endorsement by the Public Health Service or the U.S. Department of
Health and Human Services.

Additional copies of this report are available from:

National Technical Information Service, Springfield, Virginia

(703) 605-6000

You May Contact ATSDR Toll Free at

1-800-CDC-INFO

or

Visit our Home Page at: http://www.atsdr.cdc.gov

Oak Ridge Reservation (USDOE)

Final Release

PUBLIC HEALTH ASSESSMENT
Evaluation of Y-12 Mercury Releases
U.S. DEPARTMENT OF ENERGY, OAK RIDGE RESERVATION

OAK RIDGE, ANDERSON COUNTY, TENNESSEE

EPA FACILITY ID: TN1890090003

Prepared by:

Division of Community Health Investigations (proposed)

Agency for Toxic Substances and Disease Registry

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Foreword
The Agency for Toxic Substances and Disease Registry, ATSDR, was established by Congress in
1980 under the Comprehensive Environmental Response, Compensation, and Liability Act, also
known as the Superfund law. This law set up a fund to identify and clean up our country’s
hazardous waste sites. The United States Environmental Protection Agency, U.S.EPA, and the
individual states regulate the investigation and clean up of the sites.
Since 1986, ATSDR has been required by law to conduct a public health assessment at each of the
sites on the U.S.EPA National Priorities List. The aim of these evaluations is to find out if people are
being exposed to hazardous substances and, if so, whether that exposure is harmful and should be
stopped or reduced. If appropriate, ATSDR also conducts public health assessments when petitioned
by concerned individuals. Public health assessments are carried out by environmental and health
scientists from ATSDR and from the states with which ATSDR has cooperative agreements. The
public health assessment program allows the scientists flexibility in the format or structure of their
response to the public health issues at hazardous waste sites. For example, a public health assessment
could be one document or it could be a compilation of several health consultations - the structure
may vary from site to site. Nevertheless, the public health assessment process is not considered
complete until the public health issues at the site are addressed.
Exposure: As the first step in the evaluation, ATSDR scientists review environmental data to see
how much contamination is at a site, where it is, and how people might come into contact with it.
Generally, ATSDR does not collect its own environmental sampling data but reviews information
provided by U.S.EPA, other government agencies, businesses, and the public. When there is not
enough environmental information available, the report will indicate what further sampling data is
needed.
Health Effects: If the review of the environmental data shows that people have or could come into
contact with hazardous substances, ATSDR scientists evaluate whether or not these contacts may
result in harmful effects. ATSDR recognizes that children, because of their play activities and their
growing bodies, may be more vulnerable to these effects. As a policy, unless data are available to
suggest otherwise, ATSDR considers children to be more sensitive and vulnerable to hazardous
substances. Thus, the health impact to the children is considered first when evaluating the health
threat to a community. The health impacts to other high risk groups within the community (such as
the elderly, chronically ill, and people engaging in high risk practices) also receive special attention
during the evaluation.
ATSDR uses existing scientific information, which can include the results of medical, toxicologic
and epidemiologic studies and the data collected in disease registries, to determine the health effects
that may result from exposures. The science of environmental health is still developing, and
sometimes scientific information on the health effects of certain substances is not available. When
this is so, the report will suggest what further public health actions are needed.

Page | i

Conclusions: The report presents conclusions about the public health threat, if any, posed by a site.
When health threats have been determined for high risk groups (such as children, elderly, chronically
ill, and people engaging in high risk practices), they will be summarized in the conclusion section of
the report. Ways to stop or reduce exposure will then be recommended in the public health action
plan.
ATSDR is primarily an advisory agency, so usually these reports identify what actions are
appropriate to be undertaken by U.S.EPA, other responsible parties, or the research or education
divisions of ATSDR. However, if there is an urgent health threat, ATSDR can issue a public health
advisory warning people of the danger. ATSDR can also authorize health education or pilot studies
of health effects, full-scale epidemiology studies, disease registries, surveillance studies or research
on specific hazardous substances.
Community: ATSDR also needs to learn what people in the area know about the site and what
concerns they may have about its impact on their health. Consequently, throughout the evaluation
process, ATSDR actively gathers information and comments from the people who live or work near
a site, including residents of the area, civic leaders, health professionals and community groups. To
ensure that the report responds to the community’s health concerns, an early version is also
distributed to the public for their comments. All the comments received from the public are
responded to in the final version of the report.
Comments: If, after reading this report, you have questions or comments, we encourage you to send
them to us.
Letters should be addressed as follows:
Agency for Toxic Substances and Disease Registry
ATTN: Records Center
1600 Clifton Road, NE (Mail Stop F-09)
Atlanta, GA 30333

Page | ii

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Contents
I.

Summary.............................................................................................................................1

I.A. Background ....................................................................................................................1

I.B.
Overall Conclusions.......................................................................................................4

I.C.
Conclusions for Past Mercury Exposure (1950–1990)..................................................5

I.D. Conclusions for Current Exposure (1990–2009) .........................................................11

II.

Background ......................................................................................................................18

II.A. Site Description............................................................................................................18

II.B. Operational History......................................................................................................20

II.C. Characteristics of Mercury...........................................................................................23

II.D. Remedial and Regulatory History................................................................................26

II.E. Site Geology/Hydrogeology ........................................................................................33

II.E.1. Bear Creek and Upper East Fork Poplar Creek Watersheds.......................33

II.F. Land Use and Natural Resources .................................................................................36

II.G. Demographics ..............................................................................................................39

II.G.1. Counties within the Y-12 Mercury Releases Study Area ..............................41

II.G.2. Cities within the Y-12 Mercury Releases Study Area ...................................42

II.H. Summary of Public Health Activities Pertaining to Y-12 Mercury Releases..............47

II.H.1. ATSDR...........................................................................................................47

II.H.2. TDOH............................................................................................................49

II.H.3. Florida Agricultural and Mechanical University (FAMU) ..........................51

II.H.4. U.S.EPA ........................................................................................................51

II.H.5. DOE ..............................................................................................................51

III.

Evaluation of Environmental Contamination and Potential Exposure Pathways.....55

III.A. Introduction..................................................................................................................55

III.B. Evaluation of Past (1950–1990) Mercury Exposure Pathways ...................................55

III.B.1. The Oak Ridge Dose Reconstruction Project ...............................................57

III.B.2. ATSDR’s Technical Review of the Task 2 Report.........................................58

III.C. Evaluation of Current (1990–2009) Mercury Exposure Pathways..............................59

III.C.1. Exposure Evaluation.....................................................................................59

III.C.2. Evaluating Exposures ...................................................................................63

III.C.3. Comparing Estimated Doses to Health Guidelines ......................................64

IV.

Public Health Evaluation ................................................................................................72

IV.A. Past Exposure (1950–1990) .........................................................................................72

IV.A.1. Potentially Exposed Communities ................................................................72

IV.A.2. Past Air Exposure Pathway ..........................................................................75

IV.A.3. Past Surface Water Exposure Pathway ........................................................82

IV.A.4. Past Soil and Sediment Exposure Pathways.................................................89

IV.A.5. Mercury in Fish.............................................................................................98

IV.A.6. Mercury in Local Produce ..........................................................................113

IV.B. Current Exposure (1990–2009)..................................................................................115

IV.B.1. Current Exposure Pathways .......................................................................115

IV.B.2. Current Air Exposure Pathway (elemental mercury) .................................115

IV.B.3. Current Surface Water Exposure Pathway (inorganic mercury) ...............116

IV.B.4. Current Groundwater Exposure Pathway ..................................................122

Page | iii

IV.B.5. Current Soil Exposure Pathway (inorganic mercury) ................................122

IV.B.6. Current Sediment Exposure Pathway (inorganic mercury)........................125

IV.B.7. Current Biota Exposure Pathway ...............................................................131

V.

Health Outcome Data Evaluation.................................................................................144

VI.

Community Health Concerns .......................................................................................147

VII.

Child Health Considerations.........................................................................................171

VIII. Conclusions and Recommendations.............................................................................176

IX.

Public Health Action Plan .............................................................................................184

X.

Preparers of Report .......................................................................................................186

XI.

References.......................................................................................................................187

List of Appendices
Appendix A. ATSDR Glossary of Terms .................................................................................. A-1

Appendix B. Summary of Other Public Health Activities ..........................................................B-1

Appendix C. Summary Briefs and Factsheets .............................................................................C-1

Appendix D. Toxicologic Implications of Mercury Exposure .................................................. D-1

Appendix E. Task 2 Pathway Discussions...................................................................................E-1

Appendix F. Evaluation of Mercury Emissions from Selected Electricity Generating FacilitiesF-1

Appendix G. Past Exposure Pathway Parameters....................................................................... G-1

Appendix H. What You Need to Know About Mercury in Fish and Shellfish .......................... H-1

Appendix I. Peer Reviewer Comments and ATSDR Responses..................................................I-1

Appendix J. Responses to Public Comments................................................................................J-1

Page | iv

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

List of Tables
Table 1. Summary of Selected Remedies, Monitoring, and Stewardship Requirements ..............31

Table 2. Populations of Anderson, Roane, Rhea, and Meigs Counties from 1940 to 2000 ..........41

Table 3. Population of Oak Ridge from 1942 to 2000...................................................................43

Table 4. Population of Harriman, Kingston, Rockwood, and Spring City from 1940 to 2000 .....46

Table 5. 1977 and 1983 Mercury Material Balance Estimates by Y-12 Plant Staff......................57

Table 6. Comparison Values for Mercury .....................................................................................64

Table 7. Health Guidelines for the Forms of Mercury...................................................................68

Table 8. Task 2 Exposure Pathways for Which Mercury Doses were Estimated for Each

Potentially Exposed Community .................................................................................73

Table 9. Three Task 2 Air Models and Potentially Exposed Communities...................................78

Table 10. Estimated Y-12 Mercury Releases to Water..................................................................83

Table 11. Maximum Mercury Concentrations Detected in EFPC Floodplain Soil .......................92

Table 12. Mercury1 Concentrations in Fish Collected Downstream of the Y-12 Plant.................99

Table 13. Methylmercury Exposure Doses from Fish Collected Downstream of the Y-12

Plant ...........................................................................................................................101

Table 14. Types of Local Produce Tested for Mercury ...............................................................113

Table 15. Mercury Concentrations in Locally Grown Produce...................................................114

Table 16. Current Exposure Pathways Evaluated........................................................................115

Table 17. Mercury Concentrations in EFPC Surface Water........................................................117

Table 18. Inorganic Mercury Concentrations in Oak Ridge Surface Water................................119

Table 19. Mercury Concentrations in LWBR Surface Water......................................................121

Table 20. Mercury Concentrations in Oak Ridge Soil.................................................................124

Table 21. Mercury Concentrations in EFPC Sediment................................................................126

Table 22. Mercury Concentrations in Oak Ridge Sediment ........................................................127

Table 23. Mercury Concentrations in LWBR Sediment..............................................................130

Table 24. Mercury Concentrations in Fish from EFPC ...............................................................132

Table 25. Estimated Methylmercury Exposure Doses from Consuming EFPC Fish..................133

Table 26. Mercury Concentrations in Edible Plants from EFPC.................................................136

Table 27. Estimated Inorganic Mercury Exposure Doses from EFPC Vegetable Consumption 137

Table 28. Mercury Concentrations in Fish and Turtles from LWBR ..........................................140

Table 29. Estimated Methylmercury Exposure Doses for LWBR Fish and Turtles ...................141

Table 30. Community Health Concerns from the ORR Community Health Concerns

Database.....................................................................................................................149

Table 31. Estimated Inorganic Mercury Exposure Doses for Pica Children ...............................175

Page | v

List of Figures
Figure 1. Location of the Oak Ridge Reservation .........................................................................19

Figure 2. Y-12 Facility Time Line.................................................................................................22

Figure 3. Characterization of Mercury Cycling.............................................................................25

Figure 4. Mercury Concentrations at the Confluence of Upper EFPC and Lower EFPC .............30

Figure 5. Cross-sectional Diagram of Pine Ridge and Chestnut Ridge in the Y-12 Vicinity........35

Figure 6. Current Land Use Along EFPC......................................................................................38

Figure 7. Demographics for a 1-Mile and 3-Mile Radius of the Y 12 Plant .................................40

Figure 8. Population Distribution of Anderson, Roane, Rhea, and Meigs Counties from 1940

to 2000 .........................................................................................................................41

Figure 9. Surface Elevation for Scarboro ......................................................................................44

Figure 10. Population of Oak Ridge, Harriman, Kingston, Rockwood, and Spring City from

1940 to 2000 ................................................................................................................46

Figure 11. ATSDR Chemical Screening Process ..........................................................................61

Figure 12. Levels of Significant Exposure to Elemental Mercury ................................................69

Figure 13. Levels of Significant Exposure to Inorganic Mercury .................................................70

Figure 14. Levels of Significant Exposure to Organic Mercury....................................................71

Figure 15. Task 2 Potentially Exposed Communities....................................................................74

Figure 16. Task 2 Estimated Mercury Releases to Air from Y-12 Operations (1953–1962)........76

Figure 17. Task 2 Estimated Mercury Releases to EFPC..............................................................84

Figure 18. EFPC RI Sampling Strategy.........................................................................................91

Figure 19. Extent of Mercury Contamination in the EFPC Floodplain (prior to completion of

remediation in 1997) ....................................................................................................93

Figure 20. Extent of Mercury Contamination at the NOAA site (prior to completion of

remediation in 1997) ....................................................................................................94

Figure 21. Extent of Mercury Contamination at the Bruner site (prior to completion of

remediation in 1997) ....................................................................................................95

Figure 22. Past Estimated Methylmercury Exposure Doses from Eating EFPC Fish

Compared to Health Effect Levels and Health Guidelines........................................104

Figure 23. Past Estimated Methylmercury Exposure Doses from Eating Poplar Creek Fish

Compared to Health Effect Levels and Health Guidelines........................................106

Figure 24. Past Estimated Methylmercury Exposure Doses from Eating Clinch River Fish

Compared to Health Effect Levels and Health Guidelines........................................108

Figure 25. Past Estimated Methylmercury Exposure Doses from Eating Watts Bar Reservoir

Fish Compared to Health Effect Levels and Health Guidelines ................................109

Figure 26. Current Estimated Methylmercury Exposure Doses from Eating EFPC Fish and

Crayfish Compared to Health Effect Levels and Health Guidelines .........................135

Figure 27. Current Estimated Methylmercury Exposure Doses from Eating LWBR Fish and

Turtles Compared to Health Effect Levels and Health Guidelines............................143

Page | vi

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Acronyms and Abbreviations

A
AF
ALS
AOEC
AT
ATSDR
BMDL
BW
C
CDC
CEDR
CERCLA
CEW
CF
Cfs
Colex
CROET
CV
D
DARA
DGM
DHHS
DOE
ED
EF
EFPC
Elex
EMEG
ERG
FACA
FAMU
FDA
g/day
g/kg/day
GIS
IARC
IR
IRIS
kg
LOAEL
LTHA
LWBR
m3
MCLG
jglL

soil adhered
bioavailability factor
amyotrophic lateral sclerosis
Association of Occupational and Environmental Clinics
averaging time
Agency for Toxic Substances and Disease Registry
benchmark dose lower limit
body weight
concentration
Centers for Disease Control and Prevention
Comprehensive Epidemiologic Data Resource
Comprehensive Environmental Response, Compensation, and Liability Act
Clinton Engineer Works
conversion factor
cubic feet per second
column exchange
Community Reuse Organization of East Tennessee’s
comparison value
exposure dose
Disposal Area Remedial Action
dissolved gaseous mercury
U.S. Department of Health and Human Services
U.S. Department of Energy
exposure duration
exposure frequency
East Fork Poplar Creek
electrical exchange
environmental media evaluation guide
Eastern Research Group, Inc.
Federal Advisory Committee Act
Florida Agricultural and Mechanical University
U.S. Food and Drug Administration
grams per day
grams per kilogram per day
geographic information system
International Agency for Research on Cancer
intake rate
Integrated Risk Information System
kilogram
lowest-observed-adverse-effect level
lifetime health advisory
Lower Watts Bar Reservoir
cubic meter
maximum contaminant level goal
micrograms per liter
Page | vii

MGD
mg/day
mg/kg
mg/kg/day
mg/L
mg/m3
MRL
MS
NAS
NCEH
ND
NHANES
NIOSH
NOAA
NPL
NOAEL
ORAU
Orex
OREIS
ORHASP
ORR
ORRHES
OU
PCB
PDF
PEL
PHAWG
ppb
ppm
ppt
RCRA
RfD
RI
RI/FS
RMEG
ROD
RSL
SAIC
SARA
TDEC
TDOH
TSCA
TVA
UEFPC
UF
U.S.EPA
USGS

million gallons per day
milligrams per day
milligrams per kilogram
milligrams per kilogram per day
milligrams per liter
milligrams per cubic meter
minimal risk level
multiple sclerosis
National Academy of Sciences
National Center for Environmental Health
not detected
National Health and Nutrition Examination Survey
National Institute for Occupational Safety and Health
National Oceanic and Atmospheric Administration
National Priorities List
no-observed-adverse-effect level
Oak Ridge Associated Universities
organic exchange
Oak Ridge Environmental Information System
Oak Ridge Health Agreement Steering Panel
Oak Ridge Reservation
Oak Ridge Reservation Health Effects Subcommittee
operable unit
polychlorinated biphenyl
probability density function
permissible exposure limit
Public Health Assessment Work Group
parts per billion
parts per million
parts per trillion
Resource Conservation and Recovery Act
reference dose
Remedial Investigation
Remedial Investigation and Feasibility Study
reference dose media evaluation guide
Record of Decision
regional screening level
Science Applications International Corporation
Superfund Amendments and Reauthorization Act
Tennessee Department of Environment and Conservation
Tennessee Department of Health
Toxic Substances Control Act
Tennessee Valley Authority
Upper East Fork Poplar Creek
Uncertainty factor
U.S. Environmental Protection Agency
U.S. Geological Survey
Page | viii

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

VIS
VOC
X

vertical integration study
volatile organic compound
chi

Page | ix

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

I.

Summary

I.A.

Background

Introduction

The Agency for Toxic Substances and Disease Registry (ATSDR) recognizes
you want to know more about past and current exposures to mercury released
from the Y-12 plant at the Oak Ridge Reservation (ORR). We intend that this
public health assessment will provide you with the information you need to
protect your health.

Mercury in
the
environment

Mercury occurs naturally in the environment. It occurs in three forms:
elemental mercury (also referred to as metallic mercury), inorganic mercury,
and organic mercury. The form of mercury can change when combined with
certain microorganisms (e.g., bacteria, fungi) or natural environmental
processes. How you are potentially exposed and harmed by mercury depends
on the form of mercury to which you are exposed.

How you are
exposed to
mercury

The following table identifies the main exposure pathways for the three forms
of mercury.
Mercury type
Elemental
mercury

Exposure pathway
Breathing in air.
About 80% of elemental mercury enters your
bloodstream directly from your lungs, and then
rapidly spreads to other parts of your body, including
the brain and kidneys (ATSDR 1999). The primary
health concerns are nervous system and kidney
effects.

Inorganic mercury

Eating soil, sediment, surface water, or plants.
Typically, less than 10% is absorbed through the
stomach and intestines, but it has been reported that
up to 40% can be absorbed (ATSDR 1999).
Inorganic mercury enters the bloodstream and moves
to many different tissues, but will mostly accumulate
in the kidneys. The primary health concern is kidney
effects.

Organic mercury
(methylmercury)

Eating contaminated fish.
Organic mercury is readily absorbed in the
gastrointestinal tract (about 95% absorbed) and can
easily enter the bloodstream (ATSDR 1999). It
moves rapidly to various tissues including the brain.
Effects on the developing nervous system in children
are the primary health concerns.
Page | 1

ORR history In 1942, the federal government established the ORR in Tennessee’s
Anderson and Roane Counties. The ORR was part of the Manhattan Project
to research, develop, and produce special nuclear materials for nuclear
weapons. Over the years, ORR operations generated a variety of radioactive
and nonradioactive wastes. These wastes were released into the environment.
In 1989, the U.S. Environmental Protection Agency (U.S.EPA) added the
ORR to the National Priorities List. The U.S. Department of Energy (DOE) is
cleaning up the ORR under a Federal Facility Agreement with U.S.EPA and
the Tennessee Department of Environment and Conservation (TDEC).

Tennessee
Department of
Health
involvement

The Tennessee Department of Health (TDOH) conducted the Oak Ridge
Health Studies (1991–1999) to evaluate whether off-site populations were
exposed in the past. The Oak Ridge Health Studies focused reconstructing the
exposure doses of individuals to contaminants released from the beginning of
the DOE facility operations in 1943 until 1990.

ATSDR’s
involvement

ATSDR is the principal federal public health agency charged with evaluating
human health effects of exposure to hazardous substances in the environment.
Since 1992, ATSDR has worked to determine whether levels of
environmental contamination at and near the ORR present a public health
hazard to surrounding communities. ATSDR has identified and evaluated
several public health issues and has worked closely with many parties.
ATSDR has responded to requests and addressed health concerns of
community members, civic organizations, and other government agencies
surrounding ORR. ATSDR’s public health activities in the 1990s addressed
current public health issues related to Superfund cleanup activities at two offsite areas affected by ORR operations—the East Fork Poplar Creek (EFPC)
area and the Watts Bar Reservoir area.
Beginning in 2000, ATSDR initiated the formal public health assessment
process for the ORR when results of TDOH’s Oak Ridge Health Studies were
available and the Oak Ridge Reservation Health Effects Subcommittee
(ORRHES) had been established by the Centers for Disease Control and
Prevention (CDC) and ATSDR. To build upon their effort s, ATSDR
scientists reviewed and analyzed the Oak Ridge Health Studies Phase I and
Phase II screening-level evaluations of past exposure (1944 to 1990) and the
Phase II dose reconstruction reports to identify contaminants of concern
requiring further public health evaluation. ATSDR has since completed nine
chemical-specific and issue-specific public health assessments on releases of
hazardous substances requiring further public health evaluation and public
health issues of concern to the community. ATSDR scientists completed
public health assessments on uranium releases from the Y-12 plant (ATSDR
Page | 2

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

2004), radionuclide releases from White Oak Creek (ATSDR 2006a), iodine
131 releases from the X-10 site (ATSDR 2008), ORR-wide polychlorinated
biphenyl (PCB) releases (ATSDR 2009), uranium and fluoride releases from
the K-25 site (ATSDR 2010), and other topics such as contaminant releases
from the Toxic Substances Control Act (TSCA) incinerator (ATSDR 2005a)
and contaminated off-site groundwater (ATSDR 2006b). In 2007, ATSDR
screened current (1990 to 2003) environmental data to identify any other
chemicals that required further evaluation (ATSDR 2007).
In conducting its public health assessments, ATSDR scientists evaluated and
analyzed the information and findings from previous studies and
investigations. ATSDR uses the public health assessment process to evaluate
potential public health impacts of past, current, and future exposures to
environmental contamination at Superfund sites. The public health
assessment process serves as a mechanism for identifying appropriate followup public health actions for particular communities. The process also serves
as a mechanism through which the agency responds to specific community
health concerns related to hazardous waste sites.

Scope

In this public health assessment, ATSDR evaluates past (1950–1990) and
current (1990–2009) exposure to mercury released from the Y-12 plant to
determine whether exposure-related health effects were possible in off-site
residents. ATSDR evaluated potential residential exposures from 1950 to
2009 to three forms of mercury: elemental mercury, inorganic mercury, and
organic mercury. ATSDR evaluated potential exposures to Y-12 plant-related
mercury in air, soil, surface water, sediment, fish, crayfish, turtles, and
produce. The agency evaluated seven communities that were the most likely
to have been affected by Y-12 mercury releases. The studied population
included people who lived in the city of Oak Ridge, the Scarboro
neighborhood, or Wolf Valley, as well as people who lived or recreated in or
along the EFPC floodplain, Poplar Creek, Clinch River, or the Watts Bar
Reservoir.

Page | 3

I.B.

Overall Conclusions

Conclusions

Most past and current exposure pathways are not a public health hazard.
However, ATSDR identified a few pathways of potential concern.
• Family members (especially young children) may have inhaled elemental
mercury carried from the Y-12 plant by workers into their homes.
• Children who swallowed water while playing in East Fork Poplar Creek
(EFPC) during some weeks from 1956 to 1958, and adults who incidentally
swallowed water during some weeks in 1958, possibly could have been
exposed to levels of inorganic mercury that may have increased the risk of
developing renal (kidney) health effects.
• Children who accidentally swallowed soil while playing in two areas along
the EFPC floodplain before the removal of mercury-contaminated soil in
1996 and 1997, possibly could have been exposed to inorganic mercury
that may have increased the risk of developing renal (kidney) health
effects.
• Children born to or nursing from women who ate fish from waterways near
the ORR may have a small increased risk of developing subtle
neurodevelopmental health effects from exposure to organic mercury. For
this small increased risk to occur, mothers had to eat fish frequently just
before and during pregnancy, or while nursing. Also, children who ate fish
from waterways near the ORR may have a small increased risk of
developing subtle neurodevelopmental health effects.
Due to a lack of information, ATSDR cannot determine whether people living
off site could have been harmed from breathing elemental mercury from 1950
through 1963, swallowing water containing inorganic mercury from EFPC
from 1953 to 1955, and eating fish containing mercury during the 1950s and
1960s.

Page | 4

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

I.C.

Conclusions for Past Mercury Exposure (1950–1990)

Past exposure
to mercury in
the air

ATSDR concludes
• In the past (1950–1963), elemental mercury carried from the Y-12 plant by
workers into their homes could potentially have harmed their families
(especially young children), but ATSDR has no quantitative data to
evaluate the magnitude of this hazard.
• People living in the Wolf Valley area were not harmed from breathing
elemental mercury released from the Y-12 plant.
The highest annual concentration was more than 14 times lower than

ATSDR’s health guideline for elemental mercury vapor.

• After 1963, the elemental mercury released to the air from the Y-12 plant
and elemental mercury vapors released from the East Fork Poplar Creek
(EFPC) water did not harm people living off site near the ORR.
No estimated air mercury concentrations for any potentially exposed
community for any year exceeded ATSDR’s health guideline for elemental
mercury vapor.
ATSDR cannot conclude
• Whether people living off site in Oak Ridge, Scarboro, and along the EFPC
floodplain, who in the past breathed elemental mercury released to the air
from the Y-12 plant from 1950 through 1963, could have been harmed.
• Whether people living near the EFPC floodplain, who breathed elemental
mercury vapors released from the EFPC water from 1950 through 1963,
could have been harmed.

Past exposure
to mercury
from East
Fork Poplar
Creek (EFPC)
surface water

ATSDR concludes
• Children who swallowed water while playing in EFPC for a short period
(acute exposure: fewer than 2 weeks) during some weeks in 1956, 1957,
and 1958 may have an increased risk of developing renal (kidney) effects
from exposure to inorganic mercury.
The estimated exposure doses for some weeks in 1956, 1957, and 1958
were higher than ATSDR’s health guidelines (i.e., MRLs) and U.S.EPA’s
health guideline (i.e., RfD) for inorganic mercury.
• Adults who swallowed water from EFPC for a short time during some
weeks in 1958 may have an increased risk of developing renal (kidney)
effects from exposure to inorganic mercury.
The estimated exposure doses for some weeks in 1958 were higher than
ATSDR’s and U.S.EPA’s health guidelines for inorganic mercury.

Page | 5

• People who swallowed water from EFPC for a short time before 1953 or
after the summer of 1958 were not harmed from exposure to inorganic
mercury.
The estimated exposure doses were lower than ATSDR’s and U.S.EPA’s
health guidelines for inorganic mercury.
• People who swallowed water from EFPC over a longer period of time
(intermediate and chronic exposures: more than 2 weeks) were not harmed
from exposure to inorganic mercury.
The estimated exposure doses were lower than ATSDR’s and U.S.EPA’s
health guidelines for inorganic mercury.
• People who swallowed water from EFPC were not harmed from exposure
to methylmercury.
The estimated exposure doses were lower than ATSDR’s and U.S.EPA’s
health guidelines for organic mercury.
ATSDR cannot conclude
• Whether people who swallowed water from EFPC for a short time during
1953, 1954, and 1955 could have been harmed from exposure to inorganic
mercury.

Past exposure
to mercury
from EFPC
soil and
sediment

ATSDR concludes
• Children, who played in the EFPC floodplain at the National Oceanic and
Atmospheric Administration (NOAA) site and Bruner site before soil
removal activities in 1996 and 1997, may have accidentally swallowed
inorganic mercury in soil that may have increased the risk of developing
renal (kidney) effects.
The estimated exposure doses exceeded ATSDR’s health guidelines for
inorganic mercury.
• Adults are not expected to have been harmed from inorganic mercury in the
soil at the NOAA and Bruner sites before soil removal activities in 1996 and
1997.
The estimated exposure doses were below ATSDR’s health guidelines for
inorganic mercury.
• People who contacted EFPC floodplain soils in the past were not harmed
from exposure to methylmercury.
The estimated exposure doses were below ATSDR’s health guideline for
organic mercury.

Page | 6

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Past exposure
to mercury
from EFPC
fish

ATSDR concludes
• Periodically eating fish from EFPC (up to nine meals per year for adults
and up to four meals per year for children 1) in the
Eating nine fish meals
1980s did not harm people’s health from exposure
per year is a worst case
to methylmercury, including children who ate fish,
assumption for this nonnursing infants whose mothers ate fish, and fetal
productive fishing area.
exposure from mothers who ate fish.
The estimated methylmercury exposure doses were below ATSDR’s and
U.S.EPA’s health guidelines.
ATSDR cannot conclude
• Whether eating fish from EFPC during the 1950s, 1960s, and 1970s could
have harmed people’s health from exposure to methylmercury.
Note: Since the 1980s there has been a fish consumption advisory due to
mercury and PCB contaminated fish.

Past exposure
to mercury
from Poplar
Creek fish

ATSDR concludes
• Children born to or nursing from women who ate 12 fish meals per month
(i.e., the maximum consumption rate) from Poplar Creek in the 1970s,
1980s, and 1990 had an increased risk of subtle neurodevelopmental
effects from exposure to methylmercury.
The estimated methylmercury exposure doses came close to the
methylmercury dose identified by the National Academy of Sciences
(NAS) that resulted in a 5 percent increase in the incidence of abnormal
scores on the Boston Naming Test in the Faroe Islands study. The NAS
health effect level is consistent with the range identified as the benchmark
dose lower limit (BMDL05) by the U.S.EPA in the Faroe Islands study.
• Children who ate up to six meals a month (i.e., the maximum consumption
rate) of Poplar Creek fish in the 1970s, 1980s, and 1990 had an increased
risk of subtle neurodevelopmental effects.
The estimated methylmercury doses came close to the NAS health effect
level, which is associated with subtle neurodevelopmental effects.
• Children born to or nursing from women who ate approximately three
meals a month (i.e., the average consumption rate) of Poplar Creek fish in
the 1970s, 1980s, and 1990 had a small increased risk of subtle
neurodevelopmental effects. Also, children who ate about 1.5 meals a
month (i.e., the average consumption rate) of Poplar Creek fish had a small
increased risk of neurodevelopmental effects.

1

Appendix G contains detailed information on how the intake rates were derived for fish obtained from each of the
surface water bodies evaluated: EFPC, Poplar Creek, Clinch River, and Watts Bar Reservoir.
Page | 7

A few estimated methylmercury exposure doses were only slightly above
ATSDR’s and U.S.EPA’s health guidelines for methylmercury and were
not close to the NAS health effect level, which is associated with subtle
neurodevelopmental effects.
ATSDR cannot conclude
• Whether eating fish from Poplar Creek during the 1950s and 1960s could
have harmed people’s health from methylmercury exposure.
Note: Since the 1980s there has been a fish consumption advisory due to PCB
contaminated fish.

Past exposure
to mercury
from Clinch
River fish

ATSDR concludes
• Children born to or nursing from women who ate 12 fish meals per month
(three fish meals a week) (i.e., the maximum consumption rate) from the
Clinch River in the 1970s, 1980s, and 1990 had a small increased risk of
subtle neurodevelopmental effects.
The estimated methylmercury exposure doses are only slightly above
ATSDR’s and U.S.EPA’s health guidelines for methylmercury and were
not close to the NAS health effect level, which is associated with subtle
neurodevelopmental effects.
• Children who ate approximately six fish meals a month (i.e., the maximum
consumption rate) from the Clinch River in the 1970s, 1980s, and 1990 had
a small increased risk of subtle neurodevelopmental effects.
The estimated methylmercury exposure doses were only slightly above
ATSDR’s and U.S.EPA’s health guidelines for methylmercury and were
not close to the NAS health effect level, which is associated with subtle
neurodevelopmental effects.
• Children born to or nursing from women who ate up to three Clinch River
fish meals per month (i.e., the average consumption rate) were not harmed
from exposure to methylmercury.
The estimated exposure doses were below ATSDR’s and U.S.EPA’s health
guidelines.
• Children who ate less than two Clinch River fish meals a month (i.e., the
average consumption rate) were not at risk of harmful neurodevelopmental
effects.
The estimated exposure doses were below ATSDR’s and U.S.EPA’s health
guidelines.

Page | 8

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

ATSDR cannot conclude
• Whether eating fish from Clinch River during the 1950s and 1960s could
have harmed people’s health.
Note: Since the1980s there has been a fish consumption advisory due to PCB
contaminated fish.

Past exposure
to mercury
from Watts
Bar Reservoir
fish

ATSDR concludes
• Children born to or nursing from women who ate 20 fish meals per month
(i.e., the maximum consumption rate) (5 fish meals a week) from Watts
Bar Reservoir in the 1980s and 1990 had a small increased risk of subtle
neurodevelopmental effects.
The estimated exposure doses were only slightly above U.S.EPA’s health
guideline and were not close to the NAS health effect level, which is
associated with subtle neurodevelopmental effects.
• Children who ate approximately 10 fish meals a month (i.e., the maximum
consumption rate) from Watts Bar Reservoir in the 1980s and 1990 had a
small increased risk of subtle neurodevelopmental effects.
The estimated exposure doses were only slightly above U.S.EPA’s health
guideline and were not close to the NAS health effect level, which is
associated with subtle neurodevelopmental effects.
• Children born to or nursing from women who ate up to five Watts Bar
Reservoir fish meals per month (i.e., the average consumption rate) were
not harmed from exposure to methylmercury.
The estimated exposure doses were below ATSDR’s and U.S.EPA’s health
guidelines.
• Children who ate less than three Watts Bar Reservoir fish meals a month
(i.e., the average consumption rate) were not at risk of harmful
neurodevelopmental effects.
The estimated exposure doses were below ATSDR’s and U.S.EPA’s health
guidelines.
ATSDR cannot conclude
• Whether eating fish from Watts Bar Reservoir during the 1950s, 1960s,
and 1970s could have harmed people’s health.
Note: Since the1980s there has been a fish consumption advisory due to PCB
contaminated fish.

Page | 9

Past exposure
to mercury
from edible
plants

ATSDR concludes
• People who ate local produce grown in gardens in the EFPC floodplain or
in private gardens that contained mercury-contaminated soils from the
floodplain were not harmed from exposure to inorganic mercury.
The estimated exposure doses for children and adults were below

ATSDR’s health guidelines for inorganic mercury.

Page | 10

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

I.D.

Conclusions for Current Exposure (1990–2009)

Current
exposure to
mercury from
East Fork
Poplar Creek
(EFPC) air

ATSDR concludes

Current
exposure to
mercury from
Lower Watts
Bar Reservoir
(LWBR) air

ATSDR concludes

Current
exposure to
mercury from
EFPC surface
water

ATSDR concludes

• People who breathe the air near the EFPC floodplain are not being harmed
from exposure to mercury.
The concentrations of mercury in all of the EFPC ambient air samples
(collected near the areas with the highest levels of mercury contamination)
are below the ATSDR comparison value for elemental mercury in air.

• People who breathe the air near LWBR are not being harmed from
exposure to mercury.
Even though no Lower Watts Bar Reservoir (LWBR) ambient air samples
have been analyzed for mercury concentrations, the occurrence of harmful
health effects from exposure to mercury vapor from contaminated soil is
not a concern for the LWBR. The mercury contamination accumulated in
the sediments of the river channel and is now buried under cleaner
sediment and several meters of water. Additionally, the near-shore
sediment concentrations in the LWBR are much lower than those found in
the EFPC floodplain.

• Children who swallow surface water while playing in EFPC are not being
harmed from exposure to inorganic mercury. However, there is a bacterial
advisory warning people to avoid contact with the water.
Only one EFPC surface water concentration of mercury was detected
slightly above the U.S.EPA’s maximum contaminant level goal (MCLG)
for inorganic mercury. To assess the exposure further, ATSDR evaluated
two scenarios: 1) a farm family member’s exposure, and 2), a child’s
exposure if the bacterial advisory to avoid contact with the water is
ignored. The calculated mercury exposure doses for both scenarios are
below U.S.EPA’s health guideline value for chronic exposure.

Page | 11

Current
exposure to
mercury from
Oak Ridge
surface water

ATSDR concludes

Current
exposure to
mercury from
Scarboro
surface water

ATSDR concludes

Current
exposure to
mercury from
LWBR
surface water

ATSDR concludes

Current
exposure to
mercury from
EFPC soil

ATSDR concludes

• People who incidentally swallow surface water from Oak Ridge are not
being harmed from exposure to inorganic mercury.
Only one concentration of mercury in Oak Ridge surface water was higher
than U.S.EPA’s MCLG. To evaluate the exposure further, ATSDR
calculated exposure doses for adults and children using the maximum
concentration detected in Oak Ridge surface water. Both estimated doses
are below the U.S.EPA’s health guideline for chronic exposure.

• Children who swallow surface water while playing in ditches in Scarboro
are not being harmed from exposure to inorganic mercury.
Mercury has not been detected in any surface water samples collected from
the Scarboro community.

• People who incidentally swallow surface water from LWBR are not being
harmed from exposure to inorganic mercury.
All of the LWBR surface water samples are below U.S.EPA’s MCLG for
inorganic mercury.

• Children, who played in the EFPC floodplain at the NOAA and Bruner
sites before soil removal activities in 1996 and 1997, may have
accidentally swallowed inorganic mercury in soil that may have increased
the risk of developing renal (kidney) effects.
The estimated exposure doses exceeded ATSDR’s health guidelines for
inorganic mercury.
• Adults are not expected to have been harmed from the EFPC floodplain
soil at the NOAA and Bruner sites before removal activities in 1996 and
1997.
The estimated exposure doses were below ATSDR’s health guidelines.
• People who come in contact with EFPC floodplain soil after cleanup
activities are not being harmed from exposure to mercury.

Page | 12

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Floodplain soils with concentrations greater than 400 ppm of mercury were
removed in 1996 and 1997. ATSDR evaluated exposure to floodplain soils
with up to 400 ppm of mercury and determined that this clean-up level is
safe.

Current
ATSDR concludes
exposure to
• People who come in contact with Oak Ridge soil are not being harmed
mercury from
from exposure to mercury.
Oak Ridge soil
Some of the concentrations of inorganic mercury in Oak Ridge soil are
higher than ATSDR’s comparison value. To evaluate the exposure further,
ATSDR calculated exposure doses for adults and children using the
maximum inorganic mercury concentration detected in Oak Ridge soil.
Both estimated doses are well below health effect levels.

Current
exposure to
mercury from
Scarboro soil

ATSDR concludes
• People who contact Scarboro soil are not being harmed from exposure to
inorganic mercury.
All of the surface soil samples collected in Scarboro are below ATSDR’s
comparison value for inorganic mercury.

Current
exposure to
mercury from
LWBR soil

ATSDR concludes
• People who contact soil near the LWBR are not being harmed from
exposure to inorganic mercury.
No soil samples have been collected from the LWBR, but the occurrence of
harmful health effects from exposure to mercury in soil along the LWBR
shoreline is not a concern. ORR operations have not contaminated the soil
near LWBR with mercury. The mercury that ORR released into EFPC was
transported to the LWBR through Poplar Creek and the Clinch River. That
mercury accumulated in the sediments of the LWBR deep river channel,
but it was buried under cleaner sediment. Potential exposure (ingestion,
inhalation, and dermal contact) to mercury concentrations in these
subsurface sediments does not pose a health concern even if these deep
channel sediments were removed and used as surface soil on residential
properties. Additionally, the near-shore sediment mercury concentrations in
the LWBR are much lower than the comparison value for mercury in soil.

Page | 13

Current
exposure to
mercury from
EFPC
sediment

ATSDR concludes

Current
exposure to
mercury from
Oak Ridge
sediment

ATSDR concludes

Current
exposure to
mercury from
Scarboro
sediment

ATSDR concludes

Current
exposure to
mercury from
LWBR
sediment

ATSDR concludes

• People who contact EFPC sediment are not being harmed from exposure to
inorganic mercury.
Some of the concentrations of mercury in EFPC sediment are higher than
ATSDR’s comparison value for inorganic mercury. Thus to assess the
exposure further, ATSDR evaluated two scenarios: 1) a farm family
member’s exposure, and 2) a child's exposure if the bacterial advisory
warning signs are ignored. The estimated mercury exposure doses for both
scenarios are below the U.S.EPA’s health guideline value for chronic
exposure to inorganic mercury.

• People who contact Oak Ridge sediment are not being harmed from
exposure to inorganic mercury.
Some of the concentrations of mercury in Oak Ridge sediment are higher
than ATSDR’s comparison value for inorganic mercury. To evaluate the
exposure further, ATSDR calculated exposure doses for adults and children
using the maximum concentration detected in Oak Ridge sediment. Both
the estimated doses are below U.S.EPA’s health guideline value for
chronic exposure to inorganic mercury.

• People who contact Scarboro sediment are not being harmed from
exposure to inorganic mercury.
The levels of mercury in all of the sediment samples collected in Scarboro
are below ATSDR’s comparison value for inorganic mercury.

• People who contact LWBR sediment are not being harmed from exposure
to inorganic mercury.
All of the near-shore sediment samples and deep-water sediment samples
collected from the LWBR are below ATSDR’s comparison value. Still, a
few concentrations of mercury in unspecified depth sediment samples are
higher than the comparison value. To evaluate further the exposure to
sediment, ATSDR calculated exposure doses for adults and children using
the maximum concentration detected in LWBR sediment from unspecified
depths. Both the estimated doses are below the U.S.EPA’s health guideline
value for chronic exposure to inorganic mercury.
Page | 14

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

To prevent unnecessary exposures to workers and the public, ATSDR
cautions that the sediments should not be disturbed, removed, or disposed
of without careful review by the interagency working group.

Current
ATSDR concludes
exposure to
• Children born to or nursing from women who ignore the posted warning
mercury from
signs and eat one meal of fish caught from EFPC a month are not at risk of
EFPC fish and
being harmed from exposure to methylmercury. However, eating one or
shellfish
more crayfish meals a month from the
EFPC is not a productive fishing location,
EFPC floodplain increases the risk of
and a fish consumption advisory is in
subtle neurodevelopmental effects.
place. That anyone is actually eating fish
from EFPC is unlikely. The Tennessee
The estimated methylmercury
fish advisories are available at
exposure doses for eating fish are at or
http://www.tn.gov/environment/wpc/public
below ATSDR’s and U.S.EPA’s
ations/pdf/advisories.pdf.
health guidelines. The estimated
methylmercury exposure dose for eating crayfish is slightly above the
health guidelines but is not close to the NAS health effect level, which is
associated with subtle neurodevelopmental effects.

• Children who ignore the posted warning signs and eat one meal of EFPC
fish a month have a small increased risk of subtle neurodevelopmental
effects. Eating one or more crayfish meals a month from EFPC increases
that risk.
The estimated methylmercury exposure doses for eating fish are slightly
above the U.S.EPA’s health guideline but are not close to the NAS health
effect level, which is associated with subtle neurodevelopmental effects.
The estimated methylmercury exposure dose for eating crayfish comes
close to the NAS health effect level.
ATSDR recommends
• Children, pregnant women, and nursing mothers follow the fish
consumption advisory for EFPC.

Current
exposure to
mercury from
LWBR fish

ATSDR concludes
• Adults and children who eat one LWBR fish meal a month are not at risk
of developing harmful effects.
The estimated methylmercury exposure doses are below ATSDR’s and
U.S.EPA’s health guidelines.
• Children who eat fish from LWBR once a week have a small increased risk
of subtle neurodevelopmental effects.

Page | 15

The estimated methylmercury
exposure doses are slightly above
ATSDR’s and U.S.EPA’s health
guidelines but are not close to the
NAS health effect level, which is
associated with subtle
neurodevelopmental effects.

People frequently fish in LWBR. But since
1987, fishing advisories have warned people
to avoid or limit their consumption of fish due
to PCB contamination in the reservoir. ATSDR
evaluated three potential exposure scenarios:
1) adults and children eating one fish meal
with the average concentration of mercury
each month, 2) adults and children eating one
fish meal with the average concentration of
mercury each week, and 3) adults eating
about two fish meals with the average
concentration of mercury each week.

• Children born to or nursing from
women who eat one or two meals
of largemouth bass or striped bass
from LWBR a week have a small
increased risk of subtle neurodevelopmental effects. Eating catfish and
sunfish once a week is a safer alternative for pregnant and nursing women.
The estimated methylmercury exposure doses for largemouth bass and
striped bass are slightly above the U.S.EPA’s health guideline but are not
close to the NAS health effect level, which is associated with subtle
neurodevelopmental effects.
• Adults and children who eat the edible portion of turtles from LWBR once
or twice a week have a small increased risk of subtle neurodevelopmental
effects.
The estimated methylmercury exposure doses are slightly above the
U.S.EPA’s health guideline but are not close to the NAS health effect level,
which is associated with subtle neurodevelopmental effects.
ATSDR recommends
• Children, pregnant women, and nursing mothers follow the fish
consumption advisory for LWBR.

Current
exposure to
mercury from
EFPC
Vegetables

ATSDR concludes
• People who eat beets, kale, or tomatoes grown in the EFPC floodplain are
not being harmed from exposure to inorganic mercury.
Comparison values are not available for screening concentrations detected
in edible plants. Thus ATSDR used average concentrations to calculate the
estimated inorganic mercury exposure doses and evaluate exposure.
ATSDR found that the health effect levels available in the toxicological
and epidemiological literature are at least three orders of magnitude higher
than the estimated doses for adults and children eating vegetables grown in
EFPC gardens. Further, plants tend to store metals such as mercury in a
form that is not readily bioavailable to humans.

Page | 16

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Current
exposure to
mercury from
Oak Ridge
vegetables

ATSDR concludes

For more
information

Call ATSDR toll-free at 1-800-CDC-INFO if you have questions or
comments. Ask for information on the Oak Ridge Reservation site. Detailed
information about the toxicology of mercury is also available in ATSDR’s
Toxicological Profile for Mercury at
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24.

• People who eat vegetables from Oak Ridge are not being harmed from
exposure to inorganic mercury.
Within the city of Oak Ridge only four vegetable samples from one garden
were collected and analyzed for mercury. Mercury was not detected in any
of the samples.

Page | 17

II.

Background

II.A. Site Description
The Oak Ridge Reservation (ORR) is a U.S. Department of Energy (DOE) facility situated on
more than 34,000 acres in Anderson and Roane Counties in East Tennessee (Figure 1). The
Clinch River forms the southern and western boundaries of the ORR, and Poplar Creek and East
Fork Poplar Creek (EFPC) drain the property to the north and west (DOE 1997a). The ORR was
originally part of the Clinton Engineer Works (CEW), which was established by the War
Department in 1943 2 as part of the Manhattan Project. The mission of the CEW was to research,
develop, and produce special nuclear materials for nuclear weapons (ChemRisk 1993a; TDOH
2000). Four facilities were built: the Y-12 plant, the K-25 site, and the S-50 site to enrich
uranium, and the X-10 site to demonstrate processes for producing and separating plutonium
(TDOH 2000).
When the federal government established the CEW, the reservation consisted of 58,575 acres.
After World War II, the federal government conveyed 24,340 of the original 58,575 acres to
various parties, including the city of Oak Ridge and the Tennessee Valley Authority (TVA)
(ORNL 2002). DOE continues to control the remaining 34,235 acres (Jacobs Engineering Group
1996; ORNL 2002). Most of the ORR property is within the Oak Ridge city limits (EUWG
1998).
The Y-12 plant is in the eastern end of Bear Creek Valley; it is bordered on the south by
Chestnut Ridge and on the north by Bear Creek Road and Pine Ridge (ChemRisk 1999a). The
825-acre Y-12 plant is within the present-day corporate limits of the city of Oak Ridge, about 2
miles south of downtown (ChemRisk 1999a). It is less than a half-mile from the Scarboro
community. But Pine Ridge, which rises to about 300 feet above the valley floor, separates the
Y-12 plant from the main residential areas of Oak Ridge and hinders the exchange of air between
the city and the Y-12 plant (U.S. Weather Bureau 1953). The main Y-12 production area is about
0.6 miles wide and 3.2 miles long and contains roughly 240 principal buildings (ChemRisk
1999b).

2

The Tennessee project was originally called the Kingston Demolition Range. Land was acquired, trees were
cleared, and construction began in the fall of 1942. The name Clinton Engineer Works was officially adopted in
early 1943.
Page | 18

Source: ChemRisk 1999a (with modifications)

Figure 1. Location of the Oak Ridge Reservation

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Page | 19

II.B.

Operational History

The first buildings at the Y-12 plant 3 were built in 1943. They were part of the Manhattan
Project’s production-scale separation of uranium isotopes for use in the first atomic bomb. In
1950, research and pilot operations began at Y-12 to identify a viable process for large-scale
production of enriched lithium for use in hydrogen bombs (ChemRisk 1999a). In 1952, the
facilities were converted to fabricate nuclear weapon components (ChemRisk 1999a). At the end
of the Cold War, the Y-12 missions were curtailed. In 1992, the major focus of the Y-12 plant
was the remanufacture of nuclear weapon components and dismantling and storage of strategic
nuclear materials from retired nuclear weapons systems. In October 2000, oversight of the Y-12
plant passed from the DOE Oak Ridge Operations to the DOE National Nuclear Security
Administration. The National Nuclear Security Administration currently uses the Y-12 National
Security Complex as the primary storage site for highly enriched uranium. See Figure 2 for a
time line of the major processes at the Y-12 plant.
In the early 1950s, the Y-12 plant began separating high-purity lithium 6 from natural lithium to
produce enriched lithium 6 deuteride for thermonuclear weapons (i.e.,
During the Colex
hydrogen bombs) (UCCND 1983a, 1983b). During pilot scale tests
process, lithium isotopes
conducted between 1950 and 1955, alternate processes to separate
were separated by
lithium isotopes were investigated at Y-12, including Orex (organic
transferring them
exchange), Elex (electrical exchange), and Colex (column exchange)
between a water-based
(ChemRisk 1999a). Colex was determined to be the most efficient
solution of lithium
hydroxide and a solution
process for enriching lithium (DOE 1993b). Two of these processes
of lithium in mercury.
(Elex and Colex) were used in full-scale production, and both processes
used large quantities of liquid mercury (Brooks and Southworth 2011;
ChemRisk 1999a). The Colex process used large quantities of mercury as an extraction solvent.
Production-level lithium isotopic separation using the Elex process began in August 1953 and
ended in 1957. Production using the Colex process began in January 1955 and ended in May
1963 (ChemRisk 1999a). After Colex production ended, the mercury was removed from the
process-related equipment and put into storage or sent back into the commercial marketplace
(Brooks and Southworth 2011).
These dates are important for assessing mercury releases from the ORR. By far, the highest offsite releases of mercury occurred during these production years. Pilot project investigations
resulted in mercury releases to the soil, air, and water before actual production. But those
releases were minor—the quantities of materials used were relatively small. And Y-12 mercury
releases after 1963 (after the Colex process shut down) came from secondary sources such as
mercury spills in buildings and onto soils, mercury rebottling operations, and stripping
operations, that is, clean up, tear down, and removal of production equipment. Overall, with the
exception of the production years, Y-12 plant post-production activities were only responsible
for relatively small mercury releases.

3

Because this public health assessment focuses on exposure to mercury released from the Y-12 plant, the other
main facilities on ORR are not discussed in detail.
Page | 20

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Throughout the 1950s and 1960s, six pilot plants, three production facilities, and several
auxiliary support facilities used about 24 million pounds of mercury during lithium separation
processes (DOE 1993b). Most of the substantial mercury
Three major efforts have estimated Y-12
losses to the environment occurred from 1955 to 1962
mercury releases to water and air:
during the Colex production scale operations (Brooks and
� In 1977, Y-12 personnel prepared a
Southworth 2011; ChemRisk 1999a). Mercury was also
classified report called the 1977
used in small quantities in several other operations at the
Mercury Inventory Report.
Y-12 plant, at the X-10 site, and at the K-25 site. Still, Task
� In the early 1980s, the Mercury Task
2 4 found either 1) no evidence that mercury was released
Force investigated what was known
about mercury use and releases at the
from those activities or 2) if it was, that the releases were
Y-12 plant.
insignificant—in fact, they were less than 1 percent of the
�
In
the 1990s, the Task 2 team revised
releases from the lithium isotope separation processes at Ythe previous estimates of mercury
12 (ChemRisk 1999a). In any event, production of enriched
releases.
lithium stopped in 1962 (Richmond and Auerbach 1983).
In total, about 73,000 pounds of mercury was released to the air, primarily through building
ventilation systems. These ventilation systems were installed in the lithium enrichment facilities
to lower the amount of mercury inhaled by the workers (ChemRisk 1999a), and about 280,000
pounds of mercury (or about 12 cubic yards) was also released to EFPC, largely from an early
process in which mercury was washed with nitric acid (ChemRisk 1999a).

4

Task 2 of the Reports of the Oak Ridge Dose Reconstruction, Mercury Releases from Lithium Enrichment at the
Oak Ridge Y-12 Plant—a Reconstruction of Historical Releases and Off-Site Doses and Health Risks (ChemRisk
1999a) (referred to as the “Task 2 report”) describes in greater detail the history of the lithium isotope separation
process at the Y-12 plant.
Page | 21

Figure 2. Y-12 Plant Time Line

MAJOR PROCESSES
Electromagnetic Separation of U-235, 1943-48
Uranium Chemical Processing and Parts Manufacturing, 1943-present
Disposal in Boneyard/Burnyard, 1944-72
Electromagnetic Separation of Stable Isotopes, 1947-90
ELEX & COLEX Separarting Process for Lithium Isotopes (Using Mercury), 1950-63
Production of Thorium Weapon Components, 1950-75
Production of Lithium and Beryllium Weapon Components, 1950-present
Waste Disposal in S-3 Ponds, 1951-82
Disposal in Bear Creek Burial Ground, 1954-92
Waste Disposal in New Hope Pond, 1963-88

ORR ENVIRONMENTAL MONITORING DATA
1947-48, Radioactivity, Flourine, Uranium in Clinch River, Poplar Creek
1950-present, Radioactivity, Mercury in EFPC, Bear Creek
1955-57, Mercury, Manganese in Clinch River, Poplar Creek, EFPC
1959-present, Radionuclides, Metals in Clinch River
1960-64, Radionuclides, Chemicals in Clinch River, Poplar Creek
1971-present, Uranium, Radionuclides, Metals in EFPC, Poplar Creek, Bear Creek

Surface
Water

1971-90, PCBs in Bear Creek
1983, Organics, Priority Pollutants in Bear Creek
1983, VOCs, PCBs, Metals in Bear Creek
1984, Metals, VOCs, Radioactivity, Radionuclides in Clinch River, EFPC
1984-86, Mercury, Organics, in Bear Creek
1985, Herbicides, Pesticides, PCBs in Bear Creek
1986, Cs-137 in Watts Bar Reservoir
1989-90, Metals, Organics, Radionuclides, PCBs, SVOCs, Pesticides, Tritium in Clinch River, Poplar Creek
1990, Metals, Organics, Radionuclides, in Melton Hill, Norris, and Watts Bar Reservoir
1993, EFPC Remedial Investigation
1995-96, Clinch River/Watts Bar Remedial Investigations
1998, Radionuclides, metals, organics in Scarboro
2001, Radionuclides, metals, VOCs, SVOCs, pesticides, & PCBs in Scarboro

1961-present, I-131 and SR-90 in Cows' Milk within 50 miles of ORR

1948-49, Radioactivity Radionuclides in Clinch River Fish

1967-present, Mercury, PCBs, Radionuclides, in Clinch River Fish
1970-82, Mercury in EFPC, Bear Creek Fish
1974-77, Mercury in Clinch River and Poplar Creek Fish
1977, Metals, PCBs in Clinch River and Poplar Creek Fish
1977-present, Radionuclides in ORR Deer
1977-present, Radionuclides in Grass from ORR Perimeter and Remote Stations

Biota

1979, Metals in Melton Hill Reservoir and Clinch River Fish
1982, Mercury in Pasture Grass in EFPC Drainage
1982, Mercury in Cow and Horse Grazing on EFPC Floodplain
1983, Mercury in EFPC and Bear Creek Frogs and Crayfish
1983-87, Mercury in Native Vegetation and Garden Vegetables on EFPC Floodplain
1984, Mercury in EFPC and Poplar Creek Turtles
1984, Metals, PCBs, Radionuclides in Melton Hill Reservoir, EFPC,
Bear Creek, and Clinch River Fish, Frogs, Turtles, and Crayfish
1985-present, Metals and Organics in EFPC Fish
mid-80's, Metals in Deer from the EFPC Floodplain
1986, Mercury, PCBs in EFPC Fish
1986-89, Metals, Pesticides, PCBs, in Melton Hill and Watts Bar Reservoir Fish
1987-present, Radioactivity in Geese
1989, Metals, PCBs, Pesticides, SVOCs, Radionuclides in Clinch and Tennessee River Fish
1993, EFPC Remedial Investigation
1995-96, Clinch River/Watts Bar Remedial Investigations
1998, Radionuclides, metals, organics in Scarboro

1951-66, 77, Radionuclides in Clinch River and Tennessee River

2001, Radionuclides, metals, VOCs, SVOCs, pesticides, & PCBs in Scarboro

1960-64, Organics and Radioactivity in Clinch and Tennessee River
1970, Mercury in Melton Hill Reservoir, EFPC, Bear Creek
1972, Mercury in EFPC, Bear Creek
1973-74, 79, PCBs in Clinch River, EFPC, Poplar Creek

Sediment

1973-82, Metals and PCBs in Melton Hill Reservoir
1974-75, Mercury in EFPC
1975-present, Metals in Clinch River, EFPC
1981-82, Metals in Bear Creek and EFPC
1984-86, Metals, Organics, and Radionuclides in Bear Creek
1985, Herbicides, Pesticides, and PCBs in Bear Creek
1985, Metals, PCBs, Organics, and Radionuclides in Clinch River, Poplar Creek, EFPC, Bear Creek
1986, Cs-137 in Watts Bar Reservoir
1989-90, Metals, VOCs, SVOCs, PCBs, Pesticides, Tritium, Radionuclides in Clinch River, Poplar Creek
1990, Metals, Organics, Radionuclides in Melton Hill, Norris, and Watts Bar Reservoir
1993, EFPC Remedial Investigation
1995-96, Clinch River/Watts Bar Remedial Investigations

1949-present, External Gamma Radiation Measurements
1959-1968, Routine Aerial Background Surveys

1971-present, Radionuclides in Soil at Perimeter and Remote Monitoring Stations
1973-74, 1980, 1986, 1989, and 1992, Airborne Gamma Radiation Surveys

Soil

1978-79, Technetium-99 in Soils near K-25
1983-87, Metals, PCBs, and Radionuclides in EFPC Floodplain Soils
1984, Radiation Survey of the Oak Ridge Sewer Beltway
1989-90, Surface Radiation Exposures to Hunters on ORR
1993, EFPC Remedial Investigation
1998, Radionuclides, metals, organics in Scarboro
2001, Radionuclides, metals, VOCs, SVOCs, pesticides, & PCBs in Scarboro

1955-present, Particle Number, Fallout Particle Number, Beta Radioactivity, Beta Radioactivity in Rainwater, Uranium, Nickel, Lead, Chromium, Particulates (nickel, lead, chromium no longer sampled)

Air

1963-present, I-131
1975-present, Particulate Gamma Emitters, SR-90, uranium, thorium
1986-present, Mercury
1990-present, Uranium Particulates, Flourides, Particulates
1993, EFPC Remedial Investigation

Drinking
Water

1959-present, Radionuclides in Water from Clinch River Water Intakes
1981, 83, Radionuclides, Metals in Residential Well Water
1985, Radioactivity in Residential Well Water
1986, Radioactivity, Radionuclides, Inorganics in Residential Well Water
1986, 89-present, Metals, Organics, Radionuclides in Residential Drinking Water

PUBLIC HEALTH ACTIVITIES AT THE ORR
1942-93, Oak Ridge Health Studies Phase 1 Report—Dose Reconstruction Feasibility Study (10/93)
1942-90, PCBs, Phase II of Oak Ridge Health Studies Dose Reconstruction Reports (7/99)
1944-90, Uranium, Phase II of Oak Ridge Health Studies Dose Reconstruction Reports (7/99)
1944-56, White Oak Creek Releases, Phase II of Oak Ridge Health Studies Dose Reconstruction Reports (7/99)
1944-56, Iodine 131, Phase II of Oak Ridge Health Studies Dose Reconstruction Reports (7/99)
1950-63, Mercury, Phase II of Oak Ridge Health Studies Dose Reconstruction Reports (7/99)
1959, 1973, 1980, 1989, 1992, 1997, Aerial Radiological Surveys of the Scarboro Community (1998)
1980-92, Health Statistics Review of Mortality Rates (1994)
1984, Pilot Survey of Mercury Levels in Oak Ridge (10/85)
1985-95, Health Consultation on Lower Watts Bar Reservoir (2/96)
1988-90, Health Statistics Review to Address Oak Ridge Physician's Concerns (10/19/92)
1990-92, Health Consultation on Y-12 Weapons Plant
Chemical Releases into East Fork Poplar Creek (3/93)
1992, Review of Clinical Information on Persons Living in or near Oak Ridge, Tennessee (9/92)
1995, Health Consultation on Proposed Mercury Clean-Up Levels (1/96)
1997, Watts Bar Reservoir Exposure Investigation (3/98)
1998, Scarboro Community Health Investigation (7/00)

19

44

19

50

19

55

19

60

19

65

19

70

19

75

8
19

0

8
19

5

90

19

95

19

00

20

05
20

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

II.C. Characteristics of Mercury
Mercury is the only chemical evaluated by ATSDR in this public health assessment, and as such,
it is important to summarize the key characteristics of this ever-present metal to aid in the
discussion that follows in this document. The intent of this section is to provide a very brief
overview of mercury, explain general sources of mercury releases, describe the cycle of mercury
in the environment, and explain common types of mercury exposures. More detailed information
on mercury is presented in Appendix D of this public health assessment and in ATSDR’s
Toxicological Profile for Mercury (ATSDR 1999).
Mercury exists in the environment naturally, and is present in three forms: metallic (elemental)
mercury, inorganic mercury, and organic mercury. The form of mercury can change when
combined with certain microorganisms (e.g., bacteria, fungi) or natural environmental processes.
The change of one mercury form to another is referred to as “mercury methylation.” Each
mercury form is briefly summarized below.
• Metallic mercury, also called elemental mercury, is mercury in its pure form as it does not
combine with any other elements in the environment. At room temperature, metallic mercury
is a liquid, but some of it can evaporate and enter the air. Once in air, this mercury vapor can
change into other forms of mercury, and further transport to water or soil in rain or snow.
• Inorganic mercury compounds (also called mercury salts) can form when mercury mixes
with elements such as oxygen, sulfur, or chlorine. Inorganic mercury may enter air, water, or
soil from various sources (e.g., incineration of mercury-containing municipal garbage).
• Organic mercury compounds form when carbon combines with mercury. Organic mercury
can be released to air, water, or soil. Environmental microorganisms (and less commonly,
human activities) can convert inorganic mercury to methylmercury, the most common
organic mercury compound. Methylmercury can enter air, water, soil, and of greatest
concern, accumulate in the food chain.
While the purpose of this public health assessment is to only
Of all the potential mercury-related
evaluate potential exposures to mercury released from the Y-12
exposures, the most significant
plant at the ORR, it is worth noting other common sources of
health concern for people and
wildlife is exposure to mercurymercury releases. Both natural (e.g., weathering of mercurycontaminated fish. Air pollution is
containing rocks) and human activities lead to mercury releases
the main source of methylmercury
to the environment. Of the mercury released from human
contamination in fish (USGS 1995).
activities, approximately 80 percent is elemental mercury
released to air from mining, smelting, fossil fuel combustion (mainly coal), and solid waste
incineration. For reference, TVA operates two coal-burning fossil fuel plants in the Oak Ridge
area: the Bull Run Plant and Kingston Steam Plant. 5 An additional 15 percent is mercury
released to soil from municipal solid waste, fertilizers, and fungicides; and the remaining 5
percent is mercury released to water from industrial wastewater.
As shown in Figure 3, the mercury cycle is multi-faceted. The mercury cycle is characterized by
degassing of mercury from soils and surface water, followed by atmospheric transport, wet and
5

An evaluation of mercury emissions associated with the Kingston Steam Plant is presented in Appendix F. Note
that a screening modeling analysis showed that past mercury emissions from the TVA Kingston Plant almost
certainly did not have substantial air quality impacts near the Y-12 plant, even when considering a series of
health-protective assumptions. The Bull Run Plant was not built during the time evaluated, however.
Page | 23

dry deposition of mercury back to land and surface water, sorption of mercury to soil and
sediment particulates, revolatilization of mercury deposited on land and surface water back into
the atmosphere, and bioaccumulation in both terrestrial and aquatic food chains.
People can be exposed to mercury in the environment through various ways. As presented in
Section III, ATSDR evaluates potential exposures to Y-12 mercury releases throughout the
environmental mercury cycle, including air, surface water, soil, sediment, fish, and local
produce. But by far the primary health concern for mercury exposure in the general population is
associated with people eating mercury-contaminated fish. Because methylmercury accumulates
in fish, bigger and older fish tend to have the highest contaminant levels and represent the
greatest health risk.
ATSDR recommends the public follow state fish advisories and federal government
recommendations. In March 2004, the U.S.EPA and the Food and Drug Administration (FDA)
released a joint national fish advisory. The advisory acknowledged that nearly all fish and
shellfish contain traces of mercury. It emphasized that fish and shellfish are an important part of
a healthy diet, and that the risk of mercury-related health effects from eating fish and shellfish
are not a concern for most people. The advisory pointed out that the risks from mercury in fish
and shellfish depend on the mercury levels in the fish and shellfish, and the amount eaten. The
FDA and U.S.EPA advised women who might become pregnant, women already pregnant,
nursing mothers, and young children to avoid some types of fish and to eat fish and shellfish
known to have lower mercury levels (EPA 2004; FDA 2004). The National Fish Advisory is
included in Appendix H. In addition, the state of Tennessee publishes advisories specific to local
water bodies at http://www.tennessee.gov/environment/wpc/publications/pdf/advisories.pdf.

Page | 24

Source: NOAA 2008

Figure 3. Characterization of Mercury Cycling

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Page | 25

II.D. Remedial and Regulatory History
Over the years, ORR operations released a variety of radioactive and nonradioactive wastes. In
1989, U.S.EPA added ORR to the National Priorities List
This ORR Federal Facility Agreement
(NPL) (EPA 2002b). DOE is conducting clean-up activities at
was implemented on January 1, 1992.
the ORR under the Comprehensive Environmental Response,
It is a legally binding agreement to
Compensation, and Liability Act 6 and under a Federal Facility
establish timetables, procedures, and
Agreement, an Interagency Agreement with the U.S.
documentation for remediation actions
at ORR. The Federal Facility
Environmental Protection Agency (U.S.EPA) and the
Agreement is available online at
Tennessee Department of Environment and Conservation
http://www.ucor.com/ettp_ffa.html.
(TDEC). U.S.EPA and TDEC, along with the public, help
DOE with the details for remedial actions at the ORR. DOE integrates required measures from
the Corrective Action sections of the Resource Conservation and Recovery Act (RCRA) with
response actions under CERCLA. See Figure 2 for a time line of surface water, biota, sediment,
soil, air, and drinking water environmental monitoring data related to activities at the Y-12 plant.
But contaminants remain in old ORR waste sites. These sites occupy 5 to 10 percent of the
ORR’s total area. Abundant rainfall (annual average of 55 inches) and high water tables (for
example, 0 to 20 feet below the surface) contribute to leaching of these contaminants, resulting
in contaminated surface water, sediment, groundwater, and biota (EUWG 1998). Since 1986
(when initial clean-up activities commenced), DOE has initiated approximately 50 response
actions under the Federal Facility Agreement. These actions address contamination and disposal
issues on the reservation. The following remedial actions pertain to the Y-12 plant specifically
(SAIC 2007).
Upper East Fork Poplar Creek is located entirely on the site. It originates from a spring
beneath the Y-12 plant and is initially confined to a human-dug channel and flows through the
Y-12 plant along Bear Creek Valley. Contaminants released to the storm drain system
commingle and contribute to the surface water contamination. The principal contaminants
detected in the surface water are mercury and uranium. The principal contaminants in the
sediment are mercury, uranium, and PCBs.
The Upper EFPC Remedial Investigation (RI) report provides a comprehensive overview of
historical investigations of mercury fate and transport at the Y-12 plant. Residual mercury
remains in soils and storm sewers at Y-12, as well as in Upper EFPC sediments and bank soils.
How much residual mercury remains is currently unknown, but the flux of mercury from these
various sources is highly variable and dependent on a number of factors (SAIC 2007). Station
17, where Upper EFPC flows into Lower EFPC, has and will continue to monitor Y-12 plant
mercury releases. Mercury concentrations at Station 17 have decreased since 1995 (see Figure 4;
Bechtel Jacobs 2010; SAIC 2004, 2007).
U.S.EPA, TDEC, and DOE negotiated a Record of Decision (ROD) that selected a number of
different source control remedies to control the influx of mercury from the Y-12 plant into Upper
EFPC. Major actions include
•
•
•
6

The hydraulic isolation of contaminated soils in the West End Mercury Area.
The treatment of the discharge of groundwater into Upper EFPC at Outfall 51.
The removal of contaminated sediments from storm sewers, Upper EFPC, and Lake Reality.
CERCLA, also known as Superfund
Page | 26

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

• Land use controls to prevent consumption of fish from Upper EFPC and to monitor access by
workers and the public.
• Surface water monitoring.
The goal is to restore surface water in Upper EFPC to human health recreational risk-based
values where Upper EFPC flows into Lower EFPC (DOE 2002; EPA 2002a). Future planned
CERCLA actions are expected to achieve the 200 parts per trillion (ppt) performance goal for
mercury in surface water at Station 17 (SAIC 2007).
In 2006, a comprehensive Five-Year Review was performed to evaluate baseline conditions in
advance of fully implementing the remedy outlined in the Upper EFPC Phase I ROD. The
remedy is expected to be protective of human health and the environment upon completion.
Until, however, further information is obtained, a human-health protective determination cannot
be made (SAIC 2007).
Lower East Fork Poplar Creek flows north from the
Y-12 plant off site through a gap in Pine Ridge and into
the City of Oak Ridge. Lower EFPC flows through
residential and business sections of Oak Ridge to join
Poplar Creek, which flows to the Clinch River. Starting
in the early 1950s, Lower EFPC was contaminated by
releases of mercury and other contaminants.
The Remedial Investigation (SAIC 1994a) and
Feasibility Study (SAIC 1994b) (RI/FS) for Lower
EFPC were completed in 1994. Mercury was identified
as the primary contaminant of concern in the floodplain
soils (SAIC 1994a). The ROD was approved in
September 1995 (DOE 1995b), and remediation field
activities began in June 1996 (ATSDR et al. 2000). The
Remedial Investigation and Proposed Plan (DOE 2001;
SAIC 2004) ultimately led to the decision to

Lower EFPC RI/FS Conclusions
Mercury was identified as the primary
contaminant in floodplain soils, and incidental
soil ingestion was identified as the principal
exposure route.
No excessive risk associated with mercury in
surface water was found. The mercury
concentrations were less than drinking water
standards, except on occasion near the Y-12
plant.
Shallow groundwater was not being used and is
not expected to be used in the future as a
drinking water source.
Such limited exposure to contaminated stream
channel sediment reduced the human health risk
to acceptable levels.
Source: SAIC 2007

• Excavate those floodplain soils with mercury levels
higher than 400 parts per million (ppm),
• Dispose of contaminated soils in the Y-12 industrial landfill v (subtitle d landfill),
• Perform confirmatory sampling to ensure that all mercury above this level had been removed,
backfill the excavated areas with clean borrow soil and vegetating appropriately, and
• Monitor periodically to ensure the remediation’s effectiveness.
The clean-up level of 400 ppm was based on “open” land use; it protects the most sensitive
human receptor (children) via inadvertent soil ingestions and dermal contact, and considers the
specific form of mercury (mercuric sulfide) present in the EFPC floodplain soil (SAIC 2007).
The Agency for Toxic Substances and Disease Registry (ATSDR) evaluated the public health
impacts of the 400 ppm clean-up level and concluded that it was protective of public health
(ATSDR 1996a).
The excavation of floodplain soils with greater than 400 ppm of mercury was conducted in two
phases. From July 8 to September 14, 1996 (Phase I), 4,250 loose cubic meters (m3) of mercurycontaminated soils were removed from the floodplain near the National Oceanic and
Page | 27

Atmospheric Administration (NOAA) Atmospheric Diffusion Laboratory off Illinois Avenue.
From March 3 to October 24, 1997 (Phase II), an additional 29,970 loose m3 of mercurycontaminated soils were removed from the floodplain near the NOAA site and across the Oak
Ridge Turnpike from the Bruner’s Shopping Center on the Wayne Clark Property (SAIC 1994a,
2002a). Confirmatory samples were taken during both phases of the excavation to ensure that the
remediated areas were statistically below the clean-up standard (SAIC 1998). Post remediation
monitoring (mercury input, stream stability, and fish sampling) was conducted to ensure the
excavation’s effectiveness (SAIC 2002a).
In 2006, a comprehensive Five Year Review evaluated the protectiveness of the Lower EFPC
ROD (SAIC 2007). The remedy implemented for the Lower EFPC floodplain soil, groundwater,
and floodplain remains protective of human health and the environment. A second ROD, the
EFPC Surface Water and Creek Bed Sediment ROD, is planned for the future and will
investigate media the current ROD did not address (SAIC 2007).
As part of the 2006 Five Year Review, DOE reviewed land use changes along the EFPC
floodplain and the exposure factors used in the baseline risk assessment. The evaluation of land
use indicated residential use of land adjacent to the Lower EFPC floodplain increased
significantly in three locations and was consistent with the future land use projected in the 1994
RI/FS. The only exception was commercial development of the Community Reuse Organization
of East Tennessee’s (CROET) reindustrialization of the ETTP Parcel ED-1, the Horizon Center.
The key exposure factors were the mercuric sulfide bioavailability factor used to develop the
400-ppm clean level and the soil-to-vegetable biotransfer factors used to evaluate the vegetable
ingestion pathway. A search of the most current literature revealed no information that might
alter the original factors used or that might question the protectiveness of the 400-ppm mercury
level in floodplain soil (SAIC 2007).
The review concluded the following potential changes in human health exposure and toxicity
information:
• Because mercuric sulfide is stable in soil and has a low potential for biotransfer to plants, the
pathway has a lower risk than that calculated in the original baseline risk assessment.
• Dermal exposure to mercuric sulfide has the same risks as those calculated in the original
baseline risk assessment.
• Consumption of produce with mercury has the same risks as those calculated in the original
baseline risk assessment.
Changes in the Lower EFPC stream channel and floodplain were surveyed annually to evaluate
whether erosion of potentially mercury-bearing sediments was occurring and to identify areas
where sediment was being deposited in the channel and floodplain. The data indicated little
change in erosion, and deposition of mercury above the cleanup level was not occurring (SAIC
2007). Therefore, the floodplain survey was discontinued in 2004.
Since the mid-1980s, mercury concentrations in fish have been increasing at two Lower EFPC
locations (SAIC 2007). This raised concerns about the assumptions regarding the importance of
upstream industrial sources of mercury relative to floodplain or in-stream sediment sources
(Bechtel Jacobs 2010). Southworth et al. (2010) investigated the sources of mercury to EFPC
downstream from the Y-12 plant. They concluded that floodplain sources of mercury have the
potential to continue contaminating EFPC even if headwater sources are removed, although
more investigation is needed (Southworth et al. 2010). The upstream source continues to provide
Page | 28

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

sufficient mercury to account for the concentrations in fish, and will confound the ability to
determine the role of floodplain soils and stream sediments as sources until it is substantially
reduced (SAIC 2007).
Lower Watts Bar Reservoir (LWBR) extends from the confluence of the Tennessee River and
the Clinch River downstream to the Watts Bar Dam. All surface water and sediment released
from the ORR enter the LWBR (DOE 2001; DOE 2003; SAIC 2004). In 1995, a RI/FS revealed
that discharges of radioactive, inorganic, and organic pollutants from the ORR contributed to
biota, water, and sediment contamination in the LWBR (ORNL and Jacobs Engineering Group
1995). In September 1995, a ROD (DOE 1995c) identified the following contaminants of
concern: 1) mercury, arsenic, PCBs, chlordane, and aldrin in fish; 2) mercury, chromium, zinc,
and cadmium in dredged sediments and sediments used for growing food products; and 3)
manganese through ingestion of surface water (ATSDR et al. 2000; DOE 2001, 2003; SAIC
2004).
The main source of additional mercury in LWBR is related to current and historical sources from
EFPC and the Y-12 plant. But as distances from the EFPC increase, mercury concentrations in
fish decrease. As such, mercury concentrations in fish caught in LWBR are 5–10 times lower
than fish caught in EFPC (SAIC 2004).
The main threat to public health from the LWBR is related to the consumption of PCBcontaminated fish (ATSDR 1996b, 2009; DOE 2001, 2003; SAIC 2004). The remedial activities
selected for the LWBR have included using preexisting institutional controls (e.g., warning
signs) to decrease contact with contaminated sediment, fish consumption advisories printed in
the Tennessee Fish Regulations, and yearly monitoring of biota, sediment, and surface water
(ATSDR et al. 2000; DOE 1995c, 2001, 2003; EPA 2002a; SAIC 2004).
In 2006, a comprehensive Five-Year Review evaluated the protectiveness of the LWBR ROD
(SAIC 2007). The Review found that remedies in place under the LWBR ROD for the sediment
and surface water remained protective of human health and the environment. Contaminant
releases from upstream sources were reduced, which assures continued protection. Also, wellmaintained, ROD-required institutional controls remain in place (SAIC 2007).
Further detailed information on remedial and regulatory information at the ORR can be found in
Oak Ridge Health Studies Phase 1 Report: Volume II – Part A – Dose Reconstruction Feasibility
Study, Tasks 1 & 2, A Summary of Historical Activities on the Oak Ridge Reservation with
Emphasis on Information Concerning Off-Site Emission of Hazardous Material (ChemRisk
1993a); the 2004 Remediation Effectiveness Report for the U.S. Department of Energy Oak
Ridge Reservation (SAIC 2004), and Oak Ridge Reservation Annual Site Environmental Reports
(available online at http://www.ornl.gov/sci/env_rpt/). A summary of selected remedies,
monitoring, and stewardship requirements for Upper EFPC, Lower EFPC, Bear Creek Valley,
and LWBR is provided in Table 1.

Page | 29

Source: SAIC 2007

Figure 4. Mercury Concentrations at the Confluence of Upper EFPC and Lower EFPC

Page | 30

Upper EFPC

None specified

Institutional controls related to groundwater
use

None specified
None specified

� Groundwater well
sampling
� Sampling of effluent
from the treatment
system

Excavation, treatment, and disposal of leadcontaminated soil
� Removal of contaminated liquid and
sediment
� Demolition and filling of basin and sump
� Extract contaminated groundwater from
GW-845
� Treat water to reduce VOC concentrations
� Discharge treated groundwater into
UEFPC upstream of Station 17

YS-860 Firing Ranges

Y-12 Plant 9822 Sediment
Basin and Building 81-10

Y-12 Plant East End VOC
Plume

UEFPC Union Valley

None specified

Removal of contaminated piping

Building 9201-4

None specified

No further action

Abandoned Nitric Acid
Pipeline

None specified

None required

Monitoring

No further action

� Removal of mercury-containing sediment
and water from three tanks
� Two tanks abandoned in place
� One tank returned to service

Selected Remedy

Plating Shop Container
Areas

Mercury Tanks

Site

Annual property owner notification
Title searches
License agreements
Water use surveys
Notification to well drillers

Page | 31

Maintain existing institutional controls such as
license agreements with affected property
owners to restrict groundwater use

None specified

None specified

�
�
�
�
�

None specified

None specified

None specified

None required

Stewardship Requirements

Table 1. Summary of Selected Remedies, Monitoring, and Stewardship Requirements

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Source:
DARA:
EFPC:
LWBR:
RCRA:
UEFPC:
VOC:

LWBR

Hydraulic isolation
Sediment removal
Treatment of discharges
Land use controls
Surface water monitoring

Selected Remedy

� Institutional controls
� Fish consumption advisories
� Annual monitoring

� Excavation of identified floodplain soils
containing greater than 400 ppm mercury
� Confirmatory sampling
� Backfilling and revegetation
� Monitoring
� Fish bioaccumulation survey
� Institutional controls (if needed)

�
�
�
�
�

SAIC 2004, 2007
Disposal Area Remedial Action
East Fork Poplar Creek
Lower Watts Bar Reservoir
Resource Conservation and Recovery Act
Upper East Fork Poplar Creek
volatile organic compound

Lower EFPC

UEFPC Watershed

Site

� Surface water sampling
� Sediment sampling
� Fish sampling

� Surface water sampling
� Land use survey
� Stream channel survey

� Surface water sampling
� Biota sampling

Monitoring

Page | 32

� Maintain existing institutional controls to
control potential sediment-disturbing
activities
� Fish consumption advisories
� Participation in the Interagency Working
Group

DOE will monitor to detect any future
residential use of shallow groundwater and, if
found, to mitigate any risk associated with
such use.

� Property record restrictions, notices
� Zoning notices for the western Y-12 area
� Continuation of Y-12 access controls,
signage, and security patrols
� Maintenance of treatment facilities per
operating specifications
� Continuation of excavation/penetration
permit program

Stewardship Requirements

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

II.E.

Site Geology/Hydrogeology

The ORR is in the East Tennessee Valley, part of the Valley and Ridge Province of the
Appalachian Mountains. The East Tennessee Valley is bound to the west by the Cumberland
Mountains of the Appalachian Plateau Province and to the east by the Smoky Mountains of the
Blue Ridge Province. The defining characteristics of the Valley and Ridge Province are the
southwest trending series of ridges and valleys due to crustal folding and faulting due to
compressive tectonic forces. Differential weathering of the various underlying formations also
define the province.
The major hydrologic watersheds associated with the ORR are East Tennessee Technology Park
Watershed, Bethel Valley Watershed, Melton Valley Watershed, Bear Creek Valley Watershed,
and Upper EFPC Watershed (EUWG 1998).
The majority of information available concerning the geology and hydrogeology of the site
indicates that groundwater occurs as shallow flow, with short flow paths to surface water (DOE
2004; MMES 1986; ORNL 1982; SAIC 2004; USGS 1986, 1988, 1989). The fractures and
solution cavities—common in this karst region—occur in shallow (0–100 feet deep) bedrock and
significantly decrease at depth (>100 feet deep). In the aquitard formations, as much as 95
percent of all groundwater occurs in the shallow zone and discharges into local streams and
eventually into the Clinch River. In the aquifer formations—the Knox Aquifer being the most
important—solution conduits can make flow paths much deeper and longer along the strike
(DOE 2004).
An extensive interconnection between groundwater and surface water and
the ORR groundwater contamination sources are primarily in the shallow
subsurface. And core samples have shown that beneath the alluvium at the
bottom of the area stream beds a silty-clay horizon likely impedes
downward groundwater movement (USGS 1989). The incised meander of
the Clinch River in bedrock also represents a major topographic feature
that retards groundwater from passing beneath the river (ORNL 1982).

Groundwater beneath
the ORR is typically
very shallow;
approximately 95
percent of it ends up
as surface water
before leaving the site
boundary (DOE 2004).

In 2006, ATSDR conducted a public health assessment to evaluate potential community
exposures to contaminated groundwater coming from the ORR. ATSDR concluded that no
human exposures to contaminated groundwater outside the ORR boundary have occurred in the
past, are currently occurring, or are likely to occur in the future (ATSDR 2006b). See ATSDR’s
2006 Evaluation of Potential Exposures to Contaminated Off-site Groundwater at
http://www.atsdr.cdc.gov/HAC/oakridge/phact/groundwater/index.html.

II.E.1. Bear Creek and Upper East Fork Poplar Creek Watersheds
On the ORR, Bear Creek Valley comprises a large portion of the Bear Creek watershed and the
Upper EFPC watershed. Bear Creek Valley is bordered by Chestnut Ridge and Pine Ridge. The
825-acre Y-12 plant is in Bear Creek Valley, predominantly in the Upper EFPC watershed.
Figure 5 illustrates how groundwater flows along strikes in Pine Ridge and Chestnut Ridge. The
southward sloping orientation of the bed planes beneath Pine Ridge prevents groundwater from
flowing north toward Scarboro.
As is the case throughout much of the ORR, surface and groundwater are highly interconnected.
Gaining and losing reaches of Bear Creek are found along the entire Bear Creek Valley. These
reaches are often contamination sources of surface water. As they increase contaminant
Page | 33

concentrations in the shallow groundwater, the shallow groundwater increasingly contaminates
the reaches. Indeed, several large solution cavities are beneath Bear Creek, which (along certain
reaches) serve as a hydraulic drain to the Maynardville Limestone (Lemiski 1994; SAIC 1996).
Groundwater in the Upper EFPC watershed typically flows along strike from west to east in the
Maynardville Formation between 100 feet and 400 feet below ground. Groundwater flow
direction in this area is also influenced by anthropogenic structures such as pipes, drains, and
other underground structures that have created preferential flow paths for contaminated
groundwater (SAIC 2005). But the Maynardville Limestone is the primary pathway for
contaminant migration off-site from Y-12. Because of its well developed karst-system,
groundwater from adjacent formations tends to flow toward the Maynardville Limestone.
Because of the high interconnectivity with surface water, groundwater discharges at seeps and
springs constitutes much of the base flow of Scarboro Creek and Upper EFPC. Depth to
groundwater in this area is between 1 and 4 feet below ground during the winter and between 2
and 7 feet below ground in the summer (USGS 1989).
ATSDR’s 2006 Evaluation of Potential Exposures to Contaminated Off-site Groundwater
provides more detail about the hydrogeology and contamination beneath the Upper EFPC
watershed (ATSDR 2006b). See
http://www.atsdr.cdc.gov/HAC/oakridge/phact/groundwater/index.html.

Page | 34

Figure 5. Cross-sectional Diagram of Pine Ridge and Chestnut Ridge in the Y-12 Vicinity

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Page | 35

II.F.

Land Use and Natural Resources

Together with the three major DOE installations—the East Tennessee Technology Park
(formerly the K-25 site), Oak Ridge National Laboratory (formerly the X-10 site), and the Y-12
National Security Complex (formerly the Y-12 plant)—The ORR currently owns 34,235 acres,
occupying about 30 percent of the reservation. In 1980, the remaining 70 percent was established
as a National Environmental Research Park to provide protected land for environmental science
research and education and to demonstrate that energy technology development can coexist with
a quality environment. Over the past several decades large portions of the reservation have
grown into full forests. Some of this land includes areas known as “deep forest” that contain
ecologically significant flora and fauna; portions of ORR are considered biologically rich (SAIC
2002b).
The ORR also includes an area set aside for residential, commercial, and support services. The
city of Oak Ridge was created in 1942 to provide housing to the employees of the ORR and was
originally controlled by the military (Friday and Turner 2001). The self-governing portion of the
city of Oak Ridge comprises about 14,000 acres and contains housing, schools, parks, shops,
offices, and industrial areas. Some residential properties are adjacent to the ORR boundary line.
Outside the urban areas, much of the region (about 40 percent) still reflects its historical pattern
of farms and small communities (ChemRisk 1993b).
Public access is restricted at the Y-12 plant, which is entirely within the ORR “229 Boundary.”
Y-12 is “an active production and special nuclear materials management facility [and so]
additional security and access limitations apply” (DOE 2002). Out of 1,170 acres in the Upper
EFPC area, 800 are currently used for industrial purposes. This acreage includes maintenance
facilities, office space, training facilities, change houses, former Oak Ridge National Laboratory
Biology Division facilities, waste management facilities, construction contractor support areas,
and a high-security portion that supports core National Nuclear Security Administration missions
(DOE 2002).
A number of area maps indicate a wide range of land types, including “types of urban or built up
land, agricultural land, rangeland, forestland, water, and wetlands,” and uses such as “residential,
commercial, public and semi-public, industrial, transportation, communication and utility, and
extractive (e.g., mining)” (ChemRisk 1993b).
Although agriculture (beef and dairy cattle) and forestry had been the two predominant land uses
in the area around ORR, both are currently in decline. For many years, milk was produced,
bottled, and distributed locally. Corn, tobacco, wheat, and soybeans were the major crops grown
in the area. During certain periods hunters seek small game, waterfowl, and deer (ChemRisk
1993b).
EFPC originates from within the Y-12 plant boundary, flows through the city of Oak Ridge for
about 12 miles, and ultimately converges with Poplar Creek near the K-25 facility (DOE 1989).
A number of small tributaries flow into the creek and support some small aquatic life. While
people do not use the streams on the reservation, they do have access downstream from the
reservation. The area through which the Lower EFPC flows has many uses, but they can be
grouped into five major categories: residential, commercial, agricultural, open land, and DOEowned (DOE 1995b). Land use changes were evaluated during the 2006 Five Year Review
(SAIC 2007). Much of the land along the creek remains undeveloped; however, residential use of

Page | 36

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

land adjacent to the Lower EFPC floodplain has increased in the following three locations (see
Figure 6):
• Development along Wiltshire Drive (approximately 24 parcels, with 12 adjacent to the
floodplain).
• Jackson Crossing (approximately 30 parcels, with 6 adjacent to the floodplain).
• Southwood subdivision (many residential parcels, with almost half adjacent to the
floodplain).
Within the city of Oak Ridge, EFPC is too shallow for swimming, however, children
occasionally play in the creek water. The area near the confluence with Poplar Creek is deep
enough for swimming, wading, and fishing. TDEC issued a
fishing advisory for EFPC that warns the public to avoid
Fish Advisories for Waterways Near
the ORR
eating fish from the creek because of mercury and PCB
Tennessee River
contamination. They also have an advisory to avoid contact
with the water due to bacterial contamination. The presence
Catfish, striped bass, and hybrid (striped
bass-white bass) bass should not be
of bacteria in the water affects the public’s ability to safely
eaten due to elevated levels of PCBs.
swim, wade, and fish in streams and reservoirs. According
Children, pregnant women, and nursing
to TDEC, bacterial sources include failing septic tanks,
mothers should not consume white bass,
collection system failure, failing animal waste systems, or
sauger, carp, smallmouth buffalo, and
urban runoff. In 1992, some of the advisory signs along the
largemouth bass, but other people can
safely consume one meal per month of
creek were replaced and additional signs posted to warn the
these species.
public about contaminated surface water and fish (TDEC
Clinch River
1992). The state reviews and updates postings along EFPC
Striped bass should not be eaten due to
to address exposure to surface water and fish. Postings
elevated levels of PCBs. Children,
warning about the presence of bacteria may be removed in
pregnant women, and nursing mothers
the future; however, postings warning of contamination in
should not consume catfish and sauger,
fish will remain (SAIC 2007).
but other people can safely consume
one meal per month of these species.
The LWBR is downstream of the ORR and extends from the
East Fork Poplar Creek
confluence of the Clinch and Tennessee Rivers to the Watts
No fish should be eaten due to elevated
Bar Dam (DOE 1995a). The waters of the reservoir supply
mercury and PCB levels. Avoid contact
domestic water (although LWBR is not a direct source of
with the water due to bacterial
drinking water), industrial water, and irrigation for plants
contamination.
and livestock (DOE 1995c). The area around LWBR is
For the advisories, see
forested or agricultural, with moderate residential
http://www.tn.gov/environment/wpc/publi
development and little industrial development (DOE 2003).
cations/pdf/advisories.pdf.
The public has access to the LWBR, which it uses for
recreational purposes such as boating, swimming, fishing, skiing, and shoreline activities (DOE
1996, 2003). The LWBR area comprises over 47 recreational parks and facilities (including
marinas, resorts, and golf courses) (TVA 1990). In the early 1990s, the total annual visitor-days
were estimated at over 1 million, with the area from the Watts Bar Dam upstream to Kingston
receiving the most visits (TVA 1987, 1990). TDEC issued a fishing advisory that warns the
public to avoid or limit how much fish from the LWBR they eat because of elevated levels of
PCBs (ORNL and Jacobs Engineering Group 1995; SAIC 2004).

Page | 37

Source: SAIC 2007

Figure 6. Current Land Use Along EFPC

Page | 38

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

II.G. Demographics
The Y-12 mercury releases study area consists of two separate areas, with distinct exposures and
communities. The first area surrounds EFPC, which runs through the city of Oak Ridge. The
communities evaluated in this area live within the city of Oak Ridge, including the Scarboro
community and the communities living along the EFPC floodplain. The city of Oak Ridge is in
Anderson County and part of Roane County, Tennessee. The second area evaluated surrounds
the LWBR. Harriman, Kingston, Rockwood, and Spring City are the four main cities within the
reservoir area. Harriman, Kingston, and Rockwood are in Roane County, and Spring City is in
Rhea County. Meigs County is also in the area that surrounds LWBR and, therefore, is also in
the study area. Figure 7 provides the current demographics for a 1-mile and 3-mile radius of the
Y-12 plant.

Page | 39

Figure 7. Demographics for a 1-Mile and 3-Mile Radius of the Y 12 Plant

Page | 40

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

II.G.1. Counties within the Y-12 Mercury Releases Study Area
Since 1940, the populations of Anderson, Roane, Rhea, and Meigs Counties have all grown by
about 50 percent (U.S. Census Bureau 1940, 1950, 1960, 1970, 1980, 1993, 2000). Table 2
shows the population for these counties over that 60-year period, and Figure 8 shows the
population distribution for the counties over that same period.
Table 2. Populations of Anderson, Roane, Rhea, and Meigs Counties from 1940 to 2000
County

1940

1950

1960

1970

1980

1990

2000

Anderson County

26,504

59,407

60,032

60,300

67,346

68,250

71,330

Roane County

27,795

31,665

39,133

38,881

48,425

47,227

51,910

Rhea County

16,353

16,041

15,863

17,202

24,235

24,344

28,400

Meigs County

6,393

6,080

5,160

5,219

7,431

8,033

11,086

Source: U.S. Census Bureau 1940, 1950, 1960, 1970, 1980, 1993, 2000

Figure 8. Population Distribution of Anderson, Roane, Rhea, and Meigs Counties
from 1940 to 2000

Source: U.S. Census Bureau 1940, 1950, 1960, 1970, 1980, 1993, 2000

Anderson County
From 1940 to 1950, as people came to build and operate the new Y-12 facilities, the Anderson
County population more than doubled: from 26,504 to 59,407. Over the next 50 years, the county
grew steadily at the more modest rate of 20 percent to 71,330 in the year 2000 (U.S. Census
Bureau 1940, 1950, 1960, 1970, 1980, 1993, 2000). Figure 8 shows the pattern of growth. As of
2000, most residents worked in management, professional, and related fields. Anderson County

Page | 41

has 66,593 whites, 2,766 African Americans, and 828 persons of other races. Most residents are
between 40 and 44 years old, with a median age of 39.9 (U.S. Census Bureau 2000).
Roane County
Over this same 60-year period, the Roane County population has grown by 86.8 percent, as
shown in Table 2 (U.S. Census Bureau 1940, 1950, 1960, 1970, 1980, 1993, 2000). The
population declined slightly from 1960 to 1970, and between 1980 and 1990 (East Tennessee
Development District 1995; U.S. Census Bureau 1960, 1970, 1980, 1993). The county
population grew during the remaining time and reached a population of 51,910 in 2000. Figure 8
shows the population distribution of the county over time (East Tennessee Development District
1995; U.S. Census Bureau 1940, 1950, 1960, 1970, 1980, 1993, 2000).
Most of Roane County’s 2000 population is white (49,440); the rest are African American
(1,409) and other races (1,061) (U.S. Census Bureau 2000). Since the 1970s, the median age of
Roane County residents has increased from 32.1 to 40.7, suggesting that the county population is
aging (East Tennessee Development District 1995; U.S. Census Bureau 1993, 2000). The X-10
site and the K-25 site are both within Roane County (East Tennessee Development District 1995;
Jacobs EM Team 1997). Primarily because of these two facilities, between 1940 and 1990
manufacturing was the dominant occupation for Roane County residents (East Tennessee
Development District 1995; U.S. Census Bureau 1993).
Rhea County
The population of Rhea County declined between 1940 and 1960, but has increased steadily
since the 1960s (see Table 2 and Figure 8). The largest increase (40.9 percent) was between 1970
and 1980, when the number of residents increased from 17,202 to 24,235. Over the past 60 years,
the population of Rhea County has increased by nearly 75 percent (U.S. Census Bureau 1940,
1950, 1960, 1970, 1980, 1993, 2000). As of 2000, most residents worked in the manufacturing
industry. Rhea County has 27,097 whites, 580 African Americans, and 723 persons of other
races. Most residents are between the ages of 35 and 44, with a median age of 37.2 (U.S. Census
Bureau 2000).
Meigs County
Between 1940 and 1960, the population of Meigs County decreased. But the population has
nearly doubled since then—from 5,160 to 11,086 (46.5 percent) (see Table 2 and Figure 8). The
largest percentage increase in population occurred between 1970 and 1980, when the number of
residents grew from 5,219 to 7,431 (42.4 percent). Since 1940, the population of Meigs County
has grown by almost 60 percent (U.S. Census Bureau 1940, 1950, 1960, 1970, 1980, 1993,
2000). As of 2000, most residents worked in the manufacturing industry. The Meigs County
population is comprised of 10,826 whites, 138 African Americans, and 122 persons of other
races. Most residents are between the ages of 35 and 44, and the median age is 36.7 (U.S. Census
Bureau 2000).

II.G.2. Cities within the Y-12 Mercury Releases Study Area
Oak Ridge
In 1942, the city of Oak Ridge, Tennessee, was established in Anderson County for the 13,000
persons who were expected to work at the ORR (Friday and Turner 2001). By July 1944, the
population of Oak Ridge had increased to 50,000. The Oak Ridge population peaked in 1945 at
Page | 42

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

approximately 75,000 and declined to 30,229 by 1950 (see Table 3) (Oak Ridge Comprehensive
Plan 1988). For the last three census years (1980, 1990, 2000) the city population has been
between 27,000 and 28,000. In 1959, about 14,000 acres within the city of Oak Ridge became
self-governing (ChemRisk 1993b). Almost since its establishment, the city of Oak Ridge has
been one of the largest population centers in the area (ChemRisk 1993b).
Table 3. Population of Oak Ridge from 1942 to 2000
Oak Ridge

1942

1944

1945

1950

1960

1970

1980

1990

2000

13,000

50,000

75,000

30,229

27,169

28,319

27,662

27,310

27,387

Sources: ChemRisk 1993b; Oak Ridge Comprehensive Plan 1988; U.S. Census Bureau 1940, 1950, 1960, 1970,
1980, 1993, 2000

From 1940 to 1960, the city of Oak Ridge had a higher proportion of working age people and
fewer seniors than the rest of Tennessee (ChemRisk 1993b). Since 1960, however, the resident
population under age 35 and over age 55 has increased, while the population of children under
age 16 has declined (Oak Ridge Comprehensive Plan 1988). The education level of Oak Ridge
citizens is dramatically higher than in surrounding areas; Oak Ridge boasts one of the highest per
capita PhD ratios of any city in the United States (Oak Ridge Comprehensive Plan 1988).
Scarboro
The Scarboro community is within the city of Oak Ridge, outside of the EFPC floodplain (see
Figure 9). It’s about a half mile from the Y-12 plant and is separated from the Y-12 plant by Pine
Ridge. Before 1950, the area was known as the Gamble Valley Trailer Camp, and the population
was predominantly white. In 1950, Scarboro was established to provide single-family homes,
duplexes, apartments, and an elementary school to African American Oak Ridge residents
(Friday and Turner 2001). To this day, Scarboro remains predominantly African American (94
percent) (Friday and Turner 2001).
In the fall of 1999, the Joint Center for Political and Economic Studies conducted a survey of the
broader Scarboro community (Friday and Turner 2001). The staff identified 380 residences, of
which 326 were occupied. About 266 persons responded to the survey (82 percent). The report
generated from the survey is one of the few sources of detailed information available on the
Scarboro community (Friday and Turner 2001).
The Scarboro community is aging—the average respondent is almost 53 years old. Only 36
percent of participating households reported having at least one member between the ages of 18
and 34 years. About half of the households reported having one senior citizen or more, while
only 23 percent of the surveyed households reported having children. Additionally, 39 percent of
respondents were retired. As of 1999, the average length of residence in Scarboro was 29 years.
But many (82 percent) of the young adult residents (18–30 years old) moved to Scarboro after
1994. For additional details, see the Scarboro Community Assessment Report (Friday and Turner
2001).

Page | 43

Figure 9. Surface Elevation for Scarboro

Page | 44

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

EFPC Floodplain
The EFPC floodplain surrounds EFPC. Using available information, researchers found that over
the history of the ORR, approximately 10 farms were located in the floodplain (ChemRisk
1999a). The Task 2 team estimated that the total population size between 1940 and 1990 was
between 40 and 200 persons—the number in any given year was estimated to be between 10 and
50 (ChemRisk 1999a).
Harriman
The city of Harriman is located along Roane County’s Emory River, to the west of the ORR (see
Figure 1). As seen in Table 4 and Figure 10, the population of Harriman peaked between 1970
and 1980 (8,734 and 8,303, respectively) and has continued to decline since (East Tennessee
Development District 1995; U.S. Census Bureau 1940, 1950, 1960, 1970, 1980, 1993, 2000).
The median age of the population is 40.5 years; about 40 percent of the residents are between the
ages of 25 and 54 (U.S. Census Bureau 2000). About 90 percent of the population is white, 7.4
percent is African America, and a small percentage is persons of other races (U.S. Census
Bureau 2000). In 1990, Harriman had more minority residents than any other city in Roane
County (8.6 percent of the population; East Tennessee Development District 1995). In 1969, 18
of the 29 manufacturing plants in Roane County were located within the city of Harriman. By
1990, however, only 15 of 35 manufacturing plants were in Harriman (East Tennessee
Development District 1995). As of 2000, manufacturing was Harriman’s leading industry.
Kingston
The City of Kingston is in Roane County, at the confluence of the Clinch River and the
Tennessee River, southwest of the ORR (see Figure 1). The population of Kingston has grown
steadily from 1940 to 2000, except for a 0.2 percent decrease between 1980 and 1990 (see Table
4 and Figure 10) (East Tennessee Development District 1995; U.S. Census Bureau 1940, 1950,
1960, 1970, 1980, 1993, 2000). The median age of the population is 41.6 years. About 40
percent of the residents are between the ages of 25 and 54, with the greatest portion between 45
and 54 years of age (U.S. Census Bureau 2000). The majority of the population is white (93.8
percent), 3.6 percent are African American, and a small percentage consists of persons of other
races (U.S. Census 2000). Since 1990, the greatest portion of residents (26.2 percent) has been
employed in the professional services field (East Tennessee Development District 1995; U.S.
Census Bureau 2000).
Rockwood
Rockwood is situated to the southwest of ORR, northwest of the confluence of the Clinch and
Tennessee Rivers, also in Roane County. As seen in Table 4 and Figure 10, the city experienced
steady growth between 1940 and 2000, except for slight declines that occurred between 1960 and
1970, and between 1980 and 1990 (East Tennessee Development District 1995; U.S. Census
Bureau 1940, 1950, 1960, 1970, 1980, 1993, 2000). As of 2000, the median age was 42 years.
About 38 percent of the population is between the ages of 25 and 54 (U.S. Census Bureau 2000).
The majority of the population is white (92.9 percent), about 5.4 percent are African American,
and a small percentage are persons of other races (U.S. Census Bureau 2000). The largest
percentage of residents is employed in the manufacturing field. In 1969, 10 out of 29
manufacturing plants in Roane County were located in Rockwood; by 1990, Rockwood had 13
out of the 35 manufacturing plants in the county (East Tennessee Development District 1995).
Page | 45

Spring City
Spring City is in Rhea County along the Tennessee River, south of the confluence with the
Clinch River and north of the Watts Bar Dam. Between 1940 and 2000, the Spring City
population remained relatively steady, with the number of residents slowly increasing by about
25 percent (see Table 4 and Figure 10). The largest percent increase in population was seen
between 1980 and 1990, followed by the largest decrease between 1990 and 2000 (U.S. Census
Bureau 1940, 1950, 1960, 1970, 1980, 1993, 2000). The median age of the population is 44
years. About 36 percent of the residents are between the ages of 25 and 54, with the greatest
portion between 35 and 44 years of age (U.S. Census Bureau 2000). The majority of the
population is white (94.5 percent), 4.5 percent are African American, and a small percentage
consists of persons of other races (U.S. Census 2000). As of 2000, the largest percentage (31.6
percent) of residents worked in the manufacturing industry (U.S. Census Bureau 2000).
Table 4. Population of Harriman, Kingston, Rockwood, and Spring City from 1940 to 2000
City

1940

1950

1960

1970

1980

1990

2000

Harriman

5,620

6,389

5,931

8,734

8,303

7,119

6,744

Kingston

880

1,627

2,010

4,142

4,561

4,552

5,264

Rockwood

3,981

4,272

5,345

5,259

5,695

5,348

5,774

Spring City

1,569

1,725

1,800

1,756

1,951

2,199

2,025

Sources:
2000

East Tennessee Development District 1995; U.S. Census Bureau 1940, 1950, 1960, 1970, 1980, 1993,

Figure 10. Population of Oak Ridge, Harriman, Kingston, Rockwood, and Spring City
from 1940 to 2000
30,000

Number of People

25,000

20,000

15,000

10,000

5,000

0
1940

1950

Oak Ridge

Sources:
1993, 2000

1960

Harriman

1970

Kingston

1980

Rockwood

1990

2000

Spring City

East Tennessee Development District 1995; U.S. Census Bureau 1940, 1950, 1960, 1970, 1980,

Page | 46

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

II.H. Summary of Public Health Activities Pertaining to Y-12 Mercury Releases
This section describes the public health activities that pertain to Y-12 mercury releases. Several
additional public health activities conducted at the ORR by ATSDR, the Tennessee Department
of Health (TDOH), and other agencies are described in Appendix B. Summary of Other Public
Health Activities. See Figure 2 for a time line of public health activities related to the Y-12 plant.

II.H.1. ATSDR
Since 1992, ATSDR has addressed health concerns of community members, civic organizations,
and other government agencies. ATSDR has worked to determine whether levels of
environmental contamination at and near the ORR present a public health hazard. During this
time, ATSDR has identified and evaluated several public health issues and has worked closely
with many parties, including community members, civic organizations, physicians, and several
local, state, and federal environmental and health agencies. While the TDOH conducted the Oak
Ridge Health Studies to evaluate whether off-site populations have experienced exposures in the
past (1944–1990), ATSDR’s activities in the 1990s focused on current public health issues
current at that time to prevent duplication of the state’s efforts. The ATSDR ORR Web site
(http://www.atsdr.cdc.gov/HAC/oakridge/phact/index.html) highlights ATSDR’s major public
health activities at the ORR. The following paragraphs highlight major public health activities
conducted by ATSDR that pertain to Y-12 mercury releases.
Health Consultation on Y-12 Weapons Plant Chemical Releases Into East Fork Poplar Creek,
April 1993 (ATSDR 1993). This health consultation provided DOE with advice on current public
health issues related to past and present chemical releases into the creek from the Y-12 plant.
Before finalizing its remedial investigation and feasibility study on EFPC, DOE implemented
many of ATSDR’s recommendations. The EFPC Phase Ia data evaluated for this health
consultation indicate that the creek’s soil, sediment, groundwater, surface water, air, and fish are
contaminated with various chemicals. ATSDR reached the following public health conclusions:
• Soil and sediments in certain locations along the EFPC floodplain are contaminated with
levels of mercury that pose a public health concern.
• Fish in the creek contain levels of mercury and PCBs that pose a moderately increased risk of
adverse health effects to people who eat fish frequently over long periods of time.
• Shallow groundwater in a few areas along the EFPC floodplain contains metals at levels of
public health concern; however, this shallow groundwater is not used for drinking or other
domestic purposes.
Other contaminants found in soil, sediment, surface water, and fish were not detected at levels
that could make people ill. In summary, among other recommendations, ATSDR advised
continuation of the EFPC fish advisory with posting of signs, especially at the confluence of
Poplar Creek (ATSDR 1993). Access this public health consultation at
http://www.atsdr.cdc.gov/HAC/PHA/efork1/y12_toc.html. A brief summarizing the health
consultation is provided in Appendix C. Summary Briefs and Factsheets.
ATSDR Science Panel Meeting on the Bioavailability of Mercury in Soil, August 1995 (Canady
et al. 1997). The purpose of the science panel was to identify methods and strategies that would
enable health assessors to develop data-supported, site-specific estimates of the bioavailability of
inorganic mercury and other metals (arsenic and lead) from soils. The panel consisted of private
Page | 47

consultants and academicians internationally known for their metal bioavailability research.
Experts from ATSDR, the Centers for Disease Control and Prevention (CDC), U.S.EPA, and the
National Institute for Environmental Health Science also participated. ATSDR used information
obtained from the panel meeting to evaluate the EFPC clean-up level. ATSDR also used the
findings to characterize and evaluate soil containing mercury at other waste sites. Three technical
papers and an ATSDR overview paper on the findings of the panel meeting were published in
Volume 17:5 of the International Journal of Risk Analysis in 1997 (Canady et al. 1997).
Health Consultation on Proposed Mercury Clean Up Levels, January 1996 (ATSDR 1996a). In
response to a request from community members and the City of Oak Ridge, ATSDR evaluated
the public health effects of DOE’s clean-up levels of 180 milligrams per kilogram (mg/kg) and
400 mg/kg of mercury in the EFPC floodplain soil. ATSDR concluded that both clean-up levels
would be protective of public health and would pose no health threat to adults or children
(ATSDR 1996a). Access this public health consultation at
http://www.atsdr.cdc.gov/hac/pha/pha.asp?docid=1360&pg=0. Note: Floodplain soils with
mercury concentrations greater than 400 ppm were remediated in 1996 and 1997 (SAIC 1994a,
2002a).
Watts Bar Reservoir Exposure Investigation, March 1998 (ATSDR 1998). In following up on the
findings of previous studies and investigations of the Watts Bar Reservoir, including Feasibility
of Epidemiologic Studies by the TDOH, ATSDR conducted the exposure investigation in
cooperation with the TDOH and the Roane County Health Department. The 1996 exposure
investigation was conducted to measure actual PCB and mercury levels in people consuming
moderate to large amounts of fish and turtles from the Watts Bar Reservoir. The investigation
also was to determine whether these people were exposed to high levels of PCBs and mercury.
ATSDR published the following three major findings:
• The exposure investigation participants’ serum PCB levels and blood mercury levels were
very similar to levels found in the general population.
• Five of the 116 people tested (4 percent) had PCB levels higher than 20 micrograms per liter
(fg/L) or parts per billion (ppb), which is considered to be an elevated level of total PCBs.
Of the five participants who exceeded 20 fg/L, four had levels of 20–30 fg/L. One
participant had a serum PCB level of 103.8 fg/L—higher than the general population
distribution.
• Only 1 of 116 participants had an elevated blood mercury level. The participants’ blood
mercury levels were very similar to levels found in the general population (ATSDR 1998).
A brief summarizing the exposure investigation is provided in Appendix C. Summary Briefs and
Factsheets.

Page | 48

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment
Where Can I Obtain More Information on ATSDR’s Activities at the ORR?
ATSDR has conducted several analyses that are not documented here, as have other agencies that have been involved with
this site. Community members can find more information on ATSDR’s past activities in the following three ways:
1. Visit one of the records repositories. Copies of ATSDR’s publications on the ORR, along with publications from other
agencies, can be viewed in records repositories at public libraries and the DOE Information Center (located at 475 Oak
Ridge Turnpike, Oak Ridge, Tennessee; 865-241-4780). For directions to these repositories, please contact ATSDR at 1­
800-CDC-INFO (1-800-232-4636).
2. Visit the ATSDR or ORRHES Web sites. These Web sites include past publications, schedules of future events, and other
materials. ATSDR’s ORR Web site is at http://www.atsdr.cdc.gov/HAC/oakridge. The most comprehensive summary of
past activities can be found at http://www.atsdr.cdc.gov/HAC/oakridge/phact/c_toc.html.
3. Contact ATSDR directly. Residents can contact representatives from ATSDR directly by dialing the agency’s toll-free
number, 1-800-CDC-INFO (1-800-232-4636).

II.H.2. TDOH
Oak Ridge Health Studies. In 1991, DOE and the state of Tennessee entered into the Tennessee
Oversight Agreement, which allowed the TDOH to undertake a two-phase independent state
research project to determine whether past environmental releases from ORR operations harmed
people who lived nearby (ChemRisk 1999d; ORHASP 1999). Access all the technical reports
produced for the TDOH Oak Ridge Health Studies at
http://health.state.tn.us/ceds/oakridge/oridge.html.
Phase I. Phase I of the Oak Ridge Health Study is a Dose Reconstruction Feasibility Study. This
feasibility study evaluated all past releases of hazardous substances and operations at the ORR.
The objective of the study was to determine the quantity, quality, and potential usefulness of the
available information and data on these past releases and subsequent exposure pathways. Phase I
of the health studies began in May 1992 and was completed in September 1993.
The findings of the Phase I Dose Reconstruction Feasibility Study indicated that a significant
amount of information was available. Researchers could use this information to reconstruct the
past releases and potential off-site exposure doses for four hazardous substances that may have
been responsible for adverse health effects. These four substances include 1) radioactive iodine
releases associated with radioactive lanthanum processing at the X-10 site from 1944 through
1956; 2) mercury releases associated with lithium separation and enrichment operations at the
Y-12 plant from 1950 through 1963; 3) PCBs in fish from EFPC, the Clinch River, and the Watts
Bar Reservoir; and 4) radionuclides from White Oak Creek associated with various chemical
separation activities at the X-10 site from 1943 through the 1960s. A brief summarizing the
Phase I Feasibility Study is provided in Appendix C. Summary Briefs and Factsheets.
Phase II (also referred to as the Oak Ridge Dose Reconstruction). Phase II of the health studies
conducted at Oak Ridge began in mid-1994 and was completed in early 1999. Phase II was
primarily a dose reconstruction study focusing on past releases of radioactive iodine, mercury,
radionuclides from White Oak Creek, and PCBs. In addition to the full dose reconstruction
analyses, the Phase II effort also included additional detailed screening analyses for releases of
uranium and several other toxic substances that Phase I had not fully characterized. The
following paragraphs describe the significant findings for each of the substances evaluated.

Page | 49

• Radioactive iodine releases were associated with radioactive lanthanum processing at the X­
10 site from 1944 through 1956. Results indicate that children who were born in the area in
the early 1950s and who drank milk produced by cows or goats living in their yards had an
increased risk of developing thyroid cancer. The report stated that children living within a
25-mile radius of Oak Ridge were likely to have had an increased risk of more than 1 in
10,000 of developing thyroid cancer (ChemRisk 1999e).
• The study evaluated mercury releases associated with lithium
U.S.EPA’s reference
separation and enrichment operations at the Y-12 plant from 1950
dose is an estimate of
the largest amount of a
through 1963. Results indicate that depending on their activities,
substance that a person
persons living in the area during the years that mercury releases
can take in on a daily
were highest (mid-1950s to early 1960s) may have received annual
basis over their lifetime
average doses of mercury exceeding the U.S.EPA reference doses
without experiencing
(RfDs) used for evaluating potential health effects from different
adverse health effects.
mercury exposure scenarios (ChemRisk 1999a). A brief
summarizing this study is provided in Appendix C. Summary Briefs and Factsheets.
• Radionuclides associated with various chemical separation activities at the X-10 site from
1943 through the 1960s were released into White Oak Creek. Studied were eight
radionuclides (cesium 137, ruthenium 106, strontium 90, cobalt 60, cerium 144, zirconium
95, niobium 95, and iodine 131) deemed more likely than others to carry significant risks.
The results indicate that the releases caused small increases in the radiation dose of those
who ate fish from the Clinch River near the mouth of White Oak Creek. The dose
reconstruction scientists estimated that a male who ate up to 130 meals of fish from the
mouth of White Oak Creek every year for 50 years (worst-case scenario) would face an
excess cancer risk ranging from 4 to 350 in 100,000. The risk from eating fish goes down
proportionately for those who eat fewer fish and for those who eat fish taken farther
downstream (ChemRisk 1999f).
• Additional studies were conducted on PCBs in fish from EFPC, the Clinch River, and the
Watts Bar Reservoir. TDOH concluded that persons who consumed large amounts of fish
from the Clinch River and the LWBR were at risk of noncancer effects of PCBs. The studies
also concluded that three or fewer additional cases of cancer could have resulted from eating
Clinch River and Watts Bar Reservoir fish (carcinogenic risks ranged from 1 in 1,000,000 to
2 in 10,000; ChemRisk 1999c). Because, however, the estimates and modeling are
conservative, “the actual risks and expected number of cases are likely to be smaller and
could be zero” (ChemRisk 1999c). To reduce the uncertainty, TDOH also made
recommendations for further study.
• Uranium was released from various large-scale uranium operations, primarily uranium
processing and machining operations at the Y-12 plant and uranium enrichment operations at
the K-25 and S-50 plants. Because uranium was not initially given high priority as a
contaminant of concern, a Level II screening assessment for all uranium releases was
performed. Preliminary screening indices were slightly below the decision guide of one
chance in 10,000, which indicated that more work may be needed to characterize better the
uranium releases and the possible heath risk (ChemRisk 1999b).
Pilot Survey. In the fall of 1983, TDOH developed an interim soil mercury concentration for use
in environmental management decisions. CDC reviewed the methodology for the interim
Page | 50

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

mercury level in soil. CDC then recommended a pilot survey to determine whether populations
with the highest risk for mercury exposure had elevated mercury body burdens. In June and July
1984, a pilot survey was conducted to document human body levels of inorganic mercury. The
survey focused on residents of Oak Ridge with the highest potential for mercury exposure from
contaminated soil and fish. The survey also examined whether exposure to mercurycontaminated soil and fish constituted an immediate health risk to the Oak Ridge population. The
results of the pilot survey, released in October 1985, suggested that Oak Ridge, Tennessee
residents and workers were not likely at increased risk for significantly high mercury levels.
Mercury concentrations in hair and urine samples were below levels associated with known
health effects (Rowley et al. 1985).

II.H.3. Florida Agricultural and Mechanical University (FAMU)
Scarboro Community Environmental Study (FAMU 1998). In 1998, soil, sediment, and surface
water were sampled in the Scarboro community to address community concerns about
environmental monitoring in the Scarboro neighborhood. The analytical component of the study
was conducted by the Environmental Sciences Institute at Florida Agricultural and Mechanical
University (FAMU) and its contractual partners at the Environmental Radioactivity
Measurement Facility at Florida State University and the Bureau of Laboratories of the Florida
Department of Environmental Protection, and by DOE subcontractors in the Neutron Activation
Analysis Group at the Oak Ridge National Laboratory.
Organic compounds were only detected in one of the samples tested. This same sample also
contained lead and zinc at concentrations twice as high as those found in the Background Soil
Characterization Project (DOE 1993a). Mercury was found within the range given in the
Background Soil Characterization Project, and about 10 percent of the soil samples showed
evidence of uranium 235, which is associated with uranium enrichment. The final Scarboro
community Environmental Study was released in September 22, 1998, during a Scarboro
community meeting (FAMU 1998). A brief summarizing this study is provided in Appendix C.
Summary Briefs and Factsheets.

II.H.4. U.S.EPA
Scarboro Community Environmental Sampling Validation Study (EPA 2003). In 2001,
U.S.EPA’s Science and Ecosystem Division Enforcement Investigation Branch collected soil,
sediment, and surface water samples from the Scarboro community to respond to community
concerns, identify data gaps, and validate the sampling performed by FAMU in 1998 (FAMU
1998). A final report was released in April 2003 (EPA 2003). U.S.EPA concluded that the results
support the sampling performed by FAMU in 1998, and that the residents of Scarboro are not
currently exposed to harmful levels of substances in the soil, sediment, or surface water. A brief
summarizing this study is provided in Appendix C. Summary Briefs and Factsheets.

II.H.5. DOE
Mercury Inventory Report, 1977. DOE asked Union Carbide to reconstruct the historical mercury
inventory at the Y-12 plant from 1950 through 1977. Two employees spent 2 weeks gathering
information from documents and employee interviews. The classified report indicated that
550,000 pounds of mercury had been spilled or lost to the environment, and about 1.9 million
pounds were unaccounted for (Case 1977; ChemRisk 1999a).

Page | 51

Mercury Task Force, 1983. In May 1983, the Y-12 plant manager appointed the Mercury Task
Force to collect historical data (1950–1983) on mercury accountability, study mercury salvage
and recovery, and summarize mercury effects on worker health and the environment. The task
force consisted of employees who were not involved in operations when most mercury exposures
to workers and losses to the environment occurred. The classified report represents the official
statement of mercury releases from the Y-12 plant (ChemRisk 1999a).
Federal Facility Agreement, 1992. DOE is conducting clean-up activities at the ORR under a
Federal Facility Agreement—a legally binding agreement between DOE, U.S.EPA, and TDEC.
The agreement was finalized on January 1, 1992, to establish timetables, procedures, and
documentation for remediation actions at ORR. Under the Federal Facility Agreement, DOE,
U.S.EPA, and TDEC have conducted RI/FSs on the Lower EFPC Operable Unit (OU), the
LWBR OU, and the Clinch River/Poplar Creek OU. All of these OUs were placed on the NPL in
December 1989; under CERCLA an RI/FS is required for all sites on the NPL (ATSDR et al.
2000). The Federal Facility Agreement is available online at http://www.ucor.com/ettp_ffa.html.
Lower East Fork Poplar Creek Remedial Investigation/Feasibility Study, 1994 (SAIC 1994a,
1994b). The purpose of the RI/FS was to assess contamination (primarily mercury-contaminated
floodplain soils) resulting from releases since 1950 from the Y-12 plant. The objectives of the
study were to determine the extent of contamination of the EFPC floodplain, to develop a
baseline risk analysis based on the level of contaminants, and to determine whether remedial
action was required (ATSDR et al. 2000).
The findings indicated that portions of the floodplain were contaminated
with mercury. Also, floodplain soil with mercury concentrations of more
than 400 ppm would constitute an unacceptable risk to human health and
the environment. Drawing on these findings, the 1995 ROD (DOE
1995b) called for remedial action. The remedial action included
•

Excavation of four areas of the floodplain where soils had mercury
concentrations of more than 400 ppm;

• Confirmatory sampling during excavation activities to document the
removal;

A small area next to the
NOAA site was not
remediated. The area
underneath the Dean
Stallings Ford automobile
dealership parking lot
was filled. But it still
contains mercury above
400 ppm. DOE annually
visits the lot to ensure
that the land use has not
changed (SAIC 2007).

• Disposal of contaminated soil into a landfill at the Y-12 plant under a special waste permit;
• Backfilling of excavated areas, including a 0.6-acre wetland, with clean borrow soil; and
• Revegetation of the affected areas.
Remediation field activities began in June 1996 and were completed in October 1997 (ATSDR et
al. 2000).
Lower Watts Bar Reservoir Remedial Investigation/Feasibility Study, March 1995 (ORNL and
Jacobs Engineering Group 1995). The purpose of the RI/FS was to assess the level of
contamination in the Watts Bar Reservoir, to create a baseline risk analysis based on the
contaminant levels, and to establish whether remedial action was necessary. The findings of the
remedial investigation suggested that biota, sediment, and water at the Watts Bar Reservoir were
contaminated with metals, radionuclides, and organic compounds. The baseline risk analysis
suggested that protective standards for environmental and human health would not be reached if
deep channel sediments permeated with cesium 137 were dredged and placed in a residential
Page | 52

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

area, and if people consumed moderate to high quantities of fish that contained increased levels
of PCBs (ATSDR et al. 2000).
Using the RI/FS results, a ROD was prepared and finalized in September 1995 (DOE 1995c).
The ROD mandated that DOE use controls to prevent adverse effects from exposure to
contaminants in the Watts Bar Reservoir. These controls included TDEC-administered fish
consumption advisories, ongoing monitoring, and controlling activities that could disturb
sediment (ATSDR et al. 2000; DOE 1995c).
Clinch River/Poplar Creek Remedial Investigation/Feasibility Study, March 1996 (Jacobs
Engineering Group Inc. 1996). The purpose of the RI/FS was to examine the past and present
releases to off-site surface water and to establish whether remedial action was necessary
(ATSDR et al. 2000). The RI/FS found two main hazards associated with the Clinch
River/Poplar Creek OU: 1) exposure to chromium, cesium 137, mercury, and arsenic located in
deep sediment within the main river channel, and 2) exposure to mercury, chlordane, PCBs, and
arsenic in fish tissue (DOE 1997a; Jacobs Engineering Group Inc. 1996).
A baseline risk assessment was conducted as part of the RI/FS. It suggested that consumption of
certain fish contaminated with PCBs posed the greatest risk to public health. Fish contaminated
with chlordane, mercury, and arsenic presented possible health risks as well. The assessment also
determined that the consumption of any type of fish in Poplar Creek posed a health risk, as did
bass from the Clinch River below Melton Hill Dam. The risk assessment further determined that
contaminants in the buried sediments in the deep-water river channel would only present a health
risk if they were dredged; there is no current exposure to these sediments (DOE 1997a; Jacobs
Engineering Group Inc. 1996).
Again using the results of the RI/FS, another ROD was finalized in September 1997 (DOE
1997a). This ROD recommended (DOE 1997a):
•

Fish consumption advisories,

•

Controls on activities that could disrupt sediment,

•

Yearly monitoring of fish, sediment, surface water, and turtles, and

•

Surveys to assess the value of fish consumption advisories.

In February 1998, a Remedial Action Report was approved (DOE 1997b). This report
recommended that monitoring for surface water, fish, sediment, and turtles in the Clinch
River/Poplar Creek OU (ATSDR et al. 2000).
Oak Ridge Environmental Information System (OREIS), April 1999. Because of the availability
of an abundance of environmental data for the ORR, DOE created an electronic data
management system to integrate all of the data into a single database, facilitating public and
government access to environmental operations data while maintaining data quality. DOE’s
objective was to ensure that the database had long-term retention of the environmental data and
useful methods to access the information. OREIS contains data on compliance, environmental
restoration, and surveillance activities. Information from all key surveillance activities and
environmental monitoring efforts is entered into OREIS, which include but are not limited to
studies of the Clinch River embayment and the Lower Watts Bar, as well as annual site summary
reports. As new studies are completed, the environmental data are entered as well.

Page | 53

Upper East Fork Poplar Creek Record of Decision for Phase I Interim Source Control Actions,
May 2002 (DOE 2002). The ROD selected a number of different source control remedies to
control the influx of mercury from the Y-12 plant into Upper EFPC. The major actions are
• Hydraulic isolation of the West End Mercury Area (e.g., capping contaminated soils);
• Removal of contaminated sediments from storm sewers, Upper EFPC, and Lake Reality;
• Treatment of discharge from Outfall 51;
• Temporary water treatment;
• Land use controls to prevent consumption of fish from Upper EFPC and to monitor access by
workers and the public; and
• Monitoring of surface water.
The remedial action’s goal is to reduce the mass flux of mercury to Upper EFPC. Specifically,
200 ppt is the performance goal for mercury in surface water at Station 17, Building 9201-2
effluent discharge point, Outfall 550, and Outfall 551 (SAIC 2007).
2006 Remediation Effectiveness Report/Second Reservation-wide CERCLA Five-Year Review,
February 2007 (SAIC 2007). DOE conducted the second ORR-wide Five Year Review in 2006.
Five Year Reviews are required at all post-Superfund Amendments and Reauthorization Act
(SARA) sites that still have hazardous substances remaining above levels that allow for
unlimited use and unrestricted exposures. The purpose is to report on completed and ongoing
CERCLA actions and to determine whether the remedy at each site is protective of human health
and the environment. Because many of the CERCLA decisions on the ORR fall within this
definition, the ORR as a whole is subject to Five Year Reviews indefinitely. This Five Year
Review assesses an important set of key, off-site completed remedial actions (e.g., LWBR,
Clinch River/Poplar Creek, and Lower EFPC) and reviews the effects and progress of two major
watershed RODs (the Phase I ROD for Bear Creek Valley and the Interim Record of Decision
for Melton Valley) (SAIC 2007).

Page | 54

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

III. Evaluation of Environmental Contamination and Potential Exposure
Pathways
III.A. Introduction
In 2001, ATSDR scientists conducted a review and analysis of the Phase I and Phase II screening
evaluation of TDOH’s Oak Ridge Health Studies. ATSDR’s purpose was to identify
contaminants that require further public health evaluation. In the Phase I and Phase II screening
evaluation, TDOH conducted extensive reviews of available information. TDOH also conducted
qualitative and quantitative analyses of past (1944–1990) releases and off-site exposures to
hazardous substances from the entire ORR. After ATSDR’s review and analysis of TDOH’s
Phase I and Phase II screening evaluations, ATSDR scientists completed public health
assessments on
• Y-12 plant uranium releases (ATSDR 2004);
• White Oak Creek radionuclide releases (ATSDR 2006a);
• Site-wide current and future chemical exposures (ATSDR 2007);
• X-10 site iodine 131 releases (ATSDR 2008);
• X-10 site, Y-12 plant, and K-25 site PCB releases (ATSDR 2009);
• K-25 site uranium and fluoride releases (ATSDR 2010); and
• Other issues of community concern, such as contaminant releases from the Toxic Substances
Control Act (TSCA) Incinerator (ATSDR 2005a) and contaminated off-site groundwater
(ATSDR 2006b).
This public health assessment on the Y-12 mercury releases evaluates and analyzes the
information, data, and findings of previous studies and investigations of releases of mercury
from the Y-12 plant and assesses the health implications of past and current mercury exposures
to residents living near the ORR.
The public health assessment is the primary public health process ATSDR uses to evaluate
further these contaminants. The documents released to date are available at
http://www.atsdr.cdc.gov/HAC/oakridge/phact/index.html and can also be ordered through the
agency’s toll-free number, 1-800-CDC-INFO (1-800-232-4636).
III.B. Evaluation of Past (1950–1990) Mercury Exposure Pathways
Over the years, three major efforts have been made to estimate Y-12 mercury releases to water
and air. Two of them included investigations to account for past mercury inventories at the Y-12
plant. In 1977, Y-12 personnel prepared a classified report entitled the 1977 Mercury Inventory
Report (Case 1977). In the early 1980s, after the public became aware that large quantities of
mercury had been released from the Y-12 plant, DOE appointed a Mercury Task Force to
investigate what was known about mercury use and releases. The Mercury Task Force studied
the 1977 Mercury Inventory Report and released its own reports in 1983 (UCCND 1983a,
1983b). The Task 2 report documents the third major effort to estimate Y-12 mercury releases
(ChemRisk 1999a). (See Section III.B.1 for a more detailed discussion of the report.) The Task 2
report did not revisit all of the previous inventory estimates, but it revised the previous estimates
of mercury releases to the air and water. The estimates of mercury inventories and releases to air
Page | 55

and water in all three of these reports focused on the lithium enrichment production years (1953–
1963).
The 1977 and 1983 mercury inventory estimates are presented in Table 5. Table 5 does not
include the increased quantities of mercury released to the water and air that Task 2 estimated.
The Task 2 team’s estimates of the quantities of mercury lost to water and air were 40,000
pounds and 22,000 pounds greater, respectively, than the 1983 Mercury Task Force estimates
(ChemRisk 1999a).
As shown in Table 5, a large amount of the mercury originally received at the Y-12 plant is
unaccounted for. Table 5 distinguishes between what is lost and what is not accounted for. The
term “lost” includes the quantities of mercury that were estimated to have gone into the air, soil,
and water. The term “not accounted for” is arrived at by subtraction. It describes mercury
quantities received at the plant that could not be accounted for in either the lost quantities (to air,
water, and soil) or the remaining inventory of products and unused mercury. Personnel who
wrote the 1983 Mercury Task Force Report estimated that over 700,000 pounds of mercury were
lost to the environment and an additional 1,290,000 pounds of mercury were not accounted for
(UCCND 1983a, 1983b).
In interviews with former workers, the 1983 Mercury Task Force identified possible
explanations that might account for about half of the 1,290,000 pounds of mercury that was not
accounted for. 7 It estimated that perhaps 500,000 pounds of the mercury “not accounted for” was
never received, and that this discrepancy is a result of accounting errors. Mercury came into the
plant in 76-pound flasks. But the mercury was not accounted for by weight; it was accounted for
by the numbers of flasks (i.e., the amount of mercury coming into the plant was estimated by the
number of flasks times 76 pounds). People who worked at the plant said that at times flasks that
were leaking or not completely full would arrive at the plant. Thus, the 1983 Mercury Task Force
Report suggested it was likely that the accounting practice for recording the incoming amount of
mercury overestimated the true inventory. The 1983 Mercury Task Force also estimated that
another 60,000 pounds of mercury was unaccounted for in the production building walls, floors,
ceilings, and insulation (UCCND 1983a, 1983b). This rough estimate was based on a 1975
U.S.EPA study of mercury use in the chloralkali industry (Garrett 1975). The 1983 Mercury
Task Force authors emphasized that these figures were speculative.
Including the Task 2 revisions, approximately 1,230,000 pounds of mercury that were vouchered
into inventory during the lithium separation production years (1953–1963) are not accounted for.
This is still larger by more than half than the amount of mercury that Task 2 estimated was lost
to the environment (795,000 pounds; ChemRisk 1999a). Several theories might explain why the
mercury inventories have not been accounted for, and the 1983 Mercury Task Force Report
identifies some of them. Nevertheless, it’s more likely that these discrepancies will never be
confidently accounted for. More mercury might have been released to the environment than the
Task 2 team estimated.

7

The 1983 Mercury Task Force Report only presents two explanations that may account for 560,000 pounds. The
report is silent on the other 85,000 pounds that it says it identified explanations for.
Page | 56

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Table 5. 1977 and 1983 Mercury Material Balance Estimates by Y-12 Plant Staff

1977 Mercury
Inventory Report
(pounds)

1983 Mercury
Task Force Report
(pounds)

24,321,000

24,348,852

*

21,666,348

1,000

1,400

10,000

14,000

111,000

174,000

12,000

17,212

*

200,000

100,000

250,000

**

800

*

22,323,796

2,437,752

2,025,056

30,000

51,300

470,000

238,944

7,200

6,629

**

8,475

49,853

49,853

Known lost to ground, seven other spills

**

375,000

Known lost to ground, Building 81-10 operations

**

3,000

557,053

733,201

1,880,699

1,291,855

Source of Material Inventory and Losses
VOUCHERED to Y-12:
Returned unopened or rebottled and stored/sold
In lithium hydroxide tails, sold and stored
In Building 9201-5 scrap, sold
In Building 9201-5 sludge, removed and sold
As flasking overage given to GSA
In Building 9201-4 equipment, still in place
In sludges and sumps in Alpha-4 Building
In Building 9201-2 sewer pipe
ACCOUNTED FOR Total:
Known LOST and NOT ACCOUNTED FOR Total:
Known lost to air
Known lost to East Fork Poplar Creek
Known lost to New Hope Pond sediment, Chestnut Ridge
Known lost to New Hope Pond sediments now in place
Known lost to ground, Building 9201-5 spill accident

Known LOST Total:
NOT ACCOUNTED FOR Total:
Source: ChemRisk 1999a
*
These data were classified for security reasons in 1977.
**
Data not available in 1977 report.

III.B.1.

The Oak Ridge Dose Reconstruction Project

In 1991, the State of Tennessee and DOE entered into the Oak Ridge Health Agreement. The
agreement’s purpose was to investigate health risks to off-site populations from past ORRrelated releases of hazardous substances to the environment. TDOH administered The Oak Ridge
Health Agreement for the State of Tennessee. As a part of the Oak Ridge Health Agreement,
TDOH conducted the Oak Ridge Health Studies. The studies’ purpose was to evaluate whether
off-site populations were exposed to ORR-related chemical and radiological releases and to
assess the risk posed by off-site exposures. The TDOH Commissioner appointed a 12-member
panel—the Oak Ridge Health Agreement Steering Panel (ORHASP)—to direct and oversee the
Oak Ridge Health Studies and to promote community interaction and cooperation.
McLaren/Hart-ChemRisk (referred to as ChemRisk) was hired to conduct Phase I of the Oak
Page | 57

Ridge Health Studies—the feasibility study—which it did during 1992 and 1993. Using the
feasibility study, ORHASP and TDOH recommended dose reconstruction for
•
•
•
•

Radioactive iodine releases from the X-10 site (Task 1),
Mercury releases from the Y-12 plant (Task 2),
Releases of PCBs (Task 3), and
Radionuclides released from the X-10 site to the Clinch River via White Oak Creek (Task 4).

ORHASP and TDOH also recommended
•
•

Screening evaluations of Y-12 and K-25 uranium releases (Task 6) and
A screening-level evaluation of additional materials of potential concern (Task 7).

Task 5 was an additional task comprising a systematic review of historical records to support the
other six tasks. Phase II of the Oak Ridge Health Studies—the Oak Ridge Dose Reconstruction
Project—began in late 1994 and was completed in July 1999.
The Task 2 report estimated and evaluated exposures to past releases (1950–1990) of mercury
from the ORR. TDOH and ORHASP expended a great amount of work, resources, oversight, and
peer review on the Oak Ridge mercury dose reconstruction (Task 2). Drawing on the comments
from ATSDR’s technical reviewers of the mercury dose reconstruction (see Section III.B.2,
ATSDR decided that it would not attempt to reproduce the dose reconstruction work. It would
use the results of the Task 2 mercury dose reconstruction to assess past exposures to mercury for
its public health assessment.
In particular, Task 2 amassed and reviewed a large amount of data and a large number of
documents. These data and documents described mercury inventories and releases, which formed
the basis of the source terms used to estimate past environmental mercury concentrations. Thus
further investigation of archived data would not substantially improve the Task 2 estimates of the
mercury source terms. Secondly, the dispersion models used to estimate mercury concentrations
in air and water are standard models—ATSDR would use the same or similar dispersion models.
Therefore, without substantial new information about past releases of mercury, newly discovered
historical environmental sampling data or meteorological data—none of which ATSDR presently
has—ATSDR would not likely improve on the basic elements of the Task 2 mercury dose
reconstruction.

III.B.2.

ATSDR’s Technical Review of the Task 2 Report

Although source terms and dispersion models are not easily subjected to external analysis,
ATSDR can review many other assumptions go into dose estimation. In choosing to adopt the
Task 2 results for its public health assessment, ATSDR recognizes that dose reconstruction is a
technical investigation fraught with much uncertainty. Therefore, ATSDR wanted an additional
round of expert review of the Task 2 report. Rather than attempting to reproduce the work or the
results of the mercury dose reconstruction for its public health assessment, ATSDR believes that
an independent expert review of the Task 2 report assumptions offers the best insight into the
validity and usefulness of the Task 2 results for making public health decisions.

Page | 58

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

In 2001, ATSDR contracted with Eastern Research Group, Inc. (ERG) to select five expert

technical reviewers to determine whether the Task 2 report provides a foundation on which

ATSDR can base its mercury public health assessment for the
The five outside technical experts
ORR and surrounding communities. The reviewers were asked
reviewed the following documents:
to comment on the study design, methods, and completeness of
� Reports of the Oak Ridge Dose
the mercury dose reconstruction, as well as the conclusions of
Reconstruction: The Report of
the report’s authors. The reviewers read the entire dose
Project Task 2 – July 1999.
reconstruction document on mercury releases, including
� Mercury Releases from Lithium
appendices, and the appropriate sections of the steering panel
Enrichment at the Oak Ridge Y-12
document. ERG received the reviewer comments and compiled
Plant—a Reconstruction of
Historical Releases and Off-Site
and summarized them for ATSDR in June 2001.
Doses and Health Risks. Volumes
2 (main report) and 2A
(appendices). (submitted to the
Tennessee Department of Health
by ChemRisk) (ChemRisk 1999a).

In July 2003, ATSDR released the compilation and summary
of the reviewer comments to the public. The document is
titled, “Comments by Technical Reviewers on the Oak Ridge
Dose Reconstruction - Task 2 Report, Volume 2: Mercury
� Releases of Contaminants from
Releases from Lithium Enrichment at the Oak Ridge Y-12
Oak Ridge Facilities and Risks to
Plant - a Reconstruction of Historical Releases and Off-Site
Public Health, report of the Oak
Doses and Health Risks, July 2003” (ATSDR 2003). The Task
Ridge Health Agreement Steering
2 report and the Comments by Technical Reviewers report
Panel (ORHASP 1999).
were discussed in meetings of the Public Health Assessment
Work Group (PHAWG) of the Oak Ridge Reservation Health Effects Subcommittee (ORRHES)
from July through December 2003. Throughout these discussions, the PHAWG understood and
recognized the limitations and recommendations of the Task 2 report, and agreed with ATSDR's
use of the Task 2 report in this public health assessment.
III.C. Evaluation of Current (1990–2009) Mercury Exposure Pathways

III.C.1.

Exposure Evaluation

What is meant by exposure?
Exposure or contact drives ATSDR’s public health
assessments. Contaminants (chemicals or radioactive
materials) released into the environment have the potential
to cause harmful health effects. Nevertheless, a release
does not always result in exposure. People can only be
exposed to a contaminant if they come into contact with it.
If no one comes into contact with a contaminant, no
exposure occurs, and no health effects occur. Often the
public does not have access to the source area of
contamination or areas where contaminants move through
the environment. This lack of access becomes important in
determining whether people could come into contact with
the contaminants.
The route of a contaminant’s movement is the pathway.
ATSDR identifies and evaluates exposure pathways by
considering how people might come into contact with a
contaminant. An exposure pathway could involve air,

An exposure pathway has five elements:
(1) a source of contamination, (2) an
environmental medium, (3) a point of
exposure, (4) a route of human exposure,
and (5) a receptor population. The
exposure pathway is incomplete if any
one of these five elements is missing.
The source is the place where the
chemical or radioactive material was
released. The environmental media (such
as, groundwater, soil, surface water, or
air) transport the contaminants. The point
of exposure is the place where persons
come into contact with the contaminated
media. The route of exposure (for
example, ingestion, inhalation, or dermal
contact) is the way the contaminant enters
the body. The people actually exposed
are the receptor population.

Page | 59

surface water, groundwater, soil, dust, or even plants and animals. Exposure can occur by
breathing, eating, drinking, or by skin contact with the chemical contaminant.
How does ATSDR determine which exposure situations to evaluate?
ATSDR scientists evaluate site-specific conditions to determine whether people are exposed to
site-related contaminants. When evaluating exposure pathways, ATSDR identifies whether
exposure to contaminated media (soil, water, air, waste, or biota) is occurring through ingestion,
dermal (skin) contact, or inhalation.
If exposure is possible, ATSDR scientists then consider whether environmental contamination is
present at levels that might affect public health. ATSDR evaluates environmental contamination
using available environmental sampling data and, in some cases, modeling studies. ATSDR
selects contaminants for further evaluation by comparing environmental contaminant
concentrations against health-based comparison values. ATSDR develops comparison values
from available scientific literature on exposure and health effects. Comparison values are derived
for each of the media and reflect an estimated contaminant
ATSDR uses comparison
concentration not expected to cause harmful health effects for a given
values to screen
contaminant, assuming a standard daily contact rate (for example, the
chemicals that require
amount of water or soil consumed or the amount of air breathed) and
additional evaluation.
representative body weight.
Comparison values are not thresholds for harmful health effects. ATSDR comparison values
represent contaminant concentrations many times lower than levels at which no effects were
observed in studies on experimental animals or in human epidemiologic studies. If contaminant
concentrations are above comparison values, ATSDR further analyzes exposure variables (such
as site-specific exposure, duration, and frequency) for health effects, including the toxicology of
the contaminant, other epidemiology studies, and the weight of evidence. Figure 11 illustrates
ATSDR’s chemical screening process.
More information about the ATSDR evaluation process can be found in ATSDR’s Public Health
Assessment Guidance Manual (ATSDR 2005b) at
http://www.atsdr.cdc.gov/hac/PHAManual/toc.html or by contacting the agency at 1-800-CDC­
INFO (1-800-232-4636).

Page | 60

Pathway Evaluation

Preliminary
Screening

Exposure Dose
Comparison

Contaminants
of Concern

Public Health
Implications

Evaluate public
health implications

YES

Contaminants of concern

YES

Are estimated exposure
doses higher than
screening guidelines?

YES

Are the chemical
concentrations higher
than medium-specific
comparison values?

YES

Are there completed
and/or potential exposure
pathways where chemicals
have been detected?

YES

Chemicals detected in
environmental media

NO

NO

NO

Not a contaminant
of concern

Not a contaminant
of concern

Not a contaminant
of concern

Decision Diagram

Public Health Implications Evaluation—
Weight of Evidence

Screening
Guidelines

Medium-Specific
Comparison Values

Exposure Pathways

icals
Chem

Illustration

Figure 11. ATSDR Chemical Screening Process

s %VALUATE WHETHER CONTAMINANTS OF CONCERN CAN AFFECT
PUBLIC HEALTH IN THE VICINITY OF THE SITE

s 2EVIEW TOXICOLOGIC MEDICAL EPIDEMIOLOGIC AND OTHER
SCIENTIFIC DATA ON THE CONTAMINANTS OF CONCERN

s %VALUATE THE PUBLIC HEALTH IMPLICATIONS OF CONTAMINANTS
OF CONCERN IN GREATER DETAIL

s %STIMATE DOSES BASED ON SITE
SPECIFIC EXPOSURE CONDITIONS
s 5SE MORE REALISTIC EXPOSURE ASSUMPTIONS
n REALISTIC CONCENTRATIONS
n REALISTIC EXPOSURE DURATION
n REALISTIC EXPOSURE FREQUENCY
n REALISTIC EXPOSURE BIOAVAILABILITY

s "ASED ON MAXIMUM EXPOSURE CONDITIONS
n MAXIMUM CONCENTRATION DETECTED
n MAXIMUM EXPOSURE DURATION
n MAXIMUM EXPOSURE FREQUENCY
n MAXIMUM EXPOSURE BIOAVAILABILITY

s #AN OR ARE EXPOSURES OCCURRING
s )DENTIFY POTENTIAL OR COMPLETED EXPOSURE PATHWAYS

s "ASED ON THE RESULTS OF ENVIRONMENTAL INVESTIGATIONS

Criteria

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

If people are exposed, will they get sick?
Exposure does not always result in harmful health effects. The type and severity of health effects
in a person as the result of contact with a contaminant depend on several factors:
•
•
•
•

Exposure concentration (how much),
Frequency (how often) and duration of exposure (how long),
Route or pathway of exposure (breathing, eating, drinking, or skin contact), and
Multiplicity of exposure (combination of contaminants).

Once exposure occurs, characteristics such as age, sex, nutritional status, genetics, lifestyle, and
health status of the exposed person influence how that person absorbs, distributes, metabolizes,
and excretes the contaminant. Taken together, these factors and characteristics determine the
health effects that can occur as a result of exposure to a contaminant in the environment.

III.C.2.

Evaluating Exposures

ATSDR evaluated available, current data to determine whether mercury concentrations were
above ATSDR’s comparison values. ATSDR also reviewed relevant toxicologic and
epidemiologic data about mercury toxicity. It’s important to remember that exposure to a
contaminant does not always result in harmful health effects. The type and severity of health
effects expected to occur depends on the exposure concentration, the toxicity of the contaminant,
the frequency and duration of exposure, and the multiplicity of exposures.
Comparing Environmental Data to Comparison Values
ATSDR uses the term

Concentrations are compared to comparison values to determine
“conservative” to refer to values
which contaminants need to be further evaluated. Comparison
that are protective of public
health in essentially all situations.
values are concentrations derived using conservative exposure
assumptions and health-based doses. Comparison values reflect
Conservative values are
developed with assumptions that
concentrations much lower than those found to cause adverse
are more likely to overestimate
health effects. Thus, comparison values are protective of public
than underestimate actual risks.
health in essentially all exposure situations. As a result,
concentrations detected at or below ATSDR’s comparison values do not warrant health
concern. While concentrations at or below the relevant comparison value can reasonably be
considered safe, it does not automatically follow that any environmental concentration exceeding
a comparison value would be expected to produce adverse health effects. The fact that
comparison values are not thresholds of toxicity cannot be emphasized strongly enough. If
contaminant concentrations are above comparison values, ATSDR further analyzes
exposure variables (for example, duration and frequency of exposure), the toxicology of the
contaminant, other epidemiology studies, and the weight of evidence for health effects. The
likelihood that adverse health outcomes will actually occur depend on site-specific conditions
and individual lifestyle that affect the route, magnitude, and duration of actual exposure, as well
as current health condition (e.g., chronic health conditions) and genetic factors. An
environmental concentration alone will not cause an adverse health outcome.
When evaluating chemical effects of mercury exposure, ATSDR scientists used comparison
values specific to each environmental media. The comparison values used are shown in Table 6.

Page | 63

Table 6. Comparison Values for Mercury
Media

Comparison Value
mg/m3

Source

Air

0.0002

Surface Water

2 /g/L

LTHA/MCLG for inorganic mercury

Soil/Sediment

20 mg/kg

Child RMEG for mercuric chloride

Fish

0.14 mg/kg

RSL for methylmercury

EMEG:
LTHA:
MCLG:
/g/L:
mg/kg:
mg/m3:
RMEG:
RSL:

Chronic EMEG for elemental mercury

ATSDR’s environmental media evaluation guide
U.S.EPA’s lifetime health advisory
U.S.EPA’s maximum contaminant level goal
microgram per liter (parts per billion or ppb)
milligram per kilogram (parts per million or ppm)
milligram per cubic meter
ATSDR’s reference dose media evaluation guide
U.S.EPA’s regional screening level

ATSDR’s environmental media evaluation guide (EMEG) is a compilation of nonenforceable,
health-based comparison value developed for screening environmental contamination for further
evaluation. ATSDR’s reference dose media evaluation guide (RMEG) is a lifetime exposure
level at which adverse, noncarcinogenic health effects would not be expected to occur.
U.S.EPA’s regional screening level (RSL) is a health-based comparison value. Concentrations
above the RSL may warrant further investigation or site cleanup. The lifetime health advisory
(LTHA) is the concentration of a chemical in drinking water not expected to cause any adverse
noncarcinogenic health effects for a lifetime of exposure. U.S.EPA’s maximum contaminant
level goal (MCLG) is the risk-based level of a contaminant that may be present in drinking water
under the Safe Drinking Water Act. The MCLG for mercury is the same as the enforceable
maximum contaminant level (MCL).

III.C.3.

Comparing Estimated Doses to Health Guidelines

Deriving exposure doses
Exposure doses are expressed in milligrams of mercury per kilogram of
An exposure dose is the
body weight per day (mg/kg/day). When estimating exposure doses,
amount of chemical a
health assessors evaluate chemical concentrations to which people
person is exposed to over
a specified period of time.
could have been exposed, together with the length of time and the
frequency of exposure. Collectively, these factors influence a person’s
physiological response to chemical exposure and potential outcomes. Where possible, ATSDR
used site-specific information regarding the frequency and duration of exposures. When sitespecific information was not available, ATSDR employed several conservative exposure
assumptions to estimate exposures.
The following general equation was used to calculate exposure doses:
Estimated exposure dose = C × IR × EF × ED

BW × AT

Page | 64

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

where:
C =
IR =
EF =
ED =
BW =
AT =
§

Concentration of chemical in parts per million (ppm, which is also mg/kg)

Intake Rate—varies with media§

Exposure Frequency, or number of exposure events per year of exposure—

varies with media§
Exposure Duration, or the duration over which exposure occurs: adult = 70
years; child = 6 years
Body Weight: adult = 70 kg; child = 28.1 kg (mean weight of an 8-year-old
child; EPA 1997)
Averaging Time, or the period over which cumulative exposures are averaged:
adult = 70 years*365 days/year; child = 6 years*365 days/year

The intake rate and exposure frequency factors are different for each media (e.g., air, soil, water) and for different
ages among the receptor population (i.e., the people who are actually or potentially exposed). These assumptions
are described during the media-specific health evaluations.

Using health guidelines to evaluate potential health hazards
Noncancer effects
ATSDR analyzes the weight of evidence of available toxicologic, medical, and epidemiologic
data to determine whether exposures might be associated with harmful health effects. As part of
this process, ATSDR examines relevant health effects data to determine whether estimated doses
are likely to result in harmful health effects. As a first step in evaluating noncancer effects,
ATSDR compares estimated exposure doses to conservative health guideline values, including
ATSDR’s minimal risk levels (MRLs) and U.S.EPA’s reference doses (RfDs). MRLs and RfDs
are based on noncancer health effects only. Proposed MRLs undergo a rigorous scientific review
process:
• Health Effects/MRL workgroup reviews within ATSDR’s Division of Toxicology,
• External expert panel peer reviews; and
• Agency-wide MRL workgroup reviews, with participation from other federal agencies,
including U.S.EPA.
The MRLs are then submitted for public comment. MRLs are derived when data are sufficiently
reliable to identify the target organs of effect or the most sensitive health effects for a specific
duration for a given route of exposure.
Proposed RfDs also undergo rigorous internal and external peer reviews and are submitted for
agency consensus, technical editing, and quality assurance.
MRLs and RfDs are estimates of the daily human exposure to a hazardous substance likely to be
without appreciable risk of adverse noncancer health effects over a specified duration of
exposure. These substance-specific estimates, which are intended to serve as screening levels,
are used to rule out contaminants at levels that are not expected to cause adverse health effects. It
is important to note that MRLs are not intended to define clean-up or action levels. MRLs are
intended only to serve as a screening tool to help public health professionals decide where to
look more closely.

Page | 65

MRLs and RfDs are derived for hazardous substances using
the no-observed-adverse-effect level (NOAEL)/lowest­
observed-adverse-effect level (LOAEL)/uncertainty factor
approach. They are below levels that might cause adverse
health effects in the people most sensitive to such effects.
The LOAEL is the lowest tested dose of
Most MRLs and RfDs contain a degree of uncertainty
a substance in a study that has been
because of the lack of precise toxicologic information on the
reported to cause harmful (adverse)
people who might be most sensitive (for example, infants,
health effects in people or animals.
the elderly, or persons who are nutritionally or
immunologically compromised) to the effects of hazardous substances. Consistent with the
public health principle of prevention, ATSDR uses a conservative (that is, protective) approach
to address this uncertainty.
The NOAEL is the highest tested dose
of a substance in a study that has been
reported to have no harmful (adverse)
health effects on people or animals.

MRLs and RfDs are generally based on the most sensitive noncancer end point considered of
relevance to humans. Exposure to levels above the MRL or RfD does not mean that adverse
health effects will occur. Estimated doses at or less than these values are not considered of health
concern. To maximize human health protection, MRLs and RfDs have built-in uncertainty or
safety factors, making these values considerably lower than levels at which health effects have
been observed. The result is that even if a dose is higher than the MRL or RfD, it does not
necessarily follow that harmful health effects will occur.
Table 7 shows the health guidelines (MRLs and RfDs) developed for the different forms of
mercury referenced in this public health assessment. Also, see Figure 12 for levels of significant
exposure to elemental mercury, Figure 13 for levels of significant exposure to inorganic
mercury, and Figure 14 for levels of significant exposure to organic mercury. More detailed
toxicological studies and information are available in ATSDR’s Toxicological Profile for
Mercury (ATSDR 1999) and U.S.EPA’s Integrated Risk Information System (IRIS)—a database
of human health effects that could result from exposure to various substances found in the
environment (EPA 1993, 1995a, 2002c). ATSDR’s toxicological profile for mercury is available
on the Internet at http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24 or by contacting
the National Technical Information Service (NTIS) at 1-800-553-6847. IRIS is available on the
Internet at http://www.epa.gov/iris. For more information about IRIS, please call U.S.EPA’s
IRIS hotline at (202) 566-1676 or send an e-mail to [email protected]. Additional information
is provided in Appendix D. Toxicologic Implications of Mercury Exposure.

In a clinical human population study of exposure, an adverse effect is typically reported only if seen in 1 percent or more of
the study population. That does not mean that anyone who is exposed to the substance has a 1 percent chance of having a
particular adverse effect that was seen in 1 percent of the study population. It just means that that effect may be seen in an
“exposed” population of comparable size to the clinical study population.
In an epidemiological study, it takes a population of exposed individuals to determine whether an effect seen has any
statistical significance. Any health effects cannot be attributed to a single exposure dose. Therefore, ATSDR cannot predict
with any certainty whether a single person with an exposure above a health guidance value such as an MRL or RfD that is
based on a large study population will have a particular effect. It takes a substantial population to identify a causal
relationship between exposure and effect.

Page | 66

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

If health guideline values are exceeded, ATSDR examines the health effects levels discussed in
the scientific literature and more fully reviews exposure potential. ATSDR reviews available
human studies as well as experimental animal studies. This information is used to describe the
disease-causing potential of a particular chemical and to compare site-specific dose estimates
with doses shown in applicable studies to result in illness (known as the margin of exposure).
This process enables ATSDR to weigh the available evidence in light of uncertainties and offer
perspective on the plausibility of harmful health outcomes under site-specific conditions.
When comparing estimated exposure doses to actual health effects levels in the scientific
literature, ATSDR estimates doses based on more realistic, site-specific, exposure scenarios to
use for comparison. In this level of the evaluation, an average concentration is used to calculate
exposure doses to estimate a more probable exposure. This approach is taken because it is highly
unlikely that anyone would contact the maximum concentration on a daily basis and for an
extended period of time.
Cancer effects
Animal studies provide limited information about whether mercury causes cancer in humans
(ATSDR 1999). U.S.EPA has determined that mercuric chloride and methylmercury are possible
human carcinogens (EPA 2012a, 2012b). International Agency for Research on Cancer (IARC)
has determined that methylmercury compounds are possibly carcinogenic to humans (Group 2B),
and metallic mercury and inorganic mercury compounds are not classifiable as to their
carcinogenicity to humans (Group 3) (IARC 1997). The National Academy of Sciences (NAS)
concluded that studies on carcinogenic effects in humans are inconclusive (NRC 2000). Some
studies observed an increase in incidence of renal tumors in male mice from chronic exposure to
methylmercury, however, that effect was observed only at doses that were toxic to the kidney
and is thought to be secondary to cell damage and repair. Exposure to methylmercury did not
increase tumor rates in female mice or rats of either sex (NRC 2000). Therefore, the focus of
methylmercury exposure in this public health assessment will be on the most sensitive endpoint
for methylmercury toxicity (i.e., noncancer neurodevelopmental health effects). As explained
here, whether or not mercury causes cancer is still under scientific debate. However, basing the
public health evaluation of methylmercury exposure in this public health assessment on the most
sensitive endpoint of mercury exposure—neurodevelopmental effects—is likely protective of
any potential carcinogenic effects.

Page | 67

NOAEL; No renal effects were observed
in rats administered 0.23 mg/kg/day 5
days a week for 26 weeks.

The highest NOAEL is 3.0 mg/m3 in
rats. The lowest less serious LOAEL is
0.17 mg/m3 in mice (serum
antinucleolar antibodies).
Kishi et al. 1978; Warfvinge et al. 1995

Dose and Endpoint

Source

MRL NOAEL; No adverse effects were observed in over 700
mother-infant pairs exposed to doses of 0.0013 mg/kg/day in
fish for 66 months.
RfD LOAEL; A 5% increase in neurodevelopmental effects
were observed in the 7-year-old offspring of 900 mothers with
a benchmark dose lower limit (BMDL05) range of 46 to 79 ppb
methylmercury in maternal cord blood. This BMDL05 equates
to doses of 0.000857-0.001472 mg/kg/day.
NAS health effect level; A 5% increase in abnormal scores on
Boston Naming Test was observed in offspring of mothers with
a BMDL of 58 ppb methylmercury in maternal blood cord. The
BMDL equates to a dose of 0.0011 mg/kg/day.
Davidson et al. 1998; Grandjean et al. 1997; NRC 2000

LOAELs; Autoimmune effects were
observed in rats exposed to doses of
0.226, 0.317, and 0.633 mg/kg/day.
U.S.EPA notes that the oral RfD was
“arrived at from an intensive review and
workshop discussions of the entire
inorganic mercury data base.”

Andres 1984; Bernaudin et al. 1981;
Druet et al. 1978

LOAEL; Increased frequency of hand
tremors were observed in male
workers exposed to doses of 0.026
mg/m3 for about 15 years.

Fawer et al. 1983

Dose and Endpoint

Source

Page | 68

0.0003 mg/kg/day (MRL)
0.0001 mg/kg/day (RfD)

0.0003 mg/kg/day (RfD)

0.0002 mg/m3 (MRL)

Cox et al. 1989; Magos and Butler 1972

The highest NOAEL is 0.84 mg/kg/day in rats. The lowest less
serious LOAEL is 0.0012 mg/kg/day in human infants (delayed
walking, abnormal motor scores).

Not available

Chronic MRL/RfD

NTP 1993

0.002 mg/kg/day

Not available

Intermediate MRL

Cox et al. 1989 ; Yasutake et al. 1991

Fredriksson et al. 1992

Source

NTP 1993

The highest NOAEL is 24 mg/kg/day in mice. The lowest less
serious LOAEL is 0.0012 mg/kg/day in human infants (delayed
walking, abnormal motor scores).

Dose and Endpoint

Acute MRL
NOAEL; No renal effects were observed
in rats administered 0.93 mg/kg/day
once daily for 14 days, excluding
weekends.

Developmental effects in offspring

There are no NOAELs. The lowest
serious LOAEL is 0.05 mg/m3 in rats
(hyperactive offspring, significantly
impaired spatial learning).

Kidneys
Not available

Central nervous system and kidneys

Primary target organ

Ingestion (fish)

Ingestion (soil, sediment, surface water,
plants)

Organic Mercury

0.007 mg/kg/day

Inhalation (air)

Primary exposure pathway

Methylmercury

Mercuric chloride and mercuric nitrate

Inorganic Mercury

Not available

Metallic mercury

Example

Elemental Mercury

Table 7. Health Guidelines for the Forms of Mercury

Figure 12. Levels of Significant Exposure to Elemental Mercury

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Page | 69

Figure 13. Levels of Significant Exposure to Inorganic Mercury

Page | 70

Figure 14. Levels of Significant Exposure to Organic Mercury

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Page | 71

IV.

Public Health Evaluation

IV.A. Past Exposure (1950–1990)

IV.A.1.

Potentially Exposed Communities

The potentially exposed communities ATSDR used to evaluate exposures to past mercury
releases from the Y-12 operations are the same as those selected in the Task 2 report (ChemRisk
1999a), namely Wolf Valley residents, Scarboro community residents, Robertsville school
children, East Fork Poplar Creek farm families, Oak Ridge community residents (two
populations), and several fish consumer populations who ate fish from Watts Bar Reservoir,
Clinch River/Poplar Creek, and EFPC (see Table 8 and Figure 15).

Page | 72

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Mercury Species

Wolf Valley Resident

Scarboro Community
Resident

Robertsville School–
General Student

Robertsville School–
Student Recreator

EFPC Floodplain Farm
Family

Oak Ridge Community
Populations (2)

Clinch River/Poplar Creek
Fish Consumer

Watts Bar Reservoir Fish
Consumer

Table 8. Task 2 Exposure Pathways for Which Mercury Doses were Estimated

for Each Potentially Exposed Community

Inhalation

Elemental

Xa

Xb

Xc

Xc

Xc

Xc

E

E

Fruit/vegetable consumption

Inorganic

Xa

Xb

E

E

Xc

Xc

E

E

Milk consumption

Inorganic

Xa

E

E

E

Xc

E

E

E

Beef consumption

Inorganic

Xa

E

E

E

Xc

E

E

E

Inorganic

E

X

X

X

X

E

E

E

Skin contact with soil

Inorganic

E

X

X

X

X

E

E

E

Vegetable consumption

Inorganic

E

X

E

E

X

E

E

E

Milk consumption

Inorganic

E

E

E

E

X

E

E

E

Beef consumption

Inorganic

E

E

E

E

X

E

E

E

Inorganic

E

X

E

X

X

E

E

E

Inorganic

E

X

E

X

X

E

E

E

Surface water pathways
Incidental ingestion of water

Inorganic

E

X

E

X

X

E

E

E

Skin contact with water

Inorganic

E

X

E

X

X

E

E

E

Milk consumption

Inorganic

E

E

E

E

E

E

E

E

Beef consumption

Inorganic

E

E

E

E

E

E

E

E

Exposure Pathway
Air pathways

Soil pathways
Soil ingestion

Sediment pathways
Sediment ingestion
Skin contact with sediment

Fish consumption

Methylmercury

E
X
E
E
X
E
X
X
Source: ChemRisk 1999a
Xs indicate that the exposure pathways were evaluated for the potentially exposed community.
Es indicate that the exposure pathways were eliminated. Exposure pathways were eliminated if site characteristics
make past, current, and future human exposures extremely unlikely.
a
Evaluated for direct airborne releases of mercury from the Y-12 plant.
b
For 1953–1962, evaluated for both direct airborne releases of mercury from the Y-12 plant and volatilization of
mercury from EFPC; for the remaining years, evaluated for volatilization of mercury from EFPC only.
c
Evaluation for volatilization of mercury from EFPC only.

Page | 73

Figure 15. Task 2 Potentially Exposed Communities

Page | 74

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

IV.A.2.

Past Air Exposure Pathway

Task 2 Estimated Y-12 Mercury Releases to Air
When lithium separation studies began at the Y-12 plant, mercury was known to pose a health
hazard to people who inhaled mercury vapors. Y-12 personnel were concerned about indoor air
mercury concentrations; they made efforts to reduce and maintain
Airborne mercury contaminants
indoor air mercury concentrations below the acceptable worker
at the Y-12 plant may have
standard at the time (0.1 mg/m3). Engineering controls, such as the
occurred as a result of primary
installation of large high-speed exhaust fans in the buildings,
operations and accidental
releases. Information pertaining
helped to reduce indoor air mercury concentrations, but possibly
to air mercury releases is largely
increased mercury vapor releases off site. Other modifications,
based on available statistics
such as resurfacing indoor building walls to reduce microscopic
regarding process operations,
mercury adhesion and flooding building floors with water or
accidents, on-site and off-site
sodium thiosulfate solutions to suppress the vaporization of spilled
release monitoring data, and air
dispersion modeling.
mercury, would have decreased the indoor air mercury
concentrations, as well as the release of mercury to the outdoors.
Three investigation teams (1977 Mercury Task Force, 1983 Mercury Task Force, and Task 2
team) independently estimated air mercury releases from the Y-12 plant. Specifically, Task 2
studied building engineering reports that included flow and ventilation diagrams, exhaust
measurements, and information on the upgrade of ventilation systems. Task 2 also gathered
hundreds of weekly-, monthly-, and quarterly-average indoor air measurements that were only
made in some of the pilot and production buildings for a select period of time during lithium
isotope separation operations. To compensate for missing data, air concentrations and flow rates
were estimated, based on similar conditions in buildings where measurements had been made.
Task 2 identified 114 point sources that included 62 stacks, 43 fans, and 9 vents on 9 buildings.
The buildings included three main production facilities, three steam plants, a mercury storage
warehouse, a scrap metal furnace, and Building 81-10, which housed the mercury recovery
furnace. 8 A separate source term was estimated for each point source for each year that the
source was known to have been in operation (1953–1962). Air source terms are expressed in
units of mass per unit time. Task 2 estimated that a total of 73,000 pounds of mercury had been
released from Y-12 operations during the 11 years of lithium isotope separation activities (see
Figure 16). This represents a 43 percent increase over the 1983 Mercury Task Force estimates.
None of the three investigation teams estimated Y-12 air mercury releases for the years before or
after the 1953–1962 operational time period.

8

Building 81-10 was a facility at Y-12 designed to recover mercury from waste sludge materials through draining
and evaporation. Air releases from the furnace occurred because of incomplete condensation of evaporated
mercury. The furnace in Building 81-10 operated from March 1957 through July 1962, and physical separations
continued through September 1982. More than 3 million pounds of mercury were recovered from waste materials
in Building 81-10.
Page | 75

Mercury Release Estimates
(pounds)

Figure 16. Task 2 Estimated Mercury Releases to Air from Y-12 Operations (1953–1962)

25,000
20,000
15,000
10,000
5,000
0
1953

1954

1955

1956

1957

1958

1959

1960

1961

1962

Year
Source: ChemRisk 1999a

ATSDR scientists did not attempt to verify or reproduce the Task 2 air source terms—that work
is beyond the scope of this public health assessment. Consequently, the quality of the Task 2 air
mercury source terms was not evaluated. But confidence in those estimates is high: three
separate teams have studied the applicable records over the years. As a result, each team has
made contributions to our understanding of the activities at the Y-12 plant that resulted in air
mercury releases. ATSDR accepts the Task 2 air mercury source terms with one reservation—
Task 2 stated that it did not develop a source term for certain mercury spills to soil “because any
mercury runoff to EFPC within the plant boundary and before the [water] sampling location
would have been included in the mercury concentrations measured [in water] at the site
boundary.” All the mercury spills to soil, however, did not go into EFPC.
Some mercury spills to the ground were routed to the storm sewer system, which fed into EFPC.
In 1957, after the mercury recovery furnace was constructed in Building 81-10, some mercury
spills were removed and taken to the furnace. But no estimates are available of how long
mercury from any spill was on the ground and how long that mercury emitted vapors before it
was contained or removed. The percent recovery of mercury after some of the spills was low.
The 1983 Mercury Task Force estimated that 85,000 pounds of mercury were “not recovered”
after a major spill occurred outside between production buildings in 1956; and, 3,000 pounds of
mercury were lost to the ground (as of 1971) at Building 81-10. In another example, shelves
containing mercury flasks collapsed under the load inside a building and resulted in spilled
mercury. It is not known whether indoor air measurements or window exhaust estimates
reflected the effects from these types of incidents.
With no data to describe air releases from outdoor mercury spills, estimating air mercury releases
from historic on-site mercury spills is not possible. In addition, Task 2 did not estimate air
mercury releases from mercury spill to soils. The description of mercury spills suggests that they
may have been a source of substantial air mercury releases, but spill information is not sufficient
to estimate air concentrations and subsequent health effects.

Page | 76

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Finally, Task 2 and the 1983 Mercury Task Force reported air mercury releases that were not
used to develop the Task 2 source terms for Y-12 plant releases. For example, the K-25
powerhouse, near S-50, emitted 319 pounds of mercury annually from 1953 to 1961 and half that
amount in 1962 (ChemRisk 1999a). The total air mercury releases for these years is
approximately 4 percent of the total amount of the estimated air mercury releases from the Y-12
plant. Yet in individual years, the mercury released from the K-25 powerhouse was as much as
20 percent of the amount released from the Y-12 plant in 1953. The Task 2 team did not evaluate
the impact of the K-25 air mercury releases to the Task 2 potentially exposed communities,
presumably because the releases did not come from the Y-12 plant and the effect on the
potentially exposed communities was thought to be insignificant.
Mercury Concentrations in Air
Significant releases of elemental mercury to air from the Y-12 plant
occurred from 1953 to 1963, the years of production-scale lithium
separation activities. The peak Y-12 mercury releases to air occurred in
1955. Task 2 concluded that the volatilization of mercury from EFPC
could have significantly contributed to air mercury concentrations near
the EFPC floodplain. The evidence for this conclusion is the presence of
elevated mercury concentrations in tree-core samples collected in 1993,
from red cedars growing in the EFPC floodplain (near the location where
East Tulsa Road crosses EFPC).

The primary exposure
pathway to mercury in
air is the direct
inhalation of airborne
elemental (or metallic)
mercury. Other forms
of mercury are not
considered an
inhalation hazard.

Moreover, mercury evasion from water is partly a function of the concentration of mercury in the
water. The air above EFPC would have been an important source of mercury, primarily from
1953 to 1963, when the Y-12 lithium separation program was active and Y-12 mercury releases
to water were greatest. Peak Y-12 mercury releases to water occurred in 1957. Although releases
of mercury to EFPC water did not cease when the lithium separation program ended, they
decreased considerably. This was due to 1959 process changes and due to additional abatement
efforts in later years. Total mercury concentrations decreased from a high of 14.5 milligrams per
liter (mg/L) in effluent in 1958 to below 1 mg/L after 1962, and below 0.1 mg/L after 1974, 9
according to weekly measurements in EFPC at the Y-12 plant (ChemRisk 1999a).
But off-site mercury air exposures from the Y-12 plant have another important source. ATSDR
has ample anecdotal information presented in public meetings that in the past Y-12 workers
intentionally brought metallic mercury home with them (e.g., to show their children). Or they
unintentionally brought mercury home on their work boots and clothing. In either case, it is very
possibly mercury was lost or dispersed in homes and therefore posed an indoor air hazard.
ATSDR has no quantitative data to evaluate the magnitude of this hazard in the communities
surrounding the ORR. Still, elemental mercury has a high vapor pressure. And that air exposures
to elemental mercury vapor indoors can be a greater hazard than outdoor air mercury exposures
is well known today. Elemental mercury in a home is easily lost into carpeting, flooring,
furniture, drapes, and other household materials. The body of literature identifying this hazard
has grown in recent years. The possibility for adverse effects from breathing mercury vapor,
particularly among children, can be significant. ATSDR believes this exposure pathway may
have continued well beyond the years when the lithium isotope separation process ended in
1963.
9

Data through 1982, though some values are missing.
Page | 77

Three Task 2 Models
The earliest off-site ambient air mercury concentrations were measured in 1986. Therefore, no
air data are available from the years that air and water mercury releases from the Y-12 plant were
highest. To compensate for the lack of data, Task 2 modeled the average annual air mercury
concentrations for six potentially exposed communities in or near Oak Ridge (Table 9). Task 2
used three different models to estimate annual air mercury concentrations for each off-site
community, depending on its location. (See Appendix E. Task 2 Pathway Discussions for a more
detailed discussion of the three Task 2 air mercury models.)
Table 9. Three Task 2 Air Models and Potentially Exposed Communities
Potentially Exposed
Communities
Wolf Valley

U.S.EPA
Dispersion
Model

EFPC
xlQ
Volatilization
Model
Model

X

Scarboro Community

X

X

Robertsville School

X

EFPC Floodplain

X

Oak Ridge 1

X

Oak Ridge 2

X

Among the three models that Task 2 used, the U.S.EPA ISCST3 Dispersion Model and the xlQ
Model depend on the estimated air mercury releases during Y-12 operations. The third model,
EFPC Volatilization, depends on the water mercury releases during Y-12 operations. One
limitation of all three air models is that they produce average annual air mercury concentrations
that cannot be used to evaluate acute exposures. Therefore, whether spills or other activities at
the Y-12 plant resulted in mercury air plumes that caused short-term adverse health effects is
unknown. The 1983 Mercury Task Force report listed these significant mercury spills:
• In1956, an estimated 180,000–400,000 pounds of mercury spilled
• In 1966, a spill totaled 105,000 pounds of mercury
• An undetermined number of spills occurred from 1951–1955 that exceeded 100,000 pounds
of mercury (UCCND 1983a, 1983b).
These spills were not necessarily outdoors, and the mercury was not necessarily disposed of in
the environment. Some of the mercury was recovered for reuse. But information is insufficient
to determine whether any of these events—or others—could have led to air mercury
concentrations off site that resulted in short-term adverse health effects. Task 2 estimated the
average annual air mercury concentrations to evaluate chronic inhalation exposures.
U.S.EPA Dispersion Model
Of the three models, the U.S.EPA Dispersion Model used to predict air concentrations in Wolf
Valley was the most reliable. This model uses a Gaussian dispersion equation to calculate air
concentrations at a remote location from the releases. It is an appropriate model to use in
relatively flat terrain. Therefore, the selection of this model for this application appears to be
appropriate. ATSDR considers Task 2 team’s reported estimates of air mercury concentrations in
Wolf Valley resulting from this model to be reasonable.
Page | 78

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

The Task 2 estimated air mercury concentrations in Wolf Valley ranged from 0.0000008 to
0.000014 milligrams per cubic meter (mg/m3) for the 1953 through 1962 time period (ChemRisk
1999a). The peak value (0.000014 mg/m3) was in 1955. Task 2 estimated that the uncertainty
associated with the modeled air concentrations in Wolf Valley was ± 44 percent of the true
concentration values.
ATSDR compared the highest estimated mercury concentration in Wolf Valley (0.000014
mg/m3) to the ATSDR chronic inhalation MRL for elemental mercury vapor (0.0002 mg/m3).
The highest annual concentration is more than 14 times lower than the ATSDR MRL. Even with
the Task 2 uncertainty added, the upper-bound average concentration is 10 times lower than the
ATSDR MRL. ATSDR concludes, then, that the mercury concentrations in the air in Wolf
Valley were not expected to have posed a chronic public health hazard for the period of study.
ATSDR cannot evaluate or draw a conclusion about acute, short-term exposures. Task 2
conducted an analysis of mercury doses to Wolf Valley residents and reached the same
conclusion.
Chi over Q (x/Q) Model
Task 2 used the "chi over Q" (xlQ) Model and the EFPC Volatilization Model to estimate air
mercury concentrations in the Scarboro community. The xlQ Model is based on two physical
quantities: the measured air uranium concentrations in Scarboro (x) and uranium release rates
from the Y-12 plant to the air (Q). The basis of this model is the assumption that air mercury
releases from Y-12 will follow a physical pattern similar to air uranium releases from Y-12. But
no evidence supports that assumption. Specifically, ATSDR’s
Task 2 had planned to use tree-ring
evaluation of the Task 2 team’s use of this model reveals that 1)
mercury concentrations to estimate
uranium would be in the form of particulate whereas mercury
air mercury concentrations in the
would largely be in the form of vapor, 2) evidence suggests that
EFPC floodplain, but the tree core
the average mercury vapor droplet size would be much smaller
data collected in 1993 suggested
that the mercury did not stay put in
than the size of uranium particles associated with Y-12
individual rings. Therefore, Task 2
operations, and 3) it is unclear whether the xlQ "custom
could not reliably assign the
distribution” accurately depicts the relationship between the
measured mercury concentrations
mercury quantities released from Y-12 and the air mercury
in specific tree rings to specific
concentrations in Scarboro. Therefore, ATSDR does not accept
years. As a result, Task 2
abandoned its effort to estimate
that the xlQ model reliably predicted past air mercury
annual historic air mercury
concentrations in the Scarboro community. See Appendix E for
concentrations from tree core data.
more information on ATSDR’s evaluation of this model’s use in
Task 2.
EFPC Volatilization Model
Due to the volatilization of mercury from EFPC, Task 2 used the EFPC Volatilization Model to
estimate air mercury concentrations for the following potentially exposed communities: Scarboro
community, EFPC floodplain farm family, Robertsville School children, and two populations in
Oak Ridge (“Oak Ridge 1” on Louisiana Avenue and “Oak Ridge 2” on Jefferson Avenue).
The Task 2 report suggests it used the EFPC Volatilization Model because of the absence of an
adequate air dispersion model that could predict historic air mercury concentrations beyond
Scarboro. Task 2 gave as an additional reason the presence of significant mercury levels in treecore samples.

Page | 79

The EFPC Volatilization Model estimated air
mercury concentrations from the amount of mercury
released from the Y-12 plant to the creek, the
distance the mercury traveled in the water, and the
fraction of the mercury mass in the water that
volatilized into the air. The pivotal feature of the
EFPC Volatilization Model is the volatilization
fraction, which is the fraction of metallic mercury
mass in EFPC that volatilized from the water. Task 2
assumed a log triangular distribution of values, with
a minimum value a “best estimate,” and a maximum
value equal to 1, 5, and 30 percent, respectively, of
the total mercury mass released annually to the
creek. Task 2 apparently selected these values from
data collected in the 1990s. ATSDR suggests that
conditions in EFPC were too different in the 1990s
compared with the 1950s to warrant unqualified
application of those values. Task 2 did not explain
how it derived the volatilization fractions it used,
and ATSDR believes this key variable needs to be
justified. Finally, Task 2 adopted a log triangular
distribution of the volatilization fractions, also
without explanation or justification. ATSDR is not
aware of any evidence that supports the assumption
that volatilization fractions are distributed in this
way. ATSDR concludes that the EFPC Volatilization
Model is only qualitatively supported by tree-core
data, not quantitatively supported, and that the model
does not provide reliable predictions of air mercury
concentrations off site from the Y-12 plant.

Information Regarding the Tree Core Ring
Samples
1. Although the tree core data cannot establish
annual air mercury concentrations, they
indicate that air mercury concentrations were
elevated during the 1950s and 1960s,
compared with later decades in areas beyond
Scarboro. However, the tree core data cannot
indicate from where the mercury came.
2. Task 2 indicated that mercury concentrations in
the tree core ring corresponded to 1938. This
concentration was higher than in subsequent
years in a tree on the west end of the Y-12
property. It is not known whether this mercury
may have been absorbed in later years and
migrated toward the center of the tree, or
whether it was absorbed prior to the Manhattan
Project.
3. Unfortunately, the EFPC tree core samples
were all collected from red cedars in the same
vicinity of the EFPC floodplain, which is on the
eastern-most end of EFPC, near Illinois
Avenue and East Tulsa Road. This area could
have been impacted by air releases from the Y­
12 plant, or sources other than the ORR. A
more representative sampling of trees along
the EFPC floodplain might have provided
quantitative support that mercury volatilization
from EFPC declined with distance from Y-12,
and that volatilization was responsible for
increased air mercury concentrations.
Alternatively, the samples may have indicated
that additional mercury sources were affecting
the communities around the ORR.

Task 2 Results
Using a 30 percent volatilization fraction, Task 2 estimated that mercury air concentrations in the
EFPC floodplain and in Scarboro exceeded the inhalation MRL (0.0002 mg/m3) during the years
1953 through 1961, and from 1957 through 1958, respectively. Using a 5 percent volatilization
fraction, Task 2 air concentrations in the EFPC floodplain exceeded the MRL for the years 1957
and 1958, and did not exceed the MRL at all in Scarboro (ChemRisk 1999a). Using the
assumption that 1 percent of the mercury mass in EFPC volatilized from the water, none of the
estimated air mercury concentrations for any potentially exposed community exceeded the MRL
for any year. These results reflect the relative magnitude of mercury released from the Y-12 plant
to water in different years, the distance of the potentially exposed communities from the creek,
and the assumed mercury volatilization fractions. That said, the few environmental data available
do not support the key model assumptions that volatilization of mercury was proportional to
distance from the Y-12 plant and formed a log triangular distribution from 1 to 30 percent with a
“best estimate” value of 5 percent.

Page | 80

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Past Air Exposure Pathway Summary
• None of the Task 2 models are adequate for evaluating possible past, short-term (acute) air
exposures to mercury vapor.
• ATSDR believes the U.S.EPA ISCST3 Dispersion Model is an appropriate model for
estimating annual air mercury concentrations in Wolf
Mercury Emissions from Selected
Valley.
Electricity Generating Facilities
• ATSDR’s chronic inhalation mercury MRL is the basis
for evaluating the Task 2 estimated average annual air
mercury concentrations in Wolf Valley.
• ATSDR does not believe the xlQ Model or the EFPC
Volatilization Model is adequate for quantitatively
estimating annual air mercury concentrations for any
potentially exposed community.
• Elemental mercury taken into the home could have been
spilled, resulting in unsafe indoor air mercury
concentrations.
Past Air Exposure Pathway Conclusions
The following conclusions refer to the past potential for
mercury in air from the Y-12 plant to cause harm. The
conclusions are not a measure of the past occurrence of
adverse health effects. Health outcome and exposure data
are unavailable that allow for an evaluation of the actual
occurrence of adverse health effects during the 1950s and
1960s from exposure to mercury in air.
ATSDR concludes
• Elemental mercury carried from the Y-12 plant by
workers into their homes could potentially have harmed
their families—especially young children—in the past
(1950–1963).

ERG, an independent contractor for
ATSDR, evaluated whether electric
generating facilities in close proximity to
the Y-12 plant would lead to air
concentrations of health concern. ERG
concluded the following:
EPA’s “Mercury Study Report to
Congress” suggests that emissions from
coal-fired power plants have extremely
limited incremental effects on groundlevel air quality. The modeling analyses
EPA conducted on a hypothetical coalfired power plant found essentially no
ground-level impacts at locations 2.5
kilometers (km), 10 km, and 25 km
downwind.
Consistent with these general findings,
ERG’s screening modeling analysis
showed that past mercury emissions from
the Tennessee Valley Authority’s
Kingston Fossil Plant almost certainly did
not have substantial air quality impacts
(i.e., concentrations approaching the
reference concentration) near the Y 12
plant, even when considering a series of
health-protective assumptions.
A copy of ERG’s memo to ATSDR is
included in Appendix F. Evaluation of
Mercury Emissions from Selected
Electricity Generating Facilities.

• Air and water mercury releases from the Y-12 plant after 1963 are not expected to have
harmed people living off site near the ORR.
• ATSDR concludes that breathing past (1950–1963) air mercury releases from the Y-12 plant
is not expected to have harmed people living off site in the Wolf Valley area.
ATSDR cannot conclude
• Whether off-site populations breathing elemental mercury releases in the past (1950–1963)
from the Y-12 plant could have been harmed, except for the Wolf Valley area where harm is
not expected.
• Whether people living near the EFPC floodplain breathing mercury vapors from Y-12
releases to the water from 1950 through 1963 could have been harmed.

Page | 81

IV.A.3.

Past Surface Water Exposure Pathway

Y-12 Mercury Releases to Water
Unlike exposure to mercury in air, the health hazards posed by
exposure to mercury in water were generally unknown before 1970.
Therefore, during the years of lithium isotope separation operations,
Y-12 managers were not concerned that releases of mercury to
water would affect human health or the environment. From an
economic standpoint, Y-12 administrators were more concerned
about mercury losses—mercury was a valuable commodity at the time.

Mercury contamination of
water sources at the Y-12
plant may have occurred as a
result of primary operations,
waste disposal activities, or
accidental releases.

Y-12 Mercury Releases to EFPC
Y-12 mercury releases to EFPC were highest during the years when the lithium separation
program was active. Research and development for the lithium separation processes began in
1950, and full-scale production began in 1953. Water mercury releases peaked during 1957 and
1958, but some mercury continued to enter the creek after the lithium separation operations shut
down in June 1963 (WJ Wilcox, Jr., personal communication, March 17, 2005). Subsequent
sources of mercury to EFPC included on-site cleaning operations and seepage from mercury
deposits inside building walls, ducts and equipment, and under floors. Today, the Y-12 National
Security Complex continues to release very small amounts of mercury into EFPC.
Y-12 Mercury Releases to the Storm Sewer System
The primary path by which mercury entered EFPC was via the storm sewer system that ran
through the Y-12 property. The main production buildings disposed of their liquid wastes into
collection tanks, and mercury was routinely removed from them. Overflow from the collection
tanks entered the storm sewer system that led into EFPC.
In the production waste streams, mercury was in the form of dissolved inorganic mercuric ions.
During the Colex process, liquid wastes were in the form of dilute nitric acid solutions. Nitric
acid was used to remove impurities from water and mercury used in the lithium separation
process. But washing the mercury with nitric acid dissolved a substantial amount, which then
entered the storm sewer and EFPC. When the nitric acid wash procedure was modified in June
1958, the mercury released off site through the storm sewer significantly reduced.
Indoor and outdoor mercury spills were also fed into the storm sewer. Mercury spills would have
included mercuric ions in liquid solutions and liquid elemental or metallic mercury. Spills
occurred in the production buildings, between the production buildings, in the loading area,
around the Building 81-10 recovery operations, and during stripping operations (cleaning,
tearing down, or salvaging equipment).
Y-12 Mercury Releases to New Hope Pond
In 1963, New Hope Pond was created in EFPC, downstream of the Y-12 buildings on the Y-12
property. The pond was intended to serve as a mixing location to stabilize the fluctuation of pH
in the water that flowed from the Y-12 operations. Before constructing the pond, the water pH
value that led into EFPC ranged between 3 and 12. The pond served to bring the pH into
acceptable limits (6–9) to protect fish and other aquatic life, as stipulated by the State of
Tennessee. After the pond was constructed, it became a settling location for mercury, which

Page | 82

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

reduced the amount of mercury traveling off site. New Hope Pond was dredged in 1973, and
closed, cleaned, and filled in 1989 (ChemRisk 1999a; SAIC 2007).
Estimated Mercury Releases to Water
The 1983 Mercury Task Force and Task 2 scientists used measured concentrations of mercury in
water samples on the Y-12 property. They also used measurements of the storm sewer/EFPC
water flow rate to estimate mercury releases to EFPC (see Table 10).
Table 10. Estimated Y-12 Mercury Releases to Water
Terms
Equation:

Mercury
concentration
mercury mass
volume

Example:

2.22 mg/L

(1957)

(1.85 E-5 pounds/gal)

Sources of
data:

measurements from water
samples or estimated
percentage of inventories

1

multiply
x
x

Stream flow rate
volume
time
11.0

MGD1

(4.02 E-9 gal/year)

equals
=
=

water flow
measurements or
assumed default
values

Mercury released
mercury mass
time
72,211 pounds/year
(These quantities are the
source terms for
modeling water mercury
concentrations.)

11 millions of gallons per day is the average from 1955–1957 (ChemRisk 1999a).

Given that some stream flow data, mercury concentration data, or both are absent before 1956
(and both are completely absent before 1953), Task 2 estimated values for those quantities. The
period of 1950–1955 is important—not only because both lithium separation pilot operations and
full-scale production were occurring, but because formal mercury recovery operations had not
yet begun. The operations were new, many changes were made. Spills happened, and the on-site
storm sewer became the means for liquid waste disposal.
For the flow rate estimates during this early period, Task 2 used an average of flow rates
measured in later years (1955–1957). All missing flow rate values were assumed to be 11 million
gallons per day (MGD). For missing mercury concentration data, Task 2 calculated values from
concentration measurements taken in 1953 and 1954. For those years, mercury concentrations in
samples were between 2.9 percent and 7.3 percent of mercury inventories. Task 2 estimated that
mercury losses during 1950–1952 were between 3 percent and 8 percent of the mercury
inventories for those years.
Task 2 estimated that mercury releases to EFPC exceeded 10,000 pounds in 1953, and again in
years 1955–1959. During the peak years of mercury releases to EFPC, more than 72,000 pounds
and 64,000 pounds of mercury were released in 1957 and 1958, respectively. Annual releases
dropped below 1,000 pounds in 1967 (except for a small increase in 1973, probably as a result of
dredging New Hope Pond). They decreased below 100 pounds in 1975. Mercury releases to
EFPC for the years 1988–1990 were below 40 pounds per year (see Figure 17).

Page | 83

Figure 17. Task 2 Estimated Mercury Releases to EFPC

80

Mercury Releases (pounds)

80,000

Mercury Releases (pounds)

70,000
60,000
50,000
40,000
30,000

70
60
50
40
30
20
10
0
1975

1980

20,000

1985

1990

Year

10,000
0
1950

1955

1960

1965

1970

1975

1980

1985

1990

Year
To estimate mercury releases in the early 1950s, Task 2 used data from a relatively small number
of water samples and water flow measurements. The Task 2 report did not state how many
sample data were used, or how well the samples distributed over time. ATSDR does not know
the quality of the data, nor how well the “percentage of inventory” model predicted water
mercury releases for the years 1950–1952. In all likelihood, these limitations will never be
resolved.
Water Sampling at the Y-12 Plant
During the second quarter of 1953, Y-12 employees began collecting water samples to measure
mercury concentrations. The earliest available stream flow data are from 1954; but until
September 1955, the data are sporadic. Fortunately, composite water sample data are available
for the peak years, 1957 and 1958. The highest composite weekly water mercury concentration
was 14.5 mg/L, from a sample collected during the second week of May in 1958. Water samples
were collected in the storm sewer on site, downstream of the Y-12 buildings, and later from the
outlet of New Hope Pond to EFPC. Data were reported in hundreds of weekly, monthly, and
quarterly internal technical and environmental reports over the years.
Water Collection Method before 1956
Between 1953 and 1955, water samples were taken from the surface of the storm sewer stream.
Surface water samples would likely not have captured all of the elemental mercury releases, nor
would it have captured mercury attached to particulate matter—it would have sunk in the water
and followed the course at the bottom of the streambed. Sufficient anecdotal evidence is
available in both the 1983 Mercury Task Force report and the Task 2 report that elemental
Page | 84

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

mercury releases, which occurred prior to 1955, were not accounted for in the early water
measurements. In 1955, a “dipper type” sampler was installed in the storm sewer. But whether
the device would have adequately measured elemental mercury releases is not certain (ChemRisk
1999a).
Acidification of Water Samples
During the early testing period, samples were not acidified at the time of collection. An acidic
pH favors dissolved ionic mercury, and a basic pH favors undissolved, elemental mercury. Once
mercury is in the elemental form, it may evaporate, or it may volatilize from water at ambient
temperatures. Due to the nitric acid in the liquid wastes, the risk of mercury loss from the
samples would probably have been minimal. Not all of the liquid waste streams were acidic,
however.
In 1974, U.S.EPA recommended acidifying water samples collected for mercury analysis to
minimize loss of mercury from the samples due to volatilization. Y-12 staff began acidifying
water samples in the laboratory in 1977. In 1982, water samples were acidified in the field.
Samples collected before 1977 were not acidified. Reported pH measurements of composite
weekly water samples collected from June 1955 through 1959 were between 7.1 and 11.1 (i.e.,
they were all in the basic range). The basic pH favors the formation of dissolved elemental
mercury, which may escape from the water. ATSDR does not know whether the water samples
were capped or sealed prior to analysis, nor whether the absence of acidification of water
samples collected prior to 1977 significantly affected the reported mercury concentrations.
Uncertainty in the Analytical Methods
Until June 1957, Y-12 analytical chemists determined the mercury content of EFPC water using
a colorimetric technique. This method provided a detection limit of 0.1 mg/L with a relative limit
of error for a single analysis of ± 50 percent. In July 1957, the colorimetric method was replaced
by the mercurometer method, which provided a detection limit of 0.01 mg/L, with a relative limit
of error for a single analysis of ± 40 percent. In August 1967, an atomic absorption method was
adopted that provided a detection limit of 0.001 mg/L with a relative limit of error for a single
analysis of ± 20 percent (UCCND 1983a, 1983b). Note that the uncertainties in the
measurements of water mercury concentrations through mid-1967 were relatively large.
Composite Water Sampling Data
The mercury water data of greatest interest were from samples collected weekly until the end of
the lithium separation operations in June 1963. Weekly water sample data from September 1955
through November 1960 are available, with only four data points missing during this period. The
data represent averages of mercury concentrations from composite water samples collected over
the duration of a week. The data from composite water sampling are useful; they allow for a
review of mercury concentrations within the period of acute exposures (2 weeks). Nevertheless,
the data cannot indicate the maximum water mercury concentration that may have occurred
following a single large release over the course of a few hours or a day.
Missing Water Sampling Data
Gaps appear in the weekly water sampling data before September 1955 and after November
1960. Only the gaps in the earlier period, however, appear important. Data from total mercury
release estimates, as well as monthly and quarterly reports, consistently indicate that mercury
releases to EFPC after 1958 did not result in mercury concentrations at levels that would have
Page | 85

posed a public health concern. Whether high acute mercury exposures occurred between 1953
and 1955 is not known, given that the weekly water sampling data and supporting information
are incomplete.
The production scale lithium isotope separation work began using the Elex process in August
1953 and the Colex process in January 1955. (The Orex process never progressed beyond pilot
development.) These were new technologies at the time, and production start-up was marred by
difficult problems such as the loss of mercury. Estimated mercury spills before 1957 ranged from
200,000–500,000 pounds (UCCND 1983a, 1983b). Some of the spilled mercury was recovered,
though the 1983 Mercury Task Force report does not estimate how much went into the water.
From the earliest production days Y-12 managers considered mercury losses from the Colex
process “serious,” and considerable effort went into addressing them.
Fate and Transport of Mercury Releases in Water
Except for a period from 1974 through mid-1977, the
analytical data are measurements of total mercury in the
water. From January 1974 to June 1977, water samples were
filtered and analyzed for soluble mercury only. ATSDR has
a qualitative—not quantitative—knowledge of the species of
mercury in the water: through multiple physical and
chemical processes in the creek, as described below, the
mercury released from the Y-12 plant to EFPC may change
form. These uncertainties are accounted for, to the extent
possible, in the subsequent discussion on the bioavailability
of mercury.

We could not assess acute mercury
exposure because the data were not
representative of an acute exposure
scenario (0–14 days). The monthly
water sample data collection that
began in April 1954 and the quarterly
water sample data collection that
began in June 1953 were combined
averages of the weekly data. The
longer the duration over which
periodic data are averaged, the lower
the peak values. For example, the
average annual water mercury
concentrations were lower than some
of the quarterly concentrations for the
same period, and the average
quarterly concentrations were lower
than some of the monthly
concentrations.

The mercury released into the storm-sewer drainage ditch at
the Y-12 plant was primarily divalent mercuric nitrate and
elemental mercury. Mercuric nitrate is very soluble in water,
but neutralization of the acid in the creek water would have
The longer-period average mercury
formed mercuric oxide, or in the presence of sulfide ion,
water concentration values are
mercuric sulfide. Mercury also adheres to, and forms
appropriate to evaluate average longcompounds with, other inorganic and organic species,
term exposures, but not to estimate
including particulate matter and plant material. The basic pH short-term (acute) exposures.
Because not enough appropriate data
of the composite weekly water samples at Y-12 during the
are available, ATSDR scientists
1950s would have favored the formation of the oxide and
cannot determine whether short-term
sulfide salts, some of which would have precipitated out of
mercury releases to EFPC from 1953–
solution and would have been carried along in the stream.
1955 could have resulted in harmful,
acute exposures.
Some of them would have settled in the streambed or
floodplain soil and diminished the concentration of mercury
in the water. But the 1983 Mercury Task Force noted that suspended mercuric salts could have
“resolubilized” during “acid-dominated periods” when the water released to EFPC was acidic
(UCCND 1983a, 1983b). Basic pH (and warm temperatures) also would have favored
volatilization of dissolved elemental mercury to the air.
In 1995, Saouter et al. reported that water samples collected from the outlet of Reality Lake
(which fed EFPC on Y-12 property) contained approximately 83 percent mercury associated

Page | 86

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

with particulate matter and 17 percent dissolved mercury (Saouter et al. 1995). Methylmercury
was less than 0.1 percent of the total mercury concentration of 0.00175 mg/L.
Southworth et al. (2004) published data from sixteen streams and rivers
throughout the Southeast United States (including EFPC) showing that
the percent of methylmercury in water decreases with increases in the
total mercury concentration (unfiltered water samples). The total mercury
concentrations during the 1950s were thousands of times greater in EFPC
water than in the 1990s. However, the portion of dissolved and suspended inorganic mercury that
remained in the water downstream of the Y-12 plant in the 1950s and 1960s remains highly
uncertain.
The level of hazard
depends on the species
and the quantity of
mercury in the water.

The Oral Bioavailability of Mercury in EFPC
Not all of the mercury a person swallows is absorbed into the blood. Some of it passes through
the gastrointestinal tract and is eliminated in the feces. Adverse
The oral bioavailability of a
health effects associated with the ingestion of mercury depend on
substance is the fraction of
how much mercury gets into the blood, not how much mercury is
the total amount of the
swallowed. Mercury can also cause harm to the inside lining of the
substance swallowed that
stomach and intestines, but at levels much higher than those reported
is absorbed.
in EFPC. The fraction of the mercury swallowed that passes through
the lining of the stomach and intestines and enters the bloodstream is referred to as the amount
that is bioavailable. This fraction is biologically available to cause harm to the tissues and organs
inside the body through its transport in the circulatory system.
Different forms of mercury have different bioavailabilities. For organic mercury, studies in
humans regarding the oral ingestion of methylmercury bound to fish muscle protein have shown
that absorption is almost complete (95 percent) (ATSDR 1999). In
Newborn mice exhibited higher
contrast, elemental mercury absorbs poorly into the blood from the
inorganic mercury absorption
gastrointestinal tract, even when it is ingested in large quantities.
than adult mice. Similarly, the
For inorganic mercury, the highest oral bioavailability factor
stomach lining of nursing
reported in the scientific literature is 38 percent for mercuric
human infants is not fully
developed. It allows more
chloride administered in water to week-old suckling laboratory
substances, such as milk
mice (ATSDR 1999).
proteins, from the mother into
the blood. In this way, mothers
transfer nutritional and immune
proteins to their children. Yet
immature stomach linings also
make infants more vulnerable
to heavy metal poisoning than
are older children and adults.

In adult mice, the bioavailability of mercuric chloride has been
reported to be 20–25 percent. In human studies, mercuric nitrate
was reported to be 15 percent bioavailable (ATSDR 1999). In other
studies, the mercury concentration in kidneys of mercuric sulfidedosed mice was approximately 20-fold to 50-fold lower than in
mercuric chloride-dosed mice, even when significantly higher
doses of mercury were administered to the mercuric sulfide-dosed
mice, and at more frequent intervals (Paustenbach et al. 1997; Sin et al. 1983, 1990). After
identical exposures, the kidney deposition of mercury was approximately 30–60 times lower in
mice exposed to mercuric sulfide, as compared with mice exposed to mercuric chloride.
Although these studies do not measure the bioavailability of mercuric sulfide, they do show that
mercuric sulfide is absorbed from the gastrointestinal tract to a measurable extent, though likely
to a lesser extent than mercuric chloride (Schoof and Nielsen 1997). A quantitative determination
of the relative bioavailabilities of mercuric sulfide versus mercuric chloride has not been derived
Page | 87

in the available studies, nor has the relative bioavailability of mercuric sulfide in humans been
examined (ATSDR 1999). Nevertheless, because of mercury’s high water solubility, scientists
generally believe that mercuric chloride is among the most bioavailable of inorganic mercury
species. Thus an upper bound bioavailability factor for the oral ingestion of inorganic mercury in
non-nursing children and adults appears to be approximately 25 percent.
In this evaluation, ATSDR compared exposure doses with the ATSDR oral inorganic mercury
MRLs, which are based on measured exposure doses to mercuric chloride. The inorganic
mercury in EFPC water, however, is expected to be primarily mercuric nitrate. ATSDR therefore
calculated doses using the relative bioavailability of mercuric nitrate to the bioavailability of
mercuric chloride (Paustenbach et al. 1997). The oral bioavailability of mercuric nitrate in
humans has been reported as 15 percent (Rahola et al. 1973). In the dose calculations for
exposures to mercuric nitrate, ATSDR used a bioavailability factor of 0.6. Relative to mercuric
chloride, the bioavailability of mercuric nitrate is 60 percent (i.e., 0.15 - 0.25 = 0.6). See
Appendix G. Past Exposure Pathway Parameters for ATSDR’s assumptions and formulas used to
estimate exposure doses.
Past Surface Water Exposure Pathway Conclusions
ATSDR based the following conclusions on a comparison of the
calculated exposure doses with the ATSDR oral organic and
inorganic mercury MRLs. A person whose dose exceeds an MRL
may not experience adverse health effects. No health data are
available that would allow ATSDR to evaluate the actual occurrence
of adverse health effects during the 1950s and 1960s from exposure to
water in EFPC. With these points in mind, ATSDR concludes

Note that many uncertainties
are associated with the
estimated exposure doses,
and note that people vary
widely in their response to
hazardous substances. The
conclusions refer to the past
potential for mercury in
EFPC to cause harm. The
conclusions are not a
measure of the past
occurrence of adverse
health effects.

• Children who swallowed water from EFPC containing inorganic
mercury for a short period of time (acute exposure, less than 2
weeks) during some weeks in 1956, 1957, and 1958 may have an
increased risk of developing renal (kidney) effects. Adults, who
swallowed water from EFPC for a short time during some weeks
in 1958, may have an increased risk of developing renal (kidney) effects.

• Swallowing water from EFPC containing inorganic mercury for a short time before 1953, or
after the summer of 1958, is not expected to have harmed people’s health.
• Intermittently (intermediate exposure, greater than two weeks and less than a year)
swallowing water from EFPC containing inorganic mercury is not expected to have harmed
people’s health during any year.
• Swallowing water from EFPC containing mercury over a long
period of time (chronic exposure, more than a year) in the past is
not expected to have harmed people’s health.
• Swallowing water from EFPC containing methylmercury is not
expected to have harmed people’s health.
ATSDR cannot conclude whether

ATSDR concludes, from the
Task 2 water model that
long-term exposures to
mercury in EFPC water
were not a public health
hazard. ATSDR’s separate
evaluation agrees with the
Task 2 results.

• Swallowing water from EFPC containing inorganic mercury for a short time during 1953,
1954, and 1955 could have harmed people’s health.
Page | 88

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

ATSDR also examined the average annual and quarterly mercury concentrations (inorganic and
organic) in water at the Y-12 plant. These data may represent the highest mercury concentrations
in EFPC, with the possible exception of areas where mercury deposits in the EFPC floodplain
may have served as secondary sources. None of the data from water samples at Y-12 exceeded
ATSDR’s assessment of intermediate-term exposures (15–364 days) (i.e., calculated doses were
below ATSDR’s intermediate MRL). These data sets indicate that none of the mercury
concentrations in EFPC were an oral hazard to children playing in the creek.

IV.A.4.

Past Soil and Sediment Exposure Pathways

Y-12 Mercury Releases to the EFPC Floodplain
Y-12 mercury releases to water during the 1950s and 1960s resulted in significant mercury
deposits in off-site soils within the EFPC floodplain. Before 1983,
Mercury contamination of
people collected EFPC floodplain soil to supplement private gardens.
soil and sediments along the
EFPC floodplain near the YThe city of Oak Ridge personnel collected EFPC floodplain soil to
12 plant occurred primarily
backfill 10 miles of sewer line installation. These activities resulted in
as a result of mercury
distribution of mercury-contaminated soils from the EFPC floodplain
releases to surface water.
to other areas of Oak Ridge.
Sediment consists of dirt, silt, and sand that accumulate at the bottom and along the banks of
rivers, streams, and other surface water bodies. Sediment accumulates in areas where the stream
depth, breadth, or direction changes. Some reaches of EFPC have very little bottom sediment;
the stream scours the bedrock and moves the lighter weight particulate matter downstream. Thus
collection of sediment samples from all locations along EFPC is difficult. Fewer sediment
samples were collected from EFPC compared with soil samples collected from the EFPC
floodplain. 10 But compared with floodplain soil, people have less opportunity for exposure to
EFPC sediment. Mercury concentrations detected in sediment (as reflected in the sampling data)
are generally comparable to, or less than, those detected in soil. This discussion therefore
primarily focuses on mercury levels detected in soil, with less emphasis on the limited sediment
data.
During the early 1980s, the Oak Ridge Associated Universities (ORAU) and the TVA conducted
the earliest comprehensive surveys of mercury in EFPC floodplain soils and sediment. ORAU
collected more than 3,000 surface soil samples between 1983 and 1985 from the EFPC
floodplain, the Oak Ridge sewer line beltway, and private lawns and gardens in and around Oak
Ridge (Hibbitts 1984, 1986; TDHE 1983). TVA collected approximately 100 core samples in 10­
inch increments from 27 transects across the EFPC floodplain during
Transects are imaginary
1984 (SAIC 1994a). The DOE EFPC Floodplain and Sewer Line
lines that cross the
Beltway Remedial Investigation (RI) is the most recent large-scale
floodplain. They’re a
sampling effort (SAIC 1994a). This investigation is discussed in
method of plotting where
greater detail in the following section.
soil samples are collected.
The EFPC Floodplain and Sewer Line Beltway RI
In October 1990, DOE began soil and sediment sampling of the lower EFPC floodplain. DOE
reviewed earlier ORAU and TVA data. These data indicated where the mercury contamination
was most concentrated along the floodplain. The RI is the most comprehensive soil and sediment
10

There were 50 sediment samples in both the CERCLA RI Phases Ia and Ib combined.
Page | 89

investigation of mercury in the EFFC floodplain and the sewer line beltway area of Oak Ridge
(SAIC 1994a, 1994c). The RI characterizes mercury distribution in the EFPC floodplain and is
the primary source of data used to evaluate potential past mercury exposures for people living
near Lower EFPC.
The two-phase investigation comprised Phase Ia, which included more than 100 soil samples and
was designed to identify contaminants of potential concern; 11 and Phase Ib, which was designed
to establish the nature and extent of contamination. 12 Phase Ib included more than 2,600 soil
samples collected from 159 transects across the EFPC floodplain.
RI Sampling Methodology
Transects were separated at approximately 100-meter (330-foot) intervals beginning from the
confluence of EFPC with Poplar Creek and culminating at the mouth of Lake Reality on the Y­
12 property. 13 Samples were collected at the edge of the water and every 20 meters (65 feet)
along each transect, up to (or beyond) the elevation of the 100-year
Vertical Integration Study
floodplain and on both sides of the creek (see Figure 18) (SAIC
The vertical integration study
1994a). The spacing of the samples (i.e., sampling density)
(VIS) was included in the RI
collected was initially determined from a statistical analysis of the
report and examined the vertical
costs of sampling and remediation and the variation of mercury
stratification of mercury in oneconcentrations in surface soil as measured in the earlier ORAU
inch increments down to 16
inches below ground surface.
study (Hibbitts 1984, 1986; TDHE 1983).
The purpose of the study was to
examine the stratification of
mercury in the soil and the effect
which compositing the cores had
on the analytical results. Five
core samples were collected
from four locations in the
floodplain with one duplicate
sample at the Bruner site.

Most of the RI soil samples were core samples collected in depths
of 1 or 2 feet (for Phase Ia samples) or 16 inches (for Phase Ib
samples). To minimize Phase Ib costs, collection of core samples
below the first 16-inch cores was planned for every other transect.
In some cases, physical obstacles prevented deeper sampling. Each
core sample was turned into a composite (i.e., the soil was blended
into a uniform mixture) for analysis. The average mercury
concentration for that sample interval was reported.

11

In addition to mercury, many other analytes were tested in the samples.
Surface water, groundwater, air, and biota samples were also collected for the RI.
13
The total distance was approximately 23 kilometers or 14.2 miles.
12

Page | 90

Figure 18. EFPC RI Sampling Strategy

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Page | 91

Sampling Results
The data collected for the RI provided a comprehensive view of ORR mercury distribution in off-site
soils. The RI data are consistent with those collected in the earlier ORAU and TVA studies. The RI
sampling data demonstrated that mercury was present in some soils along the entire length of EFPC.
Mercury contamination did not typically extend out very far from the creek banks and rarely to the
elevation of the 100-year floodplain. Figure 19 shows the extent of mercury contamination in the EFPC
floodplain prior to remediation. The greatest deposition of mercury in the EFPC floodplain was found
in two regions: 1) behind the NOAA building at 456 South Illinois Avenue (see Figure 20) and 2) along
a reach (approximately 2,000 feet) of the creek—south of the Oak Ridge Turnpike—from about 750
feet west of Louisiana Avenue to about 1,000 feet west of Jefferson Avenue (see Figure 21). In DOE
reports, the former area is referred to as the NOAA site and the latter area is referred to as the Bruner
site. 14 These two locations contained the highest measured and the most broadly distributed 15 mercury
concentrations in the EFPC floodplain soils (see Table 11). The highest soil mercury concentrations
detected during the RI were 2,110 ppm from a 1-foot core composite sample collected from the Bruner
site and 1,590 ppm from a 16-inch core composite sample from the NOAA site (SAIC 1994a).
Table 11. Maximum Mercury Concentrations Detected in EFPC Floodplain Soil
Location
Bruner site

NOAA site

Sample type

Concentration (ppm)

Data Set

1-foot core

2,110

RI

10-inch core

1,300

TVA

16-inch core1

3,420

VIS

16-inch core

1,590

RI

Surface soil

2,400

ORAU (April 1985)

10-inch core

1,800

TVA

core1

2,870

VIS

16-inch

Sources: ChemRisk 1999a; SAIC 1994a
ppm:
parts per million (this is the same as mg/kg)
RI:
EFPC Floodplain and Sewer Line Beltway Remedial Investigation
TVA:
Tennessee Valley Authority
ORAU: Oak Ridge Associated Universities
VIS:
vertical integration study
1
These peak concentrations were found 10–11 inches and 9–10 inches below ground surface, respectively
(ChemRisk 1999a).

In 1995, during the ROD process, DOE, U.S.EPA, and TDEC established a 400-ppm remediation
(clean-up) goal for mercury in the EFPC floodplain (DOE 1995b). Most of the core mercury samples
(more than 98 percent) collected during the RI were below 400 ppm (DOE 2001; SAIC 2004). In fact,
almost all of the soil and sediment samples collected during the RI were below this concentration.
Exceptions were several samples at the NOAA site and the Bruner site, one sample near the creek in
the Grand Cove area of Oak Ridge, two samples near South Illinois Avenue northwest of Tuskegee
Drive, and three samples on DOE property—one on the Y-12 property and two core samples at the
same location on the K-25 property.
14

The Bruner site is also referred to as the Bruner’s Center site or the Bruner and Sturm properties. At the time of the RI,
the Bruner site included properties in the EFPC floodplain southeast of the Oak Ridge Turnpike. The name Bruner
referred to the owners of a shopping area on the northwest side of the Turnpike. The virtual extension of Louisiana
Avenue across the Turnpike.
15
Detected at the greatest distance from EFPC and greatest vertical depths
Page | 92

Page | 93

Figure 19. Extent of Mercury Contamination in the EFPC Floodplain (prior to completion of remediation in 1997)

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Page | 94

Figure 20. Extent of Mercury Contamination at the NOAA site (prior to completion of remediation in 1997)

Page | 95

Figure 21. Extent of Mercury Contamination at the Bruner site (prior to completion of remediation in 1997)

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Remedial Activities
Remedial activities were first initiated in 1984, when DOE removed mercury-contaminated soils
from private residences (upon request) and from the Oak Ridge sewer line beltway.
The CERCLA Lower East Fork Poplar Creek Remedial Action prompted removal of mercurycontaminated soil at the NOAA and Bruner sites (DOE 2000). The NOAA site was remediated in
1996, and the Bruner site in 1997. Remedial activities consisted of removing about 34,000 cubic
yards of mercury-contaminated soils from the NOAA and Bruner sites, transporting the
contaminated soil to the Y-12 Industrial Landfill V, and subsequently backfilling the excavated
areas with clean fill and topsoil (SAIC 2002a). Soils at the Grand Cove location and soil
northwest of Tuskegee Drive (maximum core mercury concentration = 443 ppm) were not
removed. Nearby sample concentrations were below 400 ppm and contamination in that area was
not expected to pose a public health risk.
Evaluation of Soil Mercury Data
Exposures to contaminants in soil typically occur in the top 3 inches. Still, children sometimes
dig deeper in the soil than 3 inches when playing, and adults may dig deeper when gardening or
during construction work, such as building a foundation for a bridge or some other structure. In
addition, soil below the ground surface was at one time close to or at the surface. Thus the
possibility remains that people were exposed in the past to mercury currently below the EFPC
floodplain surface. People may in the future come in contact with excavated subsurface soils or
sediments, or sediments that rise to the surface through natural processes. ATSDR scientists
assume that beginning in the early 1950s, people generally had access to soils with the highest
mercury concentrations; that is, until soil removal activities occurred in the 1980s and 1990s.
Human exposure pathways to mercury in both soil and sediment include incidental ingestion and
dermal absorption (contaminants passing through skin). Digging in the soil or playing in or near
EFPC connects people with the contamination. Incidental ingestion may occur because people
transfer soil from their hands to their mouths. Note here that dose estimates of mercury exposure
are based on a series of assumptions that account for how much mercury is in the soil, how much
soil or sediment people ingest, how much adheres to the skin, and ultimately, how much mercury
is absorbed into the bloodstream. See Appendix G. Past Exposure Pathway Parameters for
ATSDR’s assumptions and formulas used to estimate exposure doses.
In evaluating the soil and sediment data, ATSDR can eliminate from further consideration those
places along EFPC where mercury concentrations were detected below its comparison values;
these levels have not been shown to cause adverse health effects. Using the exposure dose
assumptions outlined in Appendix G. Past Exposure Pathway Parameters, mercury
concentrations at or below 2,400 ppm will result in doses at or below ATSDR’s oral mercury
MRLs (see Table 7). Using ATSDR’s dose assumptions, this site-specific comparison value
(2,400 ppm) applies to both dermal absorption and oral ingestion pathways, both inorganic and
organic mercury species in the soil and sediment, and to acute, intermediate, and chronic
exposures.
Among the reported soil and sediment data, three vertical integration study (VIS) core samples
collected at the NOAA and Bruner sites contained mercury concentrations above 2,400 ppm.
Among the three core samples, mercury exceeding 2,400 ppm was detected in six 1-inch layers
(layers were analyzed separately within each core sample). The maximum mercury concentration
Page | 96

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

reported was 3,420 ppm. None of the other soil or sediment data in the ORAU, TVA, or RI data
sets contained mercury concentrations above 2,400 ppm (ChemRisk 1999a).
The TVA and RI data sets include soil mercury concentrations in composite core samples, not in
undisturbed soil layers. The VIS data indicate the mercury concentrations varied considerably by
vertical depth, even for core samples collected near each other. The highest mercury
concentrations in each of the five VIS samples (in 1-inch layers) ranged from 1 to 4.3 times
greater than the average concentration in each of the 16-inch core composite samples collected
from the same areas. But this is a small sample set, and it contains highly variable patterns of
mercury distribution in the soil (mercury concentrated in a fairly narrow band in one sample and
mercury highly dispersed throughout the core in another). The VIS data, then, are not especially
useful for predicting when the mercury was deposited in the floodplain or what mercury
concentrations people were actually exposed to in the past.
ATSDR scientists considered that the mixing of soil within each core sample (using composite
samples) likely diluted the mercury that was concentrated in narrow bands within the cores.
During the RI, an average concentration for each core composite sample was produced rather
than a minimum and maximum range across core layers. The range would have more accurately
reflected any large differences in concentration that may have occurred across varying core
depths. ATSDR accounted for this dilution effect of composite samples by applying an adjusted
core sample value that provides an estimate of the maximum mercury concentration possibly
detected within each core sample (see Appendix E. Task 2 Pathway Discussions for more
details).
Among the adjusted RI data, 27 samples (among 2,808 data points 16) exceeded 2,400 ppm. The
range of mercury concentrations among the adjusted RI data that exceeded 2,400 ppm was from
2,491 to 8,440 ppm. Except for one sample, all were collected from the NOAA and Bruner sites.
The exception was one subsurface floodplain core sample (16–32 inches below ground surface)
collected on undeveloped DOE property on the northwest side of the Oak Ridge Turnpike
(Highway 95) east of the Horizon Center on the south side of the EFPC at a sharp bend in the
creek. The adjusted mercury concentration for this sampling location is 3,010 ppm. At the upper
end of the adjusted RI data (8,400 ppm) the estimated child exposure doses exceed ATSDR’s
inorganic mercury oral MRLs (acute = 0.007 mg/kg/day; intermediate = 0.002 mg/kg/day).
Exposure doses did not exceed the mercury MRLs in adults. Nor does the maximum adjusted
concentration (8,400 ppm) result in exposure doses to children or adults exceeding ATSDR’s
methylmercury MRL (0.0003 mg/kg/day). (See Appendix G. Past Exposure Pathway Parameters
for more details on estimated doses.)
Although childhood exposures to inorganic mercury exceed their respective MRLs at the highest
adjusted mercury concentration (8,400 ppm), the estimated dose is approximately 10 times lower
than the NOAEL of 0.23 mg/kg/day used to derive the intermediate oral inorganic mercury MRL
(the smaller of the two inorganic mercury oral MRLs) (ATSDR 1999). Using health-protective
exposure assumptions and the highest adjusted mercury concentration, health effects have not
been observed in human or animal studies at the estimated doses. However, the uncertainties in
the assumed exposure dose parameters and limitations with the studies used to derive the MRLs
16

This adjusted RI data group did not include RI sediment or sewer line beltway data, or data from the TVA or
ORAU data sets. ATSDR examined all of those data and confirmed that none would have exceeded 2,400 ppm if
they were similarly adjusted.
Page | 97

do not assure us that exposures—particularly for very young children—are safe. While the
likelihood of young children playing in the floodplain soils diminishes with decreasing age, the
risk of harm from equivalent exposures increases with decreasing age and body size. In short, the
uncertainties in both the exposure parameters and the comparison values suggest that the
mercury in the floodplain soil could have posed an oral and dermal hazard to young children.
The estimated acute mercury dose for an adult worker exposed to the upper end of the adjusted
RI data for mercury in floodplain soil on undeveloped DOE property (3,010 ppm) is
approximately 8 times lower than the acute MRL and over 250 times lower than the NOAEL.
Therefore, exposure of an adult involved in excavation, digging, and other activities that turn
over the floodplain soil in the undeveloped area of DOE property is not expected to cause
harmful health effects for a worker contacting the floodplain soil.
Past Soil and Sediment Exposure Pathway Conclusions
ATSDR concludes
• Children who played at the NOAA site and Bruner site before the soil removal activities in
1996 and 1997 could have accidentally swallowed inorganic mercury in EFPC floodplain
soils. For children, eating this soil may have an increased their risk of developing harmful
renal (kidney) effects. Adults are not expected to have been harmed.
• Accidental ingestion of methylmercury in EFPC floodplain soils in the past is not expected to
have caused harmful health effects for anyone contacting the floodplain soil.
• Adult workers involved in excavation, digging, and other activities that turn over the EFPC
floodplain soil in the undeveloped area of DOE property are not expected to be harmed from
exposure to mercury in the floodplain soil.
Past Soil and Sediment Exposure Pathway Recommendations
• DOE should maintain long-term oversight of the mercury-contaminated EFPC floodplain soil
in the undeveloped area of DOE property east of the Horizon Center. DOE should also
consider remediation of the spot or deed restrictions if the property is transferred to another
party.

IV.A.5.

Mercury in Fish

Mercury in fish and shellfish is predominantly methylmercury, with small amounts of inorganic
mercury. When elemental or inorganic mercury enters freshwater environments, some of it is
transformed into methylmercury, which accumulates in fish and seafood. It is the methylmercury
form in fish that is harmful to the developing fetus and young children. Tests for mercury in fish,
however, often measure all forms of mercury. We refer to these tests as total mercury
concentration or just mercury concentration. Identification of just the methylmercury or
inorganic mercury concentrations in fish requires specific tests.
Sampling Data
Fish downstream from the Y-12 plant were first collected and analyzed for total mercury 17 in
1970. ATSDR reviewed mercury concentrations in fish samples collected from 1970 through
1990. This data was also used by the Task 2 investigators to develop the fish mercury model.
17

Methylmercury comprises nearly 100% of the mercury in fish tissue (ChemRisk 1999a).
Page | 98

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Table 12 provides a summary of the fish data. Bolded numbers represent the maximum fish
mercury concentrations in each stream sampled: EFPC, Poplar Creek, Clinch River, and Watts
Bar Reservoir. The numbers of fish contributing to each dataset are not available from the Task 2
report; each data set specifies a different location, a different collection period, or a different fish
species (ChemRisk 1999a).
Table 12. Mercury1 Concentrations in Fish Collected Downstream of the Y-12 Plant
Location

Concentration (ppm)
Average2
Maximum

Year

No. of Data Sets

EFPC

1970

3

0.55

1.3

EFPC

1982

4

1.4

3.6

EFPC

1983

6

0.28

0.74

EFPC

1984

18

0.73

1.4

Poplar Creek

1976

6

0.5

1.4

Poplar Creek

1977

36

0.3

2.1

Poplar Creek

1982

24

0.35

1.3

Poplar Creek

1984

3

0.2

0.42

Poplar Creek

1990

3

0.49

0.88

Clinch River

1976

23

0.29

2.1

Clinch River

1977

24

0.23

1.5

Clinch River

1979

7

0.11

1.1

Clinch River

1984

7

0.24

1.2

Clinch River

1990

2

0.27

0.77

Watts Bar Reservoir

1984

6

0.14

0.45

Watts Bar Reservoir

1987

1

< 0.10

< 0.10

Watts Bar Reservoir

1990

2

0.08

0.25

Source:

ChemRisk 1999a (Refer to Appendix J Table J-3 in the Task 2 report for information regarding fish
species sampled and specific sample location.)
EFPC: East Fork Poplar Creek
ppm:
parts per million
All concentrations are reported as fresh (i.e., wet) weight.

Bolded numbers represent the highest average and maximum fish concentrations in each stream sampled.

1
Methylmercury comprises nearly 100% of the mercury in fish tissue (ChemRisk 1999a).

2
The average represents the average of the mean reported for each data set and is not weighted to reflect the

difference in sample size across the different studies.

ATSDR used the fish data from Table 12 to evaluate past exposures to methylmercury18 in fish.
ATSDR scientists considered both acute and chronic exposures to mercury in fish. For acute
exposures (eating fish for short periods of time with high mercury concentrations, fewer than 2
weeks), we used the maximum fish concentrations reported. For chronic exposures (eating fish
from the local streams over an extended period of time, more than a year), we used the highest
yearly average mercury concentrations reported in fish tissue samples collected from each of the
sampling location.
18

ATSDR assumed that the mercury measured in fish is 100% methylmercury.
Page | 99

Although the datasets are limited, the mercury concentrations detected in fish samples fall within
a relatively narrow range (range of mean values: <0.10–1.4 ppm). This suggests mercury levels
do not vary widely across the different sampling locations. But we have no way of knowing how
mercury concentrations in fish caught prior to 1970 compare with these data.
Results and Discussion: Chronic Exposures from Eating Fish
Estimating mercury intake from eating fish is uncertain. The intake varies depending on the type,
frequency, and quantity of fish eaten. Fish mercury concentrations generally decrease with
distance downstream from the Y-12 plant, while the fish consumption rates increase with
distance from the Y-12 plant. The highest mercury concentrations were in EFPC. However, the
anglers who ate fish from Poplar Creek, Clinch River, and Watts Bar Reservoir have the higher
estimated mercury doses than anglers who ate fish from EFPC; they eat more fish than anglers in
EFPC because EFPC is not a productive fishing location. See Appendix G. Past Exposure
Pathway Parameters for ATSDR’s assumptions and formulas used to estimate exposure doses.
To evaluate the long-term (chronic exposure, more than a year) methylmercury exposure to the
average individual eating fish caught downstream from the Y-12 plant, ATSDR used the average
mercury concentrations from EFPC, Poplar Creek, Clinch River, or Watts Bar Reservoir (see
bold concentrations in Table 12) and the average fish consumption rates reported in the Task 2
report (see Table G-2 and Table G-3). For EFPC, Clinch River, and Watts Bar Reservoir, the
estimated doses of the fish-eating populations are about an order of magnitude lower than both
the ATSDR chronic organic mercury MRL of 3.0 × 10-4 mg/kg/day and the U.S.EPA RfD of 1.0
× 10-4 mg/kg/day (see Table 13, Table 7, and Figure 14). The estimated doses for Poplar Creek
were above the U.S.EPA RfD, but below the ATSDR MRL (see Table 13, Table 7, and Figure
14).
To evaluate people eating the estimated maximum amount of fish from EFPC, we used the
average yearly mercury concentrations and the maximum fish consumption rates reported in the
Task 2 report to estimate methylmercury doses. The estimated exposure doses were below both
the U.S.EPA RfD and the ATSDR MRL (see Table 13, Table 7, and Figure 14).
For recreational anglers (adults and child) eating Poplar Creek, Clinch River, or Watts Bar
Reservoir fish, we also used the average yearly mercury concentrations and the maximum fish
consumption rates reported in the Task 2 report to estimate methylmercury doses. All of the
estimated doses were above the U.S.EPA RfD (see Table 13, Table 7, and Figure 14). Some
were also above the ATSDR MRL.

Page | 100

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Table 13. Methylmercury Exposure Doses from Fish Collected Downstream

of the Y-12 Plant

Exposure Doses Using
Average Concentrations
and Average Consumption
Rates1
(mg/kg/day)
Adults
Children

Exposure Doses Using
Average Concentrations
and Maximum
Consumption Rates1
(mg/kg/day)
Adults
Children

Year

Average
Concentration
(ppm)

EFPC

1970

0.55

9.4 × 10-6

1.2 × 10-5

3.1 × 10-5

3.9 × 10-5

EFPC

1982

1.4

2.4 × 10-5

3.0 × 10-5

8.0 × 10-5

1.0 × 10-4

EFPC

1983

0.28

4.8 × 10-6

6.0 × 10-6

1.6 × 10-5

2.0 × 10-5

EFPC

1984

0.73

1.3 × 10-5

1.6 × 10-5

4.2 × 10-5

5.2 × 10-5

Poplar Creek

1976

0.5

1.3 × 10-4

1.6 × 10-4

4.6 × 10-4

5.9 × 10-4

Poplar Creek

1977

0.3

7.7 × 10-5

9.6 × 10-5

2.8 × 10-4

3.5 × 10-4

Poplar Creek

1982

0.35

9.0 × 10-5

1.1 × 10-4

3.3 × 10-4

4.1 × 10-4

Poplar Creek

1984

0.2

5.1 × 10-5

6.4 × 10-5

1.9 × 10-4

2.3 × 10-4

Poplar Creek

1990

0.49

1.3 × 10-4

1.6 × 10-4

4.6 × 10-4

5.8 × 10-4

Clinch River

1976

0.29

7.5 × 10-5

9.3 × 10-5

2.7 × 10-4

3.4 × 10-4

Clinch River

1977

0.23

5.9 × 10-5

7.4 × 10-5

2.1 × 10-4

2.7 × 10-4

Clinch River

1979

0.11

2.8 × 10-5

3.5 × 10-5

1.0 × 10-4

1.3 × 10-4

Clinch River

1984

0.24

6.2 × 10-5

7.7 × 10-5

2.2 × 10-4

2.8 × 10-4

Clinch River

1990

0.27

6.9 × 10-5

8.6 × 10-5

2.5 × 10-4

3.2 × 10-4

Watts Bar Reservoir

1984

0.14

6.0 × 10-5

7.5 × 10-5

2.2 × 10-4

2.7 × 10-4

Watts Bar Reservoir

1987

< 0.10

4.3 × 10-5

5.3 × 10-5

1.6 × 10-4

2.0 × 10-4

Watts Bar Reservoir

1990

0.08

3.4 × 10-5

4.3 × 10-5

1.3 × 10-4

1.6 × 10-4

Location

1

See Table G-2 in Appendix G for average and maximum consumption rates.

Bold text indicates that the exposure dose is higher than the U.S.EPA RfD of 1.0 × 10-4 mg/kg/day.

The ATSDR chronic MRL of 3 × 10-4 mg/kg/day for ingestion of organic mercury is based on
the Seychelles Child Development Study, in which people who were exposed to 1.3 × 10-3
mg/kg/day of methylmercury from eating fish did not experience any adverse health effects
(Davidson et al. 1998) (See Table 7 and Figure 14.) Over 700 mother-infant pairs have been
followed and tested from birth through 107 months of age (Myers et al. 2009). The Seychellois
regularly consume a large quantity and variety of ocean fish, with 12 fish meals per week
representing a typical methylmercury exposure. Developing fetuses were exposed to
methylmercury in utero through maternal fish ingestion before and during pregnancy. Neonates
continued to be exposed to maternal mercury during breastfeeding (some mercury is secreted in
breast milk), and methylmercury exposure from the regular diet continued after the gradual postweaning shift to a fish diet (Davidson et al. 1998). After 66-months test results revealed no
evidence of adverse effects in offspring attributable to a mother’s chronic ingestion of low levels
of mercury (median total mercury concentration in 350 fish sampled from 25 species consumed
by the Seychellois was <1 ppm [range, 0.004–0.75 ppm]) of methylmercury in fish (Davidson et
al. 1998). After 107 months test results revealed a number of associations between postnatal
exposure and test outcomes, but the results varied. Although the authors concluded that the
Page | 101

findings were consistent with the earlier 66-month testing, they suggested that postnatal exposure
should be further studied (Myers et al. 2009). More information about the harmful effects of
methylmercury is available in ATSDR’s Toxicological Profile for Mercury (ATSDR 1999).
The U.S.EPA RfD of 1.0 × 10-4 mg/kg/day for methylmercury is based on a long-term study of
children born to women who lived on the Faroe Islands (See Table 7 and Figure 14). 19 This
population relies heavily on seafood and whales as a protein source. The investigators used
various neurological tests that monitor child development. They concluded that at birth, cord
blood mercury levels in the mother were associated with lower performance on standardized
neurobehavioral tests at age 7 years involving attention, verbal memory, confrontational naming,
and to a lesser extent visual/spatial abilities and fine-motor functions (Grandjean et al. 1997).
Follow-up studies at age 14 years showed similar findings (Debes et al. 2006). Using a
mathematical model, U.S.EPA concluded that the benchmark dose lower limit (BMDL05) range
from 46 to 79 ppb methylmercury concentration in maternal cord blood. This range of
methylmercury concentration in maternal cord blood is associated with a 5 percent increase in
the incidence of neurodevelopmental effects. This methylmercury concentration in maternal cord
blood equated to a range of 8 × 10-4 mg/kg/day to 1.5 × 10-3 mg/kg/day as a dietary intake. The
doses were divided by an uncertainty factor of 10 to arrive at the RfD of 1.0 × 10-4 mg/kg/day.
The U.S.EPA’s approach is consistent with the National Academy of Sciences (NAS)
recommendation of using the BMDL of 58 ppb methylmercury in maternal cord blood from the
Faroe Islands Study to develop the methylmercury RfD (NRC 2000) (See Table 7 and Figure
14.) The NAS concluded that the Boston Naming Test was the most sensitive and reliable at
detecting neurodevelopmental effects in the Faroe Island children (NRC 2000). The NAS
concluded that the estimated BMDL of 58 ppb of methylmercury in maternal cord blood is the
dose that resulted in a 5 percent increase in the incidence of abnormal scores on the Boston
Naming Test (a picture-naming, vocabulary test) (NRC 2000). 20 The cord blood concentration of
58 ppb methylmercury corresponds to 12 ppm methylmercury concentration in maternal hair
(NRC 2000). The associated dietary intake was calculated to be 1.1 × 10-3 mg/kg/day (NRC
2000).
None of the estimated exposure doses from fish collected downstream of the Y-12 plant were
higher than the NOAEL (1.3 × 10-3 mg/kg/day) from the Seychelles study (Davidson et al. 1998)
(Table 13, Table 7, and Figure 14). Nor were they higher than the LOAELs (8 × 10-4 mg/kg/day
to 1.5 × 10-3 mg/kg/day) from the Faroe Island study (Grandjean et al. 1997). However, some of
the doses were in the same order of magnitude as the LOAELs from the Faroe Island study.

19
20

Weaknesses in the RfD derivation process are provided in Dourson et al. (2001).

These neurodevelopmental effects were observed at a population level; not on an individual basis.

Page | 102

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment
Conclusions for Eating Fish Containing Methylmercury
Public health hazard The estimated exposure doses are above the NAS health effect level.
Increased risk

The estimated exposure doses are below the NAS health effect level. However, they are above
ATSDR’s and U.S.EPA’s health guidelines for methylmercury and come close to the NAS health effect
level.

Small increased risk The estimated exposure doses are above ATSDR’s and U.S.EPA’s health guidelines for
methylmercury. However, they are not close to the NAS health effect level.
No health hazard

The estimated exposure doses are below ATSDR’s and U.S.EPA’s health guidelines for
methylmercury.

East Fork Poplar Creek
The estimated methylmercury doses are below the U.S.EPA RfD and ATSDR MRL and are not
at levels associated with harmful effects in children or fetuses of women who consumed an
average or maximum rate of EFPC fish in 1970 and the 1980s. Figure 22 compares the estimated
exposure doses in Table 13 to the health guidelines. These estimated doses for EFPC are based
on an occasional meal of EFPC fish (approximately four meals a year for a child and nine meals
a year for an adult). Low consumption rates are used because EFPC is not a productive fishing
area.

Page | 103

Figure 22. Past Estimated Methylmercury Exposure Doses from Eating EFPC Fish

Compared to Health Effect Levels and Health Guidelines

Page | 104

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Poplar Creek
Developing fetuses were at an increased risk of subtle neurodevelopmental effects if, before and
during pregnancy, women ate approximately 12 meals per month of Poplar Creek fish caught in
the 1970s, 1980s, and 1990. In Table 13, a woman’s estimated methylmercury dose from eating
Poplar Creek fish at the maximum consumption rate approached 1.1 × 10-3 mg/kg/day (see
Figure 23). This was identified by the NAS in the Faroe Islands study as a dose that results in a 5
percent increase in the incidence of abnormal scores on the Boston Naming Test (a picturenaming, vocabulary test) (NRC 2000). The NAS effect level is consistent with the range of 8.5 ×
10-4 mg/kg/day to 1.5 × 10-3 mg/kg/day identified as the benchmark dose lower limit (BMDL05)
by the U.S.EPA. Based on the Faroe Islands study, this BMDL05 is the lowest dose that is
expected to be associated with a 5 percent increase in the incidence of neurodevelopmental
effects (NRC 2000). Possible harmful effects identified from studies of children exposed in utero
involve attention, verbal memory, confrontational naming, and to a lesser extent visual/spatial
abilities and fine-motor functions (Debes et al. 2006; Grandjean et al. 1997; NAS 2000). In
addition, even if children were not exposed in utero, some young children who frequently eat the
same fish as their mother ate are also at an increased level of risk for harmful effects. This
conclusion is somewhat uncertain because studies were not done on children not exposed in
utero; therefore, it is not known whether children are as sensitive to neurotoxic effects as fetuses.
Further, a person’s mercury response is itself somewhat uncertain. Contributing to that
uncertainty is how the body handles mercury, and the sex, genetics, health, and nutritional status
of the person who eats the fish, or how mercury is handled in the body.
Similarly, children who ate 6 meals a month (the maximum consumption rate) of Poplar Creek
fish also have estimated doses that come close to the NAS dose effect level and the EPA
BMDL05 (see Figure 23). Whether children are as sensitive to the neurotoxic effects of mercury
as the fetus is uncertain. To be protective, U.S.EPA’s and FDA’s national fish advisory includes
a warning for children as well as women who are pregnant, who plan to become pregnant, and
nursing mothers (see Appendix H).
National Fish Advisory
Women who consumed an average
In March 2004, the U.S.EPA and the FDA released a joint national fish
rate of approximately 3 meals a
advisory. It emphasized that fish and shellfish were an important part of a
healthy diet. The advisory pointed out that fish and shellfish contained
month of Poplar Creek fish in the
high-quality protein and other essential nutrients, were low in saturated
1970s, 1980s, and 1990 are at a
fat, and provided omega-3 fatty acids (a heart healthy chemical). A wellsmall increased risk of harming a
balanced diet that included a variety of fish and shellfish could contribute
developing fetus if they are
to heart health and to children's proper growth and development. The
pregnant or a baby if the mother is
advisory concluded that people, including women and young children,
should include fish or shellfish in their diets (EPA 2004; FDA 2004).
nursing. Also, children who ate
about 1.5 meals a month (average
The joint advisory acknowledged that nearly all fish and shellfish contain
traces of mercury. For most people, the risk of mercury-related health
consumption rate) of Poplar Creek
effects from eating fish and shellfish was not a concern. Yet some fish
fish have a small increased risk of
and shellfish may contain levels of mercury considered unhealthy. The
neurodevelopmental effects. Most
risks from mercury in fish and shellfish depend on the mercury levels in
of the estimated doses in Table 13
the fish and shellfish and the amount eaten. The FDA and the U.S.EPA
for these women and children are
advised women who might become pregnant, women already pregnant,
nursing mothers, and young children to avoid some types of fish and to
below the U.S.EPA RfD and
eat fish and shellfish known to have lower mercury levels (EPA 2004;
ATSDR MRL and the few doses
FDA 2004). The National Fish Advisory is included in Appendix H.
that are slightly above the RfD are
not close to the NAS dose effect level or the EPA BMDL05 (see Figure 23).
Page | 105

Figure 23. Past Estimated Methylmercury Exposure Doses from Eating Poplar Creek Fish

Compared to Health Effect Levels and Health Guidelines

Page | 106

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Clinch River
Women who consumed a maximum rate of approximately 12 meals a month of Clinch River fish
in the 1970s, 1980s, and 1990 have a small increased risk of harming a developing fetus if they
were pregnant or a baby if the mother was nursing the baby. Children who consumed an average
rate of approximately 6 meals a month of Clinch River fish also have a small increased risk of
neurodevelopmental effects. The estimated doses in Table 13 for these women and children are
only slightly above the RfD and MRL; however, these estimated doses are not close to the NAS
dose effect level or the EPA BMDL05 (see Figure 24).
The estimated doses in Table 13 for women and children who consumed 2-3 meals of Clinch
River fish a month are not at risk of harmful effects from mercury in fish. The estimated doses in
Table 13 for women and children are below the U.S.EPA RfD and ATSDR MRL (see Figure
24).
Watts Bar Reservoir
Women who consumed a maximum rate of approximately 20 meals a month of Watts Bar
Reservoir fish in the 1980s and 1990 have a small increased risk of harming a developing fetus if
they were pregnant or their baby if the mother was nursing the baby. Children who consumed an
average rate of approximately 10 meals a month of Watts Bar Reservoir also have a small
increased risk of neurodevelopmental effects. The estimated doses in Table 13 for these women
and children are only slightly above the RfD; however, these estimated doses are not close to the
NAS dose effect level or the EPA BMDL05 (see Figure 25).
The estimated doses in Table 13 for women and children who consumed 3-5 meals of Watts Bar
Reservoir fish a month are not at risk of harmful effects from mercury in fish. The estimated
doses in Table 13 for these women and children were below the U.S.EPA RfD and the ATSDR
MRL (see Figure 25).

Page | 107

Figure 24. Past Estimated Methylmercury Exposure Doses from Eating Clinch River Fish

Compared to Health Effect Levels and Health Guidelines

Page | 108

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Figure 25. Past Estimated Methylmercury Exposure Doses from Eating Watts Bar

Reservoir Fish Compared to Health Effect Levels and Health Guidelines

Page | 109

Results and Discussion: Acute Exposures from Eating Fish
To evaluate acute exposure, the maximum mercury concentration reported from the Task 2 fish
data set was used (see Table 12). It was assumed that a person would eat one fish meal consisting
of 170 grams (about 6 ounces) of fish.
The scientific literature includes one study in which the LOAEL for acute methylmercury
exposure was estimated to be 0.001 mg/kg/day. This was a study of Iraqi children born to
mothers who had consumed grain tainted with methylmercury used as a fungicide (Cox et al.
1989). The adverse affect was delayed onset of walking in young children. However, a closer
examination of the study revealed numerous shortcomings and confounding factors (Crump et al.
1995). Further, the same results were not observed in the Seychelles study used to derive the
ATSDR chronic methylmercury MRL (Davidson et al. 1998) nor in the Faroes study (Grandjean
et al. 1997) used to derive the U.S.EPA RfD for methylmercury. Neither the Seychelles study
nor other human studies examined acute methylmercury exposures.
In animal studies, neurotoxic signs, including muscle spasms, gait disturbances, flailing, and
hindlimb crossing were observed in rats after acute-duration gavage dosing with methylmercury
concentrations at doses as low as 4 mg/kg/day for 8 days (Inouye and Murakami 1975). The
authors stated the effects may not be observed until several days after dosing has stopped. It is
not clear whether 4 mg/kg/day represents an acute toxicological threshold for humans. Evidence
from the scientific literature, however, suggests that no adverse effects in rats occur at dose
levels of 2 mg/kg/day (Hughes and Annau 1976; Inouye and Murakami 1975). At the highest
mercury concentration reported in the Task 2 datasets (fish from EFPC, mercury concentration =
3.6 ppm), a child eating 2 six-ounce meals of fish per day would have a dose of 0.044
mg/kg/day, which is two orders of magnitude below these acute doses. Except for
neurodevelopmental effects observed following methylmercury exposures in utero and to
nursing babies via breast milk, the animal studies suggest exposures to older children and adults
from consuming fish from EFPC or farther downstream will not result in acute adverse health
effects.
The scientific evidence is clear that fetuses and breast feeding babies are much more sensitive to
mercury than are older children and adults. Four-month old rats were reported to exhibit
significant reduction in behavior performance tests after exposure in utero to methylmercury at
doses as low as 0.008 mg/kg/day during gestational days 6–9. Doses of 0.004 mg/kg/day did not
result in performance reduction (Bornhausen et al. 1980). A pregnant woman would not exceed
the LOAEL dose of 0.008 mg/kg/day by eating only one 6-ounce fish meal (170 grams) with a
mercury concentration of 2.8 ppm (6.9 × 10-3 mg/kg/day). And would not exceed the NOAEL
dose (0.004 mg/kg/day) by eating one meal with a mercury concentration of 1.4 ppm (3.4 × 10-3
mg/kg/day). Only eating fish from EFPC in 1982 would result in an acute exposure dose higher
than the LOAEL.

Page | 110

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment
Benefits from Fish Consumption
It is important to note that, even though there are federal and state fish advisories in place across the country, there are many
fish species in U.S. water bodies that are safe to eat. And having a healthy diet that includes lean sources of protein (such as
grilled, broiled, and baked fish) can provide health benefits. Much of the research regarding beneficial effects of consuming
fish surrounds species with higher levels of omega-3 fatty acids (e.g., sardines, mackerel, tuna, herring, trout, and salmon).
The scientific literature regarding the health benefits from eating freshwater species is not as robust as with saltwater species.
The following text provides suggestive evidence that fish consumption provides 1) beneficial developmental effects, 2)
decreased incidence of and mortality from cancer, and 3) improvements in heart health.
• Developmental Effects. Higher developmental scores were reported in children at 15 months of age from women eating
fish (omega-3 rich) one to four times per week compared to those of women who seldom ate fish. The children were
tested for social activity, vocabulary, and language; all improved with increased maternal fish consumption (Daniels et al.
2004).
• Cancer. Observations of protection against breast cancer among fisherman’s wives in Norway date back at least a
decade (Lund and Bonaa 1993). Larsson et al. (2004) reviewed studies showing that omega-3 fatty acid (fish)
consumption protects against breast cancer by several mechanisms. The incidence of both breast and colorectal cancer
is decreased proportionally to the amounts of omega-3 rich fish consumed (Caygill et al. 1996; de Deckere 1999).
• Heart Disease. One of the most serious complications of diabetes is increased risk of mortality from coronary artery
disease. But fish (omega-3 rich) intake shows significant protection, at least in women, against atherosclerosis (Connor
2004; Erkkila et al. 2004), as well as against coronary heart disease and total mortality (Hu et al. 2003). Fish intake (tuna
and other broiled or baked fish, but not fried fish) also lowers the incident risk of atrial fibrillation (Mozaffarian et al. 2004).

Conclusions for Fish
ATSDR’s conclusions refer to the potential to cause harm for methylmercury exposures (in the
past) from eating fish downstream from the Y-12 plant. Given the available information, an
evaluation of reported adverse health effects that could be attributed to methylmercury exposure
from consuming fish during the 1950s and 1960s is not possible. It is also important to
emphasize that ATSDR’s conclusions should only be interpreted as a potential for health effects
to have occurred due to methylmercury exposures in the past.
• ATSDR concludes that periodically eating fish from EFPC (up to nine meals per year) in the
1980s is not expected to have harmed people’s health, including children who ate fish,
nursing infants whose mothers ate fish, and children born to women who ate fish during
pregnancy. Intake rates of fish from EFPC are low because it is not a productive fishing area,
and the estimated methylmercury exposure doses are below both the U.S.EPA RfD and the
ATSDR MRL for methylmercury (see Figure 22).
• ATSDR concludes that eating approximately 12 fish meals per month from Poplar Creek in
the 1970s, 1980s, and 1990 may have increased the risk of subtle neurodevelopmental effects
in children who ate fish and children born to women who ate fish during pregnancy. The
estimated methylmercury exposure doses approach the dose of 1.1 × 10-3 mg/kg/day
identified by the National Academy of Sciences in the Faroe Islands study as a dose that
results in a 5 percent increase in the incidence of abnormal scores on the Boston Naming Test
(a picture-naming, vocabulary test) (NRC 2000). The NAS effect level is consistent with the
range of 8.5 × 10-4 mg/kg/day to 1.5 × 10-3 mg/kg/day identified as the BMDL05 by the
U.S.EPA in the Faroe Islands study. Similarly, children who ate up to 6 meals a month of
Poplar Creek fish also have estimated methylmercury doses that come close to the NAS dose
effect level and the EPA BMDL05 (see Figure 23).

Page | 111

• Women who consumed an average rate of approximately three meals a month of Poplar
Creek fish in the 1970s, 1980s, and 1990 are at a small increased risk of harming a
developing fetus or their nursing child. Also, children who consumed about 1.5 meals a
month (average consumption rate) of Poplar Creek fish were at a small increased risk of
neurodevelopmental effects. Most of the estimated methylmercury doses for these women
and children are below the EPA RfD and the few doses that are slightly above the RfD are
not close to the NAS dose effect level or the EPA BMDL05 (see Figure 23).
• ATSDR concludes that women eating 12 fish meals per month (3 fish meals a week) from
the Clinch River in the 1970s, 1980s, and 1990 had a small increased risk of subtle
neurodevelopmental effects in children born to women who ate fish while pregnant. Children
who ate approximately six fish meals a month from the Clinch River also had a small
increased risk of subtle neurodevelopmental effects. The estimated methylmercury exposure
doses are only slightly above the U.S.EPA RfD and ATSDR MRL and are not close to the
NAS dose effect level or the U.S.EPA BMDL05 identified in the Faroe Islands study.
Pregnant women who ate up to three Clinch River fish meals per month would not have
resulted in increased risk of harmful health effects to developing fetuses (see Figure 24) .
• ATSDR concludes that pregnant or nursing women who ate 20 fish meals per month (five
fish meals a week) from the Watts Bar Reservoir in the 1980s and in 1990 have a small
increased risk of subtle neurodevelopmental effects in the fetus or nursing child. Children
who ate approximately 10 fish meals a month from the Watts Bar Reservoir also had a small
increased risk of subtle neurodevelopmental effects. The estimated exposure methylmercury
doses are only slightly above the U.S.EPA RfD and are not close to the NAS dose effect
level or the U.S.EPA BMDL05 identified in the Faroe Islands study. Eating fewer than six
meals per month is not expected to have caused harmful health effects to a developing
fetus(see Figure 25) .
• ATSDR cannot conclude whether eating fish from EFPC, Poplar Creek, Clinch River, or
Watts Bar Reservoir during the 1950s and 1960s could have harmed people’s health (from
both acute and chronic exposures). Although mercury concentrations in water, surface
sediments, and surface soils were higher during the 1950s and 1960s than they were in later
decades, we do not have adequate data characterizing the methylmercury concentrations in
fish in those waters during the 1950s and 1960s. Earlier attempts to model the average annual
mercury concentrations in fish or exposure doses from eating fish (beginning in 1950)
included assumptions not easily verifiable and may not be appropriate for making public
health decisions.
• ATSDR cannot conclude whether eating fish from EFPC and Watts Bar Reservoir during the
1970s could have harmed people’s health (from both acute and chronic exposures). A small
number of fish samples were collected from EFPC in 1970 (after 1970, samples were not
collected again until 1982). But they are not representative of the entire decade of the 1970s.
No fish samples were collected from Watts Bar Reservoir in the 1970s. Therefore, the hazard
posed by fish consumed from either EFPC or Watts Bar Reservoir during the 1970s cannot
be evaluated.

Page | 112

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

IV.A.6.

Mercury in Local Produce

Sampling Data
ORAU evaluated mercury accumulation in vegetation between 1983 and 1987; Science
Applications International Corporation (SAIC) evaluated mercury accumulation in vegetation as
part of the EFPC RI in 1992 (ChemRisk 1999a). ORAU collected approximately 150 vegetation
samples and analyzed them for mercury. The samples were collected from a variety of locations
throughout the city of Oak Ridge and EFPC floodplain with a wide range of reported soil
mercury concentrations. SAIC collected 55 vegetation samples from the EFPC floodplain.
ORAU also collected 32 samples from plants grown in a laboratory greenhouse. Table 14 lists
the specific types of edible samples collected and analyzed for mercury.
Data from higher plants indicate that virtually no mercury is taken up from the soil into the
shoots of plants such as peas, although mercury concentrations in the roots may be significantly
elevated and reflect the mercury concentrations of the surrounding soil (Lindqvist 1991).
ATSDR assumed that the total mercury measured in fruits and vegetables is inorganic mercury.
Mercury speciation studies of plants grown in soil with inorganic mercury contamination
indicate that the mercury taken into plants is taken up as inorganic mercury (i.e., mercuric ions)
(ChemRisk 1999a).
Table 14. Types of Local Produce Tested for Mercury
Fruits and Other Vegetables

Leafy Vegetables

Root Crops

Banana Pepper

Broccoli

Beets

Bell Pepper

Cabbage

Carrots

Blackberry

Chard

Onions

Corn

Collard greens

Potatoes

Cucumber

Green beans-Pod

Radishes

Eggplant

Kale

Turnips

Grapes

Lettuce

Green Beans

Radish leaves

Okra

Spinach leaves

Pea Pods

Turnip leaf

Squash

Watercress

Strawberry
Tomato
Watermelon
Zucchini

A flowering meadow perennial called sneezeweed had the highest total mercury concentration in
vegetation across both studies (maximum = 239.4 ppm). 21 Mercury concentrations in most of the
edible produce sampled from Oak Ridge-area gardens were below 1 ppm. None of the ORAU
vegetable samples collected in the city of Oak Ridge and EFPC floodplain exceeded 1 ppm, and
21

Mercury concentrations in vegetation are reported in ppm on a dry weight basis. The sneezeweed (genus,
Helenium) samples were greenhouse samples grown in soil with soil mercury concentrations of 1,140 ppm.
Page | 113

only four of SAIC edible produce samples collected from the EFPC floodplain Bruner site
exceeded 1 ppm (ChemRisk 1999a). The highest mercury concentration in edible produce
samples from the Bruner site was 3.2 ppm in a kale leaf sample. On average, leafy vegetables
and root vegetables had similar mercury concentrations, and both had higher mercury
concentrations than fruits. The average mercury concentration was 1.6 ppm in leafy vegetables,
1.4 ppm in root vegetables, and 0.025 ppm in fruits (see Table 15).
Table 15. Mercury Concentrations in Locally Grown Produce
Edible Produce

No. of Samples

Average Hg Concentration (ppm)

Leafy vegetables

32

1.6

Fruits

72

0.025

Root vegetables

16

1.4

120

0.64

Total
Source: ChemRisk 1999a
ppm:
parts per million
Hg:
mercury

Results and Discussion for Local Produce
The data show that vegetables or fruits grown in private gardens with mercury-contaminated
floodplain soils may contain inorganic mercury. That said, whether edible vegetation is
consumed in large enough quantities or at a sufficient frequency to pose harm to people’s health
is unlikely. Based on an EPA estimated intake rate for people living in the south, adults and
children were assumed to eat 2.27 grams of homegrown vegetables per kilogram of body weight
per day (EPA 1997) (See Appendix G. Past Exposure Pathway Parameters for additional
exposure assumptions.). The estimated mercury exposure doses for children and adults are well
below the acute oral MRL (0.007 mg/kg/day) and the intermediate oral MRL (0.002 mg/kg/day).
Using the average mercury concentration of 1.6 ppm in leafy vegetables, the estimated
intermediate oral doses for children and adults are 0.0001 mg/kg/day and 0.00009 mg/kg/day,
respectively. For acute exposure, the highest concentration of 3.2 ppm mercury in edible produce
was used to estimate the acute oral doses of 0.001 mg/kg/day for children and 0.0007 mg/kg/day
for adults. This analysis suggests that the mercury in the fruits and vegetables grown in the city
of Oak Ridge and the EFPC floodplain are not expected to have harmed people’s health, even
when consumed regularly in moderate to high quantities.
Conclusions for Local Produce
ATSDR concludes
• Consuming local produce grown in mercury-contaminated gardens in the city of Oak Ridge
and the EFPC floodplain is not expected to have harmed people’s health.

Page | 114

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

IV.B. Current Exposure (1990–2009)
Because the Task 2 dose reconstruction evaluated past exposures through 1990, exposures since
1990 are evaluated as “current exposures” in this public health assessment.

IV.B.1.

Current Exposure Pathways

To evaluate current exposures, ATSDR gathered and
Note that current conditions are not likely to be
assessed available data from four main areas of interest:
different than those in the late 1990s, because
East Fork Poplar Creek, the city of Oak Ridge, the
there have been no significant mercury
Scarboro neighborhood within the city of Oak Ridge,
releases and remediation activities involving
mercury at Y-12 are being monitored.
and the Lower Watts Bar Reservoir (including the
Clinch River and Watts Bar Reservoir). The media
evaluated include air, surface water, soil, sediment, and biota (including fish and vegetables) (see
Table 16).
Table 16. Current Exposure Pathways Evaluated
East Fork
Poplar Creek

Oak Ridge

Scarboro

Lower Watts
Bar Reservoir

Elemental

X

E

E

X

Surface water pathway

Inorganic

X

X

X

X

Soil pathway

Inorganic

X

X

X

X

Sediment pathway

Inorganic

X

X

X

X

Organic

X

E

E

X

Exposure Pathway

Mercury Species

Air pathway

Biota pathways
Fish consumption

Vegetable consumption
Inorganic
X
X
E
E
Xs indicate that the exposure pathways were evaluated.

Es indicate that the exposure pathways were eliminated. Exposure pathways were eliminated if site characteristics

make past, current, and future human exposures extremely unlikely.

IV.B.2.

Current Air Exposure Pathway (elemental mercury)

Current EFPC Air
In 1993 and 1996, ATSDR evaluated ambient elemental air data from the EFPC RI (ATSDR
1993, 1996a). These data were collected before the floodplain soil was remediated. Specifically,
short-term (minutes to hours) and long-term (days to weeks) ambient air samples were collected
from three floodplain locations (NOAA, Lysimeter, and Minit Chek) with known mercury soil
contamination up to 3,000 mg/kg. Ambient mercury concentrations ranged from 0.0000059 to
0.0000109 mg/m3 using short-term monitoring and from 0.0000031 to 0.0000124 mg/m3 using
long-term monitoring (DOE 1992b; SAIC 1994c). All of the concentrations are one to two orders
of magnitude below the chronic EMEG of 0.0002 mg/m3 for mercury concentrations in air.
Before, during, and after Phase I remediation of the Lower EFPC floodplain soil, continuous
mercury air monitoring was conducted at the NOAA site, located approximately 200 meters
northeast of the excavation area (Barnett et al. 1997). Monitoring was conducted from March 10
to October 14, 1996 (Phase I excavation occurred from July 8 to September 14, 1996; SAIC
2002a). All of the concentrations were below the comparison value of 0.0002 mg/m3 for mercury
concentrations in air (the maximum concentration detected was 0.000061 mg/m3; Barnett et al.
Page | 115

1997). As expected airborne mercury after the excavation was at least three times lower than the
concentrations before and during remediation (Barnett et al. 1997).
During Phase II remediation of the Lower EFPC floodplain soil, over 10,000 ambient air samples
were collected near the Bruner site (OREIS 2009; SAIC 2002a). Monitoring was conducted from
March 12 to October 21, 1997. All of the mercury ambient air concentrations were at least 2.5
times lower than the comparison value of 0.0002 mg/m3 for mercury concentrations in air (the
maximum concentration detected was 0.00008 mg/m3; OREIS 2009).
Ambient air sampling was conducted near the areas with the highest levels of mercury
contamination. Sampling was also conducted during the summer months when increased sunlight
and temperature cause more mercury vapor to release from the soil (Barnett 1997). All of the air
samples were less than the comparison value for mercury in air. As stated earlier, health-based
comparison values reflect concentrations much lower than those that have been observed to
cause adverse health effects and are protective of public health in essentially all exposure
situations. As a result, we do not consider concentrations detected at or below ATSDR’s
inhalation comparison values to warrant health concern. Therefore, no further evaluation is
required. The air monitoring data indicate that the mercury levels in the ambient air at EFPC are
not at levels of public health concern.
Current LWBR Air
No ambient air samples have been analyzed for mercury concentrations at the LWBR. But the
occurrence of harmful health effects from exposure to mercury vapor from contaminated soil is
not a concern for the LWBR. The mercury contamination accumulated in the sediments of the
river channel (where little, if any, exposure occurs), buried under as much as 80 centimeters of
cleaner sediment (ORNL and Jacobs Engineering Group 1995). The near-shore sediment
concentrations in the LWBR (less than 1 mg/kg; ORNL and Jacobs Engineering Group 1995) are
much lower than those found in the EFPC floodplain. Thus mercury levels in the ambient air
near LWBR (if any) are not expected to be at levels of public health concern.

IV.B.3.

Current Surface Water Exposure Pathway (inorganic mercury)

Current EFPC Surface Water
In a 1993 health consultation concerning Y-12 plant releases into
EFPC, ATSDR evaluated exposures to mercury contamination in
surface water using data from a summary of the EFPC Phase Ia RI
(ATSDR 1993). Within the creek in 1991 and 1992, surface water
was sampled from five stations (the mouth of Lake Reality,
confluence of EFPC with Poplar Creek, two intermediate stations,
and an area of known high contaminant concentrations in the
floodplain soil). Mercury was only detected in one sample. The
mercury concentration was 0.72 ppb (DOE 1992a; SAIC 1994a);
below U.S.EPA’s MCLG of 2 ppb in drinking water. Therefore, no
further evaluation is required. ATSDR concluded that the levels of
mercury in the surface water do not present a public health concern.

As stated earlier, comparison
values reflect concentrations
that are much lower than
those that have been
observed to cause adverse
health effects and are
protective of public health in
essentially all exposure
situations. As a result,
concentrations detected at or
below ATSDR’s comparison
values are not considered to
be a health concern.

The OREIS Environmental Database contains almost 650 surface water samples from EFPC
(OREIS 2009). The majority of the surface water samples were collected during Phase II
remediation of the Lower EFPC floodplain soil (Phase II excavation occurred from March 3 to
Page | 116

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

October 24, 1997; SAIC 2002a). Water samples were collected in 1991–1994, 1996, 1997, and
1999–2009 from 25 different locations in the creek. Of the 647 samples collected from the EFPC
surface water, mercury was detected in only 126 samples (about 1 out of 5 samples). As shown
in Table 17, in 1992, only one mercury concentration (about 0.1 percent) was detected slightly
above U.S.EPA’s MCLG of 2 ppb for drinking water. None of the 643 water samples collected
since 1992 have exceeded the MCLG. This indicates that the vast majority of the concentrations
were detected at levels not warranting health concern.
Table 17. Mercury Concentrations in EFPC Surface Water
Year

Minimum (ppb)

Maximum
(ppb)

1991

0

0.54

0.092

3/14

1992

2.8

2.8

2.8

1/1

1993

ND

ND

ND

0/2

1994

0

0.25

0.016

6/39

1996

0.10

0.52

0.30

5/5

1997

0

0.77

0.022

30/505

1999

0.22

0.71

0.467

2/2

2000

0.03

0.5

0.19

8/8

2001

0.029

0.96

0.25

11/11

2002

0.025

0.35

0.13

8/8

2003

0.02

0.21

0.093

8/8

2004

0.024

0.45

0.16

8/8

2005

0.028

0.45

0.15

8/8

2006

0.016

0.28

0.12

8/8

2007

0.022

0.28

0.095

8/8

2008

0.017

0.46

0.13

8/8

2009

0.19

0.28

0.15

4/4

0

2.8

0.047

126/647

Overall
Source: OREIS 2009

ND:
not detected

Average (ppb) Detection Frequency

Note: remember that exceeding a comparison value does not automatically mean that the
environmental concentrations are expected to produce harmful health effects. Comparison values
are not thresholds of toxicity. They simply indicate to ATSDR that further evaluation is
warranted. Keep in mind, too, that the comparison value ATSDR is using to screen surface water
samples is a drinking water guideline based on a lifetime exposure that assumes ingesting 1 liter
(children) or 2 liters (adults) of water per day. Adults and children are unlikely to participate in
recreational activities that would involve drinking EFPC surface water, especially since signs are
posted to warn the public to avoid contact with the water because of the bacterial contamination.

Page | 117

To evaluate the potential for exposure, ATSDR calculated exposure doses using the maximum
concentration detected in the EFPC surface water (2.8 ppb; OREIS 2009) and the formula
described in Section III.C.3 Comparing Estimated Doses to Health Guidelines. Both adults and
children were assumed to ingest 0.15 liters of water/day during a 3-hour swimming event (EPA
1997) for 4 days/year (minimum value for a farm family member described in ChemRisk 1999a).
ATSDR assumed that adults weighed 70 kg and were exposed for 30 years, and children
weighed 28.1 kg and were exposed for 6 years. Using these assumptions in the exposure dose
formula, both the estimated adult dose (6.6 × 10-5 mg/kg/day) and child dose (1.6 × 10-4
mg/kg/day) were below the U.S.EPA RfD of 3.0 × 10-4 mg/kg/day for chronic exposure to
inorganic mercury. The RfD is an estimate of the daily human exposure to a hazardous substance
likely to be without appreciable risk of adverse noncancer health effects. It has built-in
uncertainty or safety factors, making it considerably lower than levels at which health effects
have been observed. Estimated doses that are less than this value are not considered of health
concern. ATSDR does not expect that exposure to EFPC surface water would cause adverse
health effects.
ATSDR also evaluated an additional exposure scenario, assuming that the posted bacterial
advisory is ignored. Children were assumed to ingest 0.15 liters/day during a 3-hour swimming
event (EPA 1997) for 18 days/year (four times per month for 3 months plus six times over the
remainder of the year). As noted earlier, ATSDR assumed that children weighed 28.1 kg and
were exposed for 6 years. This scenario produced an estimated exposure dose (1.1 × 10-4
mg/kg/day) below the RfD (3.0 × 10-4 mg/kg/day) using the average concentration (0.42 ppb). 22
Even if children ignore the bacterial advisory, slightly more frequent exposures to mercury in the
surface water are also not expected to cause harmful health effects.
Current Oak Ridge Surface Water
The OREIS Environmental Database contains 53 surface water samples from the city of Oak
Ridge (OREIS 2009). Samples were collected in 1990, 1991, 1993, 1995–2001, and 2003–2005
from 15 different locations within the city of Oak Ridge. Of the 53 samples collected, mercury
was only detected in 10 samples (19 percent). In 1993, only one sample containing mercury was
above U.S.EPA’s MCLG of 2 ppb for drinking water (see Table 18). None of the water samples
collected since 1993 have exceeded the MCLG. In fact, mercury was only detected in one
sample since 1993. This indicates that the vast majority of the concentrations were detected at
levels not warranting a health hazard.

22

By using an average concentration, ATSDR can estimate a more probable exposure. In this case, using the
average concentration is even more appropriate given that the maximum detection seems to be an outlier. The
second highest concentration was 0.96 ppb and all but one sample were detected below the conservative
comparison value of 2 ppb (OREIS 2009).
Page | 118

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Table 18. Inorganic Mercury Concentrations in Oak Ridge Surface Water

Year

Minimum (ppb)

Maximum
(ppb)

1990

0.2

0.2

0.2

1/1

1991

ND

0.3

0.15

1/2

1993

0.2

4.7

1.08

7/7

1995

ND

ND

ND

0/2

1996

ND

ND

ND

0/11

1997

ND

0.1

0.014

1/7

1998

ND

ND

ND

0/2

1999

ND

ND

ND

0/4

2000

ND

ND

ND

0/6

2001

ND

ND

ND

0/2

2003

ND

ND

ND

0/6

2004

ND

ND

ND

0/2

2005

ND

ND

ND

0/1

Overall
ND
Source: OREIS 2009

ND:
not detected

4.7

0.15

10/53

Average (ppb) Detection Frequency

To evaluate the exposure further, ATSDR calculated exposure doses using the maximum
concentration detected in Oak Ridge surface water (4.7 ppb;
Remember that the RfD is an
OREIS 2009) and the formula described in Section III.C.3
estimate of daily human
Comparing Estimated Doses to Health Guidelines. Both adults and
exposure (including sensitive
subgroups) to a hazardous
children were assumed to ingest 0.15 liters of water/day during a
substance that is likely to be
3-hour swimming event (EPA 1997) for 4 days/year (minimum
without appreciable risk of
value for a farm family member described in ChemRisk 1999a).
adverse noncancer health
As noted earlier, ATSDR assumed that adults weighed 70 kg and
effects. It has built-in
were exposed for 30 years, and children weighed 28.1 kg and were
uncertainty factors, making it
considerably lower than levels
exposed for 6 years. Using these assumptions in the exposure dose
-4
at which health effects have
formula, both the estimated adult dose (1.1 × 10 mg/kg/day) and
-4
-4
been observed. Estimated
child dose (2.7 × 10 mg/kg/day) were below the RfD of 3.0 × 10
doses that are less than this
mg/kg/day for chronic exposure to inorganic mercury. ATSDR
value are not considered a
does not expect that exposure to surface water in the city of Oak
health hazard.
Ridge would cause harmful health effects.
Current Scarboro Surface Water
In May 1998, the Environmental Sciences Institute at FAMU collected seven surface water
samples from drainage ditches in the Scarboro community. Mercury was not detected in any of
the samples (the quantitation limit was 0.1 ppb; FAMU 1998; OREIS 2009). In September 2001,
U.S.EPA collected two surface water samples from the Scarboro community to validate the 1998
FAMU results. Mercury was not detected in either sample (the detection limit was 0.029 ppb;
Page | 119

EPA 2003). Therefore, no further evaluation is required—mercury has not been detected in any
surface water samples collected from the Scarboro community. The data indicate that exposure
to the surface water in Scarboro is not at levels that could cause adverse health effects.
As mentioned earlier in the hydrogeology section, the southward sloping orientation of the bed
planes beneath Pine Ridge prevents groundwater from flowing north toward Scarboro.
Furthermore, Scarboro is located outside of the EFPC floodplain. As Figure 9 shows, the
elevation of Scarboro is greater than 50 feet higher than EFPC. Therefore, contamination from
EFPC could not have reached Scarboro.
Current LWBR (Clinch River/Watts Bar Reservoir) Surface Water
In a 1996 health consultation on LWBR, ATSDR evaluated exposures to mercury contamination
in surface water in the reservoir. ATSDR determined that the levels of mercury in the surface
water do not present a public health concern, and the reservoir is safe for swimming, skiing,
boating, and other recreational purposes (ATSDR 1996b).
To arrive at this conclusion, ATSDR used surface water data from the LWBR RI/FS (ORNL and
Jacobs Engineering Group 1995), which references data from Phase I of the Clinch River RI
(Cook et al. 1992) and the ORR Environmental Monitoring Program (Energy Systems 1993).
Mercury was not detected in any of the surface water samples analyzed (detection limits ranged
from 0.05 to 0.2 ppb; ORNL and Jacobs Engineering Group 1995). Because mercury was not
detected in the surface water and the detection limits were below U.S.EPA's MCLG of 2 ppb, no
public health concerns arise from exposure to mercury in LWBR surface water.
The OREIS Environmental Database contains 311 surface water samples from LWBR (OREIS
2009). Samples were collected in 1990 and from 1993 to 2009 from 19 different locations in the
reservoir. Mercury was only detected 5 percent of the time (OREIS 2009). As shown in Table
19, when mercury was detected, the concentrations were less than U.S.EPA's MCLG of 2 ppb for
mercury in drinking water. No further evaluation is required, and the data indicate that exposure
to mercury in the surface water in LWBR is not causing harmful health effects.

Page | 120

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Table 19. Mercury Concentrations in LWBR Surface Water

Year

Minimum (ppb)

Maximum
(ppb)

1990

ND

ND

ND

0/4

1993

ND

ND

ND

0/14

1994

ND

1.3

0.024

10/90

1995

ND

0.056

0.0056

1/10

1996

ND

ND

ND

0/11

1997

ND

ND

ND

0/15

1998

ND

ND

ND

0/14

1999

ND

ND

ND

0/16

2000

ND

ND

ND

0/26

2001

ND

ND

ND

0/13

2002

ND

ND

ND

0/28

2003

ND

0.2

0.033

3/12

2004

ND

ND

ND

0/17

2005

ND

ND

ND

0/12

2006

ND

ND

ND

0/12

2007

ND

ND

ND

0/10

2008

ND

ND

ND

0/4

2009

ND

ND

ND

0/3

Overall
ND
Source: OREIS 2009

ND:
not detected

1.3

0.0084

14/311

Average (ppb) Detection Frequency

Municipal Water Systems
Drinking water from the municipal water supply systems is safe. The City of Oak Ridge,
including Scarboro, is supplied with treated water from the
Information about Tennessee’s Safe
Clinch River (Melton Reservoir) upstream of the ORR.
Drinking Water Program can be
Rockwood and Spring City draw surface water from the Piney
found at
River and King Creek tributary embayments of the LWBR. The
http://www.tn.gov/environment/dws/.
Kingston municipal water system intake is in the Tennessee
River upstream from where the Clinch River joins with the Tennessee River to form LWBR (see
Figure 1). Harriman receives their public water supply from the Emory River, which flows into
the LWBR. In addition, these municipal water systems are required to meet specific drinking
water quality standards set by U.S.EPA. Under the authorization of the Safe Drinking Water Act,
U.S.EPA has set national health-based standards to protect drinking water and its sources. TDEC
enforces these requirements and ensures that the drinking water is safe for public consumption.
Residents who use municipal drinking water should have no health concerns about that water.

Page | 121

Seeps and Springs
In 2006, ATSDR conducted a public health assessment that evaluated potential exposures to
contaminated off-site groundwater from the ORR (ATSDR 2006b). In this assessment, ATSDR
evaluated data from seeps and springs from various sampling locations around the main ORR
facilities: near the East Tennessee Technology Park (formerly the K-25 site), near the Oak Ridge
National Laboratory (formerly the X-10 site), and near the Y-12 National Security Complex
(formerly the Y-12 plant). Elevated levels of mercury were not found in any of the seep or spring
water samples. For the complete evaluation of seeps and springs, please refer to ATSDR’s Public
Health Assessment: Evaluation of Potential Exposures to Contaminated Off-Site Groundwater
from the Oak Ridge Reservation (ATSDR 2006b) (available on the Internet at
http://www.atsdr.cdc.gov/HAC/pha/PHA.asp?docid=1371&pg=0).

IV.B.4.

Current Groundwater Exposure Pathway

In the 2006 public health assessment, ATSDR concluded that no human exposures to
contaminated groundwater outside the ORR boundary have occurred in the past, are currently
occurring, or are likely to occur in the future (ATSDR 2006b). Therefore, ATSDR does not
expect any health effects from exposure to contaminated off-site groundwater. For a complete
evaluation of groundwater, please refer to ATSDR’s Public Health Assessment: Evaluation of
Potential Exposures to Contaminated Off-Site Groundwater from the Oak Ridge Reservation
(ATSDR 2006b) (available on the Internet at
http://www.atsdr.cdc.gov/HAC/oakridge/phact/groundwater/index.html).

IV.B.5.

Current Soil Exposure Pathway (inorganic mercury)

Current EFPC Soil
EFPC Floodplain Soil (prior to remediation in 1997)
In a 1993 health consultation concerning Y-12 plant releases into EFPC, ATSDR evaluated soil
data from the EFPC Phase Ia RI (ATSDR 1993). ATSDR concluded that in some locations along
EFPC, mercury levels in the floodplain soil could pose a threat to people—especially children—
who ingest, inhale, or have dermal contact with contaminated soil while playing or fishing along
the creek’s floodplain (ATSDR 1993).
See section IV.A.4. Past Soil and Sediment Exposure Pathways for a more extensive public
health analysis of potential exposure to the EFPC floodplain soil prior to remediation of soil
containing greater than 400 ppm of mercury in 1996 and 1997. ATSDR concluded that children
who played at the NOAA site and Bruner site before the soil removal activities could have
accidentally swallowed inorganic mercury in EFPC floodplain soils, which may have increased
the risk of developing renal effects. Adults are not expected to have been harmed from exposure
to inorganic mercury in soil. Accidental ingestion of methylmercury in EFPC floodplain soils in
the past is not expected to have caused harmful health effects for anyone contacting the
floodplain soil.
ATSDR’s Evaluation of DOE’s Proposed Mercury Cleanup Level for EFPC Floodplain
Soil
In response to public comments on the 1995 Proposed Plan for East Fork Poplar Creek (DOE
1995d), DOE, U.S.EPA, and TDEC selected a remedial action to remove soils containing greater
than 400 ppm of mercury from the EFPC floodplain (DOE 1995b). This 400 ppm mercury cleanPage | 122

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

up level is higher than the original remediation goal of 50 ppm. Some community members and
organizations were concerned about this higher clean-up level and asked ATSDR to evaluate
whether the proposed clean-up level of 400 ppm in EFPC floodplain soil was protective of public
health.
To help evaluate the proposed EFPC mercury clean-up level for soil, ATSDR sponsored a
Science Panel Meeting on the Bioavailability of Mercury in Soil. The science panel convened to
identify methods and strategies for the development of data-supported, site-specific estimates of
the bioavailability of inorganic mercury and other metals from soils. Private consultants and
academicians internationally known for their metal bioavailability research were invited to the
meeting, which was held in August 1995. In addition to these members, the panel included
experts from ATSDR, CDC, U.S.EPA, and the National Institute for Environmental Health
Sciences. The science panel published four articles on bioavailability of inorganic mercury in
soil in Risk Analysis 17(5), 527-569 (Canady et al. 1997).
ATSDR analyzed the clean-up level using a worst-case scenario and a likely mercury exposure
scenario of young children in a residential setting (ATSDR 1996a). The worst-case exposure
scenario assumed a 16-kg child ingested 100 mg of soil every day. The likely exposure scenario
assumed that a 16-kg child ingested 100 mg/day, 5 days/week for 36 weeks/year. For both
exposure scenarios, estimated oral exposure doses of mercury were orders of magnitude lower
than the NOAEL and LOAEL for inorganic mercury. ATSDR also considered inhalation of
mercury vapor from the floodplain soil and determined that the level of mercury vapor in air
above floodplain soil with 400 ppm of mercury or less would be too low to be a health hazard
(ATSDR 1996a). ATSDR concluded that the clean-up level of 400 ppm of mercury in EFPC
floodplain soil is protective of public health and poses no health threat to children or adults
(ATSDR 1996a).
The excavation of floodplain soils with greater than 400 ppm of mercury was conducted in two
phases. From July 8 to September 14, 1996 (Phase I), 4,250 m3 of mercury-contaminated soils
were removed from the floodplain near the NOAA Atmospheric Diffusion Laboratory off Illinois
Avenue. From March 3 to October 24, 1997 (Phase II), an additional 29,970 loose m3 of
mercury-contaminated soils were removed from the floodplain near the NOAA site and across
the Oak Ridge Turnpike from the Bruner’s Shopping Center on the Wayne Clark Property (SAIC
1994a, 2002a). Confirmatory samples 23 were taken during both phases of the excavation to
ensure that the remediated areas contained less mercury than the clean-up standard (SAIC 1998).
Postremediation monitoring (mercury input, stream stability, and fish sampling) is conducted to
ensure the effectiveness of the excavation (SAIC 2002a). Following cleanup and removal in
1996 and 1997, mercury in EFPC is not a public health hazard.
Current Oak Ridge Soil
The OREIS Environmental Database contains over 200 soil samples from the city of Oak Ridge
(OREIS 2009). Samples were collected in 1991, 1992, 1995, 1999, and 2000 from 176 different
locations within the city. As shown in Table 20, mercury was detected in 157 samples (70
percent). Of the 224 samples collected from soil in the city of Oak Ridge, 34 samples (15
percent) were detected above the comparison value of 20 ppm (OREIS 2009).

23

Data from Phase Ia and Ib of the EFPC RI, including the confirmatory samples, appear to be included in OREIS.
Page | 123

Table 20. Mercury Concentrations in Oak Ridge Soil

Maximum
(ppm)

Year

Minimum (ppm)

1991

2.3

126

14.13

45/45

1992

ND

158

22.18

45/52

1995

ND

48.6

6.38

45/85

1999

ND

49.5

2.62

21/41

2000

0.13

0.13

1/1

10.89

157/224

Overall
ND
Source: OREIS 2009

ppm:
parts per million

0.13
158

Average (ppm) Detection Frequency

Because the comparison value was exceeded, ATSDR continued to evaluate exposures to Oak
Ridge soil. As the next step in the screening process, ATSDR
calculated exposure doses using the maximum concentration
As stated earlier, comparison
values reflect concentrations
detected in the soil (158 ppm; OREIS 2009) and the formula
much lower than those that
described in Section III.C.3. Comparing Estimated Doses to
have been observed to cause
Health Guidelines. To calculate exposure doses, an adult was
adverse health effects and are
assumed to ingest 100 mg of soil/day for 16 days/year (2 times a
protective of public health in
month for 8 months; likely scenario described in ChemRisk
essentially all exposure
situations. As a result,
1999a). A child was assumed to ingest 200 mg/day for 180
concentrations detected at or
days/year (20 times a month for 6 months). As noted earlier,
below ATSDR’s comparison
ATSDR assumed that adults weighed 70 kg and were exposed for
values are not considered a
30 years, and children weighed 28.1 kg and were exposed for 6
health concern.
years.
Using these assumptions in the exposure dose formula, the estimated adult dose (9.9 × 10-6
mg/kg/day) was below U.S.EPA’s RfD of 3.0 × 10-4 mg/kg/day for chronic exposure to
inorganic mercury. Estimated doses at or less than the RfD are not considered a health hazard.
But the child dose (5.5 × 10-4 mg/kg/day) was slightly higher than the RfD. Still, when compared
with actual health effects levels studied in the toxicological and epidemiological literature
(autoimmune effects were observed in Brown Norway rats exposed to doses of 0.226, 0.317, and
0.633 mg/kg/day [Andres 1984; Bernaudin et al. 1981; Druet et al. 1978]), the child dose is three
orders of magnitude lower. Therefore, ATSDR does not expect that exposure to mercury in Oak
Ridge soil to cause adverse health effects.
Current Scarboro Soil
In May 1998, the Environmental Sciences Institute at FAMU collected 40 surface soil samples
from the Scarboro community. Mercury concentrations ranged from 0.021 to 0.30 ppm, with a
median value of 0.11 ppm (FAMU 1998; OREIS 2009). In September 2001, U.S.EPA collected
six surface soil samples from the Scarboro community to validate the 1998 FAMU results.
Mercury concentrations ranged from 0.0432 to 0.0904 ppm, with an average concentration of
0.07 ppm (EPA 2003). All of these concentrations are below the comparison value of 20 ppm for
mercury in soil. Therefore, no further evaluation is required. The sampling data indicate that the
mercury levels in the surface soil in Scarboro are not at levels of public health hazard.
Page | 124

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Current LWBR Soil
The OREIS Environmental Database does not contain any soil samples collected from the
LWBR (OREIS 2009). Even though no data are available, the occurrence of harmful health
effects from exposure to mercury in soil along the LWBR shoreline is not a concern. Mercury
from ORR operations has not contaminated the soil near LWBR. Mercury from the ORR was
released into EFPC from the Y-12 plant and traveled to the LWBR through Poplar Creek and the
Clinch River. The mercury accumulated in the sediments of the LWBR river channel (where
little, if any, exposure would occur) and is buried under as much as 80 centimeters of cleaner
sediment and several meters of water (ORNL and Jacobs Engineering Group 1995; ATSDR
1996b). The near-shore sediment concentrations in the LWBR were less than 1 ppm—much
lower than the comparison value of 20 ppm for mercury in soil (ORNL and Jacobs Engineering
Group 1995).
In 1996, ATSDR evaluated ORR-related chemical and radiological contaminants in the surface
and deep channel sediments of the LWBR (ATSDR 1996b). Specifically, ATSDR evaluated
surface sediments in shallow areas of the reservoir using maximum concentrations of
contaminants (e.g., mercury) and worst case scenarios, including if surface sediments were
dredged and used as surface soil at residential properties. ATSDR concluded that the maximum
chemical contaminant concentrations (including mercury) would not present a public health
hazard. Additionally, ATSDR evaluated the potential exposure (ingestion, inhalation, and dermal
contact) if these subsurface sediments were removed and used as surface soil on residential
properties. ATSDR concluded that the potential exposure to mercury would not pose a health
concern, even if these deep sediments were dredged and used as residential soil. Accordingly, the
mercury levels in the soil near the LWBR are not a public health hazard.

IV.B.6.

Current Sediment Exposure Pathway (inorganic mercury)

Current EFPC Sediment
In a 1993 health consultation concerning Y-12 plant releases into
Remember, an environmental
EFPC, ATSDR evaluated sediment data from the EFPC Phase Ia RI
concentration that exceeds a
(ATSDR 1993). From Autumn 1990 to Spring 1991, nine samples
comparison value doesn’t
automatically mean harmful
were collected from seven sites within EFPC to define source
health effects. Comparison
contributions (DOE 1992a; SAIC 1994a). Phase 1b of the EFPC RI
values are not thresholds of
was conducted from August 1991 to February 1992 to determine the
toxicity. They simply indicate
extent and distribution of contaminants within the floodplain (SAIC
to ATSDR that further
1994a). Transects were established across the floodplain at 100evaluation is warranted.
meter intervals. Stream sediment samples were taken at oddnumbered transects, and every three sequential sediment samples were composited for analysis.
Investigators collected 27 sediment samples, each one representing 600 meters of the creek
(SAIC 1994a). Sediment samples from both phases ranged from 10 to 2,240 ppm, which
exceeded the comparison value of 20 ppm for mercury in sediment. But the maximum value
(2,240 ppm) appears to be an outlier; it was reportedly taken from an area with obvious creek
sediment contamination (SAIC 1994a). The second highest concentration from this dataset
appears to be 95.6 ppm, 24 which also exceeds the comparison value (SAIC 1994a). The mean
24

ATSDR does not have access to the raw data. ATSDR makes an assumption about the 2,240 ppm detection being
an outlier based on the data presented in tables within the EFPC RI (SAIC 1994a). Specifically, Table 3.19, the
results for the Phase 1a and 1b sediment sampling, does not contain this value.
Page | 125

concentration, based on a total of 35 samples (excluding the 2,240 ppm outlier) is 14.9 ppm
(SAIC 1994a). The data from the EFPC RI does not appear to be in the OREIS Environmental
Database. Because ATSDR does not have access to the raw data from this investigation, the
EFPC RI data cannot be combined with the data available in OREIS.
The OREIS Environmental Database contains 58 sediment samples from EFPC (OREIS 2009).
Samples were collected in 1990–1992, 1994, and 1996 from 38 different locations in the creek.
As shown in Table 21, mercury concentrations exceeded the comparison value of 20 ppm for
sediment. Of the 58 samples collected from the EFPC sediment, 20 samples (34 percent) were
detected above the comparison value (OREIS 2009).
Table 21. Mercury Concentrations in EFPC Sediment
Year

Minimum (ppm)

Maximum
(ppm)

Average (ppm) Detection Frequency

1990

15.4

42

28.7

2/2

1991

ND

101

17.58

13/26

1992

0.94

120

24.13

19/19

1994

0.03

1996

2.24

Overall
ND
Source: OREIS 2009

ppm:
parts per million

0.061
78.89
120

0.045

2/2

40.00

9/9

21.59

45/58

Because the comparison value was exceeded in both datasets, ATSDR continued to evaluate
exposures to EFPC sediments. Adults and children are unlikely to participate in recreational
activities in the EFPC sediments, especially since signs are posted to warn the public to avoid
contact with the creek’s surface water because of the bacterial contamination. In 1992, some of
the advisory signs along the creek were replaced and additional signs were posted (TDEC 1992).
However, to evaluate the potential for exposure, ATSDR calculated exposure doses using the
maximum concentration detected in the sediments (2,240 ppm; SAIC 1994a) and the formula
described in Section III.C.3. Comparing Estimated Doses to Health Guidelines. Specifically,
ATDSR assumed that adults weighed 70 kg and were exposed to the maximum concentration for
30 years, and children weighed 28.1 kg and were exposed to the maximum concentration for 6
years. To calculate exposure doses, an adult was assumed to ingest 50 mg of sediment/day for 4
days/year (minimum value for a farm family member described in ChemRisk 1999a). A child
was assumed to ingest 100 mg/day of sediment for 4 days/year (minimum value for a farm
family member described in ChemRisk 1999a). ATSDR assumed that adults weighed 70 kg and
were exposed for 30 years, and children weighed 28.1 kg and were exposed for 6 years.
Using these assumptions in the exposure dose formula, both the estimated adult dose (1.8 × 10-5
mg/kg/day) and child dose (8.7 × 10-5 mg/kg/day) were below U.S.EPA’s RfD of 3.0 × 10-4
mg/kg/day for chronic exposure to inorganic mercury. Remember that the RfD is an estimate of
the daily human exposure to a hazardous substance that is likely to be without appreciable risk of
adverse noncancer health effects. Estimated doses below these values are not considered of
health concern. Furthermore, ATSDR used the maximum concentration (2,240 ppm) (most likely
an outlier) to calculate these exposure doses. The levels that people are actually being exposed to
Page | 126

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

are expected to be much lower. Exposures to EFPC sediments are not expected to cause harmful
health effects.
ATSDR also evaluated an additional exposure scenario: assuming the posted bacterial advisory
to avoid contact with the water is ignored. Children were assumed to ingest 100 mg/day of
sediment for 18 days/year (four times per month for 3 months plus six times over the remainder
of the year). As noted earlier, ATSDR assumed that children weighed 28.1 kg and were exposed
for 6 years. Using the maximum concentration (2,240 ppm; SAIC 1994a), this scenario produced
an estimated exposure dose (3.9 × 10-4 mg/kg/day) slightly above the RfD (3.0 × 10-4
mg/kg/day). As stated earlier, however, ATSDR believes that the maximum concentration from
the EFPC RI is an outlier. If this data point is removed and the dose is recalculated using the
second highest concentration (120 ppm from the OREIS database), the resulting exposure dose
(2.1 × 10-5 mg/kg/day) is lower than the RfD for chronic exposure to inorganic mercury. Thus,
even if the bacterial advisory for water is ignored, more frequent exposures to mercury in the
sediments are not expected to cause harmful health effects for children.
Current Oak Ridge Sediment
The OREIS Environmental Database contains 36 sediment samples from the city of Oak Ridge
(OREIS 2009). Samples were collected in 1990, 1991, 1993, 1995, and 1997–2001 from 15
different locations within the city. As shown in Table 22, mercury was detected in 30 samples
(83 percent). Of the 36 samples collected from sediment in the city of Oak Ridge, 6 samples (17
percent) were detected above the comparison value of 20 ppm (OREIS 2009).
Table 22. Mercury Concentrations in Oak Ridge Sediment
Year

Minimum (ppm)

Maximum
(ppm)

Average (ppm) Detection Frequency

1990

34.4

34.4

34.4

1/1

1991

20.4

35.7

30.57

3/3

1993

0.096

6.6

1.64

7/7

1995

ND

31.8

6.19

7/11

1997

ND

0.93

0.47

1/2

1998

0.29

0.37

0.33

2/2

1999

0.12

0.25

0.18

6/6

2000

ND

0.35

0.18

1/2

2001

0.12

0.17

0.15

2/2

5.80

30/36

Overall
ND
Source: OREIS 2009

ppm:
parts per million

35.7

Comparison value exceedences caused ATSDR to continue its evaluation of exposures to Oak
Ridge sediment. As the next step in the screening process, ATSDR calculated exposure doses
using the maximum concentration detected in the sediment (35.7 ppm; OREIS 2009) and the
formula described in Section III.C.3. Comparing Estimated Doses to Health Guidelines. To
calculate exposure doses, an adult was assumed to ingest 50 mg of sediment/day for 24 days/year
Page | 127

(4 times per month for 4 months plus two times a month for 4 months). A child was assumed to
ingest 100 mg/day of sediment for 32 days/year (6 times a month for 4 months plus 2 times per
month for 4 months). ATSDR assumed that adults weighed 70 kg and were exposed for 30 years,
and children weighed 28.1 kg and were exposed for 6 years.
Using these assumptions in the exposure dose formula, both the estimated adult dose (1.7 × 10-6
mg/kg/day) and child dose (1.1 × 10-5 mg/kg/day) were below U.S.EPA’s RfD of 3.0 × 10-4
mg/kg/day for chronic exposure to inorganic mercury. Estimated doses below the RfD are not
considered to be a health hazard. ATSDR does not expect that exposure to mercury in the
sediment in the City of Oak Ridge would cause adverse health effects.
Current Scarboro Sediment
In May 1998, the Environmental Sciences Institute at FAMU collected nine sediment samples
from drainage ditches in the Scarboro community. Mercury concentrations ranged from 0.018 to
0.12 ppm, with an average of 0.05 ppm (FAMU 1998; OREIS 2009). In September 2001,
U.S.EPA collected two sediment samples from the Scarboro community to validate the 1998
FAMU results. Mercury was detected at concentrations of 0.0271 and 0.0393 ppm (EPA 2003).
All of these concentrations are at least two orders of magnitude below the comparison value of
20 ppm for mercury in sediment. No further evaluation is required—the sampling data indicate
that the mercury levels in Scarboro sediment are not at levels of public health concern.
Current LWBR Sediment
Mercury from the ORR was released into EFPC from the Y-12 plant and traveled to the LWBR
through Poplar Creek and the Clinch River. The mercury accumulated in the deep sediments of
the LWBR river channel, buried under as much as 80 centimeters of cleaner sediment and
several meters of water (ORNL and Jacobs Engineering Group 1995). Exposure to sediments in
the deep channel, therefore, is not expected. On the other hand, exposure to sediment in shallow,
near-shore areas is more likely. ATSDR thus evaluated these exposure scenarios separately,
except when the depths of the sediment sampling were unspecified.
Shallow, near-shore sediment
For several months every winter, sediments in shallow areas along the LWBR are above the
water line. In a 1996 LWBR health consultation, ATSDR evaluated exposures to mercury
contamination in surface sediments in the reservoir using maximum concentrations and worstcase scenarios (ATSDR 1996b). ATSDR assumed children could be exposed to mercury in the
shallow sediments while swimming or fishing in the reservoir or if surface sediments were
dredged and used for surface soil at residential properties. ATSDR determined that the levels of
mercury in the surface sediments did not present a public health concern.
ATSDR used near-shore sediment data from the LWBR RI/FS (ORNL and Jacobs Engineering
Group 1995), which references data from TVA’s Recreation Area Sampling Study (TVA 1991).
In May and June1990, the TVA sampled near-shore sediments from recreational areas along the
LWBR. Five sediment samples were collected from each recreational area, which were then
combined to make one composite sample for analysis (TVA 1991). Mercury was only detected
in three of the 12 composite samples in concentrations of 0.15 ppm 25 (the detection limit was 0.1
ppm; TVA 1991). These concentrations are two orders of magnitude below the comparison value
25

These data appear to be included in the OREIS database.
Page | 128

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

of 20 ppm for mercury in sediment. Therefore, no further evaluation is required—the sampling
data indicated that the mercury levels in the shallow sediments in LWBR were not at levels of
public health concern. ATSDR does not expect current conditions to be different from those in
the 1990s, because there have been no significant mercury releases and the deep channel
sediments have not been disturbed.
Deep channel sediments
As stated earlier, people are not directly exposed to the highest concentrations of mercury in the
subsurface sediments; these deposits are found in deep channels where contaminants are covered
by 40 to 80 centimeters of sediment and several meters of water (ORNL and Jacobs Engineering
Group 1995). In a 1996 health consultation, ATSDR evaluated potential exposure a child might
receive if the subsurface sediments were removed from the deep reservoir channels and used as
surface soil in residential properties (ATSDR 1996b). ATSDR determined that the levels of
mercury in the deep channel sediments do not present a public health concern.
ATSDR used deep-water sediment data from the LWBR RI/FS (ORNL and Jacobs Engineering
Group 1995), which references mercury data from a 1986 study in which two core samples from
the LWBR were analyzed (TVA 1986) and a 1992 study in which four core samples from the
LWBR were analyzed (Cook et al. 1992). Mercury was detected in concentrations ranging from
1 to 3 ppm (ORNL and Jacobs Engineering Group 1995). These concentrations are six to 20
times lower than the 20-ppm comparison value for mercury in sediment. No further evaluation is
required—the sampling data indicate that the mercury levels in the deep channel sediments in
LWBR are not at levels of public health concern.
Unspecified sediment depths
The OREIS Environmental Database contains 140 sediment samples from the LWBR (OREIS
2009). In 1990, from 1993 to 2002, and in 2004, samples were collected from 43 different
reservoir locations. The depths of the sediment samples are not clear. As shown in Table 23, in
1990 and 2002, maximum mercury concentrations exceeded the comparison value of 20 ppm for
sediment. Yet the average mercury concentrations were below the comparison value. Of the 140
samples collected from the LWBR sediment, only six samples (about 4 percent) were detected
above the comparison value (OREIS 2009). This indicates that the vast majority of the
concentrations were detected at levels that do not warrant health concern.

Page | 129

Table 23. Mercury Concentrations in LWBR Sediment

Year

Minimum (ppm)

1990

0.061

1993

1.4

1994

0.05

1995

ND

1996

Maximum
(ppm)

Average (ppm) Detection Frequency
11.76

39/39

6.4

2.35

16/16

12.3

1.77

42/42

1.21

0.60

4/5

0.11

6.2

1.48

6/6

1997

0.52

0.52

0.52

1/1

1998

0.57

0.59

0.58

2/2

1999

0.24

4.5

1.42

6/6

2000

0.09

2.79

1.57

6/6

2001

0.17

1.05

0.55

5/5

2002

0.08

42.2

6.15

8/8

2004

ND

11.4

3.6

3/4

4.78

138/140

Overall
ND
Source: OREIS 2009

ppm:
parts per million

160

160

Nevertheless, because the comparison value was exceeded, ATSDR further evaluated exposures
to LWBR sediments. As the next step in the screening process, ATSDR calculated exposure
doses using the maximum concentration detected in the unspecified sediments (160 ppm; OREIS
2009) and the formula described in Section III.C.3. Comparing Estimated Doses to Health
Guidelines. For exposure purposes, ATSDR assumed that all the unspecified depth samples were
shallow, near-shore sediments—that is, that they were accessible.
LWBR is a high-use recreational area. Not only do people live in the vicinity of the reservoir, but
people from outside the area visit the many parks and recreational facilities (TVA 1987, 1990).
People, particularly children, who fish, play, hike, or swim along the reservoir may be exposed to
mercury through ingestion of sediment from inadvertent hand-to-mouth activities. Young
children have the greatest risk of exposure to mercury. Given that children play in the dirt and
engage in frequent hand-to-mouth activity and often mouth objects, they are likely to have the
most frequent and longest duration exposure to LWBR near-shore sediments.
To calculate exposure doses. ATSDR assumed an adult ingested 50 mg of sediment/day for 24
days/year (four times per month for 4 months plus two times a month for 4 months). We
assumed a child ingested 100 mg/day for 32 days/year (six times a month for 4 months plus two
times per month for 4 months). ATSDR assumed that adults weighed 70 kg and were exposed
for 30 years, and children weighed 28.1 kg and were exposed for 6 years.
Using these assumptions in the exposure dose formula (see Section III.C.3. Comparing
Estimated Doses to Health Guidelines), both the estimated adult dose (7.5 × 10-6 mg/kg/day) and
child dose (5.0 × 10-5 mg/kg/day) were well below U.S.EPA’s RfD of 3.0 × 10-4 mg/kg/day for
chronic exposure to inorganic mercury. Remember that estimated doses at or less than the RfD
Page | 130

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

are not considered of health concern. Furthermore, ATSDR used the maximum concentration
(160 ppm) to calculate these exposure doses, but the vast majority of the samples (96 percent)
were detected below the conservative comparison value of 20 ppm. Exposures to LWBR
sediments are not expected to cause harmful health effects.
Still, to prevent unnecessary exposures to workers and the public, ATSDR cautions that the
sediments should not be disturbed, removed, or disposed of without careful review by the
interagency working group (DOE, TDEC, U.S.EPA, TVA, and the U.S. Army Corps of
Engineers). Established in 1991, the interagency working group coordinates and reviews
permitting and other use activities that could result in the disturbance, resuspension, removal,
disposal—or a combination thereof—of contaminated sediments in the Watts Bar Reservoir
(DOE 1995c; SAIC 2004).

IV.B.7.

Current Biota Exposure Pathway

Current EFPC Biota
EFPC Fish (methylmercury)
In a 1993 health consultation concerning Y-12 plant releases into EFPC, ATSDR evaluated a
summary of the November, 1990, and May, 1991, fish data from EFPC compiled by the DOE
Biological Monitoring and Abatement Program (ATSDR 1993). Concentrations of mercury in
fish fillets ranged from 0.08 to 1.31 ppm 26 (DOE 1992a; ORNL 1992). This exceeded the
comparison value of 0.14 ppm for fish samples. ATSDR concluded that the levels of mercury
found in fish from EFPC were at levels of public health concern (ATSDR 1993).
The OREIS Environmental Database contains 430 samples from redbreast sunfish, rock bass,
largemouth bass, and crayfish collected from seven locations in EFPC (OREIS 2009). Redbreast
sunfish were collected in 1991 and 1995 through 2001 and 2004–2008; rock bass were collected
in 2004, 2006, 2008, and 2009; largemouth bass were collected in 1995; and crayfish were
collected in 1991. As shown in Table 24, mercury was detected in all 430 fish and crayfish
samples above the comparison value of 0.14 ppm (OREIS 2009). Remember this does not
automatically mean that an environmental concentration exceeding a comparison value is
expected to produce harmful health effects. Comparison values are not thresholds of toxicity.
They simply indicate a need for further evaluation.

26

These data appear to be included in the OREIS database.
Page | 131

Table 24. Mercury Concentrations in Fish from EFPC

Species

Portion

Minimum
(ppm)

Maximum
(ppm)

Average
(ppm)

Detection
Frequency

Largemouth Bass (Hg)

Muscle

0.51

0.61

0.56

2/2

Redbreast Sunfish (Hg)

Fillet/Muscle

0.37

1.8

0.87

167/167

Redbreast Sunfish (Hg)

Whole body

0.59

2.5

1.4

8/8

Redbreast Sunfish (Hg)

Unknown

0.35

1.6

0.86

120/120

Redbreast Sunfish (MeHg)

Muscle

0.50

1.5

0.92

24/24

Redbreast Sunfish (MeHg)

Unknown

0.19

1.6

0.63

36/36

Rock Bass (Hg)

Muscle

0.64

1.58

1.0

67/67

Crayfish (Hg)

Whole body

0.51

6.6

3.3

6/6

0.19

6.6

—

430/430

Overall
Source: OREIS 2009
ppm:
parts per million

Some of the fish samples were analyzed specifically for methylmercury and other samples were
analyzed for total mercury (OREIS 2009). In fish tissue, mercury is present predominantly as
methylmercury (about 85 percent; Jones and Slotten 1996). Methylmercury is the organic form
of mercury and is much more harmful via the oral route than the elemental and inorganic forms
(ATSDR 1999). Thus ATSDR took a conservative approach and assumed that all the total
mercury detected in the fish was methylmercury.
Because the comparison value was exceeded, ATSDR continued to evaluate mercury exposures
from eating EFPC fish. That anyone is actually eating fish from EFPC is unlikely. EFPC is not a
productive fishing location, and a fish consumption advisory is in place. Nevertheless, ATSDR
evaluated a potential exposure scenario and assumed people would ignore the advisory.
To evaluate this potential exposure scenario, ATSDR calculated exposure doses using the
average concentration detected in the EFPC fish fillet and muscle samples 27 and the formula
described in Section III.C.3. Comparing Estimated Doses to Health Guidelines. ATSDR assumed
that both adults and children ate one 8-ounce fish meal each month (12 meals/year = 7.5
grams/day). As noted earlier, ATSDR assumed that adults weighed 70 kg and were exposed for
30 years, and children weighed 28.1 kg and were exposed for 6 years.
Using these assumptions in the exposure dose formula, some of the estimated doses from eating
EFPC fish once a month were above both the ATSDR MRL for methylmercury (3.0 × 10-4
mg/kg/day) and the U.S.EPA RfD for methylmercury (1.0 × 10-4 mg/kg/day) (see Table 25).
Remember that calculated exposure doses higher than the health guidelines do not automatically
mean harmful health effects. They are instead an indication that ATSDR should examine further
the harmful effect levels reported in the scientific literature and more fully review exposure
potential. Therefore, ATSDR compared these potential exposure doses with actual health effects
levels in the toxicological and epidemiological literature.
27

It is standard protocol to analyze fillets/edible portions when evaluating human health concerns.
Page | 132

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Table 25. Estimated Methylmercury Exposure Doses from Consuming EFPC Fish

Species

Average
Concentration
(ppm)

Estimated Exposure Doses
(mg/kg/day)
adult
child

Largemouth Bass (Hg in muscle)

0.56

6.0 × 10-5

1.5 × 10-4

Redbreast Sunfish (Hg in fillet/muscle)

0.87

9.3 × 10-5

2.3 × 10-4

Redbreast Sunfish (MeHg in muscle)

0.92

9.9 × 10-5

2.5 × 10-4

Rock Bass (Hg in muscle)

1.0

1.1 × 10-4

2.7 × 10-4

Crayfish (Hg in whole body)

3.3

3.5 × 10-4

8.8 × 10-4

mg/kg/day: milligrams per kilograms per day
ppm:
parts per million
Bold text indicates that the exposure dose is higher than the U.S.EPA RfD of 1.0 × 10 -4 mg/kg/day.

The ATSDR chronic MRL of 3 × 10-4 mg/kg/day for ingestion of organic mercury is based on
the Seychelles Child Development Study, in which people who were exposed to 1.3 × 10-3
mg/kg/day of methylmercury in their food did not experience any adverse health effects
(NOAEL; Davidson et al. 1998). The U.S.EPA RfD of 1 × 10-4 mg/kg/day is based on the Faroe
Islands study, in which maternal dietary intakes of 8 × 10-4 mg/kg/day to 1.5 × 10-3 mg/kg/day
were associated with performance on standardized neurobehavioral tests involving effects on
attention, memory, confrontational naming, and to a lesser extent visual/spatial abilities and finemotor functions in children (LOAELs; Debes et al. 2006; Grandjean et al. 1997; NRC 2000).
These U.S.EPA benchmark dose lower limits (BMDL05) are expected to be associated with a 5
percent increase in the incidence of neurodevelopmental effects in children exposed in utero. The
U.S.EPA RfD is consistent with the approach used by the NAS which identified a dose of 1.1 ×
10-3 mg/kg/day as a dose that results in a 5 percent increase in the incidence of abnormal scores
on the Boston Naming Test (a picture-naming, vocabulary test) (NRC 2000). 28
Women who ate one meal a month of EFPC fish in the 1990s and 2000s were not at risk of
harming a developing fetus if they were pregnant. The estimated doses in Table 25 for women
are at or below the U.S.EPA RfD and are not at levels associated with harmful effects in the
fetus. However, the estimated exposure doses for children eating fish from EFPC once a month
are slightly above the U.S.EPA RfD, but are not close to the NAS dose effect level or the EPA
BMDL05. Figure 26 compares the estimated exposure doses in Table 25 to the health effect
levels and health guidelines. Whether children are as sensitive to the neurotoxic effects of
mercury as is the fetus is uncertain. Even if children were not exposed in utero, some young
children who frequently eat the same fish as their mother ate are also at an increased level of risk
for harmful effects. This conclusion is somewhat uncertain, primarily because a person’s
mercury response is itself somewhat uncertain. Contributing to that uncertainty is how the body
handles mercury, and the sex, genetics, health, and nutritional status of the person who eats the
fish, or how mercury is handled in the body.
Only the estimated methylmercury dose for children eating one meal a month of crayfish from
the EFPC is above the lowest LOAEL (8 × 10-4 mg/kg/day) from the Faroe Island study and
28

These neurodevelopmental effects were observed at a population level; not on an individual basis.
Page | 133

comes close to the NAS dose effect level. Therefore, children who ignore the posted EFPC
advisory (no fishing and no contact with water) may be at risk of subtle neurodevelopmental
effects if they eat one crayfish meal a month. Pregnant women who ate one crayfish meal a
month have a small increased risk of harming a developing fetus because the estimated
methylmercury dose is slightly above the U.S.EPA RfD, but not close to the NAS dose effect
level or the EPA BMDL05. Figure 26 compares the estimated exposure doses in Table 25 to the
health effect levels and health guidelines. However, it is highly unlikely for pregnant women and
young children to eat one meal a month of EFPC crayfish because of the posted advisory and
EFPC is not a productive fishing location.

Page | 134

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Figure 26. Current Estimated Methylmercury Exposure Doses from Eating EFPC Fish and

Crayfish Compared to Health Effect Levels and Health Guidelines

Page | 135

EFPC Vegetables (inorganic mercury)
The OREIS Environmental Database contains 16 samples of beet, kale (cabbage), and tomato
collected from two locations in the EFPC floodplain in 1992 (OREIS 2009). Mercury was
detected in 12 of the 16 samples (75 percent). See Table 26 for a summary of the mercury
concentrations detected in each type of plant.
Table 26. Mercury Concentrations in Edible Plants from EFPC
Species

Portion

Minimum (ppm) Maximum (ppm) Average (ppm)

Detection
Frequency

Beet

Root

0.63

2.7

1.3

4/4

Kale

Leaves

0.13

3.2

0.80

7/7

Tomato

Fruit

ND

0.42

—

1/5

ND

3.2

—

12/16

Overall
Source: OREIS 2009
ppm:
parts per million

Comparison values are not available for inorganic mercury concentrations detected in edible
plants. Thus to further evaluate any edible plant exposure, ATSDR calculates exposure doses.
The exposure doses for eating plants are calculated slightly different from the other media
because a body weight factor is already incorporated into the intake rate. Therefore, ATSDR
calculated exposure doses using the maximum concentration detected in the plants (3.2 ppm;
OREIS 2009) and the following formula:
ED = Conc x IR x AF 29
ED: exposure dose
Conc: concentration
IR: intake rate
AF: bioavailability factor

According to U.S.EPA’s Exposure Factors Handbook people living in the South eat 2.27 grams
of homegrown vegetables per kilogram of body weight per day (g/kg/day) (EPA 1997). The total
survey population used to calculate this intake rate (IR) included adults and children (EPA
1997). As with the past exposure evaluation, ATSDR assumed the oral bioavailability factors
(AF) of inorganic mercury in produce are 15 percent for children and 10 percent for adults (see
Appendix G. Past Exposure Pathway Parameters).
The resulting exposure doses are 7.3 × 10-4 mg/kg/day for adults and 1.1 × 10-3 mg/kg/day for
children, above the RfD of 3.0 × 10-4 mg/kg/day for chronic exposure to inorganic mercury.
Mercury exposures through eating vegetables from EFPC gardens were then further evaluated
using a more realistic exposure scenario—average concentrations to calculate the exposure
doses. By using average concentrations, ATSDR can estimate a more probable exposure.
ATSDR used the same equation and assumptions as above but substituted the average mercury
concentration for each species for the maximum concentration (see Table 27 for the estimated
exposure doses). ATSDR then compared these potential exposure doses to actual health effects
levels in the toxicological and epidemiological literature (EPA 2012a).
29

2.27 g/kg/day was converted to 0.00227 kg/kg/day to allow the units to cancel in the formula.
Page | 136

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Table 27. Estimated Inorganic Mercury Exposure Doses from EFPC Vegetable

Consumption

Species

Average
Concentration (ppm)

Estimated Exposure Doses (mg/kg/day)
Adults

Children

Beet (root)

1.3

3.0 × 10-4

4.4 × 10-4

Kale (leaves)

0.80

1.8 × 10-4

2.7 × 10-4

Tomato (fruit)

0.42

9.5 × 10-5

1.4 × 10-4

mg/kg/day: milligrams per kilogram per day
ppm:
parts per million

The RfD for inorganic mercury was “arrived at from an intensive review and workshop
discussions of the entire inorganic mercury data base” (EPA 2012a). It is based on a back
calculation from U.S.EPA’s recommended drinking water equivalent level (DWEL). This level
is based on three studies in which autoimmune effects were observed in rats exposed to doses of
0.226, 0.317, and 0.633 mg/kg/day (Andres 1984; Bernaudin et al. 1981; Druet et al. 1978).
These health effect levels are at least three orders of magnitude higher than the estimated doses
for adults and for children eating vegetables grown in EFPC gardens. Furthermore, plants tend to
store metals such as mercury in a form not readily bioavailable to humans (ATSDR 2001).
ATSDR does not expect that eating beets, kale, or tomatoes grown in the EFPC floodplain would
cause harmful health effects.
Current Oak Ridge Biota
Oak Ridge Vegetables (inorganic mercury)
The OREIS Environmental Database contains only four vegetable samples (three kale samples
and one tomato sample) from the city of Oak Ridge (OREIS 2009). In 1992, samples were
collected from one garden within the city. Mercury was not detected in any of the samples. The
vegetable data, although minimal, indicate that eating garden vegetables grown in the city of Oak
Ridge is not likely to cause harmful health effects.
Current LWBR Biota
LWBR Fish (methylmercury)
In a 1996 health consultation on LWBR, ATSDR evaluated exposures to mercury contamination
in fish from the reservoir 30 (ATSDR 1996b). ATSDR determined that the levels of mercury in
the fish did not present a public health concern. To arrive at this conclusion, ATSDR evaluated
the available data using a worst-case scenario that assumed a 70-kg adult ate one 8-ounce fish
meal containing the maximum concentration of mercury every week for 30 years (ATSDR
1996b).

30

Fish samples were collected prior to the floodplain remediation.
Page | 137

In September 1997, ATSDR conducted an exposure investigation to quantify actual exposures
from eating moderate to large amounts of fish and turtles from LWBR (ATSDR 1998).
Preliminary information about consumption eligibility and willingness to participate was
collected from more than 550 potential participants who
Since 1987, fishing advisories
volunteered information. About 80 percent of the potential
for LWBR have been posted
participants did not eat enough fish from LWBR to be included in
warning people to avoid or limit
the exposure investigation. ATSDR chose to measure blood
their consumption of fish due to
mercury levels from 116 of the participants who during the past
PCB contamination in the
reservoir (ORNL and Jacobs
year reported eating one or more turtle meals; six or more meals of
Engineering Group 1995).
catfish and striped bass; nine or more meals of white, hybrid, or
smallmouth bass; or 18 or more meals of largemouth bass, sauger,
or carp. The participants consisted of 58.6 percent male and 41.4 percent female with an age
range from 6 to 88 years and a mean age of 52.2 years. About 80 percent of the participants ate
fish from LWBR for six or more years and 65 percent ate fish for more than 11 years. The
estimated average daily fish and turtle consumption rate for the participants was 66.5 grams per
day (g/day) (ATSDR 1998).
For the 116 participants, the total mercury levels in blood ranged from nondetectable to 20 jglL.
Eighty-nine persons had nondetectable levels of mercury in their blood (the detection limit was 3
jglL). The median value was below the detection limit and the arithmetic mean of the total
mercury detections was 5.2 jglL. Organic mercury levels in blood ranged from nondetectable to
11 jglL. One hundred and twelve participants (out of 116) had nondetectable levels of organic
mercury in their blood (the detection limit was 3 jglL). The arithmetic mean of the organic
mercury detections was 6 jglL. The ATSDR scientist concluded in the 1998 exposure
investigation that only 1 of 116 participants had an elevated blood mercury level and that the
overall exposure investigation participants’ blood mercury levels were very similar to levels
found in the general population (ATSDR 1998).
In this public health assessment on Y-12 mercury releases, ATSDR further analyzed the
exposure investigation results by comparing the total blood mercury data to the total blood
mercury data from the National Health and Nutrition Examination Survey (NHANES). We
wanted to determine if the 116 exposure investigation participants eating moderate to high
amounts of LWBR fish were exposed to elevated levels of mercury. The CDC’s National Center
for Health Statistics began conducting the NHANES in 1999, to obtain health and nutritional
related data from a nationally representative sample of adults and children in the United States in
two-year cycles. The survey combines interviews and physical examinations and includes the
measurement of 219 chemicals in people’s blood or urine. The Fourth National Report on
Human Exposure to Environmental Chemicals 2009 and the Updated Tables, February 2011
(CDC 2011) provide the most comprehensive assessment of nationally-representative
biomonitoring data of environmental chemical exposure in the U.S. population. The report and
tables are available at CDC's website http://www.cdc.gov/exposurereport/. The NHANES
biomonitoring studies provide physicians and public health officials with reference ranges that
can be used to determine whether people have been exposed to higher levels of mercury than are
found in the general population (CDC 2009). The 2011 Updated Tables presents the 95th
percentile of total blood mercury data and 95 percent confidence interval for the U.S. population
from the 2003–2004, 2005–2006, and 2007–2008 NHANES survey periods (CDC 2011). Based
on the total blood mercury data from the NHANES, except for the one elevated exposure
investigation blood mercury level of 20 μg/L, the distribution of total blood mercury from the
Page | 138

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

1998 exposure investigation of moderate to high consumers of LWBR fish is similar to the
distribution of total blood mercury for the U.S. population.
In 1996, TDEC conducted a screening study to determine the mercury levels in turtles from the
LWBR and the Clinch River (TDEC 1997). Muscle tissue from 13 common snapping turtles was
analyzed for mercury content. Mercury concentrations ranged from 0.1 to 0.35 ppm, with an
average of 0.19 ppm 31 (TDEC 1997). These levels are slightly above the comparison value of
0.14 ppm for fish. TDEC noted, however, that the mercury concentrations were below FDA’s
action level of 1 ppm for methylmercury in fish.
In 2005, DOE collected three common snapping turtles from Brashear Island (CRM 11,
downstream of Poplar Creek) to monitor mercury levels. Composited mercury concentrations
were “relatively high” in both muscle (0.465 ppm) and liver tissue (3.341 ppm), and much lower
in fat (0.048 ppm). The 2005 samples were similar to, or slightly less than those collected from
the same locations in 2000 (SAIC 2007).
The OREIS Environmental Database contains over 387 samples from channel catfish,
unspecified catfish species, largemouth bass, striped bass, gizzard shad, bluegill sunfish,
unidentified sunfish species, and red-eared sliders 32 collected every year from 1992 to 2009,
from 14 locations in the LWBR (OREIS 2009). As shown in Table 28, many of the maximum
detected concentrations exceeded the comparison value of 0.14 ppm for fish samples. Of the 387
fish samples collected from the LWBR, 214 samples (55 percent) were detected above the
comparison value (OREIS 2009).

31
32

These data do not appear to be included in the OREIS database.
Note that the red-eared slider is not one of three species that are legal to harvest: common snapping, midland
smooth softshell, and Eastern spiny softshell (TDEC 1997). That anyone is eating this particular turtle species is
unlikely. But with no other turtle sampling data available, ATSDR used red-eared sliders as a representative
species.
Page | 139

Table 28. Mercury Concentrations in Fish and Turtles from LWBR

Species

Portion

Minimum
(ppm)

Maximum
(ppm)

Average
(ppm)

Detection
Frequency

Channel Catfish

Fillet/muscle

ND

0.48

0.19

40/41

Channel Catfish

Unknown

ND

1.1

0.28

33/39

Channel Catfish

Whole body

0.10

0.58

0.32

8/8

Catfish, Unspecified Species

Fillet

0.05

0.51

0.19

16/16

Catfish, Unspecified Species

Unknown

0.053

0.36

0.17

4/4

Gizzard Shad

Whole body

0.047

0.054

0.051

3/3

Largemouth Bass

Fillet/muscle

ND

0.78

0.33

39/40

Largemouth Bass

Unknown

ND

0.77

0.27

46/54

Largemouth Bass

Whole body

0.13

0.4

0.3

6/6

Striped Bass

Fillet

0.14

0.52

0.29

4/4

Striped Bass

Unknown

0.093

0.14

0.11

2/2

Striped Bass

Whole body

0.13

0.54

0.28

7/7

Bluegill Sunfish

Unknown

ND

0.45

0.087

33/52

Bluegill Sunfish

Muscle

0.069

0.24

0.12

35/35

Sunfish species

Fillet

ND

0.53

0.14

58/59

Sunfish species

Unknown

0.069

0.16

0.11

4/4

Red-eared Slider (turtle)

Muscle

0.058

0.40

0.26

6/6

Red-eared Slider (turtle)

Whole body

0.061

1.07

0.55

7/7

ND

1.1

—

351/387

Overall
Source: OREIS 2009
ppm:
parts per million

All of the fish and turtles samples from LWBR were analyzed for total mercury (OREIS 2009).
In fish tissue, about 85 percent of mercury is methylmercury (Jones and Slotten 1996). Again,
methylmercury is the organic form and is much more harmful than the elemental and inorganic
forms (ATSDR 1999). To remain conservative, ATSDR assumed that all the total mercury
detected in the fish and turtles was methylmercury.
Because the comparison value was exceeded, ATSDR continued to evaluate exposures to eating
fish and turtles from the LWBR. People frequently fish in the reservoir. But since 1987, fishing
advisories have warned people to avoid or limit their consumption of fish due to PCB
contamination in the reservoir (ORNL and Jacobs Engineering Group 1995). To evaluate
exposure to mercury through eating fish and turtles from the reservoir, ATSDR calculated

Page | 140

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

exposure doses using the average concentration detected in fillet and muscle samples 33 and the
formula described in Section III.C.3. Comparing Estimated Doses to Health Guidelines. ATSDR
evaluated three potential intake rates. The first scenario assumed that both adults and children ate
one 8-ounce fish meal each month (12 meals/year = 7.5 grams/day). The second assumed that
both adults and children at one 8-ounce fish meal each week (52 meals/year = 32 grams/day).
The third assumed adults ate 66.5 grams of fish/day (about two 8-ounce fish meals each week),
which is the self-estimated consumption based on frequency and meal size for moderate to high
consumers of LWBR fish (ATSDR 1998). Turtle consumption is not well documented. For the
sake of this evaluation, ATSDR assumed the same consumption rates applied to turtles as to fish,
although this likely overestimates actual turtle consumption. ATSDR assumed that adults
weighed 70 kg and were exposed for 30 years, and children weighed 28.1 kg and were exposed
for 6 years.
The estimated adult and child doses from eating LWBR fish and turtles once a month were
below both the U.S.EPA RfD (1.0 × 10-4 mg/kg/day) and the ATSDR MRL (3.0 × 10-4
mg/kg/day) for methylmercury (see Table 29). All of the child and some of the adult estimated
exposure doses from eating fish and turtles for the second and third consumption scenarios (one
8-ounce fish meal each week and two 8-ounce fish meals each week) were above both the
ATSDR MRL and U.S.EPA RfD (see Table 29). Therefore, ATSDR compared these potential
exposure doses to actual health effects levels in the toxicological and epidemiological literature.
Table 29. Estimated Methylmercury Exposure Doses for LWBR Fish and Turtles

Species

Average
Concentration
(ppm)

Estimated Exposure Doses (mg/kg/day)
Moderate to
Eating fish once
Eating fish once
high
a month
a week
consumption
(7.5 g/day)
(32 g/day)
(66.5 g/day)
adult
child
adult
child
adult

Channel Catfish (fillet/muscle)

0.19

2.0 × 10-5

5.1 × 10-5

8.7 × 10-5

2.2 × 10-4

1.8 × 10-4

Catfish, Unspecified Species (fillet)

0.19

2.0 × 10-5

5.1 × 10-5

8.7 × 10-5

2.2 × 10-4

1.8 × 10-4

Largemouth Bass (fillet/muscle)

0.33

3.5 × 10-5

8.8 × 10-5

1.5 × 10-4

3.8 × 10-4

3.1 × 10-4

Striped Bass (fillet)

0.29

3.1 × 10-5

7.7 × 10-5

1.3 × 10-4

3.3 × 10-4

2.8 × 10-4

Bluegill Sunfish (muscle)

0.12

1.3 × 10-5

3.2 × 10-5

5.5 × 10-5

1.4 × 10-4

1.1 × 10-4

Sunfish species (fillet)

0.14

1.5 × 10-5

3.7 × 10-5

6.4 × 10-5

1.6 × 10-4

1.3 × 10-4

Red-eared Slider (muscle)

0.26

2.8 × 10-5

6.9 × 10-5

1.2 × 10-4

3.0 × 10-4

2.5 × 10-4

Bold text indicates that the exposure dose is higher than the U.S.EPA RfD of 1.0 × 10-4 mg/kg/day.
g/day:
grams per day
mg/kg/day: milligrams per kilogram per day
ppm:
parts per million

The ATSDR chronic MRL of 3 × 10-4 mg/kg/day for ingestion of organic mercury is based on
the Seychelles Child Development Study, in which people who were exposed to 1.3 × 10-3
mg/kg/day of methylmercury from eating fish did not experience any adverse health effects
33

It is standard protocol to analyze fillets/edible portions when evaluating human health concerns.
Page | 141

(NOAEL; Davidson et al. 1998). The U.S.EPA RfD of 1 × 10-4 mg/kg/day for mercury is based
on the Faroe Islands study, in which maternal dietary intakes of 8 × 10-4 mg/kg/day to 1.5 × 10-3
mg/kg/day were associated with effects associated with performance on standardized
neurobehavioral test involving attention, verbal memory, confrontational naming, and to a lesser
extent visual/spatial abilities and fine-motor functions in children born to women who lived on
the Faroe Islands (LOAELS; Debes et al. 2006; Grandjean et al. 1997). These U.S.EPA
BMDL05 are expected to be associated with a 5 percent increase in the incidence of
neurodevelopmental effects in children exposed in utero. The U.S.EPA RfD is consistent with
the approach used by the NAS which identified a dose of 1.1 × 10-3 mg/kg/day as a dose that
results in a 5 percent increase in the incidence of abnormal scores on the Boston Naming Test (a
picture-naming, vocabulary test) (NRC 2000).
In Table 29, the estimated methylmercury doses for adults and children from eating one meal a
month (12 meals/year) of LWBR fish and turtles are below U.S.EPA RfD of 1 × 10-4 mg/kg/day
and ATSDR's MRL of 3 × 10-4 mg/kg/day and are; therefore, not at levels that would cause
harmful effects in children or fetuses. Figure 27 compares the estimated exposure doses in Table
29 to the health guidelines.
Some of the estimated doses in Table 29 for adults who eat one meal a week (52 meals a year)
and two meals a week (104 meals a year) of LWBR fish and turtles are at levels near or slightly
above the U.S.EPA RfD; however, these estimated doses are not close to the NAS dose effect
level or the EPA BMDL05 (see Figure 27). Pregnant women who eat one and two meals of
largemouth bass, striped bass, or turtles from LWBR a week have a small increased risk of
harming a developing fetus. Possible subtle neurodevelopmental effects identified from studies
of children exposed in utero involve attention, verbal memory, confrontational naming, and to a
lesser extent visual/spatial abilities and fine-motor functions (Debes et al. 2006; Grandjean et al.
1997; NAS 2000). Eating catfish and sunfish once a week is a safer alternative for pregnant
women.
The estimated doses in Table 29 for children eating one meal a week of LWBR fish and turtles
are slightly above the U.S.EPA RfD but are not close to the NAS dose effect level or the EPA
BMDL05 (see Figure 27). Therefore, children who eat up to one LWBR fish meal a week have a
small increased risk of subtle neurodevelopmental effects. Whether children are as sensitive to
the neurotoxic effects of mercury as is the fetus is uncertain. Even if children were not exposed
in utero, some young children who frequently eat the same fish as their mother ate are also at an
increased level of risk for harmful effects.

Page | 142

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Figure 27. Current Estimated Methylmercury Exposure Doses from Eating LWBR Fish

and Turtles Compared to Health Effect Levels and Health Guidelines

Page | 143

V.

Health Outcome Data Evaluation

Health outcome data measures disease occurrence in a population. Common sources of health
outcome data are existing databases (cancer registries, birth defects registries, and death
certificates) that measure morbidity (disease) or mortality (death). Health outcome data can
provide information on a community’s general health status: where, when, and what types of
diseases occur and to whom they occur. Public health officials use health outcome data to look
for unusual patterns or trends in disease occurrence by comparing disease occurrences in
different populations over periods of years. These health outcome data evaluations are
descriptive epidemiologic analyses. They are also exploratory; they provide additional
information about human health effects and are useful in that they help identify the need for
public health intervention activities such as community health education. But health outcome
data cannot—and are not meant to—establish cause-and-effect between environmental exposures
to hazardous materials and adverse health effects in a community.
ATSDR scientists generally consider health outcome data evaluation when they see an
association between 1) a reasonable expectation of adverse health effects and 2) observed levels
of contaminant exposure. In this public health assessment on Y-12 mercury releases, ATSDR
scientists determined that because of past mercury released from the Y-12 plant, potential past
off-site exposures were possible.
Criteria for Conducting a Health Outcome Data Evaluation
To determine whether to use health outcome data in the public health assessment process,
ATSDR scientists consult epidemiologists, toxicologists, environmental scientists, and
community involvement specialists. But ultimately the following criteria, based only on sitespecific exposure considerations, determine whether a public health assessment should include a
health outcome data evaluation.
• Does the site include at least one current (or past) potential or completed exposure pathway?
• Can the period of exposure be determined?
• Can the population that was or is being exposed be quantified?
• Are the estimated exposure doses(s) and the duration(s) of exposure sufficient for a plausible,
reasonable expectation of health effects?
• Are health outcome data available at a geographic level or with enough specificity to be
correlated to the exposed population?
• Do the validated data sources or databases have information on the specific health
outcome(s) or disease(s) of interest—for example, are the outcome(s) or disease(s) likely to
occur from exposure to the site contaminants—and are those data accessible?
Using the findings of the exposure evaluation in this public health assessment, ATSDR identified
the following completed past exposure pathways to Y-12 mercury.
• In the past (1950–1963), family members could have inhaled elemental mercury carried from
the Y-12 plant by workers on their clothes into their homes.

Page | 144

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

• Children ingesting inorganic mercury in EFPC surface water during some weeks in 1956,
1957, and 1958, and adults ingesting inorganic mercury in EFPC surface water during some
weeks in 1958, may have an increased risk of developing renal (kidney) effects.
• Children accidentally swallowing inorganic mercury in EFPC floodplain soils at the NOAA
site and Bruner site before soil removal activities in 1996 and 1997 may have an increased
risk of developing renal (kidney) effects.
• Children born to or nursing from women who periodically ate 12 meals of fish per month
from Poplar Creek in the 1970s, 1980s, and 1990 were exposed to organic mercury at levels
that may have increased the risk of subtle neurodevelopmental effects in these children. Also,
in the 1970s, 1980s, and 1990, children who ate six meals a month of Poplar Creek fish have
an increased risk of subtle neurodevelopmental effects.
ATSDR then used the above criteria to determine whether any of these completed exposure
pathways would support inclusion of health outcome evaluations in this public health
assessment. ATSDR was not able to sufficiently quantify the exposed population or document
the dose and duration of past exposures sufficiently to identify observable health effects for any
of these completed exposure pathways.
In the mid-1990s, ATSDR documented the completed exposure pathway to mercury via
ingestion of fish (ATSDR 1998). ATSDR conducted an exposure investigation to quantify actual
exposures from eating moderate to large amounts of fish and turtles from LWBR. ATSDR’s
exposure investigation determined the body burden or the actual amount of mercury at a specific
time, in the bodies of 116 people who ate moderate to large amounts of fish from the Watts Bar
Reservoir. For the 116 participants, the total mercury levels in blood ranged from nondetectable
to 20 jglL. Eighty-nine persons had nondetectable levels of mercury in their blood (the detection
limit was 3 jglL). The median value was below the detection limit and the arithmetic mean of
the total mercury detections was 5.2 jglL (ATSDR 1998).
In this public health assessment on Y-12 plant mercury releases, ATSDR analyzed the exposure
investigation results by comparing the total blood mercury data to the total blood mercury data
from the NHANES to determine if the 116 exposure investigation participants eating moderate to
high amounts of LWBR fish were exposed to elevated levels of mercury. The CDC’s National
Center for Health Statistics began conducting the NHANES in 1999, to obtain health and
nutritional related data from a nationally representative sample of adults and children in the
United States in two-year cycles. The Updated Tables, February 2011 presents the 95th
percentile of total blood mercury data and 95 percent confidence interval for the U.S. population
from the 2003–2004, 2005–2006, and 2007–2008 NHANES survey periods (CDC 2011). Based
on the total blood mercury data from the NHANES, except for the one elevated blood mercury
level of 20ug/L, the distribution of total blood mercury from the 1998 exposure investigation of
moderate to high consumers of LWBR fish is similar to the distribution of total blood mercury
for the U.S. population. Because the level of mercury exposure via ingestion of moderate to high
amounts of LWBR fish in the mid-1990s is similar to the level expected in the general
population and is not expected to cause measurable health effects, no further analysis of health
outcome data is appropriate for this exposure pathway.

Page | 145

Given the lack of documentation for any of the other completed exposure pathways, no further
analysis of health outcome data is appropriate. Analysis of site-related health outcome data is not
scientifically reasonable unless the level of estimated exposure is adequately documented to meet
the criteria to conduct a health outcome evaluation. ATSDR cannot make such an exposure
estimate. Thus the requirement is complete to consider analysis of site-related health outcome
data on the basis of exposure.
In addition, many validated health outcome databases or data sources on the public generally are
not available. Especially those with data or information on the known specific health effect
(subtle neurodevelopmental effects involving attention, verbal memory, confrontational naming,
and to a lesser extent visual/spatial abilities and fine-motor functions [Debes et al. 2006;
Grandjean et al. 1997; NAS 2000], and renal effects [Andres 1984; Bernaudin et al. 1981; Druet
et al. 1978]) associated with low level environmental exposure to elemental mercury, inorganic
mercury, and organic mercury.

Page | 146

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

VI.

Community Health Concerns

Responding to community health concerns is an essential part of ATSDR’s overall mission and
commitment to public health. ATSDR actively gathers comments and other information from
those who live or work near the ORR. ATSDR is particularly interested in hearing from area
residents, civic leaders, health professionals, and community groups. ATSDR is addressing these
community health concerns in the ORR public health assessments that are related to those
concerns.
To improve the documentation and organization of community health concerns at the ORR,
ATSDR developed a Community Health Concerns Database specifically designed to compile
and track community health concerns related to the site. The database allows ATSDR to record,
track, and respond appropriately to all community concerns, and also to document ATSDR’s
responses to these concerns.
Since 2001, ATSDR compiled more than 2,500 community health concerns obtained from the
ATSDR/ORRHES community health concerns comment sheets, from written correspondence,
phone calls, newspapers, comments made at public meetings (ORRHES and work group
meetings), and surveys conducted by other agencies and organizations. These concerns were
organized in a consistent and uniform format and imported into the database.
The community health concerns addressed in this public health assessment are those concerns in
the database related to mercury releases from the Y-12 plant. Table 30 contains the actual
comments and ATSDR’s responses, and is organized according to category.
Concerns about cancer
Area residents have also voiced concerns about cancer. 34 Those living in the communities
surrounding the ORR have expressed many concerns to the ORRHES about a perceived increase
in cancer in areas surrounding the ORR. A 1993 TDOH survey of eight counties surrounding the
ORR indicated that cancer was a concern more than twice as much as any other health issue. The
survey also showed that 83 percent of the surveyed population in the surrounding counties
believed examining the actual occurrence of disease among Oak Ridge area residents was very
important.
ORRHES thus requested that ATSDR conduct an assessment of health
”Cancer incidence”
outcome data (cancer incidence) in the eight counties surrounding the
refers to newly
ORR. ATSDR conducted an assessment of cancer incidence using data
diagnosed cases of
cancer reported to
already collected by the Tennessee Cancer Registry (ATSDR 2006c). This
the Tennessee
assessment is a descriptive epidemiologic analysis providing a general
Cancer Registry.
picture of cancer occurrence in each of the eight counties. The
assessment’s purpose was to provide citizens living in the ORR area with information regarding
cancer rates in their county compared with those in the state of Tennessee as a whole. This
evaluation only examines cancer rates at the population level—not at the individual level. It is
not designed to evaluate specific associations between adverse health outcomes and documented
human exposures, and it does not—and cannot—establish cause and effect.

34

Note that the Department of Health and Human Services (DHHS) and IARC have not classified mercury as to its
human carcinogenicity. U.S.EPA has determined that mercury chloride and methylmercury are possible human
carcinogens (ATSDR 1999).
Page | 147

The cancer incidence assessment results were released in 2006. They indicated that when
compared with cancer incidence rates for the state of Tennessee generally, both higher and lower
rates of certain cancers occurred in some of the counties examined. But no consistent cancer
occurrence pattern was identified. The reasons for the increases and decreases of certain cancers
are unknown. ATSDR’s Assessment of Cancer Incidence in Counties Adjacent to Oak Ridge
Reservation is available online at
http://www.atsdr.cdc.gov/HAC/oakridge/phact/cancer_oakridge/index.html.
In addition, over the last 20 years, local, state, and federal health agencies have conducted public
health activities to address and evaluate public health issues and concerns related to chemical and
radioactive substances released from the ORR. For more information, please see the
Compendium of Public Health Activities at
http://www.atsdr.cdc.gov/HAC/oakridge/phact/c_toc.html.

Page | 148

A Subcommittee member is concerned with the loss of
2,025,056 pounds of mercury.

DOE probably knew that mercury was being released
but did not report it.

3

2

Concerned about past mercury releases in the direction
of Oliver Springs.

The concentration of mercury in the air should be
measured, so air samples should be taken also.

How was it shown that mercury was taken up by the part
of the plant above ground?

The concentration of mercury in plants should be
measured.

Mercury sampling

1

Mercury releases

Comment

Page | 149

� ATSDR concludes that elemental mercury carried from the Y-12 plant by workers on their clothes into their
homes could potentially have harmed their families (especially young children) in the past.
� ATSDR cannot conclude whether off-site populations breathing mercury releases in the past from the Y-12
plant for a short time could have been harmed because there are no data from outdoor mercury spills.
� ATSDR concludes that breathing past (1950–1963) air mercury releases from the Y-12 plant is not expected
to have harmed people living in the Wolf Valley area.
� ATSDR cannot conclude whether people living off site near the ORR breathing mercury released to the air
from the Y-12 plant from 1950 through 1963, could have been harmed.

Mercury levels in air have been estimated and measured. ATSDR evaluated the available models and data in
this public health assessment. See Section IV.A.2 for the past evaluation and Section 0 for the current
evaluation.

� ATSDR concludes that eating local produce grown in gardens in the EFPC floodplain or in private gardens
which contain mercury-contaminated soils from the floodplain is not expected to harm people’s health in the
past.
� ATSDR concludes that currently eating beets, kale, or tomatoes grown in the EFPC floodplain is not expected
to harm people’s health.
� ATSDR concludes that currently eating vegetables from Oak Ridge is not expected to harm people’s health.

Plants have been analyzed for mercury. Mercury was detected in above ground portions of the plants. In this
public health assessment, ATSDR evaluated eating plants in the past (see Section IV.A.6) and present (see
Section IV.B.7).

� In 1977, Y-12 personnel prepared a classified report called the 1977 Mercury Inventory Report.
� DOE appointed a Mercury Task Force to investigate what was known about mercury use and releases at the
Y-12 plant. The Mercury Task Force released its report in 1983 (UCCND 1983a, 1983b). The Mercury Task
Force studied the 1977 Mercury Inventory Report and adjusted many of its estimates.
� The Task 2 report also estimated Y-12 mercury releases. Task 2 did not revisit all of the previous inventory
estimates, but it revised the previous estimates of mercury releases to the air and water.

Three major efforts have been made to estimate Y-12 mercury releases to water and air over the years (see
Section III.B for more details). The estimates of mercury inventories and releases to air and water in all three of
these reports focused on the lithium enrichment production years (1953–1963).

ATSDR’s Response

Table 30. Community Health Concerns from the ORR Community Health Concerns Database

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

ATSDR should get a topographical map that shows the
ridges and valleys as well as the burial ground locations
and underground locations where water could have been
contaminated by mercury.

5

Concerns about homogenizing soil samples.

7

The DOE analyses are not valid; they did not take core
samples.

A Subcommittee member asked how Task 2 accounted
for the transfer of mercury from the upper layers to the
lower layers.

Concerns of higher concentrations of mercury missing
when homogenizing soil samples.

Concerns about deep channel and shallow sediment
sampling.

6

Concerned about mercury burial grounds and
underground locations where water could have been
contaminated by mercury.

Concerned about elevated levels of mercury that have
been shown in lab tests.

4

Comment

Page | 150

Within this public health assessment, ATSDR scientists considered that the mixing of soil within each core
sample (i.e., using composite samples) likely diluted the mercury that was concentrated in narrow bands within
the cores. ATSDR accounted for this dilution effect of composite samples by applying an adjusted core sample
value, which provides an estimate of the maximum mercury concentration that may have been detected within
each core sample. For further explanation, see Section IV.A.4.

To prevent unnecessary exposures to workers and the public, ATSDR cautions that the sediments should not be
disturbed, removed, or disposed of without careful review by the interagency working group.

ATSDR specifically evaluated mercury levels in both deep channel and shallow sediment in LWBR in this public
health assessment. ATSDR concludes that coming in contact with mercury in LWBR sediment is not expected to
harm people’s health. All of the near-shore sediment samples and deep-water sediment samples collected from
the LWBR were less than the comparison values. However, a few concentrations of mercury in unspecified depth
sediment samples were higher than the comparison value. To evaluate the exposure to sediment further, ATSDR
calculated exposure doses for adults and children using the maximum concentration detected in LWBR sediment
from unspecified depths. Both the estimated doses were below the health guideline value for chronic exposure.

ATSDR obtained topographical maps of the entire ORR area from the U.S. Geological Survey (USGS). ATSDR
conducted a separate public health assessment devoted solely to evaluating potential exposures to
contaminated off-site groundwater from the ORR (ATSDR 2006b). ATSDR concluded that no human exposures
to contaminated groundwater outside the Y-12 boundary have occurred in the past, are currently occurring, or
are likely to occur in the future. For a complete evaluation of groundwater, please refer to ATSDR’s 2006 Public
Health Assessment: Evaluation of Potential Exposures to Contaminated Off-Site Groundwater from the Oak
Ridge Reservation available at The website has been changed in the text to
http://www.atsdr.cdc.gov/HAC/pha/PHA.asp?docid=1371&pg=0.

This public health assessment reviews and evaluates the level of mercury found in the off-site air, surface water,
soil, sediment, fish, and vegetation. See Section VIII for ATSDR’s conclusions and recommendations.

� ATSDR cannot conclude whether people living near the EFPC floodplain breathing mercury vapors from water
released from the Y-12 plant from 1950 through 1963, could have been harmed.
� ATSDR concludes that air and water mercury releases from the Y-12 plant after 1963, are not expected to
have harmed people living off site near the ORR.
� ATSDR concludes that currently breathing air near EFPC is not expected to harm people’s health.
� ATSDR concludes that currently breathing air near LWBR is not expected to harm people’s health.

ATSDR’s Response

Concerns that the maximum mercury concentration of
Poplar Creek Mile 5 is 20 to 40 times higher than any
other number.

Buyers of new homes near EFPC are unaware of the
possible risk of contamination due to mercury.

Concerned about the mercury pathway for children
playing in the creek near Jefferson Circle, closer to
where East Fork pond was.

11

Concerned about mercury exposure from Y-12.

Mercury was discharged to protect workers but crossed
over the hills and exposed the residents.

10

9

Potential exposure to mercury

8

Comment

ATSDR’s Response

Page | 151

� Children who swallowed water from EFPC for a short time during some weeks in 1956, 1957, and 1958, could
have experienced harmful health effects.
� There was not enough information to determine whether swallowing water from EFPC during 1953, 1954, and
1955 could have harmed children.
� Children who swallowed water from EFPC before 1953, or after the summer of 1958, are not expected to
have experienced harmful health effects.
� Swallowing water from EFPC over a long time period in the past is not expected to have caused harmful
health effects for children.
� Children who played at the NOAA site and Bruner site prior to the soil removal activities in 1996 and 1997,
may have accidentally eaten inorganic mercury in EFPC floodplain soils that could have caused harmful
health effects.
� Accidentally eating methylmercury in EFPC floodplain soils in the past is not expected to have caused harmful
health effects for children playing in the floodplain soil.
� Children who currently swallow surface water while playing in EFPC are not expected to experience harmful
health effects.
� Children who currently contact EFPC sediment while playing are not expected to experience harmful health
effects.

ATSDR specifically evaluated whether children would have been in the past (see Sections IV.A.3 and IV.A.4) or
are currently (see Sections IV.B.3 and IV.B.6) being harmed by playing in EFPC in this public health assessment.

While exposures in the past might have caused harmful health effects, the current levels of mercury found in the
EFPC and LWBR surface water and sediment are not at levels expected to cause harmful health effects.

This entire public health assessment discusses residents’ potential past and current exposure to mercury
released from the Y-12 plant into off-site air, surface water, soil, sediment, fish, and vegetation. See Section VIII
for ATSDR’s conclusions about whether past or current exposures caused harmful health effects.

To answer this concern, ATSDR looked at surface water samples collected from PCM 5.1 and PCM 5.5 (located
downstream of where EFPC flows into Poplar Creek, but before entering the K-25 complex). One hundred
samples were analyzed for mercury in 1993 and 1994. Concentrations ranged from 0.0002 to 0.67 ppb. The
average concentration was 0.14 ppb. The maximum mercury concentration detected in EFPC was 2.8 ppb in
1992. The next highest number detected in EFPC was 0.96 ppb in 2001.Therefore, the maximum mercury
concentration detected at Poplar Creek mile 5 is not 20-40 times higher than any other sample. Further, the
concentrations at PCM 5.1 and PCM 5.5 are well below the comparison value of 2 ppb for surface water, and
therefore, do not warrant a health concern.

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

One person said that as a child he played in Poplar
Creek and the Clinch River. He was never told that
poisons such as toxic mercury were contaminating local
creeks and streams. He said that he is scared and angry
now that he knows that the “beautiful silver was really
mercury and that it was hurting us.” He reports having
multiple health problems including weight gain; edema;
loss of hearing, balance, and vision; rashes; fatigue;
headaches; dizziness; sinus and kidney problems; and
joint and muscle pain.

13

There are sick people who are still alive that were
exposed to mercury.

In 1988, a woman was boating in the Clinch River about
12 miles downstream of the K-25 complex. While
pushing the boat off the bank, her leg sunk into the
sediment. There was a “shiny layer of stuff that looked
something like tarnished silver" up to her knee. Even
after intense scrubbing it took a week for the substance
to finally shed off. It was never determined what the
substance was. But the woman suspects it might have
been mercury released from the ORR. Ever since the
incident, a rash appears unpredictably where the mud
once caked her leg. Doctors cannot explain it, leaving
her to guess what substances still lay claim to her skin.

12

Potential health effects from mercury

Comment

Page | 152

� Children who swallowed water from EFPC for a short time during some weeks in 1956, 1957, and 1958, could
have experienced harmful health effects.
� There was not enough information to determine whether swallowing water from EFPC during 1953, 1954, and
1955 could have harmed children.
� Children who swallowed water from EFPC before 1953, or after the summer of 1958, are not expected to
have experienced harmful health effects.
� Swallowing water from EFPC over a long time period in the past is not expected to have caused harmful
health effects for children.
� Children who played at the NOAA site and Bruner site prior to the soil removal activities in 1996 and 1997,
may have accidentally eaten inorganic mercury in EFPC floodplain soils that could have caused harmful
health effects.
� Accidentally eating methylmercury in EFPC floodplain soils in the past is not expected to have caused harmful
health effects for children playing in the floodplain soil.

ATSDR specifically evaluated childhood exposures to mercury released into EFPC in the past. EFPC drains into
Poplar Creek and the Clinch River after it enters the ORR.

In addition to contact dermatitis resulting from dermal exposure to mercury, metallic mercury can also become a
vapor. Breathing in these vapors might cause fever, fatigue, neuropsychiatric disturbances (e.g., memory loss,
irritability, or depression), increased blood pressure, numbness, and discolored hands and feet. For more
information, see ATSDR’s Mercury and Your Health Web site at http://www.atsdr.cdc.gov/mercury/.

It is very difficult to assess what substance this woman might have been exposed to. Mercury exists in three
main forms—metallic mercury, inorganic mercury, and organic mercury. Metallic mercury is a shiny, silver-white
liquid metal. Most inorganic and organic mercury compounds are powders or crystals. One organic mercury
compound, dimethylmercury, is a colorless liquid. Using her description, metallic mercury could be the substance
the woman was exposed to. Unless the skin is damaged, very little metallic mercury is absorbed through the
skin. Furthermore, metallic mercury is unlikely to adhere to the skin. Dermal exposure to metallic mercury can
cause contact dermatitis. Only long-term dermal exposures have resulted in more serious health effects in
people. For more information, see ATSDR’s Mercury and Your Health Web site at
http://www.atsdr.cdc.gov/mercury/.

ATSDR’s Response

ATSDR's minimal risk level for elemental mercury should
be examined more closely because it is lower that EPA's
reference dose.

Concerned about methylmercury accumulation in the
central nervous system and its possible connection to
Alzheimer's disease.

Concerned that mercury poisoning is being
misdiagnosed, sometimes as Multiple Sclerosis.

Concerned about mercury entering the lymph system.

16

17

18

Page | 153

Mercury can enter the lymphatic system, which may play an important role in the transport of mercury to target
organs (Hansen and Danscher 1995).

� ATSDR’s Toxicological Profile for Mercury at http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24
� ATSDR’s Mercury and Your Health at http://www.atsdr.cdc.gov/mercury/
� ATSDR’s Medical Management Guidelines for Mercury at
http://www.atsdr.cdc.gov/mmg/mmg.asp?id=106&tid=24
� ATSDR’s ToxFAQs for Mercury at http://www.atsdr.cdc.gov/toxfaqs/tf.asp?id=113&tid=24
� U.S.EPA’s Mercury Web site at http://www.epa.gov/mercury/
� CDC’s Emergency Preparedness and Response: Mercury at http://www.bt.cdc.gov/agent/mercury/

The symptoms of MS (numbness, fatigue, blindness, paralysis) are very similar to the symptoms of mercury
poisoning. If you think your symptoms could be caused by exposure to mercury, you should ask your doctor to
test your blood, urine, or hair for mercury (as appropriate). These tests can tell you if you have been in contact
with mercury. But they cannot show the kind of health effects you might experience, or whether you will become
sick. The following are some resources for additional information:

This is a topic that is currently being researched. Some scientists found higher mercury concentrations in brain
regions and blood of some patients with Alzheimer's disease (e.g., Mutter et al. 2007. Mercury and Alzheimer's
disease. Fortschr Neurol Psychiatr. 2007 Sep; 75(9):528-38). The current evidence seems to be suggestive, but
not conclusive.

MRLs undergo a rigorous review process—Health Effects/MRL workgroup reviews within ATSDR’s Division of
Toxicology; expert panel of external peer reviews; and agency-wide MRL workgroup reviews, with participation
from other federal agencies, including U.S.EPA. They are also submitted for public comment before being
finalized. Table 7 provides the health guidelines ATSDR uses for mercury.

� Inhalation is the most typical route of exposure to metallic mercury. The primary target organ from prolonged
exposure to low concentrations is the central nervous system. Exposure to high concentrations can produce
effects in the central nervous system and kidneys.
� Ingestion is the most typical route of exposure to inorganic mercury. The primary target organs are the
kidneys.
� Ingestion of fish is the most typical route of exposure to organic mercury. The primary concern is
developmental effects in offspring.

Is it appropriate to add the ingestion and inhalation
mercury doses together?

Concerned about the additive doses of mercury.

The three types of mercury are evaluated separately because the routes of exposure and health effects are
different for each. Therefore, separate doses are calculated for each. Table 7 provides the health guidelines
ATSDR uses for mercury.

ATSDR’s Response

The Dose Reconstruction may have underestimated the
effects of mercury because it considered the three
species of mercury separately.

15

14

Comment

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

No one measured the mercury in the fish or in the
sediment until 1985, which they then tried to correlate
that measurement to DOE plants in other areas to
estimate the fish mercury content. Nearly everyone who
ate those fish had a higher dose than the minimum risk
level.

Nobody measured the mercury content of fish in the
1960s.

Concerned about the Clinch River containing fish that
are contaminated with mercury.

Concerned that mercury from East Fork Poplar Creek is
being transferred into the fish population.

21

22

23

Many people were constantly eating fish that might have
had high levels of mercury.

During the fish studies did you test for mercury?

20

19

Mercury in fish

Comment

Page | 154

In this PHA, ATSDR specifically evaluated mercury in EFPC fish for the past evaluation (see Section IV.A.5) and
current evaluation (see Section IV.B.7). People should heed the consumption advisories.

In this PHA, ATSDR specifically evaluated mercury in the Clinch River/Poplar Creek fish for the past evaluation
(see Section IV.A.5) and the Clinch River/LWBR fish for the current evaluation (see Section IV.B.7). People
should heed the consumption advisories.

This is true. Fish were first analyzed for mercury in 1970. Earlier attempt to model the average annual mercury
concentrations in fish or exposure doses from eating fish (beginning in 1950) included assumptions that could not
be easily verified and may not be appropriate for making public health decisions. Because of this, ATSDR
believes that the data are not adequate to characterize the mercury concentrations in fish prior to 1970.
Therefore, ATSDR cannot conclude whether eating fish before 1970 could harm people’s health (see Section
IV.A.5).

Fish downstream from the Y-12 plant were first collected and analyzed for mercury in 1970. In this PHA, ATSDR
reviewed mercury concentrations in fish samples collected from 1970 through 2009. See Section IV.A.5 for the
past evaluation and Section IV.B.7 for the current evaluation. People should heed the consumption advisories.

� The participants’ blood mercury levels are similar to the distribution of total blood mercury for the U.S.
population.
� Only one of 116 participants had an elevated total blood mercury level.

Also, ATSDR conducted the Watts Bar Exposure Investigation (ATSDR 1998) to measure actual mercury levels
in the blood of people consuming moderate to large amounts of fish and turtles from the Watts Bar Reservoir,
and to determine whether these people were being exposed to high levels of mercury. A brief summarizing the
exposure investigation is provided in Appendix C. Summary Briefs and Factsheets.

ATSDR specifically evaluated mercury in fish during this PHA. See Section IV.A.5 for the past evaluation and
Section IV.B.7 for the current evaluation. People should heed the consumption advisories. The advisories are
available at http://www.tennessee.gov/environment/wpc/publications/pdf/advisories.pdf.

Yes. Under the Federal Facility Agreement, DOE, U.S.EPA, and TDEC collected fish and sampled them for
mercury during the LWBR RI/FS and the Clinch River/Poplar Creek RI/FS. The OREIS database contains the
results of hundreds of fish samples collected during several studies that were analyzed for mercury.

ATSDR’s Response

Concerned that concentrations of mercury in fish of
upper East Fork Poplar Creek are not decreasing.

Concerned that the concentrations of mercury in fish are
increasing at a greater rate in the fish that are further
downstream in East Fork Poplar Creek.

Did ATSDR come to the conclusion that there was no
danger from eating one fish for anything other than
PCBs when that was all you tested for?

24

25

26

Comment

ATSDR’s Response

Page | 155

ATSDR scientists completed public health assessments on Y-12 plant uranium releases (ATSDR 2004); White
Oak Creek radionuclide releases (ATSDR 2006a); X-10 site iodine 131 releases (ATSDR 2008); X-10 site, Y-12
plant, and K-25 site PCB releases (ATSDR 2009); K-25 site uranium and fluoride releases (ATSDR 2010); Y-12
plant mercury releases (ATSDR 2011); and other issues of community concern, such as contaminant releases
from the TSCA Incinerator (ATSDR 2005a) and contaminated off-site groundwater (ATSDR 2006b).

TDOH conducted the Oak Ridge Health Studies, which included extensive reviews of available information and
qualitative and quantitative analyses of past (1944 to 1990) releases and off-site exposures to hazardous
substances from the entire ORR, including fish from nearby waterways. ATSDR scientists reviewed and
analyzed TDOH’s Oak Ridge Health Studies to identify contaminants that required further public health
evaluation.

During the 2007 Evaluation of Current (1990 to 2003) and Future Chemical Exposures in the Vicinity of the Oak
Ridge Reservation (ATSDR 2007), ATSDR evaluated over 16,000 fish samples that were analyzed for 147
different chemicals. Separate public health assessments were written to evaluate mercury and PCBs in fish.
ATSDR’s public health assessments can be found at http://www.atsdr.cdc.gov/HAC/oakridge/phact/index.html.

To answer this concern, ATSDR looked at redbreast sunfish collected from Station EFK 6.3 (located near I-95
right before EFPC re-enters the ORR). Samples were collected from 1985 to 2007 and analyzed for mercury.
The average concentration across all the years is 0.76 ppm. The maximum concentration (1.72 ppm) was
detected in November 1998. The minimum concentration (0.21 ppm) was detected in November 2004. The
average concentrations for each year sampled range from 0.4 ppm in 1986 to 1.1 ppm in 2001. ATSDR plotted
the data with a trend line. There is an increase (roughly 0.3 ppm) in mercury concentrations from 1985 to 2007.
DOE is monitoring the increase in mercury bioaccumulation, and continuing efforts to identify the cause (see
Bechtel Jacobs 2010, SAIC 2007, and Southworth et al. 2010).

Upper EFPC is located on site within the Y-12 plant. To answer this concern, ATSDR looked at redbreast sunfish
collected from Station EFK 24.2 (located on site within the Y-12 plant complex). Samples were collected in 1991,
1992, and from 1996 to 2008 and analyzed for mercury. The average concentration across all the years is 0.6
ppm. The maximum concentration (1.59 ppm) was detected in June 2000. The minimum concentration (0.07
ppm) was detected that same day. The average concentrations for each year sampled range from 0.54 ppm in
1991 to 0.77 ppm in 1996. ATSDR plotted the data with a trend line. There is a very slight decrease (roughly 0.05
ppm) in mercury concentrations from 1991 to 2008.

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Did the mercury project ever get completed? What
happened?

ORHASP has recognized that mercury speciation is still
a problem but are not going to address it. We must have
independent analysis and research performed by both
minority and majority universities.

A community member would like to see the new study
that states that mercury is less harmful than previously
thought.

27

28

29

EFPC cleanup

Comment

Page | 156

This comment is in reference to raising the proposed EFPC floodplain soil cleanup level from 50 ppm to 400
ppm. The cleanup level was changed because additional testing of the soil (see previous comment) determined
that the type of mercury present is less absorbed into the body and less toxic than the original cleanup level
assumed.

ATSDR convened a science panel meeting on the bioavailability of mercury in soil in August 1995. The purpose
of the science panel was to identify methods and strategies that would enable health assessors to develop datasupported, site-specific estimates of the bioavailability of inorganic mercury and other metals (arsenic and lead)
from soils. The panel consisted of private consultants and academicians internationally known for their metal
bioavailability research along with experts from ATSDR, CDC, U.S.EPA, and the National Institute for
Environmental Health Science. ATSDR used information obtained from the panel meeting to evaluate the EFPC
clean-up level. ATSDR also used the findings to characterize and evaluate soil containing mercury at other waste
sites. Three technical papers and an ATSDR overview paper on the findings of the panel meeting were published
in the International Journal of Risk Analysis in 1997 (Volume 17:5).

In April 1993, DOE released the EFPC RI, which evaluated the extent and level of contamination in the 100-year
EFPC floodplain (SAIC 1994a). In June 1994, DOE released an addendum to the RI, which presented the results
of mercury speciation studies in the EFPC floodplain soil (SAIC 1994c). In the addendum, DOE stated that
several different analytical methods indicated that mercuric sulfide and metallic mercury are likely to be the
dominant inorganic mercury forms present and that mercuric chloride (the most easily absorbed and the most
toxic inorganic form of mercury) is a minor component of the total mercury in the EFPC floodplain soils (SAIC
1994c). Based on these two reports, DOE selected and U.S.EPA and TDEC approved a remedial action to
remove soils containing greater than 400 ppm of mercury (DOE 1995b).

Yes in 1996 and 1997. EFPC floodplain soils with concentrations greater than 400 ppm of mercury were
removed from the floodplain near the NOAA Atmospheric Diffusion Laboratory off Illinois Avenue in 1996, and
from the floodplain near the NOAA site and across the Oak Ridge Turnpike from the Bruner’s Shopping Center
on the Wayne Clark Property in 1997. Close to 35,000 cubic meters (m3) of soil were removed. Confirmatory
samples were taken to ensure that the remediated areas were below the clean-up standard of 400 ppm.
Postremediation monitoring was conducted to ensure the effectiveness of the excavation (SAIC 1994a, 1998,
2002a).

ATSDR’s Response

If mercury still leaches out from East Fork Poplar Creek,
it may be possible that amounts above trace amounts of
mercury are going into the Clinch River.

Sediment disturbances could be causing mercury levels
to rise downstream.

31

32

Concerned about the mercury burial ground on Hampton
Road in Scarboro.

34

The known burial sites and seepage points for any type
of mercury should be documented and pointed out on
maps like the map in the field office, which shows the
extent of mercury contamination in EFPC 100 year
floodplain.

When considering changes in soil and water composition
over time, do ATSDR's public health conclusions apply
to children who lived in the Scarboro community in the
past?

33

Scarboro

Raising the allowable mercury level in residential areas
from 50 ppm to 400 ppm appeared to have been an
example of special interest.

30

Comment

ATSDR’s Response

Page | 157

ATSDR’s 2006 Public Health Assessment: Evaluation of Potential Exposures to Contaminated Off-Site
Groundwater from the Oak Ridge Reservation evaluated the possibility of someone coming in direct contact with
groundwater at seeps or springs in Union Valley (ATSDR 2006b). Since the land overlying the known extent of

There is no evidence of any mercury burial grounds in Scarboro. FAMU conducted the Scarboro Community
Environmental Study in 1998 (FAMU 1998) and U.S.EPA conducted the Scarboro Community Environmental
Sampling Validation Study in 2001 (EPA 2003). Neither study found elevated levels of mercury in Scarboro soil,
sediment, or surface water.

Yes. As part of the public health assessment process, ATSDR evaluates whether people were exposed in the
past, are currently being exposed, or will be exposed in the future. ATSDR is committed to evaluating the special
interests of children at sites such as the ORR.

To answer this concern, ATSDR looked at surface water collected from Station EFK 6.3 (located near I-95 right
before EFPC re-enters the ORR). Forty-four samples were collected from 2000 to 2009 and analyzed for
mercury. The average concentration across all the years is 0.12 ppb. The maximum concentration (1.3 ppb) was
detected in June 2008. The minimum concentration (0.009 ppb) was detected in December 2000. ATSDR plotted
the data with a trend line. There is a slight increase (roughly 0.05 ppb) in mercury concentrations from 2000 to
2009. However, all of these concentrations are well below the comparison value of 2 ppb for surface water, and
therefore, do not warrant a health concern.

It is possible that levels of mercury above trace amounts are reaching the Clinch River. However, the mercury
levels are so low they would not be a health concern. ATSDR evaluated 647 surface water samples from EFPC
in this PHA. Samples were collected in 1991–1994, 1996, 1997, and 1999–2009 from 25 different locations in the
creek. Mercury was only detected in 126 samples (about 1 out of 5 samples). Only one mercury concentration
(about 0.1 percent) was detected slightly above the comparison value of 2 ppb for surface water in 1992 (prior to
the 1996 and 1997 cleanup). This indicates that the vast majority of the concentrations were detected at levels
not warranting health concern.

ATSDR’s science panel meeting on the bioavailability of mercury in soil consisted of private consultants and
academicians internationally known for their metal bioavailability research along with experts from ATSDR, CDC,
U.S.EPA, and the National Institute for Environmental Health Science.

In response to a request from community members and the city of Oak Ridge, ATSDR evaluated the public
health impact of DOE’s clean-up level of 400 mg/kg of mercury in the EFPC floodplain soil. ATSDR concluded
that the clean-up level of 400 mg/kg of mercury in the soil of the EFPC floodplain would be protective of public
health and pose no health threat to adults or children (ATSDR 1996a). The public health consultation discussing
this concern can be accessed at http://www.atsdr.cdc.gov/hac/pha/pha.asp?docid=1360&pg=0.

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Groundwater flows from the Y-12 plant to Scarboro.

It is generally believed by most people who live in
Tennessee and perhaps the nation that the Scarboro
neighborhood in Oak Ridge, Tennessee is contaminated
with mercury... The data showed very high levels of
mercury contamination in several areas of Oak Ridge;
however, the media primarily focused attention on
mercury contamination in the Scarboro neighborhood
(where no significant mercury was ever found).

35

36

Small streams near Y12 should be sampled for possible
contamination. One example is Mill Branch near the
South Hill Golf Course.

The springs along the north side of Pine Ridge are
contaminated.

The community needs the data from the secret wellmonitoring done since the 1980s. The community needs
the data from the surface and groundwater studies at Y­
12 and this data directly impacts the surrounding
residents.

Y-12 used carcinogenic chemicals and we know that the
surface and the ground water at Oak Ridge interchange.

Comment

Page | 158

� Mercury has not been detected in any surface water samples collected in the Scarboro community.
� All of the surface soil and sediment samples collected in Scarboro were less than the comparison value, and
therefore, not a health hazard.

� Estimated past air concentrations in Scarboro were below the comparison value.
ATSDR specifically evaluated current surface water, soil, and sediment data collected by FAMU and U.S.EPA in
the Scarboro community.

Because of its proximity to the floodplain, Scarboro was identified as a potentially exposed community in the
Task 2 report (ChemRisk 1999a). ATSDR specifically evaluated past exposures to Scarboro residents.

The highest levels of mercury were found in the EFPC floodplain soil. The NOAA and Bruner sites were the only
areas along the floodplain that contained mercury at levels above health concern (see Figure 19). The
contaminated soil was removed in the 1990s.

ATSDR’s 2006 Public Health Assessment: Evaluation of Potential Exposures to Contaminated Off-Site
Groundwater from the Oak Ridge Reservation evaluated this potential exposure scenario (ATSDR 2006b). The
Y-12 plant plume flows east-northeast along strike, paralleling the underlying geology. Current DOE plume
mapping indicates that the plume is entirely in the Maynardville Limestone (part of the Conasauga Group), an
aquifer formation with relatively high hydraulic conductivity. The Scarboro community is located on the Rome
formation that consists of low-conductivity shales and siltstones. It is unlikely that water will migrate from areas
with higher hydraulic conductivity to those with less.

Scarboro is located outside of the EFPC floodplain. As Figure 9 shows, the elevation of Scarboro is higher than
the floodplain. Therefore, the contamination from EFPC could not have reached Scarboro.

Data collected for the EFPC Floodplain and Sewer Line Beltway Remedial Investigation (RI) provided a
comprehensive view of the distribution of mercury in off-site soils (SAIC 1994a). The RI data are consistent with
those collected in the earlier ORAU and TVA studies. The RI sampling data demonstrated that mercury was
present in some soils along the entire length of EFPC. Mercury contamination did not typically extend out very far
from the creek banks and rarely to the elevation of the 100-year floodplain. The greatest deposition of mercury in
the EFPC floodplain was found at the NOAA site and Bruner site. The contamination was removed from these
areas in 1996 and 1997, respectively.

the contaminant plume is zoned as "Industrial District 2", it is unlikely that individuals will come in contact with
springs or seeps in this area. Also, most groundwater surfaces as diffuse discharge directly into Scarboro Creek.
Indeed, groundwater constitutes the baseflow for Scarboro Creek in Union Valley (see Figure 11 in the
Groundwater PHA). So, it is unlikely that individuals will come into direct contact with groundwater in seeps and
springs before dilution with surface water occurs. This public health assessment can be found at
http://www.atsdr.cdc.gov/HAC/pha/PHA.asp?docid=1371&pg=0.

ATSDR’s Response

Scarboro is the most contaminated residential area.

38

Concerned about the health of people in Scarboro.

Concerned about Scarboro community health.

An Oak Ridge resident explained that residents here
need and want to know if health problems in Scarboro
have any link to its location just over a ridge from the
nuclear reservation’s Y-12 plant.

The city should cover the contaminated ditches.

We know the soil is contaminated and want someone to
prove it. (Just tell us the truth.)

Y-12 and the surrounding area are very contaminated.

Uranium, mercury, and PCBs have been detected in
Scarboro.

37

Comment

ATSDR’s Response

Page | 159

� None of the soil, sediment, or surface water samples collected from the Scarboro community contained
chemicals at levels posing a public health hazard.
Also, FAMU (1998) and EPA (2003) are two community specific studies conducted to evaluate contamination in
Scarboro. U.S.EPA concluded that the residents of Scarboro are not currently being exposed to harmful levels of
substances in the soil, sediment, or surface water. Summaries of these studies are provided in Appendix C.
Summary Briefs and Factsheets.

As stated in previous responses to comments, several agencies, including ATSDR, U.S.EPA, FAMU, and DOE,
have assessed environmental contamination in Scarboro and evaluated exposures to Scarboro residents. In
addition to the public health assessments for groundwater, uranium, mercury, and PCBs, ATSDR conducted a
Scarboro-specific public health evaluation during its Public Health Assessment: Evaluation of Current (1990 to
2003) and Future Chemical Exposures in the Vicinity of the Oak Ridge Reservations.

� ATSDR concluded that uranium in Scarboro was not and is not at levels causing harmful health effects in the
2004 Public Health Assessment for Y-12 Uranium Releases (ATSDR 2004).
� In this public health assessment, ATSDR found that mercury levels were either not detected or too low in the
Scarboro community to be of health concern.
� ATSDR found that PCBs in EFPC sediment and associated floodplain soil near the Scarboro region were at
levels too low to affect the most sensitive residents (children playing there on a daily basis) in the 2009 Public
Health Assessment for Polychlorinated Biphenyl (PCB) Releases (ATSDR 2009).
Further, FAMU collected soil and sediment from Scarboro and analyzed 10 percent of the samples for 150
organic and inorganic chemicals in 1998. ATSDR evaluated these data and determined that none of the
chemicals detected (over 100 chemicals were not detected) were at concentrations that would cause harmful
health effects from exposure to the soil or sediment.

ATSDR specifically evaluated exposures to uranium, mercury, and PCBs in the Scarboro community in several
public health assessments. ATSDR’s public health assessments can be found at
http://www.atsdr.cdc.gov/HAC/oakridge/phact/index.html.

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Scarboro children suffer from too much asthma.

40

An Anderson County commissioner and another member

But the numbers are simply far too high to ignore, even if
they are not scientific. These incidents of problems are
far too high to be an accident. Because of Scarboro's
proximity to the reservation and the exposures that we
know exist there, a few of us have been calling for a
symptomatic survey for quite a long time. This reinforces
what we've been saying all along, that there might be a
cluster of problems in Scarboro. This certainly warrants
the need for members of the medical and scientific
community to go into Scarboro to define what type of
pattern truly exists.

The director of a Scarboro day care center for children
ages 2–12 said that “about half of the children here have
upper respiratory problems. It does make me concerned
to have that many.”

The media reported that there were an unusually large
number of children with various illnesses (allergies,
asthma, ear infections, and respiratory problems) in the
Scarboro community.

Vegetables grown in Scarboro are not safe to eat and
changed color.

39

Comment

Page | 160

In January 1999, a team of physicians representing CDC, TDOH, the Oak Ridge medical community, and the
Morehouse School of Medicine, thoroughly reviewed the findings of the physical examinations and the
community survey. Of the 23 children who were examined, 22 had evidence of some form of respiratory illness
(reported during the nurse interview or discovered during the doctor’s examination). Overall, the children
appeared healthy and no problems that needed urgent management were identified. Several children had mild

After a review of the information obtained in the health investigation survey, 36 children, including those identified
in the media report, were invited to receive a physical examination. These examinations were conducted in
November and December 1998 to confirm the results of the community survey, to establish whether children with
respiratory illnesses were getting the medical care they needed, and to determine whether the children reported
in the newspaper to have respiratory medical problems really had these problems.

In September 1998, CDC released the preliminary results of the survey. The asthma rate was 13 percent among
children in Scarboro, compared to national estimates of 7 percent among all children aged 0–18 years and 9
percent among African American children aged 0–18 years. The Scarboro rate was, however, within the range of
rates from 6 to 16 percent reported in similar studies throughout the United States. The wheezing rate among
children in Scarboro was 35 percent, compared to international estimates that range from 1.6 to 36.8 percent.

The Scarboro Community Health Investigation, which included a community health survey and a follow-up
medical evaluation of children less than 18 years of age, was coordinated by TDOH to investigate a reported
excess of respiratory illness among children in the Scarboro community (Johnson et al. 2000). This investigation
was mainly designed to measure the rates of common respiratory illnesses among children who reside in
Scarboro, compare these rates with national rates, and determine if there were any unusual characteristics of
these illnesses. The investigation was not designed to find what caused the illnesses.

� ATSDR concluded that none of the chemicals were detected in vegetables at levels causing harmful health
effects during its Public Health Assessment: Evaluation of Current (1990 to 2003) and Future Chemical
Exposures in the Vicinity of the Oak Ridge Reservations (ATSDR 2007).
� ATSDR concluded that uranium was not and is not causing harmful health effects to Scarboro residents who
ate garden vegetables in the 2004 Public Health Assessment for Y-12 Uranium Releases (ATSDR 2004).
� ATSDR found that eating vegetables grown in EFPC floodplain soil was not expected to harm people’s health
during the 2009 Public Health Assessment for Polychlorinated Biphenyl (PCB) Releases (ATSDR 2009).
� Within this public health assessment, ATSDR concluded that eating vegetables currently grown in the EFPC
floodplain or Oak Ridge is not expected to harm people’s health. ATSDR concluded that eating local produce
grown in gardens in the EFPC floodplain or in private gardens which contain mercury-contaminated soils from
the floodplain would not have harmed people’s health in the past.

ATSDR specifically looked at exposures from eating the edible portion of garden vegetables during several of its
public health assessments. ATSDR’s public health assessments can be found at
http://www.atsdr.cdc.gov/HAC/oakridge/phact/index.html.

ATSDR’s Response

“Of course the number of sick children alarms me, but
we still don't have any answers of why,” said a Scarboro
resident whose four grandchildren are among those
suffering respiratory problems. “We have been
microscoped, dissected, you name it, and we're still
waiting for answers, even though everybody knows there
is something bad going on in Scarboro.”

A Scarboro mother has two daughters and a son with
respiratory conditions. Referring to her son, she said “but
with my son, no matter what we do, he still has breathing
troubles. It's like he can get bad off if the wind just
changes direction, so being so close to the reservation
does concern me. Scarboro is our home, the only place
we've ever lived. We love it here and would hate to
leave. But sometimes, I wonder if I am killing my children
by living here.”

of the government-sponsored panel studying Oak Ridge
area health problems said that an independent
investigation is needed. “I am surprised that the numbers
of ill children the newspaper found is that large, even in a
random sample, I had heard from parents that the
children did have some problems, and some of the
parents suspect it might be caused by being so close to
the reservation, especially since all of the recent studies
have started coming out. And these type of respiratory
problems can play a role in learning disabilities, which is
another large concern of mine. I think that a study is the
only way we can be sure about what we seem to be
seeing in Scarboro. We need experts to come in and
study the population, now, so we can know what we
need to do to help.”

Comment

ATSDR’s Response

Page | 161

Dr. Redd, Chief of the Air Pollution and Respiratory Health Branch of the CDC, explained in the Y-12 Uranium
Releases Video that “We worked for several months with community members to refine the questions that would
be asked in the survey. We conducted the survey, and we followed the survey up with physical examinations of
children who had asthma or had symptoms consistent with asthma. And these examinations were conducted by
Knoxville pediatricians. We then reviewed the results of these examinations, and that review included local
physicians and an allergist from Morehouse University. Then we reported the results of these investigations back
to the community. The results were that we found a higher rate of asthma in the community than the national
average. It wasn’t a vastly elevated rate, and it was a rate that might be found in many communities in the United
States.” The video can be viewed on ATSDR’s Oak Ridge Reservation: Public Health web site at:
http://www.atsdr.cdc.gov/HAC/oakridge/index.html.

A more detailed discussion of the Scarboro Community Health Investigation is provided in Appendix B. Summary
of Other Public Health Activities.

respiratory illnesses at the time of the examination; only one child had findings of an abnormality of the lungs at
the time of the examination. None of the children had wheezing. The examinations did not indicate any unusual
pattern of illness among children in Scarboro. The illnesses that were detected were not more severe than would
be expected and were typical of those that might be found in any community. The findings of examinations
essentially confirmed the results of the community health survey. The results of the review were presented on
January 7, 1999, at a community meeting in Scarboro (Johnson et al. 2000).

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

There is a high rate of cancer deaths in Scarboro.

“There is something very, very wrong with the health of
the people here,” said a pastor in Scarboro. “We need
the authority of your leadership and your authentic
support to help us address what those problems are and
what role the contamination in this community has
played in those problems over the years. Unexplained
health problems ranging from cancer to neurological
disorders appear to plague far too many of the
approximately 650 men, women, and children living
here, less than a mile from the Y-12 plant.”

A recent news article has citizens concerned about the
safety of the Scarboro community.

41

42

43

Comment

Page | 162

Since 1998, the Joint Center for Political and Economic Studies (with the support of DOE’s Oak Ridge
Operations) has worked with the Scarboro community to help residents express their economic, environmental,
health, and social needs. In 1999, the Joint Center for Political and Economic Studies conducted a survey of the
Scarboro community to identify the residents’ environmental and health concerns. The surveyors attempted to
elicit responses from the entire community, but achieved an 82% response rate. Because Scarboro is a small
community, the community assessment provided new information about the area and its residents that would not
be available from sources that evaluate more populated areas, such as the Bureau of the Census. In addition,
the assessment identified Scarboro’s strengths and weaknesses, and illustrated the relative unimportance of
environmental and health issues among residents in comparison to other community concerns. The assessment
showed that environmental and health issues were not a priority among Scarboro residents, as the community
was more concerned about crime, security, children, and economic development. The Joint Center for Political
and Economic Studies recommended an increase in active community involvement in city and community
planning (Friday and Turner 2001).

Since this comment was made in 1997, the pastor’s opinion changed. He said the following during an interview
on the Y-12 Uranium Releases Video “We just feel that this is the place to be. Oak Valley is the place to be.
Scarboro is the place to be. It's healthy. It is safe. It is fun to live here.” The video can be viewed on ATSDR’s
Oak Ridge Reservation: Public Health web site at: http://www.atsdr.cdc.gov/HAC/oakridge/index.html.

� The results of the assessment of cancer incidence indicated both higher and lower rates of certain cancers in
some of the counties examined when compared to cancer incidence rates for the State of Tennessee. Most of
the cancers in the eight-county area occurred at expected levels, and no consistent pattern of cancer
occurrence was identified. The reasons for the increases and decreases of certain cancers are unknown.

ATSDR conducted an assessment of cancer incidence using data already collected by the Tennessee Cancer
Registry. This assessment is a descriptive epidemiologic analysis that provides a general picture of the
occurrence of cancer in each of the eight counties. The purpose of this evaluation was to provide citizens living in
the ORR area with information regarding cancer rates in their county compared to the State of Tennessee.
ATSDR’s ORR Assessment of Cancer Incidence is available online at
http://www.atsdr.cdc.gov/HAC/oakridge/phact/cancer_oakridge/index.html.

ATSDR’s Response

44

The work groups and the subcommittee need to consider
the elements arsenic, cadmium, lead, nickel, mercury,
cobalt and strontium in the PHA.

The public has not been reassured that they have not
been exposed to carcinogenic levels of uranium, fluorine,
nickel, arsenic, mercury, chromium, neptunium,
plutonium, or beryllium.

Other chemicals of concern

Comment

ATSDR’s Response

Page | 163

� Tasks 1 and 2 of the TDOH Oak Ridge Dose Reconstruction Feasibility Study (Phase I) determined that
cobalt-57 was not a contaminant of concern (ChemRisk 1993a).

� Tasks 3 and 4 of the TDOH Oak Ridge Dose Reconstruction Feasibility Study (Phase I) further evaluated
chromium (ChemRisk 1993c).
� Task 7 of the TDOH Oak Ridge Dose Reconstruction (Phase II) concluded that past hexavalent chromium
releases do not warrant a high priority for further evaluation (ChemRisk 1999g).
� ATSDR evaluated chromium during the 2007 Evaluation of Current (1990 to 2003) and Future Chemical
Exposures in the Vicinity of the Oak Ridge Reservation (ATSDR 2007).
Cobalt

� ATSDR evaluated cadmium during the 2007 Evaluation of Current (1990 to 2003) and Future Chemical
Exposures in the Vicinity of the Oak Ridge Reservation (ATSDR 2007).
Chromium

� Tasks 3 and 4 of the TDOH Oak Ridge Dose Reconstruction Feasibility Study (Phase I) further evaluated
beryllium (ChemRisk 1993c).
� Task 7 of the TDOH Oak Ridge Dose Reconstruction (Phase II) concluded that past beryllium releases do not
warrant a high priority for further evaluation (ChemRisk 1999g).
� ATSDR evaluated beryllium during the 2007 Evaluation of Current (1990 to 2003) and Future Chemical
Exposures in the Vicinity of the Oak Ridge Reservation (ATSDR 2007).
Cadmium

� Tasks 3 and 4 of the TDOH Oak Ridge Dose Reconstruction Feasibility Study (Phase I) (ChemRisk 1993c)
and Task 7 of the TDOH Oak Ridge Dose Reconstruction (Phase II) (ChemRisk 1999g) further evaluated
arsenic.
� ATSDR evaluated arsenic during the 2007 Evaluation of Current (1990 to 2003) and Future Chemical
Exposures in the Vicinity of the Oak Ridge Reservation (ATSDR 2007).
Beryllium

Arsenic

Both TDOH and ATSDR have conducted public health evaluations of these chemicals in numerous reports.
ATSDR’s public health assessments can be found at http://www.atsdr.cdc.gov/HAC/oakridge/phact/index.html.
Additionally, ATSDR conducted health consultations on EFPC (ATSDR 1993) and LWBR (ATSDR 1996b), which
evaluated current exposures to all of these chemicals. TDOH’s Oak Ridge Health Studies can be found at
http://health.state.tn.us/CEDS/OakRidge/ORidge.html. Section II.H.2 of this public health assessment provides a
summary of TDOH’s Oak Ridge Health Studies.

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Comment

Page | 164

� Tasks 3 and 4 of the TDOH Oak Ridge Dose Reconstruction Feasibility Study (Phase I) further evaluated
nickel (ChemRisk 1993c).
� Task 7 of the TDOH Oak Ridge Dose Reconstruction (Phase II) concluded that past nickel releases do not
warrant a high priority for further evaluation (ChemRisk 1999g).
� ATSDR evaluated nickel during the 2007 Evaluation of Current (1990 to 2003) and Future Chemical
Exposures in the Vicinity of the Oak Ridge Reservation (ATSDR 2007).

� Tasks 3 and 4 of the TDOH Oak Ridge Dose Reconstruction Feasibility Study (Phase I) further evaluated
neptunium (ChemRisk 1993c).
� Task 7 of the TDOH Oak Ridge Dose Reconstruction (Phase II) concluded that past neptunium-237 releases
do not warrant a high priority for further evaluation (ChemRisk 1999g).
Nickel

� Tasks 3 and 4 of the TDOH Oak Ridge Dose Reconstruction Feasibility Study (Phase I) identified mercury as
one of the highest priority contaminants for further study (ChemRisk 1993c).
� Task 2 of the TDOH Oak Ridge Dose Reconstruction (Phase II) specifically addresses mercury releases from
lithium enrichment at the Y-12 plant (ChemRisk 1999a).
� ATSDR specifically addressed exposures to mercury in this public health assessment.
Neptunium

� Tasks 3 and 4 of the TDOH Oak Ridge Dose Reconstruction Feasibility Study (Phase I) further evaluated lead
(ChemRisk 1993c).
� Task 7 of the TDOH Oak Ridge Dose Reconstruction (Phase II) concluded that further evaluation of blood
lead concentrations may not be warranted (ChemRisk 1999g).
� ATSDR evaluated lead during the 2007 Evaluation of Current (1990 to 2003) and Future Chemical Exposures
in the Vicinity of the Oak Ridge Reservation (ATSDR 2007).
Mercury

� Tasks 1 and 2 of the TDOH Oak Ridge Dose Reconstruction Feasibility Study (Phase I) determined that
fluorine was not a contaminant of concern (ChemRisk 1993a).
� ATSDR released the K-25 and S-50 Uranium Fluoride Releases Public Health Assessment to address
releases of fluorine in 2010 (ATSDR 2010).
Lead

� ATSDR evaluated cobalt during the 2007 Evaluation of Current (1990 to 2003) and Future Chemical
Exposures in the Vicinity of the Oak Ridge Reservation (ATSDR 2007).
Fluorine

ATSDR’s Response

45

ATSDR needs to look at the big picture-many elements
are used.

There could be many contaminants that were released
that DOE did not report.

Comment

ATSDR’s Response

Page | 165

Several agencies, including U.S.EPA, TDOH, TDEC, FAMU, and ATSDR, independently conducted numerous
evaluations for potential contaminants in the environment surrounding ORR. Section II.H and Appendix B.
Summary of Other Public Health Activities in this public health assessment summarize all the public health
activities that have been conducted for the ORR.

� Preliminary investigations conducted during Tasks 3 and 4 of the TDOH Oak Ridge Dose Reconstruction
Feasibility Study (Phase I) did not identify uranium as one of the highest priority contaminants for further study
(ChemRisk 1993c).
� However, after examining the Phase I findings, several ORHASP members and former ORR uranium facility
employees suggested a more detailed investigation of past uranium emissions and potential exposures. As a
result, task 6 of the TDOH Oak Ridge Dose Reconstruction (Phase II) was initiated to specifically address
uranium releases from the ORR (ChemRisk 1999b).
� ATSDR released two public health assessments dealing with exposures to uranium—Y-12 Uranium Releases
(ATSDR 2004) and K-25 and S-50 Uranium Fluoride Releases (ATSDR 2010).

� Tasks 3 and 4 of the TDOH Oak Ridge Dose Reconstruction Feasibility Study (Phase I) further evaluated
strontium-89, -90 (ChemRisk 1993c). It was not identified as a contaminant that warranted further study.
Uranium

� Tasks 3 and 4 of the TDOH Oak Ridge Dose Reconstruction Feasibility Study (Phase I) determined that
plutonium was not a contaminant of concern (ChemRisk 1993c).
Strontium

Plutonium

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Some suspicious cases that have occurred need to be
studied.

The long-term synergistic effects of multiple
combinations are not known.

Concerned about the synergistic effects of multiple
exposures to multiple contaminants.

47

Concerned about mercury testing/screening.

Should be tests available for mercury like there is for
beryllium.

Will ATSDR study people who have been found to have
mercury poisoning?

Residents should be tested for elements in a cuttingedge treatment center that is set up specifically for
affected Oak Ridge residents.

Additional public health activities

46

Comment

Page | 166

� Only 5 of the 116 people tested (4 percent) had PCB levels that were higher than 20 bg/L, which is
considered to be an elevated level of total PCBs. Of the five participants who exceeded 20 bg/L, four had
levels of 20–30 bg/L. Only one participant had a serum PCB level of 103.8 bg/L, which is higher than the
general population distribution.

� The participants’ serum PCB levels and blood mercury levels are very similar to levels found in the general
population.
� Only one of 116 participants had an elevated total blood mercury level.

Using the findings of ATSDR’s 1996 Health Consultation on LWBR (ATSDR 1996b), ATSDR conducted the
Watts Bar Exposure Investigation (ATSDR 1998) to measure actual PCB and mercury levels in people
consuming moderate to large amounts of fish and turtles from the Watts Bar Reservoir, and to determine whether
these people were being exposed to high levels of PCBs and mercury. A brief summarizing the exposure
investigation is provided in Appendix C. Summary Briefs and Factsheets.

The TDOH Oak Ridge Health Studies and ATSDR’s public health assessments on the ORR do not indicate there
is a need for follow-up public health activities such as testing at community health centers.

The interactions of carcinogens are more difficult to quantify at environmental doses because a large study group
(humans or animals) is needed for statistical significance at the lower doses observed from environmental
exposure. In the mid-1970s, under contract to the National Cancer Institute, 12 chemicals were tested in 918
pair-wise tests in over 14,500 rats (Gough 2002). Dose levels were expected to produce tumors in 20 to 80
percent of the exposed animals. The results of that study produced no convincing evidence for synergistic
carcinogen interactions while 20 possible cases of antagonism were observed (Gough 2002). In an animal study,
Takayama et al. (1989) reported that 40 substances tested in combination at 1/50 of their CELs resulted in an
increase in cancer. However, Hasegawa et al. (1994) reported no increase in cancer when dosing animals at
1/100 of the CELs for 10 compounds. It should be noted that typical environmental exposures to chemicals
(noncarcinogens and carcinogens) are more than 1,000 times below laboratory-induced health effect thresholds.

ATSDR has reviewed the scientific literature on chemical interactions. Several animal and human studies
(Berman et al. 1992; Caprino et al. 1983; Drott et al. 1993; Harris et al. 1984) have reported thresholds for
interactions. Studies have shown that exposure to a mixture of chemicals is unlikely to produce adverse health
effects as long as components of that mixture are detected at levels below the NOAEL for individual compounds
(Feron et al. 1995; Seed et al. 1995). Additionally, Jonker et al. (1990) and Groten et al. (1991) demonstrated the
absence of interactions at doses tenfold or more below effect thresholds. In two separate subacute toxicity
studies in rats (Groten et al. 1997; Jonker et al. 1993), adverse effects disappeared altogether as the dose was
decreased to below the threshold level. Other studies have provided evidence that exposure to chemical
mixtures, in which the chemicals were administered at doses near their individual thresholds, can produce
additive toxic effects.

ATSDR’s Response

A credible procedure for identifying mercury poisoning
should be identified and laid out for ORRHES and the
community.

Many people cannot afford the costs of Chelation
therapy.

48

49

Comment

ATSDR’s Response

Page | 167

Chelation therapy is only warranted when someone has a clear case of acute elemental mercury exposure, and
is symptomatic. The decision to chelate should be made by a physician. Chelation therapy becomes less
effective as the time since exposure increases. For more information, see ATSDR’s Medical Management
Guidelines for Mercury at http://www.atsdr.cdc.gov/mmg/mmg.asp?id=106&tid=24.

� ATSDR’s Toxicological Profile for Mercury at http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24
� ATSDR’s Mercury and Your Health at http://www.atsdr.cdc.gov/mercury/
� ATSDR’s Medical Management Guidelines for Mercury at
http://www.atsdr.cdc.gov/mmg/mmg.asp?id=106&tid=24
� ATSDR’s ToxFAQs for Mercury at http://www.atsdr.cdc.gov/toxfaqs/tf.asp?id=113&tid=24
� U.S.EPA’s Mercury Web site at http://www.epa.gov/mercury/
� CDC’s Emergency Preparedness and Response: Mercury at http://www.bt.cdc.gov/agent/mercury/

� If you breathed in metallic mercury vapors, you might have a fever, fatigue, irritated eyes or lungs, chest
tightness, memory loss, being sick to your stomach, increased blood pressure, numbness, discolored hands
and feet, renal damage, and chronic central nervous system effects.
� If you were exposed to high amounts of inorganic mercury, you might be severely sick to your stomach, have
bloody diarrhea, memory loss, increased blood pressure, numbness, discolored hands and feet, and renal
damage.
� If you were exposed to organic mercury, you might have headaches, sight and hearing loss, numbness, loss
of muscle control, and difficulty speaking. Children might experience developmental effects.
The following Web sites provide additional information:

There are many sources of information for identifying whether someone has been exposed to mercury. Doctors
can test how much mercury is in blood or urine. Methylmercury can also be tested in hair. These tests can tell
you if you have been in contact with mercury. But they cannot show the kind of health effects you might
experience, or whether you will become sick.

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

50

� In 1999, the Oak Ridge Reservation Health Effects Subcommittee (ORRHES) was established. The
subcommittee consisted of people who represented diverse interests, expertise, backgrounds, and
communities, as well as liaison members from federal and state agencies. It was created to provide a forum
for communication and collaboration between the citizens and the agencies that are evaluating public health
issues and conducting public health activities at the ORR. To help ensure citizen participation, the meetings of
the subcommittee’s work groups were open to the public and everyone could attend and present their ideas
and opinions.
� From 2001 to 2005, ATSDR maintained a field office in the city of Oak Ridge. The office was opened to
promote collaboration between ATSDR and the communities surrounding the ORR by providing community
members with opportunities to become involved in ATSDR’s public health activities at the ORR
� ATSDR created the Oak Ridge Reservation: Public Health web site at
http://www.atsdr.cdc.gov/HAC/oakridge/index.html.
� ATSDR collected and documented health concerns and issues in the ATSDR Community Health Concerns
Database for the ORR. This database allowed ATSDR to record, tract, and address community concerns
obtained from written correspondence, phone calls, newspapers, comments made at public meetings
(ORRHES and workgroup meetings), and individuals stopping by the ATSDR Oak Ridge Field Office. ATSDR
addressed the community health concerns in the public health assessments.
� ATSDR released all of the public health assessments for public comment, and held several public availability
sessions throughout the ORR area.
� ATSDR held physician and community education programs to address health issue and concerns.
� ATSDR held Epidemiology Workshops to explain the science of epidemiology and to assist community
members develop the skills needed to review and evaluate scientific reports.
� Numerous press releases, fact sheets, two videos, and presentations were made to keep the community
informed of ATSDR’s activities.
A local African American pastor said the following during an interview on the Y-12 Uranium Releases Video “We
have several members in our congregation have worked with all of the surveys that have been done over the last
10 years…and all of that has proven to be quite helpful. We’ve been very much a part of all the committees, the
subcommittees…I am confident that we have played the kind of role in this – in gathering this data and getting it
forward that will prove to be useful.” The video can be viewed on ATSDR’s Oak Ridge Reservation: Public Health
web site at: http://www.atsdr.cdc.gov/HAC/oakridge/index.html.

Pollution, environmental contamination, and
environmental health issues appear to concern fewer
Scarboro residents than other matters. Only 9% of
respondents raised these concerns in response to an
open question regarding concerns about the Scarboro
community. Many of these respondents wanted better
information and communication about environmental
pollution and environmental health issues.

Page | 168

ATSDR has worked closely with members of the ORR community, including African American Scarboro
residents, throughout the entire public health assessment process.

ATSDR’s Response

Scarboro residents and other Afro-Americans do not
participate for fear of retaliation.

Miscellaneous

Comment

ATSDR and ORRHES seem to only look at old data and
studies and put new labels on them, the groups do not
help anyone or do anything new.

EFPC has been identified by TDEC as the most
contaminated creek in Tennessee according to the Oak
Ridger newspaper.

51

52

Comment

ATSDR’s Response

Page | 169

TDEC issued advisories for EFPC because of bacterial contamination in the water, as well as mercury and PCB
contamination in fish tissue. The presence of bacteria in the water affects the public’s ability to safely swim,
wade, and fish in streams and reservoirs. According to TDEC, bacterial sources include failing septic tanks,
collection system failure, failing animal waste systems, or urban runoff. Within the State of Tennessee about 147
river miles are posted due to bacterial contamination. Please see the posted advisories at
http://www.tennessee.gov/environment/wpc/publications/pdf/advisories.pdf. Note that EFPC is not a productive
fishing location.

� Uranium releases from the Y-12 plant (ATSDR 2004)
� Contaminant releases from the Toxic Substances Control Act (TSCA) Incinerator (ATSDR 2005a)
� Off-site groundwater (ATSDR 2006b)
� Radionuclide releases from White Oak Creek (ATSDR 2006a)
� Current and future chemical exposures (ATSDR 2007)
� Iodine 131 releases from the X-10 site (ATSDR 2008)
� PCB releases (ATSDR 2009)
� Uranium and fluoride releases from the K-25 site (ATSDR 2010)
� Mercury releases from the Y-12 plant (ATSDR 2011)
In conducting these PHAs, ATSDR scientists evaluate and analyze the information and findings from previous
studies and investigations to assess the public health implications of past and current exposure. When there is a
data gap, ATSDR may conduct an Exposure Investigation. For the ORR, ATSDR conducted the Watts Bar
Exposure Investigation (ATSDR 1998) to measure actual mercury and PCB levels in people consuming
moderate to large amounts of fish and turtles from the Watts Bar Reservoir, and to determine whether these
people were being exposed to high levels of mercury and PCBs. A brief summarizing the exposure investigation
is provided in Appendix C. Summary Briefs and Factsheets.

To expand on the efforts of TDOH, ATSDR scientists conducted a review and a screening analysis of TDOH’s
Phase I and Phase II screening-level evaluation of past exposure to identify contaminants of concern for further
evaluation. Based on this review, ATSDR scientists have completed public health assessments on the following:

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Any program that is set up should not be associated with
any contractor or else the community will not trust it.

54

Any DOE-controlled study will lack credibility.

It will be difficult to separate workers from community
members because people often fit into both roles. How
do we separate exposures as either off-site or on-site
when the exposure could have came from either place?

53

Comment

Page | 170

ATSDR is a federal public health agency of the U.S. Department of Health and Human Services. It is a separate
agency from U.S.EPA and DOE. As the lead agency within the Public Health Service for implementing the
health-related provisions of CERCLA, ATSDR is charged under the Superfund Act to assess the presence and
nature of health hazards at specific Superfund sites, to help prevent or reduce further exposure and the illnesses
that result from such exposures, and to expand the knowledge base about health effects from exposure to
hazardous substances.

The Comprehensive Epidemiologic Data Resource (CEDR) is a public-use database that contains information
pertinent to health-related studies performed at the ORR and other DOE sites. DOE provides this easily
accessible, public-use repository of data (without personal identifiers) collected during occupational and
environmental health studies of workers at DOE facilities and nearby community residents. This large resource
organizes the electronic files of data and documentation collected during these studies and makes them
accessible on the Internet at https://www.orau.gov/cedr/. Most of CEDR’s large data collection pertains to about
50 epidemiologic studies of workers at various DOE sites. Of particular interest to Tennessee residents is an
additional feature of CEDR that provides searchable text for about 1,800 original government documents (now
declassified) used by the TDOH scientists for the Oak Ridge Dose Reconstruction.

ATSDR’s ORR public health assessments only evaluate off-site exposures to contaminant releases from the
ORR. They do not evaluate any exposures potentially occurring on site at the reservation, including exposures to
workers and other individuals who may contact contaminants while at the ORR.

ATSDR’s Response

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

VII. Child Health Considerations
ATSDR recognizes that infants and children can be more sensitive to environmental exposure
than adults in communities faced with contamination of their water, soil, air, or food. Children
are not small adults; a child’s exposure can differ from an adult’s in many ways. Developing
fetuses, infants, and children have unique vulnerabilities. This sensitivity is a result of 1)
children’s higher probability of exposure to certain media because they crawl on the floor, put
things in their mouths, play closer to the ground, and spend more time outdoors; 2) children’s
shorter height, which means that they can breathe dust, soil, and vapors close to the ground; and
3) children’s generally smaller stature, which means childhood exposure will result in higher
doses of chemical exposure per body weight (i.e., a child drinks more liquid, eats more food, and
breathes more air per unit of body weight than an adult). Very young children and infants are
also more susceptible because their organs are not fully matured. Also, young children have less
ability to avoid hazards because they lack knowledge and depend on adults for decisions. As part
of ATSDR’s Child Health Initiative, ATSDR is committed to evaluating the special interests of
children at sites such as the ORR.
These behaviors can result in longer exposure durations and higher intake rates. Children grow
and develop rapidly in the first few months and years of life. In critical
Methylmercury is the
periods of development before they are born, and in the early months
form of mercury most
after birth, fetuses and children are particularly sensitive to the harmful
commonly associated
effects of metallic mercury and methylmercury on the nervous system
with a risk for
(ATSDR 1999). As with mercury vapors, exposure to methylmercury is
developmental effects.
more dangerous for young children than for adults, because more
methylmercury easily passes into the developing brain of young children and may interfere with
the development process. During critical periods of structural and functional development in
both prenatal and postnatal life, children are especially vulnerable to the toxic effects of mercury
(ATSDR 1999).
Methylmercury eaten or swallowed by a pregnant woman or metallic mercury that enters her
body from breathing contaminated air can also pass into the fetus. Inorganic mercury and
methylmercury can also pass from a mother’s body into breast milk and into a nursing infant.
The amount of mercury in the milk will vary, depending on the degree of exposure and the
amount of mercury that enter the nursing woman’s body. There are significant benefits to breast
feeding, so any concern that a nursing woman may have about mercury levels in her breast milk
should be discussed with her doctor. Methylmercury can also accumulate in an unborn baby’s
blood to a concentration higher than the concentration in the mother (ATSDR 1999).

Methylmercury Exposures in Children
Several human studies have evaluated the neurological effects of methylmercury exposure in
children.
• A long-term human study of children from the Faroe Islands, a small group of islands in the
North Atlantic Ocean affiliated with Denmark, began in 1986 and focused on children born
to women who lived on the islands. This population relies heavily on seafood and whales as a
protein source. The investigators used various tests that monitor child development. They
concluded that at birth, cord blood mercury levels in the mother were associated with
harmful effects in children at age 7 years involving language, attention and memory, and to a

Page | 171

lesser extent visual/spatial and motor functions (Grandjean et al. 1997). Follow-up studies at
age 14 years showed similar findings (Debes et al. 2006).
• In 1978, New Zealand was the site of another human study. It focused on 61 children who
were exposed in utero to high mercury levels that resulted from their mother’s consumption
of four or more fish meals a week. If the authors omitted one outlier, the data showed a
decrease in children’s intelligence quotient (IQ) at age 6 with increasing exposure to
methylmercury as measured by their mother’s hair mercury levels at birth (Crump et al.
1998).
• The third study came from the Republic of Seychelles, where 85 percent of the population
relied on local seafood for protein. Average ocean fish consumption in this population was
12 meals a week (Davidson et al. 1998). The Seychelles study initially did not find harmful
effects in children as they grew older. In one recent publication, the investigators reported
that two of 21 endpoints (one positive and one negative) were associated with prenatal
methylmercury exposure. The authors stated that these outcomes were probably due to
chance and conclude that their data did not support a neurodevelopment risk from prenatal
methylmercury exposure from eating fish (Myers et al. 2003). In another paper, the authors
reported that they found several associations between postnatal methylmercury exposure and
children’s developmental endpoints. However, the investigators concluded that no consistent
pattern of associations emerged to support a causal relationship (Myers et al. 2009).
Past Evaluation (1950–1990)
During the past evaluation, ATSDR specifically addressed childhood sensitivity to mercury in
the air, surface water, soil and sediment, fish, and edible plants.
• Exposure to elemental mercury carried from the Y-12 plant by workers into their homes
could potentially have harmed their families (especially young children) in the past (1950–
1963).
• Air and water mercury releases from the Y-12 plant after 1963, are not expected to have
harmed children living off site near the ORR. But insufficient information is available for
ATSDR to determine whether releases from 1950 through 1963 could have caused harmful
health effects.
• Breathing past (1950–1963) air mercury releases from the Y-12 plant is not expected to have
harmed children living in the Wolf Valley area.
• Children who swallowed water from EFPC for a short time during some weeks in 1956,
1957, and 1958, may have an increased risk of developing renal (kidney) effects from
exposure to inorganic mercury.
• Children who swallowed water containing mercury from EFPC before 1953, or after the
summer of 1958, are not expected to have experienced harmful health effects.
• Children who swallowed water from EFPC over a long time period in the past are not
expected to have experienced harmful health effects from mercury exposure.
• Children who played at the NOAA site and Bruner site prior to the soil removal activities in
1996 and 1997, may have accidentally swallowed inorganic mercury in EFPC floodplain
soils that may have increased the risk of developing renal (kidney) effects.
Page | 172

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

• Accidentally swallowing methylmercury in EFPC floodplain soils in the past is not expected
to have caused harmful health effects for children playing in the floodplain soil.
• Children who periodically ate fish from EFPC (up to four meals from EFPC per year) and
children born to or nursing from women who ate EFPC fish in the 1980s are not expected to
have experienced harmful health effects.
• Children born to or nursing from women who ate 12 meals a month (3 meals a week) of fish
from Poplar Creek in the 1970s, 1980s, and 1990 have an increased risk of subtle
neurodevelopmental effects.
• Children who ate six meals a month of fish from Poplar Creek in the 1970s, 1980s, and 1990
have an increased risk of subtle neurodevelopmental effects.
• Children born to or nursing from women who ate three meals a month (average consumption
rate) of Poplar Creek fish in the 1970s, 1980s, and 1990 had a small increased risk of subtle
neurodevelopmental effects.
• Children who ate about 1.5 Poplar Creek fish meals a month in the 1970s, 1980s, and 1990
have a small increased risk of neurodevelopmental effects.
• Children born to or nursing from women who ate 12 meals a month (3 meals a week) of fish
from Clinch River in the 1970s, 1980s, and 1990 have a small increased risk of developing
subtle neurodevelopmental effects.
• Children who ate six meals a month of fish from Clinch River in the 1970s, 1980s, and 1990
have a small increased risk of subtle neurodevelopmental effects.
• Children born to or nursing from women who ate up to three Clinch River fish meals per
month are not expected to have been harmed.
• Children who ate less than two Clinch River fish meals a month are not at risk of harmful
neurodevelopmental effects.
• Children born to or nursing from women who ate 20 meals a month (5 meals a week) from
Watts Bar Reservoir in the 1980s and 1990 have a small increased risk of developing subtle
neurodevelopmental effects.
• Children who ate 10 meals a month of fish from Watts Bar Reservoir in the 1980s and 1990
have a small increased risk of subtle neurodevelopmental effects.
• Children born to or nursing from women who ate up to five Watts Bar Reservoir fish meals
per month are not expected to have been harmed.
• Children who ate less than three Watts Bar Reservoir fish meals a month are not at risk of
harmful neurodevelopmental effects.
• Eating produce grown in the city of Oak Ridge and the EFPC floodplain in private gardens
that contain mercury-contaminated soils is not expected to have harmed people’s health.
Insufficient information is available to determine whether
• Children who swallowed water containing mercury from EFPC during 1953, 1954, and 1955
could have been harmed.
Page | 173

• Children who ate fish from EFPC and Watts Bar Reservoir during the 1950s, 1960s, and
1970s could have been harmed by methylmercury.
• Children who ate fish from Poplar Creek and Clinch River during the 1950s and 1960s could
have been harmed by methylmercury.
Current Evaluation (1990–2009)
During the current evaluation, ATSDR specifically addressed childhood sensitivity to mercury
from exposures through breathing the air; incidentally ingesting surface water, soil, and
sediment; and eating fish, crayfish, turtles, and vegetables.
• None of the ambient air samples detected mercury at levels of public health concern for
children, or for fetuses and nursing infants.
• The majority of the surface water samples either did not detect mercury or found mercury
well below levels of health concern for children, fetuses of pregnant women, or infants of
nursing mothers incidentally ingesting (or being exposed to) the surface water.
• Children, who played in the EFPC floodplain at the NOAA and Bruner sites before soil
removal activities in 1996 and 1997, may have accidentally swallowed inorganic mercury in
soil that may have increased the risk of developing renal (kidney) effects. Children who
come in contact with EFPC floodplain soil after cleanup activities are not being harmed from
exposure to mercury.
• Incidentally ingesting mercury in the soil
around the ORR is not expected to cause
harmful health effects for non-pica
children, or for fetuses and nursing
infants.
• Incidentally ingesting mercury in the
sediment around the ORR is not expected
to cause harmful health effects for nonpica children, fetuses, or nursing infants.
• Children born to or nursing from women
who ignore the posted warning signs and
eat one meal of fish caught from EFPC a
month are not at risk of being harmed
from exposure to methylmercury.
However, eating one or more crayfish
meals a month from the EFPC floodplain
increases the risk of subtle
neurodevelopmental effects.

Fish Advisories for Waterways near the ORR
Tennessee River
Catfish, striped bass, and hybrid (striped bass-white bass)
bass should not be eaten due to elevated levels of PCBs.
Children, pregnant women, and nursing mothers should not
consume white bass, sauger, carp, smallmouth buffalo, and
largemouth bass, but other people can safely consume one
meal per month of these species.
Clinch River
Striped bass should not be eaten due to elevated levels of
PCBs. Children, pregnant women, and nursing mothers
should not consume catfish and sauger, but other people can
safely consume one meal per month of these species.
East Fork Poplar Creek
No fish should be eaten due to elevated mercury and PCB
levels. Avoid contact with the water due to bacterial
contamination.
For the advisories, see
http://www.tennessee.gov/environment/wpc/publications/pdf/
advisories.pdf.

• Children who ignore the posted warning signs and eat one meal of EFPC fish a month have a
small increased risk of subtle neurodevelopmental effects. Eating one or more crayfish meals
a month from EFPC increases that risk.
• Eating one or two meals of largemouth bass, striped bass, and turtles a week from LWBR can
cause children, fetuses of pregnant women, and nursing infants to have a small increased risk
Page | 174

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

of subtle neurodevelopmental effects. Children who eat one LWBR fish meal a month are not
at risk of developing harmful effects. Children, pregnant women, and nursing mothers should
heed the fish consumption advisories for LWBR.
• Eating beets, kale, or tomatoes grown in the EFPC floodplain and eating garden vegetables
grown in the city of Oak Ridge are not likely to cause harmful health effects for children,
fetuses, and nursing children.

Pica Children
One additional assessment ATSDR conducts is to evaluate hazards to children displaying pica
behavior (a craving for nonnutritive substances like soil). Information on the incidence of soil
pica behavior is limited. A study described in U.S.EPA’s Exposure Factors Handbook (EPA
1997) showed that the incidence of soil pica behavior was approximately 16 percent among
children from a rural black community in Mississippi. This behavior, however, was described as
a cultural practice among the community surveyed. Thus that community may not represent the
general population. In five other studies, only one child out of more than 600 ingested an amount
of soil significantly greater than the range of other children. Although these studies did not
include data for all populations and represented short-term ingestion only, the assumption
remains that the incidence rate of child pica behavior in the general population is low.
Little information is available on the amount of soil ingested (measured in mg/day) by children
with pica behavior (EPA 1997). Intake rates between 1,000 and 10,000 mg/day have been used
to estimate exposure doses for pica children. In this health assessment, ATSDR assumed a
soil/sediment intake rate of 5,000 mg/day for 52 days per year (once a week) to represent pica
behavior in children aged 1 to 3 years of age (weighing 10 kg). ATSDR considers this a healthprotective assumption that likely overestimates soil/sediment consumption. In the case of pica
behavior, estimated exposure doses were calculated using the maximum surface soil or sediment
concentration detected in an area of likely exposure (see Table 31). ATSDR then compared these
doses to acute health effect levels—this exposure pattern can be episodic and short-term.
Table 31. Estimated Inorganic Mercury Exposure Doses for Pica Children
Location
EFPC
Oak Ridge
Scarboro
LWBR
Sources:
ppm:

Maximum Concentrations (ppm)
Soil
Sediment
3,420
158
0.3
Soil was not sampled.

2,240
35.7
0.12
160

Estimated Exposure Doses (mg/kg/day)
Soil
Sediment
2.4 × 10-1

1.6 × 10-1

1.1 × 10-2

2.5 × 10-3

All concentrations were below the comparison
value of 20 ppm.
Not available
1.1 × 10-2

OREIS 2009; SAIC 1994a
parts per million

All of the estimated exposure doses for potential pica child exposures are below the health effect
levels available in the toxicological and epidemiological literature (the acute MRL is based on a
study in which no renal effects were observed in rats administered 0.93 mg/kg/day once daily for
14 days; NTP 1993). ATSDR does not expect that children exhibiting pica behavior would
experience adverse health effects from exposure to the current levels of mercury in soil/sediment
around the ORR.

Page | 175

VIII. Conclusions and Recommendations
Past Evaluation (1950–1990)

Air (elemental mercury)
ATSDR concludes
• Elemental mercury carried from the Y-12 plant by workers into their homes could potentially
have harmed their families (especially young children) in the past (1950–1963 ), but ATSDR
has no quantitative data to evaluate the magnitude of this hazard.
• Elemental mercury releases into the air from the Y-12 plant after 1963 are not expected to
have harmed people living off site near the ORR. No estimated air mercury concentrations
for any potentially exposed community for any year exceeded ATSDR’s health guideline for
elemental mercury vapor.
• Elemental mercury vaporizing into the air from the water released from the Y-12 plant after
1963 is not expected to have harmed people living off site near the ORR. No estimated air
mercury concentrations exceeded ATSDR’s health guideline for elemental mercury vapor.
• Breathing elemental mercury from past (1950–1963) airborne releases from the Y-12 plant is
not expected to have harmed people living in the Wolf Valley area. The highest annual
concentration was more than 14 times lower than ATSDR’s health guideline for elemental
mercury vapor.
ATSDR cannot conclude
• Whether people living off site near the ORR who breathed airborne releases of elemental
mercury from the Y-12 plant from 1950 through 1963 could have been harmed.
• Whether people living near the EFPC floodplain who breathed elemental mercury vapors
from Y-12 releases to the water from 1950 through 1963 could have been harmed.

Surface Water (inorganic mercury)
ATSDR concludes
• Children who swallowed water from EFPC containing mercury for a short period of time
(acute exposure: less than 2 weeks) during some weeks in 1956, 1957, and 1958 may have an
increased risk of developing renal (kidney) effects. The estimated exposure doses for some
weeks in 1956, 1957, and 1958 were higher than ATSDR’s health guidelines (i.e., MRLs)
and U.S.EPA’s health guideline (i.e., RfD) for inorganic mercury.
• Adults who swallowed water from EFPC containing mercury for a short time during some
weeks in 1958 may have an increased risk of developing renal (kidney) effects. The
estimated exposure doses for some weeks in 1958 were higher than ATSDR’s and
U.S.EPA’s health guidelines for inorganic mercury.
• Swallowing water from EFPC containing mercury for a short time before 1953 or after the
summer of 1958 is not expected to have harmed people’s health. The estimated exposure
doses were lower than ATSDR’s and U.S.EPA’s health guidelines for inorganic mercury.

Page | 176

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

• Intermittently (intermediate exposure: more than 2 weeks and less than 1 year) swallowing
water from EFPC containing inorganic mercury is not expected to have harmed people’s
health during any year. The estimated exposure doses were lower than ATSDR’s and
U.S.EPA’s health guidelines for inorganic mercury.
• Swallowing water from EFPC containing mercury contamination over a long period of time
(chronic exposure: more than 1 year) in the past is not expected to have harmed people’s
health during any year. The estimated exposure doses were lower than ATSDR’s and
U.S.EPA’s health guidelines for inorganic mercury.
• Swallowing water from EFPC containing methylmercury is not expected have harmed
people’s health during any year. The estimated exposure doses were lower than ATSDR’s
and U.S.EPA’s health guidelines for organic mercury.
ATSDR cannot conclude
• Whether swallowing water from EFPC containing mercury for a short time during 1953,
1954, and 1955 could have harmed people’s health.

Soil and Sediment (inorganic mercury)
ATSDR concludes
• Children who played at the NOAA site and Bruner site before soil removal activities in 1996
and 1997 may have accidentally swallowed inorganic mercury in EFPC floodplain soils that
may have increased the risk of developing renal (kidney) effects. The estimated child
exposure doses exceeded ATSDR’s health guidelines for inorganic mercury. Adults are not
expected to have been harmed. The estimated adult exposure doses were below ATSDR’s
health guidelines for inorganic mercury.
• Methylmercury in EFPC floodplain soils in the past is
not expected to have caused harmful health effects
for anyone contacting the floodplain soil. The
estimated exposure doses were below ATSDR’s
health guideline for organic mercury.

Due to other contamination in the fish,
people should heed the fish
consumption advisories. For the
advisories, go to
http://www.tennessee.gov/environmen
t/wpc/publications/pdf/advisories.pdf.

• Adult workers involved in excavation, digging, and
other activities that turn over the EFPC floodplain soil in the undeveloped area of DOE
property are not expected to be harmed from exposure to mercury in the floodplain soil. The
estimated exposure dose was below ATSDR’s acute health guideline for inorganic mercury.

Fish (methylmercury)
ATSDR concludes
• Periodically eating methylmercury-contaminated fish from EFPC (up to nine meals per year
for adults and four meals per year for children) in the 1980s is not expected to have harmed
people’s health, including children who ate fish, nursing infants whose mothers ate fish, and
children born to women who ate fish. The estimated methylmercury exposure doses were
below ATSDR’s and U.S.EPA’s health guidelines.

Page | 177

• Children born to or nursing from women who ate approximately 12 fish meals per month
from Poplar Creek in the 1970s, 1980s, and 1990 have an increased risk of subtle
neurodevelopmental effects from exposure to methylmercury. The estimated methylmercury
exposure doses came close to the NAS health effect level, which is associated with subtle
neurodevelopmental effects.
• Children who ate up to six meals a month of Poplar Creek fish in the 1970s, 1980s, and 1990
have an increased risk of subtle neurodevelopmental effects from exposure to
methylmercury. The estimated methylmercury exposure doses came close to the NAS health
effect level, which is associated with subtle neurodevelopmental effects.
• Children born to or nursing from women who ate approximately three meals a month of
Poplar Creek fish in the 1970s, 1980s, and 1990 have a small increased risk of subtle
neurodevelopmental effects. A few estimated methylmercury exposure doses were only
slightly above ATSDR’s and U.S.EPA’s health guidelines for methylmercury and were not
close to the NAS health effect level.
• Children who ate about 1.5 meals a month of Poplar Creek fish in the 1970s, 1980s, and 1990
have a small increased risk of neurodevelopmental effects. A few estimated methylmercury
exposure doses were only slightly above ATSDR’s and U.S.EPA’s health guidelines for
methylmercury and were not close to the NAS health effect level.
• Children born to or nursing from women who ate 12 fish meals per month (three fish meals a
week) from the Clinch River in the 1970s, 1980s, and 1990 have a small increased risk of
subtle neurodevelopmental effects. The estimated methylmercury exposure doses are only
slightly above ATSDR’s and U.S.EPA’s health guidelines for methylmercury and were not
close to the NAS health effect level.
• Children born to or nursing from women who ate up to three Clinch River fish meals per
month were not harmed from exposure to methylmercury. The estimated exposure doses
were below ATSDR’s and U.S.EPA’s health guidelines.
• Children who ate approximately six fish meals a month from the Clinch River in the 1970s,
1980s, and 1990 have a small increased risk of subtle neurodevelopmental effects. The
estimated methylmercury exposure doses were only slightly above ATSDR’s and U.S.EPA’s
health guidelines for methylmercury and were not close to the NAS health effect level.
• Children who ate less than two Clinch River fish meals a month are not at risk of harmful
neurodevelopmental effects. The estimated exposure doses were below ATSDR’s and
U.S.EPA’s health guidelines.
• Children born to or nursing from women who ate 20 fish meals per month (5 fish meals a
week) from Watts Bar Reservoir in the 1980s and 1990 have a small increased risk of subtle
neurodevelopmental effects. The estimated exposure doses were only slightly above
U.S.EPA’s health guideline and were not close to the NAS health effect level
• Children born to or nursing from women who ate up to five Watts Bar Reservoir fish meals
per month were not harmed from exposure to methylmercury. The estimated exposure doses
were below ATSDR’s and U.S.EPA’s health guidelines.

Page | 178

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

• Children who ate approximately 10 fish meals a month from Watts Bar Reservoir in the
1980s and 1990 have a small increased risk of subtle neurodevelopmental effects. The
estimated exposure doses were only slightly above U.S.EPA’s health guideline and were not
close to the NAS health effect level.
• Children who ate less than three Watts Bar Reservoir fish meals a month are not at risk of
harmful neurodevelopmental effects. The estimated exposure doses were below ATSDR’s
and U.S.EPA’s health guidelines.
ATSDR cannot conclude
• Whether eating fish from EFPC, Poplar Creek, Clinch River, or Watts Bar Reservoir during
the 1950s and 1960s could have harmed people’s health.
• Whether eating fish from EFPC and Watts Bar Reservoir during the 1970s could have
harmed people’s health.

Edible Plants (inorganic mercury)
ATSDR concludes
• Eating local produce grown in gardens in the EFPC floodplain or in private gardens that
contain mercury-contaminated soils from the floodplain would not have harmed people’s
health in the past. The estimated exposure doses for children and adults were below
ATSDR’s health guidelines for inorganic mercury.

Current Evaluation (1990–2009)

Air (elemental mercury)
ATSDR concludes
• Breathing air near EFPC is not expected to harm people’s health. All of the EFPC ambient
air sample elemental mercury results (collected near the areas with the highest level of
contamination during the summer) were less than the comparison value for elemental
mercury in air.
• Breathing air near LWBR is not expected to harm people’s health. Despite a lack of analysis
of LWBR ambient air samples for elemental mercury concentrations, the occurrence of
harmful health effects from exposure to mercury vapor from contaminated soil is not a
concern for the LWBR. The mercury contamination accumulated in the sediments of the
deep river channel; the contamination is buried under cleaner sediment. The near-shore
sediment concentrations in the LWBR are much lower than those found in the EFPC
floodplain.

Surface Water (inorganic mercury)
ATSDR concludes
• Accidentally swallowing surface water from EFPC is not expected to harm people’s health.
Only one EFPC surface water mercury concentration was detected slightly above the
mercury comparison value. To assess the exposure further, ATSDR evaluated two scenarios:
1) a farm family member’s exposure and 2) a child’s exposure if the bacterial advisory to
Page | 179

avoid contact with water is ignored. The calculated inorganic mercury exposure doses for
both scenarios were below the chronic exposure health guideline value.
• Accidentally swallowing surface water from Oak Ridge is not expected to harm people’s
health. Only one concentration of mercury in Oak Ridge surface water was higher than the
comparison value. To evaluate the exposure further, ATSDR calculated inorganic mercury
exposure doses for adults and children using the maximum concentration detected in Oak
Ridge surface water. Both estimated inorganic mercury doses were below the chronic
exposure health guideline value.
• Accidentally swallowing surface water from Scarboro ditches will not harm people’s health.
Mercury has not been detected in any surface water samples collected from the Scarboro
community.
• Accidentally swallowing surface water from LWBR is not expected to harm people’s health.
All of the LWBR surface water samples were less than the mercury comparison value.

Soil (inorganic mercury)
ATSDR concludes
• Floodplain soils with concentrations greater than 400 ppm of mercury were removed in 1996
and 1997. Children who played in the EFPC floodplain at the NOAA and Bruner sites before
soil removal activities, may have incidentally swallowed inorganic mercury in soil that may
have increased the risk of developing renal (kidney) effects. Adults are not expected to have
been harmed. ATSDR evaluated exposure to floodplain soils with up to 400 ppm of
inorganic mercury and determined that this clean-up level is safe. People who come in
contact with EFPC floodplain soil after cleanup activities are not being harmed from
exposure to mercury.
• Coming in contact with mercury in Oak Ridge soil is not expected to harm people’s health.
Some of the concentrations of inorganic mercury in Oak Ridge soil were higher than
ATSDR’s comparison value. To evaluate the exposure further, ATSDR calculated inorganic
mercury exposure doses for adults and children using the maximum concentration detected in
Oak Ridge soil. Both the estimated inorganic mercury doses were well below health effect
levels.
• Coming in contact with mercury in Scarboro soil is not expected to harm people’s health. All
of the surface soil samples collected in Scarboro had mercury concentrations that were less
than ATSDR’s comparison value.
• Coming in contact with mercury in the soil near the LWBR is not expected to harm people’s
health. The soil near LWBR has not been contaminated with mercury from ORR operations.
Mercury from the ORR was released into EFPC and traveled through Poplar Creek and the
Clinch River to the LWBR. The mercury accumulated in LWBR deep river channel
sediments, buried under cleaner sediment. Potential exposure (ingestion, inhalation, and
dermal contact) to mercury concentrations in these subsurface sediments does not pose a
health concern even if these deep channel sediments were removed and used as surface soil
on residential properties. The near-shore sediment mercury concentrations in the LWBR
were much lower than the comparison value for mercury in soil. Despite the absence of soil

Page | 180

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

samples collected from the LWBR, the occurrence of harmful health effects from exposure to
mercury in soil along the LWBR shoreline is not a concern.

Sediment (inorganic mercury)
ATSDR concludes
• Coming in contact with mercury in EFPC sediment is not expected to harm people’s health.
Some of the concentrations of mercury in EFPC sediment were higher than the comparison
value. To assess the exposure further, ATSDR evaluated two scenarios: 1) a farm family
member’s exposure and 2) a child’s exposure if the bacterial advisory to avoid contact with
the water is ignored. The calculated exposure doses for both scenarios were below the health
guideline value for chronic exposure to inorganic mercury.
• Coming in contact with mercury in Oak Ridge sediment is not expected to harm people’s
health. Some of the concentrations of mercury in Oak Ridge sediment were higher than the
comparison value. To evaluate the exposure further, ATSDR calculated exposure doses for
adults and children using the maximum concentration detected in Oak Ridge sediment. Both
the estimated doses were below the health guideline value for chronic exposure to inorganic
mercury.
• Coming in contact with mercury in Scarboro sediment is not expected to harm people’s
health. All of the sediment samples collected in Scarboro had mercury concentrations that
were less than the comparison value.
• Coming in contact with mercury in LWBR sediment is not expected to harm people’s health.
All of the near-shore sediment samples and deep-water sediment samples collected from the
LWBR had mercury concentrations that were less than the comparison values. A few
concentrations of mercury in unspecified depth sediment samples, however, were higher than
the comparison value. To evaluate further the exposure to sediment, ATSDR calculated
inorganic mercury exposure doses for adults and children using the maximum concentration
detected in LWBR sediment from unspecified depths. Both the estimated inorganic mercury
doses were below the health guideline value for chronic exposure.

Biota (methylmercury and inorganic mercury)
ATSDR concludes
• EFPC is not a productive fishing location, and a fish consumption advisory is in place. That
anyone is actually eating fish from EFPC is unlikely. Nevertheless, ATSDR evaluated a
potential exposure scenario and assumed people would ignore the posted advisory. ATSDR
assumed that both adults and children ate one 8-ounce fish meal each month.
o Children born to or nursing from women who eat fish are not at risk of developing
harmful effects. The estimated methylmercury exposure doses for eating fish are at or
below ATSDR’s and U.S.EPA’s health guidelines. However, eating crayfish increases
the risk for children born to or nursing from women who ignore the posted warning signs.
The estimated methylmercury exposure dose for eating crayfish is slightly above the
health guidelines but is not close to the NAS health effect level.

Page | 181

o Children who eat fish have a small

increased risk of subtle

neurodevelopmental effects. The estimated

methylmercury exposure doses for eating

fish are slightly above the U.S.EPA’s

health guideline but are not close to the

NAS health effect level. Eating crayfish

increases that risk. The estimated

methylmercury exposure dose for eating

crayfish comes close to the NAS health

effect level, which is associated with

subtle neurodevelopmental effects.

Fish Advisories for Waterways near the ORR
Tennessee River
Catfish, striped bass, and hybrid (striped bass-white
bass) bass should not be eaten due to elevated levels
of PCBs. Children, pregnant women, and nursing
mothers should not consume white bass, sauger, carp,
smallmouth buffalo, and largemouth bass, but other
people can safely consume one meal per month of
these species.
Clinch River
Striped bass should not be eaten due to elevated
levels of PCBs. Children, pregnant women, and
nursing mothers should not consume catfish and
sauger, but other people can safely consume one
meal per month of these species.

• People frequently fish in LWBR. But since
1987, fishing advisories have warned people
East Fork Poplar Creek
to avoid or limit their consumption of fish due
No fish should be eaten due to elevated mercury and
to PCB contamination in the reservoir.
PCB levels. Avoid contact with the water due to
ATSDR evaluated three potential exposure
bacterial contamination.
scenarios: 1) adults and children eating one
For the advisories, see
fish meal with the average concentration of
http://www.tennessee.gov/environment/wpc/publicatio
mercury each month, 2) adults and children
ns/pdf/advisories.pdf.
eating one fish meal with the average
concentration of mercury each week, and 3) adults eating about two fish meals with the
average concentration of mercury each week.
o Adults and children who eat one LWBR fish meal a month are not at risk of developing
harmful effects. The estimated methylmercury exposure doses are below ATSDR’s and
U.S.EPA’s health guidelines.
o Children who eat fish from LWBR once a week have a small increased risk of subtle
neurodevelopmental effects from methylmercury. The estimated methylmercury exposure
doses are slightly above ATSDR’s and U.S.EPA’s health guidelines but are not close to
the NAS health effect level.
o Children born to or nursing from women who eat one or two meals of largemouth bass
and striped bass, a week have a small increased risk of subtle neurodevelopmental
effects. The estimated methylmercury exposure doses for largemouth bass and striped
bass are slightly above the U.S.EPA’s health guideline but are not close to the NAS
health effect level. Eating catfish or sunfish once a week is a safer alternative.
o Adults and children who eat the edible portion of turtles from LWBR once or twice a
week have a small increased risk of subtle neurodevelopmental effects. The estimated
methylmercury exposure doses are slightly above the U.S.EPA’s health guideline but are
not close to the NAS health effect level.
• Eating beets, kale, or tomatoes grown in the EFPC floodplain is not expected to harm
people’s health. Comparison values are not available for screening concentrations detected in
edible plants. ATSDR thus further evaluated exposure to eating them by calculating
inorganic mercury exposure doses using the average concentrations. The health effect levels
available in the toxicological and epidemiological literature are at least three orders of
Page | 182

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

magnitude higher than the estimated inorganic mercury doses for adults and children eating
vegetables grown in EFPC gardens. And plants tend to store metals such as mercury in a
form that is not readily bioavailable to humans.
• Eating vegetables from Oak Ridge is not expected to harm people’s health. Only four
vegetable samples were collected and analyzed for mercury from one garden within the city
of Oak Ridge. Mercury was not detected in any of the samples.
Recommendations
• DOE should maintain long-term oversight of the elevated mercury-contaminated soil in the
undeveloped area of DOE property at the spot along the EFPC floodplain east of the Horizon
Center and, if the property is transferred to another party, consider remediation of the spot or
deed restrictions.
• To prevent unnecessary exposures to workers and the public, ATSDR cautions that the
LWBR sediments not be disturbed, removed, or disposed of without careful review by the
interagency working group.
• People, particularly children, pregnant women, and nursing mothers, should heed the fish
consumption advisories in waterways near the ORR.

Page | 183

IX.

Public Health Action Plan

The public health action plan for the ORR contains a description of actions taken at the site and
those to be taken at the site following the completion of this public health assessment. The
purpose of the public health action plan is to ensure that this health assessment not only identifies
potential and ongoing public health hazards, but also provides a plan of action designed to
mitigate and prevent adverse human health effects resulting from exposure to harmful substances
in the environment. The following public health actions at the ORR are completed or ongoing:
Completed Actions
• Section II.H contains a summary of public health activities pertaining to Y-12 plant mercury
releases. Several additional public health activities conducted at the ORR by ATSDR,
TDOH, and other agencies are described in Appendix B. Summary of Other Public Health
Activities.
• In 1991, TDOH began a two-phase research project to determine whether environmental
releases from the ORR harmed people who lived nearby. Phase I focused on assessing the
feasibility of doing historical dose reconstruction and identifying contaminants most likely to
have public health effects (e.g., ChemRisk 1993a, 1993c). Phase II efforts included full dose
reconstruction analyses of iodine 131 (ChemRisk 1999e), mercury (ChemRisk 1999a), PCBs
(ChemRisk 1999c), radionuclides (ChemRisk 1999f), and uranium (ChemRisk 1999b), as
well as a more detailed health effects screening analysis for releases of technetium-99,
beryllium compounds, and several other toxic substances (ChemRisk 1999g). Phase II was
completed in January 2000.
• In 2004, ATSDR released the final ORR Public Health Assessment for Y-12 Uranium
Releases (ATSDR 2004). The document is available from
http://www.atsdr.cdc.gov/HAC/oakridge/phact/y12/index.html.
• In 2005, ATSDR released the final ORR Public Health Assessment for the TSCA Incinerator
(ATSDR 2005a). The document is available from
http://www.atsdr.cdc.gov/HAC/oakridge/phact/tsca/index.html.
• In 2006, ATSDR released the final ORR Public Health Assessment for Contaminated Offsite Groundwater Exposures (ATSDR 2006b). The document is available from
http://www.atsdr.cdc.gov/HAC/pha/PHA.asp?docid=1371&pg=0.
• In 2006, ATSDR released the final ORR Public Health Assessment for White Oak Creek
Radionuclide Releases (ATSDR 2006a). The document is available from
http://www.atsdr.cdc.gov/HAC/oakridge/phact/white_oak/index.html.
• In 2007, ATSDR released the final ORR Public Health Assessment for the Evaluation of
Current (1990 to 2003) and Future Chemical Exposures in the Vicinity of the Oak Ridge
Reservation (ATSDR 2007). The document is available from
http://www.atsdr.cdc.gov/HAC/oakridge/phact/screening/index.html.
• In 2008, ATSDR released the final ORR Public Health Assessment for Iodine 131 Releases
(ATSDR 2008). The document is available from
http://www.atsdr.cdc.gov/HAC/oakridge/phact/iodine/index.html.

Page | 184

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

• In 2009, ATSDR released the final ORR Public Health Assessment for Polychlorinated
Biphenyl (PCB) Releases (ATSDR 2009).
• In 2010, ATSDR released the final ORR Public Health Assessment for K-25 and S-50
Uranium Fluoride Releases (ATSDR 2010).
Ongoing Actions
• On public request, ATSDR will evaluate whether providing additional environmental health
education materials would help community members understand this public health
assessment’s findings and implications.

Page | 185

X.

Preparers of Report

Jack Hanley, M.P.H.
Environmental Health Scientist
Division of Community Health Investigations (DCHI)
Agency for Toxic Substances and Disease Registry
William H. Taylor, PhD, DABT
CAPT, US Public Health Service
Formerly of the Agency for Toxic Substances and Disease Registry

Page | 186

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

XI.

References

Andres P. 1984. IgA-IgG disease in the intestine of Brown Norway rats ingesting mercuric
chloride. Clin. Immunol. Immunopathol. 30: 488-494. As cited in US Environmental Protection
Agency. 2012a. Integrated risk information system. Mercuric chloride (HgCl2) (CASRN 7487­
94-7). Available at: http://www.epa.gov/iris/subst/0692.htm. Last accessed 10 January 2012.
[ATSDR 1991] Agency for Toxic Substances and Disease Registry. 1991. Case studies in
environmental medicine: cyanide toxicity. Atlanta: US Department of Health and Human
Services. November 1991.
[ATSDR 1993] Agency for Toxic Substances and Disease Registry. 1993. Health consultation
for US DOE Oak Ridge Reservation: Y-12 Weapons Plant Chemical Releases Into East Fork
Poplar Creek, Oak Ridge, Tennessee. April 5, 1993.
[ATSDR 1996a] Agency for Toxic Substances and Disease Registry. 1996a. Health consultation
for US DOE Oak Ridge Reservation: proposed mercury clean-up level for the East Fork Poplar
Creek floodplain soil, Oak Ridge, Anderson County, Tennessee. Atlanta: US Department of
Health and Human Services. January 1996.
[ATSDR 1996b] Agency for Toxic Substances and Disease Registry. 1996b. Health
Consultation: US DOE Oak Ridge Reservation (Lower Watts Bar Reservoir Operable Unit). Oak
Ridge, Anderson County, Tennessee. US Department of Health and Human Services; Atlanta,
Georgia. February 1996.
[ATSDR 1998] Agency for Toxic Substances and Disease Registry. 1998. Exposure
Investigation for US DOE Oak Ridge Reservation: Serum PCB and Blood Mercury Levels in
Consumers of Fish and Turtles from Watts Bar Reservoir. Oak Ridge, Anderson County,
Tennessee. US Department of Health and Human Services; Atlanta, Georgia. March 5, 1998.
[ATSDR 1999] Agency for Toxic Substances and Disease Registry. 1999. Toxicological profile
for mercury. US Department of Health and Human Services; Atlanta, Georgia. March 1999.
Available at: http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24 . Last accessed 11
January 2012.
[ATSDR et al. 2000] Agency for Toxic Substances and Disease Registry (ATSDR), National
Center for Environmental Health (NCEH), National Institute for Occupational Safety and Health
(NIOSH), Tennessee Department of Health (TDOH), Tennessee Department of Environment and
Conservation (TDEC), US Department of Energy (DOE). 2000. Compendium of public health
activities at the US Department of Energy. Atlanta: US Department of Health and Human
Services. Available at: http://www.atsdr.cdc.gov/HAC/oakridge/phact/c_toc.html. Last accessed
11 January 2012.
[ATSDR 2001] Agency for Toxic Substances and Disease Registry. 2001. Summary report for
the ATSDR expert panel meeting on tribal exposures to environmental contaminants in plants.
Division of Health Assessment and Consultation, Office of Tribal Affairs; Atlanta, Georgia.
March 23, 2001.
[ATSDR 2003] Agency for Toxic Substances and Disease Registry. 2003. Comments by
technical reviewers on the Oak Ridge dose reconstruction - task 2 report, volume 2: mercury
releases from lithium enrichment at the Oak Ridge Y-12 plant - a reconstruction of historical

Page | 187

releases and off-site doses and health risks. Atlanta: US Department of Health and Human
Services. July 2003.
[ATSDR 2004] Agency for Toxic Substances and Disease Registry. 2004. Public health
assessment: Y-12 uranium releases, Oak Ridge Reservation (USDOE), Oak Ridge, Anderson
County, Tennessee. US Department of Health and Human Services; Atlanta, Georgia. January
2004. Available at: http://www.atsdr.cdc.gov/HAC/PHA/oakridgey12/oak_toc.html. Last
accessed 29 December 2011.
[ATSDR 2005a] Agency for Toxic Substances and Disease Registry. 2005a. Public health
assessment: TSCA incinerator, U.S. Department of Energy Oak Ridge Reservation, Oak Ridge,
Anderson County, Tennessee. US Department of Health and Human Services: Atlanta, Georgia.
December 2005. Available at http://www.atsdr.cdc.gov/hac/pha/pha.asp?docid=1377&pg=0.
Last accessed 29 December 2011.
[ATSDR 2005b] Agency for Toxic Substances and Disease Registry. 2005b. ATSDR Public
Health Assessment Guidance Manual. US Department of Health and Human Services; Atlanta,
Georgia. January 2005. Available at: http://www.atsdr.cdc.gov/HAC/PHAManual/toc.html. Last
accessed 11 January 2012.
[ATSDR 2006a] Agency for Toxic Substances and Disease Registry. 2006a. Public health
assessment: White Oak Creek radionuclide releases, Oak Ridge Reservation (USDOE), Oak
Ridge, Roane County, Tennessee. US Department of Health and Human Services: Atlanta,
Georgia. August 2006. Available at: http://www.atsdr.cdc.gov/hac/PHA/OakRidge0806­
TN/whiteoakcreek/woc_toc.html. Last accessed 29 December 2011.
[ATSDR 2006b] Agency for Toxic Substances and Disease Registry. 2006b. Public health
assessment: Evaluation of Potential Exposures to Contaminated Off-Site Groundwater from the
Oak Ridge Reservation (USDOE). US Department of Health and Human Services; Atlanta,
Georgia. July 2006. Available at: http://www.atsdr.cdc.gov/hac/pha/pha.asp?docid=1371&pg=0.
Last accessed 29 December 2011.
[ATSDR 2006c] Agency for Toxic Substances and Disease Registry. 2006c. Health consultation:
assessment of cancer incidence in counties adjacent to Oak Ridge Reservation, U.S. Department
of Energy, Oak Ridge, Anderson County, Tennessee. Atlanta: US Department of Health and
Human Services. October 2006. Available at:
http://www.atsdr.cdc.gov/hac/oakridge/phact/cancer_oakridge/index.html. Last accessed 10
January 2012.
[ATSDR 2006d] Agency for Toxic Substances and Disease Registry. 2006d. Public health
statement for cyanide. Atlanta: US Department of Health and Human Services. July 2006.
Available at: http://www.atsdr.cdc.gov/phs/phs.asp?id=70&tid=19. Last accessed 11 January
2012.
[ATSDR 2007] Agency for Toxic Substances and Disease Registry. 2007. Public health
assessment: Evaluation of current (1990 to 2003) and future chemical exposures in the vicinity of
the Oak Ridge Reservation, U.S. Department of Energy Oak Ridge Reservation, Oak Ridge,
Anderson County, Tennessee. US Department of Health and Human Services: Atlanta, Georgia.
January 2007. Available at: http://www.atsdr.cdc.gov/HAC/PHA/oakridge013107­
TN/ceo_toc.html. Last accessed 29 December 2011.

Page | 188

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

[ATSDR 2008] Agency for Toxic Substances and Disease Registry. 2008. Public health
assessment: Iodine-131 releases, Oak Ridge Reservation (USDOE), Oak Ridge, Anderson
County, Tennessee. US Department of Health and Human Services: Atlanta, Georgia. March
2008. Available at: http://www.atsdr.cdc.gov/hac/oakridge/phact/iodine/index.html. Last
accessed 29 December 2011.
[ATSDR 2009] Agency for Toxic Substances and Disease Registry. 2009. Public Health
Assessment: Polychlorinated Biphenyl (PCB) Releases: Oak Ridge Reservation (USDOE). US
Department of Health and Human Services; Atlanta, Georgia. March 2009.
[ATSDR 2010] Agency for Toxic Substances and Disease Registry. 2010. Public health
assessment: K-25 and S-50 uranium and fluoride releases, Oak Ridge Reservation (USDOE),
Oak Ridge, Roane County, Tennessee. US Department of Health and Human Services: Atlanta,
Georgia. September 2010.
Barnett MO, Owens JG, Lindberg SE, Turner RR. 1997. Mercury Concentrations in Air During
the Phase I Remediation of the Lower East Fork Poplar Creek Floodplain at the Oak Ridge Y-12
Plant, Oak Ridge, Tennessee. Prepared by the Environmental Sciences Division at the Oak Ridge
National Laboratory. Prepared for the US Department of Energy. January 1997.
[Bechtel Jacobs 2010] Bechtel Jacobs Company LLC. 2010. 2010 Remediation Effectiveness
Report for the US Department of Energy Oak Ridge Reservation, Oak Ridge, Tennessee: Data
and Evaluations. Prepared for the US Department of Energy. March 2010.
[Bechtel Jacobs 2011] Bechtel Jacobs Company LLC. 2011. 2011 Remediation Effectiveness
Report for the US Department of Energy Oak Ridge Reservation, Oak Ridge, Tennessee: Data
and Evaluations. DOE/OR/01-2505&D1. Prepared for the US Department of Energy. March
2011. Available at
http://www.oakridge.doe.gov/External/LinkClick.aspx?fileticket=mX1QzFhvlpM%3D&tabid=3
25&mid=1118. Last accessed 16 January 2012.
Benson M, Lyons W, Scheb JM. 1994. Report of knowledge, attitudes and beliefs survey of
residents of an eight-county area surrounding Oak Ridge, Tennessee. Prepared for Tennessee
Department of Health, Division of Epidemiology, the Oak Ridge Health Agreement Steering
Panel (ORHASP), and the Oak Ridge Reservation Local Oversight Committee (LOC).
Knoxville: University of Tennessee. August 12, 1994
Berman E, House DE, Allis JW, Simmons JE. 1992. Hepatotoxic interactions of ethanol with
allyl alcohol or carbon tetrachloride in rats. J Toxicol Environ Health 37(1):161–176.
Bernaudin JF, Druet E, Druet P and Masse R. 1981. Inhalation or ingestion of organic or
inorganic mercurials produces auto-immune disease in rats. Clin. Immunol. Immunopathol. 20:
129-135. As cited in US Environmental Protection Agency. 2012a. Integrated risk information
system. Mercuric chloride (HgCl2) (CASRN 7487-94-7). Available at:
http://www.epa.gov/iris/subst/0692.htm. Last accessed 10 January 2012.
Blade LM, Worthington KA. 1996. Health hazard evaluation report, HETA 96-0071-2584,
Lockheed Martin Energy Systems, Inc., U.S. Department of Energy Oak Ridge K-25 Site, Oak
Ridge, Tennessee. Cincinnati, OH: National Institute for Occupational Safety and Health
Publications Office. Available at: http://www.cdc.gov/niosh/hhe/reports/pdfs/1996-0071­
2584.pdf. Last accessed 11 January 2012.
Page | 189

Bornhausen M, Musch MR, Greim H. 1980. Operant behavior performance changes in rats after
prenatal methylmercury exposure. Toxicol Appl Pharmacol 56:305-316. As cited in Agency for
Toxic Substances and Disease Registry. 1999. Toxicological profile for mercury. Atlanta: US
Department of Health and Human Services; March. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
Brooks SC, Southworth GR. 2011. History of mercury use and environmental contamination at
the Oak Ridge Y-12 Plant. Environmental Pollution 159:219–228.
Canady RA, Hanley JE, Susten AS. 1997. ATSDR Science Panel on the Bioavailability of
Mercury in Soils: Lessons Learned. Risk Analysis 17(5):527–532.
Caprino L, Borrelli F, Antonetti F, Cantelmo A. 1983. Sex-related toxicity of somatostatin and
its interaction with pentobarbital and strychnine. Toxicol Letters 17:145–149.
Case 1977. Mercury inventory at Y-12 plant 1950 through 1977. Y/AD-428. June 9, 1977.
Caygill CP, Charlett A, and Hill MJ. 1996. Fat, fish, fish oil and cancer. Br J Cancer 74(1):159–
164.
[CDC 2009] Centers for Disease Control and Prevention. 2009. National Report on Human
Exposure to Environmental Chemicals Fact Sheet: Mercury. Atlanta: US Department of Health
and Human Services; November. Available at:
http://www.cdc.gov/exposurereport/Mercury_FactSheet.html. Last accessed 25 May 2011.
[CDC 2011] Centers for Disease Control and Prevention. 2011. Fourth National Report on
Human Exposure to Environmental Chemicals, Updated Tables, February 2011. Atlanta: US
Department of Health and Human Services; February. Available at:
http://www.cdc.gov/exposurereport/pdf/Updated_Tables.pdf. Last accessed 25 May 2011.
ChemRisk. 1993a. Oak Ridge health studies, phase 1 report. Volume II–part a–dose
reconstruction feasibility study. tasks 1 & 2: a summary of historical activities on the Oak Ridge
Reservation with emphasis on information concerning off-site emissions of hazardous materials.
Oak Ridge: Oak Ridge Health Agreement Steering Panel and Tennessee Department of Health.
ChemRisk. 1993b. Oak Ridge health studies, phase 1 report. Volume II– part c–dose
reconstruction feasibility study. Task 5: A summary of information concerning historical
locations and activities of populations potentially affected by releases from the Oak Ridge
Reservation. Oak Ridge: Tennessee Department of Health and the Oak Ridge Health Agreement
Steering Panel.
ChemRisk. 1993c. Oak Ridge health studies, phase 1 report. Volume II– part b–dose
reconstruction feasibility study. Tasks 3 & 4: identification of important environmental pathways
for materials released from the Oak Ridge Reservation. Oak Ridge: Tennessee Department of
Health and the Oak Ridge Health Agreement Steering Panel.
ChemRisk. 1999a. Mercury releases from Lithium Enrichment at the Oak Ridge Y-12 plant–a
reconstruction of historical releases and off-site doses and health risks, task 2 report of the Oak
Ridge Dose reconstruction, volumes 2 and 2a. Oak Ridge: Tennessee Department of Health.
Available at: http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 27 March 2007.
ChemRisk. 1999b. Uranium releases from the Oak Ridge Reservation–a review of the quality of
historical effluent monitoring data and a screening evaluation of potential off-site exposures, task
Page | 190

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

6 report of the Oak Ridge Dose reconstruction, volume 5. Oak Ridge: Tennessee Department of
Health. Available at: http://health.state.tn.us/ceds/oakridge/Uranium.pdf. Last accessed 27 March
2007.
ChemRisk. 1999c. PCBs in the environment near the Oak Ridge Reservation, a reconstruction of
historical doses and health risks, task 3. Reports of the Oak Ridge dose reconstruction, volume 3.
Nashville TN: Tennessee Department of Health, Division of Communicable and Environmental
Disease Services. July 1999. Available at: http://health.state.tn.us/ceds/oakridge/PCB.pdf. Last
accessed 27 March 2007.
ChemRisk. 1999d. Oak Ridge Dose Reconstruction Project Summary Report. Reports of the Oak
Ridge Dose Reconstruction, Volume 7. Tennessee Department of Health. July 1999. Available
at: http://health.state.tn.us/ceds/oakridge/ProjSumm.pdf. Last accessed 27 March 2007.
ChemRisk. 1999e. Iodine-131 releases from radioactive lanthanum processing at the X-10 site in
Oak Ridge, Tennessee (1944-1956) – an assessment of quantities released, off-site radiation
doses, and potential excess risks of thyroid cancer, task 1. Reports of the Oak Ridge dose
reconstruction, volume 1. Oak Ridge: Tennessee Department of Health. July 1999. Available at
http://health.state.tn.us/ceds/oakridge/oridge.html. Last accessed 4 January 2012.
ChemRisk. 1999f. Radionuclide releases to the Clinch River from White Oak Creek on the Oak
Ridge Reservation—an assessment of historical quantities released, off-site radiation doses, and
health risks, task 4. Reports of the Oak Ridge dose reconstruction, volume 4. Oak Ridge:
Tennessee Department of Health. July 1999. Available at
http://health.state.tn.us/ceds/oakridge/oridge.html. Last accessed 4 January 2012.
ChemRisk. 1999g. Screening-level evaluation of additional potential materials of concern, task
7. Reports of the Oak Ridge dose reconstruction, volume 6. Oak Ridge: Tennessee Department
of Health. July 1999. Available at http://health.state.tn.us/ceds/oakridge/oridge.html. Last
accessed 10 January 2012.
Connor WE. 2004. Will the dietary intake of fish prevent atherosclerosis in diabetic women? Am
J Clin Nutr 80(3):535–536.
Cook RB, Holladay SK, Adams SM, Hook LA, Beauchamp JJ, Levine DA, Bevelhimer MS,
Longman RC, Blaylock BG, McGinn CW, Brandt CC, Skiles JL, Ford CJ, Suter GW, Frank ML,
Williams LF, Gentry MJ. 1992. Phase 1 data summary report for the Clinch River remedial
investigation: Health risk and ecological risk screening assessment. ORNL/ER-155. Oak Ridge
National Laboratory, Oak Ridge, Tennessee. As cited in Oak Ridge National Laboratory and
Jacobs Engineering Group Inc. 1995. Remedial Investigation/Feasibility Study Report for Lower
Watts Bar Reservoir Operable Unit. Prepared for the US Department of Energy, Office of
Environmental Management. March 1995.
Cox C, Clarkson TW, Marsh DO, Amin-Zaki L, Tikriti Sa’adoun, Myers GG. 1989. Doseresponse analysis of infants prenatally exposed to methyl mercury: An application of a single
compartment model to single-strand hair analysis. Environ Res 49(2):318-332. As cited in
Agency for Toxic Substances and Disease Registry. 1999. Toxicological profile for mercury.
Atlanta: US Department of Health and Human Services; March. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
Crump K, Viren J, Silvers A, Clewell H 3rd, Gearhart J, Shipp A. 1995. Reanalysis of doseresponse data from the Iraqi methylmercury poisoning episode. Risk Analysis 1(4):523-532. As
Page | 191

cited in Agency for Toxic Substances and Disease Registry. 1999. Toxicological profile for
mercury. Atlanta: US Department of Health and Human Services; March. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
Crump KS, Kjellström T, Shipp AM, Silvers A, Stewart A. 1998. Influence of prenatal exposure
upon scholastic and psychological test performance: Benchmark analysis of a New Zealand
Cohort. Risk Anal 18(6):701–13.
Daniels JL, Longnecker MP, Rowland AS, Golding J. 2004. Fish intake during pregnancy and
early cognitive development of offspring. Epidemiology 15(4):394–402.
Davidson PW, Myers GJ, Cox C, Axtell C, Shamlaye C, Sloane-Reeves J, Cernichiari E,
Needham L, Choi A, Wang Y, Berlin M, Clarkson TW. 1998. Effects of prenatal and postnatal
methylmercury exposure from fish consumption on neurodevelopment: Outcomes at 66 months
of age in the Seychelles child development study. JAMA 280(8):701-707. As cited in Agency for
Toxic Substances and Disease Registry. 1999. Toxicological profile for mercury. US Department
of Health and Human Services; Atlanta, Georgia. March 1999. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
de Deckere EA. 1999. Possible beneficial effect of fish and fish n-3 polyunsaturated fatty acids
in breast and colorectal cancer. Eur J Cancer Prev 8(3):213–221.
Debes F, Budtz-Jorgensen E, Weihe P, White RF, Grandjean P. 2006. Impact of prenatal
methylmercury exposure on neurobehavioral function at age 14 years. Neurotoxicol Tereatol
28:363–75.
[DHHS and DOL 1978] US Department of Health and Human Services and US Department of
Labor. 1978. Occupational Health Guideline for Inorganic Mercury. September 1978. Available
at: http://www.cdc.gov/niosh/docs/81-123/pdfs/0383.pdf.
[DOE 1989] US Department of Energy. 1989. Oak Ridge Reservation environmental report for
1988. Vol. 1: Narrative, Summary, and Conclusions. Oak Ridge, Tennessee: US Department of
Energy, Office of Scientific and Technical Information.
[DOE 1992a] US Department of Energy. 1992a. Letter to Robert L. Williams, Agency for Toxic
Substances and Disease Registry, from Clayton S. Gist, Environmental Restoration Integration
Branch, regarding the summary of East Fork Poplar Creek Phase Ia Data. August 21, 1992.
[DOE 1992b] US Department of Energy. 1992b. Mercury in ambient air over flood plain of East
Fork Poplar Creek. Oak Ridge, Tennessee. DE-AC05-84OR21400.
[DOE 1993a] US Department of Energy. 1993a. Final report on the background soil
characterization project at the Oak Ridge Reservation, Oak Ridge, Tennessee. Volume 1- Results
of field sampling program. DOE/OR/01-1175/V1. October 1993.
[DOE 1993b] US Department of Energy. 1993b. Openness Press Conference Fact Sheets.
December 7, 1993. As cited in ChemRisk. 1999a. Mercury releases from Lithium Enrichment at
the Oak Ridge Y-12 plant–a reconstruction of historical releases and off-site doses and health
risks, task 2 report of the Oak Ridge Dose reconstruction, volumes 2 and 2a. Oak Ridge:
Tennessee Department of Health. Available at:
http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 12 January 2012.

Page | 192

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

[DOE 1995a] US Department of Energy. 1995a. Proposed Plan for Lower Watts Bar Reservoir.
Oak Ridge, Tennessee. DOE/OR-01-1294&D5. March 1995.
[DOE 1995b] US Department of Energy. 1995b. Record of Decision for Lower East Fork Poplar
Creek, Oak Ridge, Tennessee. US Department of Energy, Office of Environmental Management.
July 1995.
[DOE 1995c] US Department of Energy. 1995c. Record of Decision for Lower Watts Bar
Reservoir, Oak Ridge, Tennessee. Prepared by Jacobs ER Team. September 1995. Available at:
http://www.epa.gov/superfund/sites/rods/fulltext/r0495249.pdf.
[DOE 1995d] US Department of Energy. 1995d. Proposed Plan, East Fork Poplar Creek—Sewer
Line Beltway, Oak Ridge, Tennessee. DOE/OR/02-1209&D3.
[DOE 1996] US Department of Energy. 1996. Environmental Restoration Activities at Oak
Ridge Operations Office. US Department of Energy, Office of Environmental Management.
March 1996.
[DOE 1997a] US Department of Energy. 1997a. Record of Decision for the Clinch River/Poplar
Creek Operable Unit, Oak Ridge, Tennessee. Prepared by Jacobs EM Team. September 1997.
Available at: http://www.epa.gov/superfund/sites/rods/fulltext/r0497075.pdf.
[DOE 1997b] US Department of Energy. 1997b. Remedial action report for Clinch River/Poplar
Creek in East Tennessee. DOE/OR-102-1627&03. Oak Ridge, TN. February 1998.
[DOE 2000] US Department of Energy. 2000. Remedial action report on the Lower East Fork
Poplar Creek project, Oak Ridge, Tennessee. DOE/OR/01-1680&D5. U.S. Department of
Energy, Office of Environmental Management.
[DOE 2001] US Department of Energy. 2001. Overview of CERCLA actions at off-site
locations. environmental management program fact sheet. September 2001.
[DOE 2002] US Department of Energy. 2002. Record of Decision for Phase I interim source
control actions in the Upper East Fork Poplar Creek Characterization Area, Oak Ridge,
Tennessee. US Department of Energy, Office of Environmental Management. May 2002.
[DOE 2003] US Department of Energy. 2003. Lower Watts Bar Reservoir Remedial Action.
Environmental Management Program Fact Sheet. April 2003. [DOE 2004] US Department of
Energy. 2004. Oak Ridge Reservation Groundwater Strategy. May, 2004. DOE/OR/01­
2069&D2.
Dourson ML, Wullenweber AE, Poirier KA. 2001. Uncertainties in the reference dose for
methylmercury. Neurotoxicology 22(5):677-689.
Drott P, Meurling S, Gebre-Medhin M. 1993. Interactions of vitamins A and E and retinolbinding protein to healthy Swedish children—evidence of thresholds of essentiality and toxicity.
Scand J Clin Lab Invest 53:275–280.
Druet, P., E. Druet, F. Potdevin and C. Sapin. 1978. Immune type glomerulonephritis induced by
HgCl2 in the Brown Norway rat. Ann. Immunol. 129C: 777-792. As cited in US Environmental
Protection Agency. 2012a. Integrated risk information system. Mercuric chloride (HgCl2)
(CASRN 7487-94-7). Available at: http://www.epa.gov/iris/subst/0692.htm. Last accessed 10
January 2012.

Page | 193

East Tennessee Development District. 1995. 1990 Census Summary Report for Roane County.
December 1995.
Energy Systems (Martin Marietta Energy Systems). 1993. Oak Ridge Reservation environmental
monitoring report for 1992. EH/ESH-31/ (2 vols). Oak Ridge National Laboratory, Oak Ridge,
Tennessee. As cited in Oak Ridge National Laboratory and Jacobs Engineering Group Inc. 1995.
Remedial Investigation/Feasibility Study Report for Lower Watts Bar Reservoir Operable Unit.
Prepared for the US Department of Energy, Office of Environmental Management. March 1995.
[EPA 1993] US Environmental Protection Agency. 1993. Reference Dose (RfD): Description
and Use in Health Risk Assessments, Background Document 1A. March 15, 1993. Available at:
http://www.epa.gov/iris/rfd.htm.
[EPA 1995a] US Environmental Protection Agency. 1995a. Toxicological review of mercuric
chloride. Integrated Risk Information System. Washington, DC. May 1, 1995.
[EPA 1995b] US Environmental Protection Agency. 1995b. User’s guide for the industrial
source complex (ISC) dispersion models. ISCST3 version 96113. USEPA-454/B-95-003.
Research Triangle Park, NC. March 1995.
[EPA 1997] US Environmental Protection Agency. 1997. Exposure Factors Handbook.
Washington, DC. August 1997. Available at:
http://cfpub.epa.gov/ncea/cfm/recordisplay.cfm?deid=12464#Download. Last accessed 11
January 2012.
[EPA 2002a] US Environmental Protection Agency. 2002a. Tennessee NPL/NPL caliber cleanup
site summaries. US DOE Oak Ridge Reservation, Oak Ridge, Anderson County, Tennessee.
Available at: http://www.epa.gov/region4/waste/npl/npltn/oakridtn.htm (last updated 10/15/02;
accessed 10/16/2002).
[EPA 2002b] US Environmental Protection Agency. 2002b. NPL site narrative for Oak Ridge
Reservation (USDOE), Oak Ridge, Tennessee. Available at:
http://www.epa.gov/oerrpage/superfund/sites/npl/nar1239.htm (last updated 10/4/02; accessed
11/26/2002).
[EPA 2002c] US Environmental Protection Agency. 2002c. US EPA’s Process for IRIS
Assessment Development and Review. May 2002. Available at:
http://www.epa.gov/iris/process.htm.
[EPA 2003] US Environmental Protection Agency. 2003. September 2001 Sampling Report for
the Scarboro Community, Oak Ridge, Tennessee. Athens, Georgia. April 2003.
[EPA 2004] US Environmental Protection Agency. 2004. Consumption advice, joint federal
advisory for mercury in fish. Available at:
http://www.epa.gov/waterscience/fish/advice/factsheet.html. Last accessed 4 May 2011.
[EPA 2012a] US Environmental Protection Agency. 2012a. Integrated risk information system.
Mercuric chloride (HgCl2) (CASRN 7487-94-7). Available at:
http://www.epa.gov/iris/subst/0692.htm. Last accessed 10 January 2012.
[EPA 2012b] US Environmental Protection Agency. 2012b. Integrated risk information system.
Methylmercury (MeHg) (CASRN 22967-92-6). Available at:
http://www.epa.gov/iris/subst/0073.htm. Last accessed 10 January 2012.
Page | 194

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Erkkila AT, Lichtenstein AH, Mozaffarian D, Herrington DM. 2004. Fish intake is associated
with a reduced progression of coronary artery atherosclerosis in postmenopausal women with
coronary artery disease. Am J Clin Nutr 80(3):626–632.
[EUWG 1998] End Use Working Group. 1998. Final report of the Oak Ridge Reservation End
Use Working Group.[FAMU 1998] Florida Agricultural and Mechanical University. 1998.
Scarboro Community Environmental Study.
Fawer RF, de Ribaupierre Y, Guillemin MP, Berode M, Lob M. 1983. Measurement of hand
tremor induced by industrial exposure to metallic mercury. Br J Ind Med 40:204-208. As cited in
Agency for Toxic Substances and Disease Registry. 1999. Toxicological profile for mercury.
Atlanta: US Department of Health and Human Services; March. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
[FDA 2011] US Food and Drug Administration. 2011. Fish and Fishery Products Hazards and
Controls Guidance. Fourth Edition. November 2011. Available at:
http://www.fda.gov/Food/GuidanceComplianceRegulatoryInformation/GuidanceDocuments/Sea
food/FishandFisheriesProductsHazardsandControlsGuide/default.htm.
[FDA 2004] Food and Drug Administration. 2004. What you need to know about mercury in fish
and shellfish: 2004 EPA and FDA advice for women who might become pregnant, women who
are pregnant, nursing mothers, and young children. Available at:
http://www.fda.gov/Food/FoodSafety/Product­
SpecificInformation/Seafood/FoodbornePathogensContaminants/Methylmercury/ucm115662.ht
m. Last accessed 11 January 2012.
Feron VJ, Groten JP, van Zorge JA, Cassee FR, Jonker D, van Bladeren PJ. 1995. Toxicity
studies in rats of simple mixtures of chemicals with the same or different target organs. Toxicol
Lett 82–83:505–512.
Fredriksson A, Dahlgren L, Danielsson B, Eriksson P, Dencker L, Archer T. 1992. Behavioral
effects of neonatal metallic mercury exposure in rats. Toxicology 74(2-3):151-160. As cited in
Agency for Toxic Substances and Disease Registry. 1999. Toxicological profile for mercury.
Atlanta: US Department of Health and Human Services; March. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
Friday JC, Turner RL. 2001. Scarboro community assessment report. Joint Center for Political
and Economic Studies. August 2001.
Garrett D. 1975. Materials Balance and Technology Assessment of Mercury and Its Compounds
of National and Regional Bases, EPA Study 560/3-75-007, October 1975.
Gough M. 2002. Antagonism—no synergism—in pairwise tests of carcinogens in rats. Regul
Toxicol Pharmacol 35:383–392.
Grandjean P, Weihe P, White RF, Debes F, Araki S, Yokoyama K, Murata K, Sørensen N, Dahl
R, Jørgensen PJ. 1997. Cognitive deficit in 7-year-old children with prenatal exposure to
methylmercury. Neurotoxicol Teratol 19(6):417-428. As cited in Agency for Toxic Substances
and Disease Registry. 1999. Toxicological profile for mercury. Atlanta: US Department of
Health and Human Services; March. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.

Page | 195

Groten JP, Schoen ED, van Bladeren PJ, Kuper CF, van Zorge JA, Feron VJ. 1997. Subacute
toxicity of a mixture of nine chemicals in rats: detecting interactive effects with a fractionated
two-level factorial design. Fundam Appl Toxicol 36(1):15–29.
Groten JP, Sinkeldam EJ, Muys T, Luten JB, van Bladeren PJ. 1991. Interaction of dietary Ca, P,
Mg, Mn, Cu, Fe, Zn and Se with the accumulation and oral toxicity of cadmium in rats. Food
Chem Toxicol 29(4):249–258.Hansen JC, Danscher G. 1995. Quantitative and qualitative
distribution of mercury in organs from arctic sledgedogs: An atomic absorption
spectrophotometric and histochemical study of tissue samples from natural long-termed high
dietary organic mercury-exposed dogs for Thule, Greenland. Pharmacol Toxicol 77(3):189-195.
As cited in Agency for Toxic Substances and Disease Registry. 1999. Toxicological profile for
mercury. US Department of Health and Human Services; Atlanta, Georgia. March 1999.
Available at: http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11
January 2012.
Harris LW, Lennox WJ, Talbot BG, Anderson DR, Swanson DR. 1984. Toxicity of
anticholinesterases: interactions of pyridostigmine and physostigmine with soman. Drug Chem
Toxicol 7:507–526.
Hasegawa R, Miyata E, Futakuchi M, Hagiwara A, Nagao M, Sugimura T, Ito N. 1994.
Synergistic enhancement of hepatic foci development by combined treatment of rats with 10
heterocyclic amines at low doses. Carcinogenesis 15(5):1037–1041.
Henke KR, Kuhnel V, Stepan DJ, Fraley RH, Robinson CM, Chralton DS, Gust HM, Bloom NS.
1993. Topical report: Critical review of mercury contamination issues relevant to manometers at
natural gas industry sites. GRI-93/0117. Chicago, IL: Gas Research Institute. As cited in
ChemRisk. 1999a. Mercury releases from Lithium Enrichment at the Oak Ridge Y-12 plant–a
reconstruction of historical releases and off-site doses and health risks, task 2 report of the Oak
Ridge Dose reconstruction, volumes 2 and 2a. Oak Ridge: Tennessee Department of Health.
Available at: http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 12 January
2012.
Hibbitts HW. 1984. Transmittal of environmental sampling data for mercury. February–
December 1984. As cited in ChemRisk. 1999a. Mercury releases from Lithium Enrichment at the
Oak Ridge Y-12 plant–a reconstruction of historical releases and off-site doses and health risks,
task 2 report of the Oak Ridge Dose reconstruction, volumes 2 and 2a. Oak Ridge: Tennessee
Department of Health. Available at: http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last
accessed 12 January 2012.
Hibbitts HW. 1986. Transmittal of environmental sampling data for mercury. As cited in
ChemRisk. 1999a. Mercury releases from Lithium Enrichment at the Oak Ridge Y-12 plant–a
reconstruction of historical releases and off-site doses and health risks, task 2 report of the Oak
Ridge Dose reconstruction, volumes 2 and 2a. Oak Ridge: Tennessee Department of Health.
Available at: http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 12 January
2012.
Hu FB, Cho E, Rexrode KM, Albert CM, Manson JE. 2003. Fish and long-chain omega-3 fatty
acid intake and risk of coronary heart disease and total mortality in diabetic women. Circulation
107(14):1852–1857.

Page | 196

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Hughes JA, Annau Z. 1976. Postnatal behavioral effects in mice after prenatal exposure to
methylmercury. Pharmacol Biochem Behav 4:385-391. As cited in Agency for Toxic Substances
and Disease Registry. 1999. Toxicological profile for mercury. Atlanta: US Department of
Health and Human Services; March. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
[IARC] International Agency for Research on Cancer. 1997. IARC monographs on the
evaluation of carcinogenic risks to humans. Volume 58: beryllium, cadmium, mercury, and
exposures in the glass manufacturing industry. Summary of data reported and evaluation;
mercury and mercury compounds. Last updated August 22, 1997. Available at:
http://monographs.iarc.fr/ENG/Monographs/vol58/volume58.pdf. Last accessed 10 January
2012.
Inouye M, Murakami U. 1975. Teratogenic effect of orally administered methylmercuric
chloride in rats and mice. Congenital Anomalies 15:1-9. As cited in Agency for Toxic
Substances and Disease Registry. 1999. Toxicological profile for mercury. Atlanta: US
Department of Health and Human Services; March. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
Jacobs EM Team. 1997. Record of decision for the Clinch River/Poplar Creek operable unit, Oak
Ridge, Tennessee. Prepared for the US Department of Energy, Office of Environmental
Management. Oak Ridge, Tennessee. September 1997. Available at:
http://www.epa.gov/superfund/sites/rods/fulltext/r0497075.pdf.
Jacobs Engineering Group Inc. 1996. Remedial Investigation/Feasibility Study of the Clinch
River/Poplar Creek Operable Unit. Prepared for the US Department of Energy, Office of
Environmental Management. March 1996. Available at:
http://www.osti.gov/dublincore/gpo/servlets/purl/226399-5omhIT/webviewable/226399.pdf.
Johnson CA, Erwin PC, Redd SC, Robinson AJ, Ball L, Moore W. 2000. An Analysis of
Respiratory Illnesses Among Children in the Scarboro Community, Oak Ridge, Tennessee.
Scarboro Community Environmental Justice Council, Centers for Disease Control, Tennessee
Department of Health. July 2000.
Jones AB and Slotten DG. 1996. Mercury Effects, Sources, and Control Measures. A Special
Study of the San Francisco Estuary Regional Monitoring Program. San Francisco Estuary
Institute, Richmond, CA.
Jonker D, Woutersen RA, van Bladeren PJ, Til HP, Feron VJ. 1990. 4-week oral toxicity study
of a combination of eight chemicals in rats: comparison with the toxicity of the individual
compounds. Food Chem Toxicol 28(9):623–631.
Jonker D, Woutersen RA, van Bladeren PJ, Til HP, Feron VJ. 1993. Subacute (4-wk) oral
toxicity of a combination of four nephrotoxins in rats: comparison with the toxicity of the
individual compounds. Food Chem Toxicol. 31(2):125–136.
Kishi R, Hashimoto K, Shimizu S, Kobayashi M. 1978. Behavioral changes and mercury
concentrations in tissues of rats exposed to mercury vapor. Toxicol Appl Pharmacol 46:555-566.
As cited in Agency for Toxic Substances and Disease Registry. 1999. Toxicological profile for
mercury. Atlanta: US Department of Health and Human Services; March. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
Page | 197

Kostial K, Kello D, Jugo S, Rabar I, Maljkovi6. 1978. Influence of age on metal metabolism and
toxicity. Environ Health Perspect 25:81-86. As cited in Agency for Toxic Substances and
Disease Registry. 1999. Toxicological profile for mercury. US Department of Health and Human
Services; Atlanta, Georgia. March 1999. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
Larsson SC, Kumlin M, Ingelman-Sundberg M, and Wolk A. 2004. Dietary long-chain n-3 fatty
acids for the prevention of cancer: a review of potential mechanisms. Am J Clin Nutr 79(6):935–
945.
Lemiski, PJ. 1994. Geological Mapping of the Oak Ridge K-25 Site, Oak Ridge, TN.
Environmental Sciences Division, Oak Ridge National Laboratory and Department of Geological
Sciences, University of Tennessee, Knoxville, TN. January 1994.
Lindberg SE, Kim K-H, Meyers TP, Owens JG. 1995. A micrometeorological gradient approach
for quantifying air/surface exchange of mercury vapor: Tests over contaminated soils. Environ
Sci Technol 29:126–135. As cited in ChemRisk. 1999a. Mercury releases from Lithium
Enrichment at the Oak Ridge Y-12 plant–a reconstruction of historical releases and off-site doses
and health risks, task 2 report of the Oak Ridge Dose reconstruction, volumes 2 and 2a. Oak
Ridge: Tennessee Department of Health. Available at:
http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 12 January 2012.
Lindberg SE, Turner RR, Meyers TP, Taylor GE, Jr., Schroeder WH. 1991. Atmospheric
concentrations and deposition of mercury to a deciduous forest at Walker Branch watershed,
Tennessee, USA. Water, Air, Soil Pollut 56:577–594. As cited in ChemRisk. 1999a. Mercury
releases from Lithium Enrichment at the Oak Ridge Y-12 plant–a reconstruction of historical
releases and off-site doses and health risks, task 2 report of the Oak Ridge Dose reconstruction,
volumes 2 and 2a. Oak Ridge: Tennessee Department of Health. Available at:
http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 12 January 2012.
Lindqvist O. 1991. Mercury in the Swedish environment: 8. Mercury in terrestrial systems.
Water, Air, Soil Pollution 55(1-2):73-100. As cited in Agency for Toxic Substances and Disease
Registry. 1999. Toxicological profile for mercury. US Department of Health and Human
Services; Atlanta, Georgia. March 1999. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
Lund E, Bonaa KH. 1993. Reduced breast cancer mortality among fishermen’s wives in Norway.
Cancer Causes Control 4(3):283–287.
Magos L, Butler WH. 1972. Cumulative effects of methylmercury dicyandiamide given orally to
rats. Food Cosmet Toxicol 10:513-517. As cited in Agency for Toxic Substances and Disease
Registry. 1999. Toxicological profile for mercury. Atlanta: US Department of Health and Human
Services; March. Available at: http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last
accessed 11 January 2012.
Mangano JJ. 1994. Cancer mortality near Oak Ridge, Tennessee. Int J Health Serv 24(3):521–
533.
Miettinen JK. 1973. Absorption and elimination of dietary (Hg++) and methylmercury in man.
In: Miller MW, Clarkson TW, eds. Mercury, mercurial, and mercaptans. Springfield, IL, C.C.
Thomas. As cited in Agency for Toxic Substances and Disease Registry. 1999. Toxicological
profile for mercury. US Department of Health and Human Services; Atlanta, Georgia. March
Page | 198

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

1999. Available at: http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed
11 January 2012.
[MMES 1986] Martin Marietta Energy Systems, Inc. 1986. Environmental Surveillance of the
Oak Ridge Reservation and Surrounding Environs During 1985. Oak Ridge, Tennessee. ORNL­
6271. April 1986.
Mozaffarian D, Psaty BM, Rimm EB, Lemaitre RN, Burke GL, Lyles MF, Lefkowitz D,
Siscovick DS. 2004. Fish intake and risk of incident atrial fibrillation. Circulation 110(4):368–
373.
Myers GJ, Davidson PW, Cox C, Shamlaye CF, Palumbo D, Cernichiari E, Sloane-Reeves J,
Wilding GE, Kost J, Huang LS, Clarkson TW. 2003. Prenatal methylmercury exposure from
ocean fish consumption in the Seychelles child development study. The Lancet 361:1686–692.
Myers GJ, Thurston SW, Pearson AT, Davidson PW, Cox C, Shamlaye CF, Cernichiari E,
Clarkson TW. 2009. Postnatal exposure to methyl mercury from fish consumption: A review and
new data from the Seychelles Child Development Study. NeuroToxicology 30(3):338-349.
[NOAA] National Oceanic and Atmospheric Administration. 2008. Seeking a better
understanding of atmospheric mercury: the Air Resources Laboratory measures and models
atmospheric mercury to provide essential information to policy-makers and planners. April 7,
2008. Available at: http://www.oar.noaa.gov/spotlite/2008/spot_mercury.html. Last accessed on
16 January 2012.
[NRC 2000] National Research Council. 2000. Toxicological Effects of Methylmercury.
Washington (DC): National Academy of Sciences, National Research Council. Pubs: National
Academy Press.
[NTP 1993] National Toxicology Program. 1993. Toxicology and carcinogenesis studies of
mercuric chloride (CAS no. 7487-94-7) in F344/N rats and B6C3F1 mice (gavage studies).
National Toxicology Program, US Department of Health and Human Services, Public Health
Service, National Institutes of Health, Research Triangle Park, NC. NTP TR 408. NIH
publication no. 91-3139. As cited in Agency for Toxic Substances and Disease Registry. 1999.
Toxicological profile for mercury. Atlanta: US Department of Health and Human Services;
March. Available at: http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed
11 January 2012.
Oak Ridge Comprehensive Plan. 1988. Comprehensive plan including 1988 update. Available
from the Oak Ridge Reading Room, Oak Ridge Public Library, Oak Ridge Tennessee.
[OREIS 2009] Oak Ridge Environmental Information System. 2009. Environmental Database.
Last accessed December 21, 2009.
[ORHASP 1999] Oak Ridge Health Agreement Steering Panel. 1999. Releases of contaminants
from Oak Ridge facilities and risks to public health. Final report of the ORHASP. December
1999.
[ORNL 1982] Oak Ridge National Laboratory. 1982. Environmental Analysis of the Operation
of Oak Ridge National Laboratory (X-10 Site). Prepared for the US Department of Energy.
ORNL-5870. November 1982.

Page | 199

[ORNL 1992] Oak Ridge National Laboratory. 1992. Letter to Robert Williams, Agency for
Toxic Substances and Disease Registry, from G.R. Southworth, Biomonitoring Group
Environmental Sciences, regarding PCB and mercury monitoring in fish in East Fork Poplar
Creek and Bear Creek in Oak Ridge, Tennessee. July 30, 1992.
[ORNL 2002] Oak Ridge National Laboratory. 2002. Oak Ridge National Laboratory Land and
Facilities Plan. Prepared for the US Department of Energy. ORNL/TM/2002-1. August 2002.
Available at: http://www.y12sweis.com/draftrefpdfs/RM%2086-%20ORNL%202002.pdf.
[ORNL and Jacobs Engineering Group 1995] Oak Ridge National Laboratory and Jacobs
Engineering Group, Inc. 1995. Remedial Investigation/Feasibility Study Report for Lower Watts
Bar Reservoir Operable Unit. Prepared for the US Department of Energy, Office of
Environmental Management. March 1995.
Paustenbach DJ, Bruce GM, Chrostowski P. 1997. Current views on the oral bioavailability of
inorganic mercury in soil: Implications for health risk assessments. Risk Anal 17(5):533-44. As
cited in Agency for Toxic Substances and Disease Registry. 1999. Toxicological profile for
mercury. US Department of Health and Human Services; Atlanta, Georgia. March 1999.
Available at: http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11
January 2012.
Rahola T, Hattula, T, Korolainen A, Miettinen JK. 1973. Elimination of free and protein-bound
ionic mercury 203Hg2+ in man. Ann Clin Res 5:214-219. As cited in Agency for Toxic
Substances and Disease Registry. 1999. Toxicological profile for mercury. US Department of
Health and Human Services; Atlanta, Georgia. March 1999. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24 . Last accessed 11 January 2012.
Revis NW, Osborne TR, Holdsworth G, Hadden C. 1989. Distribution of mercury species in soil
from a mercury-contaminated site. Water Air Soil Pollut 45(1-2):105-114. As cited in Agency
for Toxic Substances and Disease Registry. 1999. Toxicological profile for mercury. US
Department of Health and Human Services; Atlanta, Georgia. March 1999. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
Richmond CR and Auerbach SI. 1983. Testimony of a joint hearing of the Subcommittee on
Energy Research and Production and the Subcommittee on the Investigations and Oversight of
the US Office of Science and Technology Committee on the Impact of Mercury Releases at the
Oak Ridge Complex: Summary of actions and activities related to mercury releases in the Oak
Ridge area from DOE/UCC-ND operated facilities. As cited in ChemRisk. 1999a. Mercury
releases from Lithium Enrichment at the Oak Ridge Y-12 plant–a reconstruction of historical
releases and off-site doses and health risks, task 2 report of the Oak Ridge Dose reconstruction,
volumes 2 and 2a. Oak Ridge: Tennessee Department of Health. Available at:
http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 12 January 2012.
Roman H, Walsh T, Coull B, et al. 2011. Evaluation of the cardiovascular effects of
methylmercury exposures: current evidence supports development of a dose response functions
for regulatory benefits analysis. Environmental Health Perspectives 119 (5): 607-614.
Rowley DL, Turri P, Paschal DC. 1985. A pilot survey of mercury levels in Oak Ridge,
Tennessee. Center for Disease Control and Prevention and Tennessee Department of Health and
Environment. October 1985.

Page | 200

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

[SAIC 1994a] Science Applications International Corporation. 1994a. East Fork Poplar Creek –
Sewer Line Beltway Remedial Investigation Report. Oak Ridge, Tennessee. Prepared for the US
Department of Energy. DOE/OR/02-1119&D2.
[SAIC 1994b] Science Applications International Corporation. 1994b. Feasibility Study for the
Lower East Fork Poplar Creek – Sewer Line Beltway. Oak Ridge, Tennessee. Volumes 1 and 2.
Prepared for the US Department of Energy. DOE/OR/02-1185&D2.
[SAIC 1994c] Science Applications International Corporation. 1994c. Addendum to the East
Fork Poplar Creek-Sewer Line Beltway Remedial Investigation Report. Oak Ridge, Tennessee.
DOE/OR/02-1119&D2/A1/R1. June 1994.
[SAIC 1996] Science Applications International Corporation. 1996. White Oak Creek
Watershed: Melton Valley Area Remedial Investigation Report, at Oak Ridge National
Laboratory, Oak Ridge, TN. Volume I: Main Text. Prepared for the US Department of Energy.
DOE/OR/01-1546/V1&D1. November 1996.
[SAIC 1998] Science Applications International Corporation. 1998. Phase II Confirmatory
Sampling Data Report, Lower East Fork Poplar Creek, Oak Ridge Y-12 Plant, Oak Ridge,
Tennessee. Prepared for the US Department of Energy. January 1998.
[SAIC 2002a] Science Applications International Corporation. 2002a. 2002 Remediation
Effectiveness Report for the US Department of Energy, Oak Ridge Reservation, Oak Ridge,
Tennessee. Prepared for the US Department of Energy. March 2002.
[SAIC 2002b] Science Applications International Corporation. 2002b. Land use technical report.
Science Applications International Corporation. September 2002.
[SAIC 2004] Science Applications International Corporation. 2004. 2004 Remediation
Effectiveness Report for the US Department of Energy, Oak Ridge Reservation, Oak Ridge,
Tennessee. Prepared for the US Department of Energy. March 2004.
[SAIC 2005] Science Applications International Corporation. 2005. 2005 Remediation
Effectiveness Report for the US Department of Energy Oak Ridge Reservation, Oak Ridge,
Tennessee. Prepared for the US Department of Energy. March 2005.
[SAIC 2007] Science Applications International Corporation. 2007. 2006 Remediation
Effectiveness Report/Second Reservation-wide CERCLA Five-Year Review for the US
Department of Energy Oak Ridge Reservation, Oak Ridge, Tennessee. Prepared for the US
Department of Energy. February 2007.
Sanders, M. 1975.Analysis of Long-term Data on Uranium in Air. Report Y-1992. September
29, 1975.
Saouter E, Gillman M, Turner R, Barkary T. 1995. Development and field validation of a
microcosm to simulate the mercury cycle in a contaminated pond. Environ. Toxicol. Chem.
14(1):69-77. As cited in ChemRisk. 1999a. Mercury releases from Lithium Enrichment at the
Oak Ridge Y-12 plant–a reconstruction of historical releases and off-site doses and health risks,
task 2 report of the Oak Ridge Dose reconstruction, volumes 2 and 2a. Oak Ridge: Tennessee
Department of Health. Available at: http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last
accessed 12 January 2012.

Page | 201

Schoof RA, Nielsen JB. 1997. Evaluation of methods for assessing the oral bioavailability of
inorganic mercury in soil. Risk Anal 17(5):545-55. As cited in Agency for Toxic Substances and
Disease Registry. 1999. Toxicological profile for mercury. US Department of Health and Human
Services; Atlanta, Georgia. March 1999. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
Seed J, Brown RP, Olin SS, Foran JA. 1995. Chemical mixtures: current risk assessment
methodologies and future directions. Regul Toxicol Pharmacol 22:76–94.
Shacklette HT and Boerngen JG. 1984. Element concentrations in soils and surficial materials of
the conterminous United States. US Geological Survey Professional Paper 1270. US
Government Printing Office, Washington. 1984.
Sheppard SC, Evenden WG, Schwartz WJ. 1995. Heavy metals in the environment, ingested
soil: bioavailability of sorbed lead, cadmium, cesium, iodine, and mercury. J Environ Qual
24:498–505. As cited in Paustenbach DJ, Bruce GM, Chrostowski P. 1997. Current views on the
oral bioavailability of inorganic mercury in soil: Implications for health risk assessments. Risk
Anal 17(5):533-44.
Sin YM, Lim YF, Wong MK. 1983. Uptake and distribution of mercury in mice from ingesting
soluble and insoluble mercury compounds. Bull Environ Contam Toxicol 31(5):605-612. As
cited in Agency for Toxic Substances and Disease Registry. 1999. Toxicological profile for
mercury. US Department of Health and Human Services; Atlanta, Georgia. March 1999.
Available at: http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24 . Last accessed 11
January 2012.
Sin YM, Teh WF, Wong MK, Reddy PK. 1990. Effect of mercury on glutathione and thyroid
hormones. Bull Environ Contam Toxicol 44(4):616-622. As cited in Agency for Toxic
Substances and Disease Registry. 1999. Toxicological profile for mercury. US Department of
Health and Human Services; Atlanta, Georgia. March 1999. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January
2012.Southworth G, Greeley M, Peterson M, and Lowe K. 2010. Sources of Mercury to East
Fork Poplar Creek Downstream from the Y-12 National Security Complex: Inventories and
Export Rates. Prepared by Oak Ridge National Laboratory for Bechtel Jacobs Company.
February 2010.
Southworth GR, Peterson MJ, and Bogle MA. 2004. Bioaccumulation Factors for Mercury in
Stream Fish. Environmental Practice. 6:135-143.
Takayama S, Hasegawa H, Ohgaki H. 1989. Combination effects of forty carcinogens
administered at low doses to male rats. Jpn J Cancer Res 80(8):732–736.
[TDEC 1992] Tennessee Department of Environment & Conservation. 1992. Letter to Robert C.
Williams, Agency for Toxic Substances and Disease Registry, from Earl Leming, DOE
Oversight Division, regarding posting East Fork Poplar Creek and Poplar Creek. August 21,
1992.
[TDEC 1997] Tennessee Department of Environment & Conservation, DOE Oversight Division.
1997. Report on Turtle Sampling in Watts Bar Reservoir and the Clinch River. Authors:
Henninger J, Kinsall J, Lindbom R, Peryam J, Vanaudenhove J, Zarbo A. May 1997.

Page | 202

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

[TDHE 1983] Tennessee Department of Health and Environment. 1983. Miscellaneous
correspondence to Oak Ridge community concerning DOE mercury analyses in the Oak Ridge
Vicinity. January–November 1983. As cited in ChemRisk. 1999a. Mercury releases from
Lithium Enrichment at the Oak Ridge Y-12 plant–a reconstruction of historical releases and offsite doses and health risks, task 2 report of the Oak Ridge Dose reconstruction, volumes 2 and
2a. Oak Ridge: Tennessee Department of Health. Available at:
http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 27 March 2007.
[TDOH 1996] Tennessee Department of Health, East Region. 1996. A health assessment of the
East Tennessee Region. Second Edition. October 1996.
[TDOH 2000] Tennessee Department of Health. 2000. Contaminant releases and public health
risks: results of the Oak Ridge health agreement studies. July 2000.
Thapa PB. 1996. ORHASP: Feasibility of epidemiologic studies. Final report, executive
summary. Nashville, TN: Department of Preventive Medicine, Vanderbilt University School of
Medicine. July 3, 1996.
Thomas S, Frank L, Paine A. 1997. Special Report: An investigation into illnesses around the
nation’s nuclear weapons sites: 16 Sick Kids on One Street Alarms Officials and Residents in
Oak Ridge. The Tennessean. November 9. 1997.
[TVA 1985a] Tennessee Valley Authority. 1985a. Instream contaminant study, task 2: sediment
characterization, volume I. Knoxville, TN: Office of Natural Resources and Economic
Development. April 1985. As cited in ChemRisk. 1999a. Mercury releases from Lithium
Enrichment at the Oak Ridge Y-12 plant–a reconstruction of historical releases and off-site doses
and health risks, task 2 report of the Oak Ridge Dose reconstruction, volumes 2 and 2a. Oak
Ridge: Tennessee Department of Health. Available at:
http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 12 January 2012.
[TVA 1985b] Tennessee Valley Authority. 1985b. Instream contaminant study, task 3: sediment
transport. Knoxville, TN: Office of Natural Resources and Economic Development. August
1985. As cited in ChemRisk. 1999a. Mercury releases from Lithium Enrichment at the Oak
Ridge Y-12 plant–a reconstruction of historical releases and off-site doses and health risks, task
2 report of the Oak Ridge Dose reconstruction, volumes 2 and 2a. Oak Ridge: Tennessee
Department of Health. Available at: http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last
accessed 12 January 2012.
[TVA 1986] Tennessee Valley Authority. 1986. Heavy metals and PCB concentrations in
sediments from selected TVA reservoirs – 1982. Office of Natural Resources and Economic
Development. Chattanooga, Tennessee. TVA/ONRED/AW 86/35. As cited in Oak Ridge
National Laboratory and Jacobs Engineering Group Inc. 1995. Remedial
Investigation/Feasibility Study Report for Lower Watts Bar Reservoir Operable Unit. Prepared
for the US Department of Energy, Office of Environmental Management. March 1995.
[TVA 1987] Tennessee Valley Authority. 1987. Watts Bar Reservoir land management plan
(final draft). As cited in Oak Ridge National Laboratory and Jacobs Engineering Group Inc.
1995. Remedial Investigation/Feasibility Study Report for Lower Watts Bar Reservoir Operable
Unit. Prepared for the US Department of Energy, Office of Environmental Management. March
1995.

Page | 203

[TVA 1990] Tennessee Valley Authority. 1990. Tennessee River and reservoir system operation
and planning review. Final environmental impact statement. TVA/RDG/EQS-91/1. As cited in
Oak Ridge National Laboratory and Jacobs Engineering Group Inc. 1995. Remedial
Investigation/Feasibility Study Report for Lower Watts Bar Reservoir Operable Unit. Prepared
for the US Department of Energy, Office of Environmental Management. March 1995.
[TVA 1991] Tennessee Valley Authority. 1991. Results of Sediment and Water Sampling for
Inorganic, Organic, and Radionuclide Analysis at Recreation Areas and Water Intakes – Norris,
Melton Hill, and Watts Bar Lakes. Data Report. October 1991.
[UCCND 1983a] Union Carbide Company Nuclear Division. 1983a. The 1983 Mercury Task
Force. Mercury at Y-12: A Study of Mercury Use at the Y-12 Plant, Accountability, and Impact
on Y-12 Workers and the Environment -1950-1983. Y/EX-21/del rev. August 18, 1983.
[UCCND 1983b] Union Carbide Company Nuclear Division. 1983b. The 1983 Mercury Task
Force. Mercury at Y-12: A Summary of the 1983 UCCND Task Force Study. Y/EX-23.
November 1983.
US Census Bureau. 1940. Sixteenth census of the United States: 1940 population. Volume 1:
Number of inhabitants. Available from the Tennessee State Library and Archives, Nashville,
Tennessee.
US Census Bureau. 1950. Census of population: 1950. Volume 1: Number of inhabitants.
Available from the Tennessee State Library and Archives, Nashville, Tennessee.
US Census Bureau. 1960. Census of population: 1960. Volume 1: Characteristics of the
population, part A, number of inhabitants. Available from the Tennessee State Library and
Archives, Nashville, Tennessee.
US Census Bureau. 1970. 1970 census of population—number of inhabitants, Tennessee.
Volume 1: Part 44. Available from the Tennessee State Library and Archives, Nashville,
Tennessee.
US Census Bureau. 1980. 1980 census of population—number of inhabitants, Tennessee.
Volume 1: Part 44. Available from the Tennessee State Library and Archives, Nashville,
Tennessee.
US Census Bureau. 1993. 1990 Census of Population and Housing, Population and Housing Unit
Counts, United States. US Department of Commerce, Economics and Statistics Administration.
August 1993. Available at: http://www.census.gov/prod/cen1990/cph2/cph-2-1-1.pdf.
US Census Bureau. 2000. Census of Population, Housing Unit, Area, and Density: 2000.
American FactFinder. Washington DC: US Department of Commerce. Available at:
http://factfinder.census.gov/servlet/GCTTable?ds_name=DEC_2000_SF1_U&geo_id=04000US
47&_box_head_nbr=GCT-PH1&format=ST-2.
[USGS 1967] United States Geological Survey. 1967. Hydrologic data for the Oak Ridge area
Tennessee. Geological Survey Water-Supply Paper 1839-N. Washington, DC: United States
Government Printing Office. As cited in ChemRisk. 1999a. Mercury releases from Lithium
Enrichment at the Oak Ridge Y-12 plant–a reconstruction of historical releases and off-site doses
and health risks, task 2 report of the Oak Ridge Dose reconstruction, volumes 2 and 2a. Oak
Ridge: Tennessee Department of Health. Available at:
http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 12 January 2012.
Page | 204

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

[USGS 1986] United States Geological Survey. 1986. Preliminary evaluation of the Knox Group
in Tennessee for receiving injected wastes. Water Resources Investigations Report 85-4304.
Prepared in cooperation with the US Environmental Protection Agency. Nashville, TN.
[USGS 1988] United States Geological Survey. 1988. Hydrology of the Melton Valley
radioactive waste burial grounds at Oak Ridge National Laboratory, Tennessee. US Geological
Survey Open File Report 87-686. Prepared in cooperation with the US Department of Energy.
Knoxville, TN.
[USGS 1989] United States Geological Survey. 1989. An investigation of shallow ground-water
quality near East Fork Poplar Creek, Oak Ridge, Tennessee. Water Resources Investigations
Report 88-4219. Prepared in cooperation with the US Department of Energy. Nashville, TN.
[USGS 1995] United States Geological Survey. 1995. Mercury contamination of aquatic
ecosystems. Fact sheet FS-216-95. Madison, WI: U.S. Department of Interior. Available at:
http://water.usgs.gov/wid/FS_216-95/FS_216-95.pdf. Last accessed 18 January 2012.
US Weather Bureau. 1953. A meteorological survey of the Oak Ridge area: Final report covering
the period 1948–1952. US Atomic Energy Commission, Oak Ridge Operations. Oak Ridge,
Tennessee. Report ORO-99. November 1953. As cited in ChemRisk. 1999a. Mercury releases
from Lithium Enrichment at the Oak Ridge Y-12 plant–a reconstruction of historical releases and
off-site doses and health risks, task 2 report of the Oak Ridge Dose reconstruction, volumes 2
and 2a. Oak Ridge: Tennessee Department of Health. Available at:
http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 12 January 2012.
Van Cleave ML. 1992. Memorandum to Dr. Mary Yarbough from Mary Layne Van Cleave. Dr.
Reid’s concerns about health status in Oak Ridge. Tennessee Department of Health. October 19,
1992.
Warfvinge K, Hansson H, Hultman P. 1995. Systemic autoimmunity due to mercury vapor
exposure in genetically susceptible mice: Dose-responses studies. Toxicol Appl Pharm 132:299­
309. As cited in Agency for Toxic Substances and Disease Registry. 1999. Toxicological profile
for mercury. Atlanta: US Department of Health and Human Services; March. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.
Wren C. 1996. Review of mercury levels in fish, wáter, and sediments. Progress report #2 for
Oak Ridge dose reconstruction Project. Guelph, Ontario, Candada: Ecological Services for
Planning, Ltd. August 1996. As cited in ChemRisk. 1999a. Mercury releases from Lithium
Enrichment at the Oak Ridge Y-12 plant–a reconstruction of historical releases and off-site doses
and health risks, task 2 report of the Oak Ridge Dose reconstruction, volumes 2 and 2a. Oak
Ridge: Tennessee Department of Health. Available at:
http://health.state.tn.us/ceds/oakridge/Mercury1.pdf. Last accessed 12 January 2012.
Yasutake A, Hirayama Y, Inouye M. 1991. Sex differences of nephrotoxicity by methylmercury
in mice. In: Bach PH, et al., eds. Nephrotoxicity: Mechanisms, early diagnosis, and therapeutic
management. Fourth International Symposium on Nephrotoxicity, Guilford, England, UK, 1989.
New York, NY: Marcel Dekker, Inc., 389-396. As cited in Agency for Toxic Substances and
Disease Registry. 1999. Toxicological profile for mercury. Atlanta: US Department of Health
and Human Services; March. Available at:
http://www.atsdr.cdc.gov/toxprofiles/tp.asp?id=115&tid=24. Last accessed 11 January 2012.

Page | 205

APPENDICES

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Appendix A. ATSDR Glossary of Terms
The Agency for Toxic Substances and Disease Registry (ATSDR) is a federal public health
agency with headquarters in Atlanta, Georgia, and 10 regional offices in the United States.
ATSDR’s mission is to serve the public by using the best science, taking responsive public
health actions, and providing trusted health information to prevent harmful exposures and
diseases related to toxic substances. ATSDR is not a regulatory agency, unlike the U.S.
Environmental Protection Agency (U.S.EPA), which is the federal agency that develops and
enforces environmental laws to protect the environment and human health. This glossary defines
words used by ATSDR in communications with the public. It is not a complete dictionary of
environmental health terms. If you have questions or comments, call the agency’s toll-free
telephone number, 1-800-CDC-INFO (1-800-232-4636).
Absorption
The process of taking in. For a person or an animal, absorption is the process of a substance
getting into the body through the eyes, skin, stomach, intestines, or lungs.
Acute
Occurring over a short time [compare with chronic].
Acute exposure
Contact with a substance that occurs once or for only a short time (up to 14 days) [compare with
intermediate duration exposure and chronic exposure].
Adverse health effect
A change in body function or cell structure that might lead to disease or health problems
Aerobic
Requiring oxygen [compare with anaerobic].
Ambient
Surrounding (for example, ambient air).
Anaerobic
Requiring the absence of oxygen [compare with aerobic].
Analytic epidemiologic study
A study that evaluates the association between exposure to hazardous substances and disease by
testing scientific hypotheses.
Background level
An average or expected amount of a substance or radioactive material in a specific environment,
or typical amounts of substances that occur naturally in an environment.
Biota
Plants and animals in an environment. Some of these plants and animals might be sources of
food, clothing, or medicines for people.
Body burden
The total amount of a substance in the body. Some substances build up in the body because they
are stored in fat or bone or because they leave the body very slowly.

A-1

Cancer
Any one of a group of diseases that occur when cells in the body become abnormal and grow or
multiply out of control.
Cancer risk
A theoretical risk for getting cancer if exposed to a substance every day for 70 years (a lifetime
exposure). The true risk might be lower.
Carcinogen
A substance that causes cancer.
Central nervous system
The part of the nervous system that consists of the brain and the spinal cord.
CERCLA [see Comprehensive Environmental Response, Compensation, and Liability Act of
1980]
Chronic
Occurring over a long time [compare with acute].
Chronic exposure
Contact with a substance that occurs over a long time (more than 1 year) [compare with acute
exposure and intermediate duration exposure]
Comparison value (CV)
Calculated concentration of a substance in air, water, food, or soil that is unlikely to cause
harmful (adverse) health effects in exposed people. The CV is used as a screening level during
the public health assessment process. Substances found in amounts greater than their CVs might
be selected for further evaluation in the public health assessment process.
Completed exposure pathway [see exposure pathway].
Comprehensive Environmental Response, Compensation, and Liability Act of 1980
(CERCLA)
CERCLA, also known as Superfund, is the federal law that concerns the removal or cleanup of
hazardous substances in the environment and at hazardous waste sites. ATSDR, which was
created by CERCLA, is responsible for assessing health issues and supporting public health
activities related to hazardous waste sites or other environmental releases of hazardous
substances. This law was later amended by the Superfund Amendments and Reauthorization Act
(SARA).
Concentration
The amount of a substance present in a certain amount of soil, water, air, food, blood, hair, urine,
breath, or any other media.
Contaminant
A substance that is either present in an environment where it does not belong or is present at
levels that might cause harmful (adverse) health effects.
Dermal
Referring to the skin. For example, dermal absorption means passing through the skin.
Dermal contact
Contact with (touching) the skin [see route of exposure].

A-2

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Detection limit
The lowest concentration of a chemical that can reliably be distinguished from a zero
concentration.
Disease registry
A system of ongoing registration of all cases of a particular disease or health condition in a
defined population.
DOE
United States Department of Energy.
Dose (for chemicals that are not radioactive)
The amount of a substance to which a person is exposed over some time period. Dose is a
measurement of exposure. Dose is often expressed as milligram (amount) per kilogram (a
measure of body weight) per day (a measure of time) when people eat or drink contaminated
water, food, or soil. In general, the greater the dose, the greater the likelihood of an effect. An
“exposure dose” is how much of a substance is encountered in the environment. An “absorbed
dose” is the amount of a substance that actually got into the body through the eyes, skin,
stomach, intestines, or lungs.
Dose-response relationship
The relationship between the amount of exposure [dose] to a substance and the resulting changes
in body function or health (response).
Environmental media
Soil, water, air, biota (plants and animals), or any other parts of the environment that can contain
contaminants.
Environmental media and transport mechanism
Environmental media include water, air, soil, and biota (plants and animals). Transport
mechanisms move contaminants from the source to points where human exposure can occur. The
environmental media and transport mechanism is the second part of an exposure pathway.
Epidemiology
The study of the distribution and determinants of disease or health status in a population; the
study of the occurrence and causes of health effects in humans.
Exposure
Contact with a substance by swallowing, breathing, or touching the skin or eyes. Exposure may
be short-term [acute exposure], of intermediate duration, or long-term [chronic exposure].
Exposure assessment
The process of finding out how people come into contact with a hazardous substance, how often
and for how long they are in contact with the substance, and how much of the substance they are
in contact with.
Exposure-dose reconstruction
A method of estimating the amount of people’s past exposure to hazardous substances. Computer
and approximation methods are used when past information is limited, not available, or missing.
Exposure investigation
The collection and analysis of site-specific information and biologic tests (when appropriate) to
determine whether people have been exposed to hazardous substances.
A-3

Exposure pathway
The route a substance takes from its source (where it began) to its end point (where it ends), and
how people can come into contact with (or get exposed to) it. An exposure pathway has five
parts: a source of contamination (such as an abandoned business); an environmental media and
transport mechanism (such as movement through groundwater); a point of exposure (such as a
private well); a route of exposure (eating, drinking, breathing, or touching), and a receptor
population (people potentially or actually exposed). When all five parts are present, the exposure
pathway is termed a completed exposure pathway.
Exposure registry
A system of ongoing follow up of people who have had documented environmental exposures.
Feasibility study
A study by U.S.EPA to determine the best way to clean up environmental contamination. A
number of factors are considered, including health risk, costs, and what methods will work well.
Geographic information system (GIS)
A mapping system that uses computers to collect, store, manipulate, analyze, and display data.
For example, GIS can show the concentration of a contaminant within a community in relation to
points of reference such as streets and homes.
Grand rounds
Training sessions for physicians and other health care providers about health topics.
Groundwater
Water beneath the earth’s surface in the spaces between soil particles and between rock surfaces
[compare with surface water].
Hazard
A source of potential harm from past, current, or future exposures.
Hazardous waste
Potentially harmful substances that have been released or discarded into the environment.
Health consultation
A review of available information or collection of new data to respond to a specific health
question or request for information about a potential environmental hazard. Health consultations
are focused on a specific exposure issue. Health consultations are therefore more limited than a
public health assessment, which reviews the exposure potential of each pathway and chemical
[compare with public health assessment].
Health education
Programs designed with a community to help it know about health risks and how to reduce these
risks.
Health investigation
The collection and evaluation of information about the health of community residents. This
information is used to describe or count the occurrence of a disease, symptom, or clinical
measure and to evaluate the possible association between the occurrence and exposure to
hazardous substances.
Health promotion
The process of enabling people to increase control over, and to improve, their health.
A-4

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Health statistics review
The analysis of existing health information (i.e., from death certificates, birth defects registries,
and cancer registries) to determine if there is excess disease in a specific population, geographic
area, and time period. A health statistics review is a descriptive epidemiologic study.
Incidence
The number of new cases of disease in a defined population over a specific time period [contrast
with prevalence].
Ingestion
The act of swallowing something through eating, drinking, or mouthing objects. A hazardous
substance can enter the body this way [see route of exposure].
Inhalation
The act of breathing. A hazardous substance can enter the body this way [see route of exposure].
Intermediate duration exposure
Contact with a substance that occurs for more than 14 days and less than a year [compare with
acute exposure and chronic exposure].
Lowest-observed-adverse-effect level (LOAEL)
The lowest tested dose of a substance that has been reported to cause harmful (adverse) health
effects in people or animals.
Medical monitoring
A set of medical tests and physical exams specifically designed to evaluate whether an
individual’s exposure could negatively affect that person’s health.
Metabolism
The conversion or breakdown of a substance from one form to another by a living organism.
Metabolite
Any product of metabolism.
mg/kg
Milligram per kilogram.
mg/m3
Milligram per cubic meter; a measure of the concentration of a chemical in a known volume (a
cubic meter) of air, soil, or water.
Migration
Moving from one location to another.
Minimal risk level (MRL)
An ATSDR estimate of daily human exposure to a hazardous substance at or below which that
substance is unlikely to pose a measurable risk of harmful (adverse), noncancerous effects.
MRLs are calculated for a route of exposure (inhalation or oral) over a specified time period
(acute, intermediate, or chronic). MRLs should not be used as predictors of harmful (adverse)
health effects [see reference dose].
Morbidity
State of being ill or diseased. Morbidity is the occurrence of a disease or condition that alters
health and quality of life.

A-5

Mortality
Death. Usually the cause (a specific disease, a condition, or an injury) is stated.
National Priorities List for Uncontrolled Hazardous Waste Sites (National Priorities List or
NPL)
U.S.EPA’s list of the most serious uncontrolled or abandoned hazardous waste sites in the United
States. The NPL is updated on a regular basis.
National Toxicology Program (NTP)
Part of the Department of Health and Human Services. NTP develops and carries out tests to

predict whether a chemical will cause harm to humans.

No-observed-adverse-effect level (NOAEL)

The highest tested dose of a substance that has been reported to have no harmful (adverse) health

effects on people or animals.

NPL [see National Priorities List for Uncontrolled Hazardous Waste Sites]

Pica
A craving to eat nonfood items, such as dirt, paint chips, and clay. Some children exhibit picarelated behavior.
Plume
A volume of a substance that moves from its source to places farther away from the source.
Plumes can be described by the volume of air or water they occupy and the direction they move.
For example, a plume can be a column of smoke from a chimney or a substance moving with
groundwater.
Point of exposure
The place where someone can come into contact with a substance present in the environment
[see exposure pathway].
Population
A group or number of people living within a specified area or sharing similar characteristics
(such as occupation or age).
ppb
Parts per billion.
ppm
Parts per million.
Prevalence
The number of existing disease cases in a defined population during a specific time period
[contrast with incidence].
Prevention
Actions that reduce exposure or other risks, keep people from getting sick, or keep disease from
getting worse.
Public availability session
An informal, drop-by meeting at which community members can meet one-on-one with ATSDR
staff members to discuss health and site-related concerns.

A-6

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Public comment period
An opportunity for the public to comment on agency findings or proposed activities contained in
draft reports or documents. The public comment period is a limited time period during which
comments will be accepted.
Public health action
A list of steps to protect public health.
Public health advisory
A statement made by ATSDR to U.S.EPA or a state regulatory agency that a release of
hazardous substances poses an immediate threat to human health. The advisory includes
recommended measures to reduce exposure and reduce the threat to human health.
Public health assessment (PHA)
An ATSDR document that examines hazardous substances, health outcomes, and community
concerns at a hazardous waste site to determine whether people could be harmed from coming
into contact with those substances. The PHA also lists actions that need to be taken to protect
public health [compare with health consultation].
Public health statement
The first chapter of an ATSDR toxicological profile. The public health statement is a summary
written in words that are easy to understand. The public health statement explains how people
might be exposed to a specific substance and describes the known health effects of that
substance.
Public meeting
A public forum with community members for communication about a site.
Radionuclide
Any radioactive isotope (form) of any element.
RCRA [see Resource Conservation and Recovery Act (1976, 1984)]
Receptor population
People who could come into contact with hazardous substances [see exposure pathway].
Reference dose (RfD)
A U.S.EPA estimate, with uncertainty or safety factors built in, of the daily lifetime dose of a
substance that is unlikely to cause harm in humans.
Registry
A systematic collection of information on persons exposed to a specific substance or having
specific diseases [see exposure registry and disease registry].
Remedial investigation
The CERCLA process of determining the type and extent of hazardous material contamination at
a site.
Resource Conservation and Recovery Act (1976, 1984) (RCRA)
This Act regulates management and disposal of hazardous wastes currently generated, treated,
stored, disposed of, or distributed.
RfD [see reference dose]

A-7

Risk
The probability that something will cause injury or harm.
Risk reduction
Actions that can decrease the likelihood that individuals, groups, or communities will experience
disease or other health conditions.
Risk communication
The exchange of information to increase understanding of health risks.
Route of exposure
The way people come into contact with a hazardous substance. Three routes of exposure are
breathing [inhalation], eating or drinking [ingestion], or contact with the skin [dermal contact].
Safety factor [see uncertainty factor]
SARA [see Superfund Amendments and Reauthorization Act]
Sample
A portion or piece of a whole. A selected subset of a population or subset of whatever is being
studied. For example, in a study of people the sample is a number of people chosen from a larger
population [see population]. An environmental sample (for example, a small amount of soil or
water) might be collected to measure contamination in the environment at a specific location.
Sample size
The number of units chosen from a population or an environment.
Solvent
A liquid capable of dissolving or dispersing another substance (for example, acetone or mineral
spirits).
Source of contamination
The place where a hazardous substance comes from, such as a landfill, waste pond, incinerator,
storage tank, or drum. A source of contamination is the first part of an exposure pathway.
Special populations
People who might be more sensitive or susceptible to exposure to hazardous substances because
of factors such as age, occupation, sex, or behaviors (for example, cigarette smoking). Children,
pregnant women, and older people are often considered special populations.
Statistics
A branch of mathematics that deals with collecting, reviewing, summarizing, and interpreting
data or information. Statistics are used to determine whether differences between study groups
are meaningful.
Substance
A chemical.
Superfund [see Comprehensive Environmental Response, Compensation, and Liability Act of
1980 (CERCLA) and Superfund Amendments and Reauthorization Act (SARA)]

A-8

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Superfund Amendments and Reauthorization Act (SARA)
In 1986, SARA amended the Comprehensive Environmental Response, Compensation, and
Liability Act of 1980 (CERCLA) and expanded the health-related responsibilities of ATSDR.
CERCLA and SARA direct ATSDR to look into the health effects from substance exposures at
hazardous waste sites and to perform activities including health education, health studies,
surveillance, health consultations, and toxicological profiles.
Surface water
Water on the surface of the earth, such as in lakes, rivers, streams, ponds, and springs [compare
with groundwater].
Survey
A systematic collection of information or data. A survey can be conducted to collect information
from a group of people or from the environment. Surveys of a group of people can be conducted
by telephone, by mail, or in person. Some surveys are done by interviewing a group of people.
Toxic agent
Chemical or physical (for example, radiation, heat, cold, microwaves) agents that, under certain
circumstances of exposure, can cause harmful effects to living organisms.
Toxicological profile
An ATSDR document that examines, summarizes, and interprets information about a hazardous
substance to determine harmful levels of exposure and associated health effects. A toxicological
profile also identifies significant gaps in knowledge on the substance and describes areas where
further research is needed.
Toxicology
The study of the harmful effects of substances on humans or animals.
Uncertainty factor
Mathematical adjustments for reasons of safety when knowledge is incomplete. For example,
factors used in the calculation of doses that are not harmful (adverse) to people. These factors are
applied to the lowest-observed-adverse-effect-level (LOAEL) or the no-observed-adverse-effect­
level (NOAEL) to derive a minimal risk level (MRL). Uncertainty factors are used to account for
variations in people’s sensitivity, for differences between animals and humans, and for
differences between a LOAEL and a NOAEL. Scientists use uncertainty factors when they have
some, but not all, the information from animal or human studies to decide whether an exposure
will cause harm to people [also sometimes called a safety factor].
U.S.EPA
United States Environmental Protection Agency.
Volatile organic compounds (VOCs)
Organic compounds that evaporate readily into the air. VOCs include substances such as
benzene, toluene, methylene chloride, and methyl chloroform.
Other glossaries and dictionaries:
U.S. Environmental Protection Agency (http://www.epa.gov/OCEPAterms/)

National Library of Medicine (NIH) (http://www.nlm.nih.gov/medlineplus/mplusdictionary.html)

A-9

For more information on the work of ATSDR, please contact:
Office of Policy and External Affairs

Agency for Toxic Substances and Disease Registry

1600 Clifton Road, N.E. (MS E-60)

Atlanta, GA 30333

Telephone: (404) 498-0080

A-10

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Appendix B. Summary of Other Public Health Activities
Summary of the Agency for Toxic Substances and Disease Registry (ATSDR) Activities
Health Consultation on the Lower Watts Bar Reservoir (LWBR), February 1996. ATSDR
concluded that polychlorinated biphenyls (PCBs) detected in fish from LWBR pose a public
health concern. Frequent and long-term ingestion of fish from the reservoir poses a moderately
increased risk of cancer. It could also increase the possibility of developmental effects in infants
whose mothers consume fish regularly during gestation and while nursing. ATSDR found that
current contaminant levels in the reservoir surface water and sediment are not a public health
concern. The reservoir is safe for swimming, skiing, boating, and other recreational purposes.
Additionally, water from the municipal water systems is safe to drink. ATSDR also reported that
U.S. Department of Energy’s (DOE) selected remedial actions would protect public health.
These actions include maintaining the fish consumption advisories; continuing environmental
monitoring; implementing institutional controls to prevent disturbance, resuspension, removal, or
disposal of contaminated sediment; and providing community and health professional education
regarding PCB contamination (ATSDR 1996b).
Community and Physician Education, September 1996. To follow up on the recommendations in
the ATSDR LWBR Health Consultation, ATSDR developed community and physician
education programs on PCBs in the Watts Bar Reservoir. At a community health education
meeting in Spring City, TN on September 11, 1996, Daniel Hryhorczuk, MD, MPH, ABMT, of
the Great Lakes Center, University of Illinois at Chicago, presented on the health risk associated
with PCBs in fish. On September 12, 1996, health care providers in the vicinity of the LWBR
met for a physician and health professional education meeting at the Methodist Medical Center
in Oak Ridge. ATSDR, in collaboration with local citizens, organizations, and state officials,
developed an instructive brochure on the Tennessee Department of Environment and
Conservation’s (TDEC) fish consumption advisories for the Watts Bar Reservoir (ATSDR et al.
2000).
Coordination with other parties. Since 1992, ATSDR has consulted regularly with
representatives of other parties involved with the Oak Ridge Reservation (ORR). Specifically,
ATSDR has coordinated efforts with the Tennessee Department of Health (TDOH), TDEC, the
National Center for Environmental Health (NCEH), the National Institute for Occupational
Safety and Health (NIOSH), and DOE. This effort led to the establishment of the Public Health
Working Group in 1999, which further led to the establishment of the Oak Ridge Reservation
Health Effects Subcommittee (ORRHES). ATSDR also provided some assistance to TDOH in its
study of past public health issues. ATSDR has also obtained and interpreted studies prepared by
academic institutions, consulting firms, community groups, and other parties (ATSDR et al.
2000).
Oak Ridge Reservation Health Effects Subcommittee. In 1999, ATSDR and the Centers for
Disease Control and Prevention (CDC), under authority of the Federal Advisory Committee Act
(FACA), established the ORRHES as a subcommittee of the U.S. Department of Health and
Human Services’ Citizens Advisory Committee on Public Health Service Activities and
Research at DOE sites. The subcommittee comprised people with diverse interests, expertise,
backgrounds, and communities, as well as liaison members from federal and state agencies. It
became a forum for communication and collaboration between the citizens and those agencies
that evaluate public health issues and conduct public health activities at the ORR. To help ensure

B-1

citizen participation, the meetings of the subcommittee’s work groups were open to the public.
Everyone was invited to attend and present ideas and opinions. The subcommittee
• Served as a citizen advisory group to CDC and ATSDR and made recommendations on
matters related to public health activities and research at the ORR.
• Allowed citizens to collaborate with agency staff members and to learn more about the public
health assessment process and other public health activities.
• Helped to prioritize the public health issues and community concerns evaluated by ATSDR.
ATSDR Field Office. From 2001 to 2005, ATSDR maintained a field office in the city of Oak
Ridge. Office staff promoted collaboration between ATSDR and the communities surrounding
the ORR. Staff for example provided community members with opportunities to become
involved in ATSDR’s public health activities at the ORR.
Clinical Laboratory Analysis. In June 1992, an Oak Ridge physician reported to the TDOH and
the Oak Ridge Health Agreement Steering Panel (ORHASP) that approximately 60 of his
patients may have been exposed, either occupationally or from the environment, to several heavy
metals. The physician felt that these exposures had resulted in a number of adverse health
outcomes. Such outcomes included but were not limited to increased incidence of cancer,
chronic fatigue syndrome, neurological diseases, autoimmune disease, and bone marrow damage.
In 1992 and 1993, ATSDR and NCEH assisted with clinical laboratory support by NCEH’s
Environmental Health Laboratory for patients the Oak Ridge physician referred to Howard
Frumkin, M.D., Dr.PH., Emory University School of Public Health.
Because of patient-to-physician and physician-to-physician confidentiality, results of the clinical
analysis have not been released to public health agencies. Dr. Frumkin, however, recommended
(in an April 26, 1995 letter to the TDOH Commissioner) that one should “not evaluate the
patients seen at Emory as if they were a cohort for whom group statistics would be meaningful.
This was a self-selected group of patients, most with difficult to answer medical questions (hence
their trips to Emory), and cannot in any way be taken to typify the population at Oak Ridge. For
that reason, I have consistently urged [physician name], each of the patients, and officials of the
CDC and the Tennessee Health Department, not to attempt group analyses of these patients.”
Review of Clinical Information on Persons Living In or Near Oak Ridge. In addition to the above
Clinical Laboratory Analysis, an ATSDR physician reviewed the clinical data and medical
histories provide by the Oak Ridge physician on 45 of his patients. The purpose of this review
was to evaluate clinical information on persons tested for heavy metals and to determine whether
exposure to metals was related to these patients’ illnesses. ATSDR concluded that this case
series did not provide sufficient evidence to associate low levels of metals with these diseases.
TDOH came to the same conclusion. ATSDR sent a copy of its review to the Oak Ridge
physician in September 1992.
Health Professional Education on Cyanide. In 1996, a physician education program provided
information regarding the health effects of possible cyanide intoxication. The program was
intended to assist community health care providers in responding to health concerns expressed
by employees working at the East Tennessee Technology Park (formerly the K-25 facility).
ATSDR provided the local physicians with copies of the ATSDR Case Studies in Environmental
Medicine publication “Cyanide Toxicity” (ATSDR 1991), the NIOSH final health hazard
evaluation (Blade and Worthington 1996), and the ATSDR public health statement for cyanide

B-2

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

(ATSDR 2006d). Further, ATSDR instituted a system through which local physicians could
make patient referrals to the Association of Occupational and Environmental Clinics (AOEC).
Finally, ATSDR conducted an environmental health education session for physicians at the
Methodist Medical Center in Oak Ridge, Tennessee. The medical staff grand rounds provided
the venue for conducting this session. The workshop focused on providing local physicians and
other health care providers with information to help them diagnose chronic and acute cyanide
intoxication and to answer patient questions.
Workshops on Epidemiology. At the request of ORRHES members, ATSDR held two workshops
on epidemiology for the subcommittee. The first epidemiology workshop was presented at the
June 2001 ORRHES meeting. Ms. Sherri Berger and Dr. Lucy Peipins of ATSDR’s Division of
Health Studies provided an epidemiology overview. The second epidemiology workshop was
presented at the December 2001 ORRHES meeting and was designed to help subcommittee
members develop the skills needed to review and evaluate scientific reports. At the August 28,
2001, meeting of the Public Health Assessment Work Group (PHAWG), Dr. Peipins guided the
work group and community members through a systematic, scientific approach as they critiqued
a report by J. Mangano entitled “Cancer Mortality Near Oak Ridge, Tennessee” (Mangano
1994). Using the PHAWG critique, the ORRHES made the following conclusions and
recommendation to ATSDR.
• The Mangano paper is not an adequate, science-based explanation of any alleged anomalies
in cancer mortality rates of the off-site public.
• The Mangano paper fails to establish that radiation exposures from the ORR are the cause of
any such alleged anomalies of cancer mortality rates in the public generally.
• The ORRHES recommends to the ATSDR exclusion of the Mangano paper from
consideration in the ORR public health assessment process.
Health Education Needs Assessment. Throughout the public health assessment process, ATSDR
staff members have gathered concerns from people in the communities around the ORR.
Through a cooperative agreement with ATSDR, AOEC began a community health education
needs assessment in 2000 to aid in developing a community health education action plan. George
Washington University and MCP Hahnemann University are conducting the assessment for the
AOEC. The needs assessment will help in planning, implementing, and evaluating the health
education program for the site. It will also help health educators identify key people, cultural
norms, attitudes, beliefs, behaviors, and practices in the community—information that will aid in
developing effective health education activities. Information on the needs assessment was
presented at several ORRHES meetings.
Site visits. To better understand site-specific exposure conditions, ATSDR scientists have
conducted site visits to the ORR and visited surrounding areas numerous times since 1992. The
site visits included guided tours of the ORR operation areas, as well as tours of the local
communities to identify how community members might come into contact with environmental
contamination.
Summary of TDOH Activities
The Oak Ridge Health Agreement Steering Panel (ORHASP) is a panel of experts and local
citizens. They were appointed to direct and oversee the Oak Ridge Health Studies and provide
liaison with the community. Drawing on the findings of the Oak Ridge Health Studies and what

B-3

is generally known about the health risks posed by exposures to various toxic chemicals and
radioactive substances, ORHASP concluded that past releases from ORR were likely to have
affected the health of some people. Two groups most likely to have been harmed were 1) local
children who drank milk produced by a “backyard” cow or goat in the early 1950s and 2) fetuses
of women who in the 1950s and early 1960s routinely ate fish from contaminated creeks and
rivers downstream of ORR. For additional information on the ORHASP findings, please see the
final report of the ORHASP titled Releases of Contaminants from Oak Ridge Facilities and Risks
to Public Health (ORHASP 1999).
Feasibility of Epidemiologic Studies. TDOH and ORHASP contracted with a physician from
Vanderbilt University’s Department of Preventive Medicine to explore the feasibility of
initiating analytical (for example, case-control or cohort) epidemiological studies. These studies
would address potential health concerns in the off-site populations surrounding the ORR. A
study was released in July 1996 (Thapa 1996). It concluded that the feasibility and desirability of
initiating future analytical epidemiologic studies would be significantly influenced by the
findings of the dose reconstruction studies. Those studies would clarify the extent and magnitude
of releases and possible human exposure from past releases of radioactive iodine, mercury,
PCBs, uranium, and other radionuclides, including cesium 137 (ATSDR et al. 2000).
Public Meetings. Between January 1992 and December 1999, TDOH and ORHASP held open
meetings in Oak Ridge (more than 40 meetings), Nashville (5 meetings), Harriman (2 meetings),
and Knoxville (3 meetings). In addition, the ORHASP held two meetings in the Scarboro area to
update the residents on Phase II of the Oak Ridge Health Studies. The first meeting was held at
the Oak Valley Baptist Church in November 1995; the second meeting was held at the Scarboro
Community Center in September 1997 (ATSDR et al. 2000).
Health Statistics Review. In June 1992, an Oak Ridge physician reported to TDOH and ORHASP
that he believed approximately 60 of his patients had experienced occupational and
environmental exposures to several heavy metals. The physician suggested these exposures had
resulted in increased cancer, immunosuppression, chronic fatigue syndrome, neurologic diseases,
autoimmune disease, bone marrow damage, and hypercoagulable state including early
myocardial infarctions and stroke. In 1992, the TDOH conducted a health statistics review to
compare cancer incidence rates for the period of 1988 to 1990 for counties surrounding the ORR
to rates from the rest of the state. Review findings are in a TDOH memorandum dated October
19, 1992, from Mary Layne Van Cleave to Dr. Mary Yarbrough (Van Cleave 1992). The
memorandum details an Oak Ridge physician’s concerns about the health status of Oak Ridge
area residents. Also available from TDOH are the minutes and handouts from a December 14,
1994 presentation given by Ms. Van Cleave at the ORHASP meeting.
Health Statistics Review. In 1994, local residents reported many community members with
amyotrophic lateral sclerosis (ALS) and multiple sclerosis (MS). TDOH in consultation with
Peru Thapa, MD, MPH, from the Vanderbilt University School of Medicine, conducted a health
statistics review of mortality rates for ALS, MS, and other selected health outcomes. The August
18, 1994 ORHASP meeting minutes discuss this review.
TDOH found that because ALS and MS are not reportable diseases, it is impossible to calculate
reliable incidence rates. Mortality rates for the period of 1980 to 1992 were reviewed for the 10
counties surrounding the ORR and compared with mortality rates for the state of Tennessee. On
August 18, 1994, at the ORHASP public meeting, TDOH reported the following results.

B-4

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

• No significant ALS mortality differences surfaced in any of the counties in comparison to the
rest of the state.
• For Anderson County, the rate of age-adjusted deaths from chronic obstructive pulmonary
disease was significantly higher than rates in the rest of the state. But rates for total deaths,
deaths from stroke, deaths from congenital anomalies, and deaths from heart disease were
significantly lower for the period from 1979 to 1988. No significant differences surfaced in
the rates of deaths due to cancer for all sites in comparison with rates in the rest of state.
Rates of deaths from uterine and ovarian cancer were significantly higher than the rates in the
rest of the state. The rate of deaths from liver cancer was significantly lower in comparison to
the rest of the state.
• For Roane County for the period 1979–1988, the rates of total deaths and deaths from heart
disease were significantly lower than the rates in the rest of the state. Although the total
cancer death rate was significantly lower than the rate in the rest of the state, the rate of
deaths from lung cancer was significantly higher than the rate in the rest of the state. Rates of
deaths from colon cancer, female breast cancer, and prostate cancer were also significantly
lower than the rates in the rest of the state.
• For Knox County, the rates for total deaths and deaths from heart disease were significantly
lower than the rates in the rest of the state. TDOH found no significant difference in the total
cancer death rate in comparison to the rest of the state.
• TDOH found no significant exceedances for any cause of mortality studied in Knox, Loudon,
Rhea, and Union counties in comparison to the rest of the state.
• Rates of total deaths were significantly higher in Campbell, Claiborne, and Morgan counties
in comparison with the rest of the state.
• Cancer mortality was significantly higher in Campbell County in comparison to the rest of
the state. The excess in number of deaths from cancer appeared to be attributed to the earlier
part of period 1980–1985; the rate of deaths from cancer was not higher in Campbell County
in comparison with the rest of the state for the periods 1986–1988 and 1989–1992.
• From 1980 to 1982, cancer mortality was significantly higher in Meigs County in comparison
with the rest of the state. This excess in cancer deaths did not persist from 1983 to 1992.
Knowledge, Attitude, and Beliefs Study. TDOH coordinated a study in an eight-county area
surrounding Oak Ridge, Tennessee. The study’s purpose was to 1) investigate public perceptions
and attitudes about environmental contamination and public health problems related to the ORR,
2) ascertain the public’s level of awareness and assessment of ORHASP, and 3) make
recommendations for improving public outreach programs. The report was released in August
1994 (Benson et al. 1994). Following is a summary of the findings.
• A majority of the respondents regard their local environmental quality as better than the
national environmental quality. Most rate the quality of the air and their drinking water as
good or excellent. Almost half rate the local groundwater as good or excellent.
• A majority of the respondents think that activities at the ORR created some health problems
for people living nearby. A majority think that activities at ORR created health problems for
people who work at the site. Most feel that researchers should examine the actual occurrence

B-5

of disease among Oak Ridge residents. Twenty-five percent know of a specific local
environmental condition they believe has adversely affected public health, but many of these
appear to be unrelated to ORR. Less than 0.1 percent has personally experienced a health
problem that they attribute to the ORR.
• About 25 percent have heard of the Oak Ridge Health Study. Newspapers are the primary
source of information about the study. Roughly 33 percent rate the performance of the study
as good or excellent, and 40 percent think the study will improve public health. Also, 25
percent feel that communication about the study has been good or excellent.
Health Assessment. TDOH’s East Tennessee Region conducted a health assessment of the East
Tennessee region to evaluate the health status of the population, assess the availability and use of
health services, and develop priorities in resource allocation. In December 1991, the East
Tennessee Region released the first edition of A Health Assessment of the East Tennessee
Region, which included data generally from 1986 to 1990. The second edition, released in 1996,
included data generally from 1990 through 1995 (TDOH 1996). A copy of the document is
available from the TDOH East Tennessee Region.
Presentation. At the February 16, 1995 ORHASP public meeting, Dr. Joseph Lyon of the
University of Utah presented to ORHASP and to the public multiple studies related to fallout
from the Nevada Test Site, including the study of leukemia and thyroid disease. TDOH
sponsored the presentation.
Summary of TDEC Activities
Watts Bar Reservoir and Clinch River Turtle Sampling Survey, May 1997. For several years,
TDEC issued fish consumption advisories for the Watts Bar Reservoir warning of PCB
contamination in fish. Because of the concern regarding PCBs in fish and the recognition that
people were also eating turtles from the reservoirs, TDEC sampled snapping turtles from the
Watts Bar Reservoir and Clinch River to determine the body burdens of contaminants in the
turtles. Many agencies were consulted and involved in the project, including ATSDR, DOE,
TDOH, Tennessee Valley Authority (TVA), and the Tennessee Wildlife Resources Agency.
The results of the survey indicate that turtles in the Watts Bar Reservoir and Clinch River do
accumulate PCBs and other contaminants. Using data from the fish consumption advisories for
the area, PCB concentrations in turtle tissue were found at levels of concern for human
consumption. But as with fish, most of the PCB contamination was found in fat tissue. Methods
of food preparation, therefore, especially tissue selection, can greatly affect the amount of PCBs
consumed with the turtle meat (ATSDR et al. 2000; TDEC 1997).
Summary of Joint Center for Political and Economic Studies Activities
Scarboro Community Assessment Report. In 1999, the Joint Center for Political and Economic
Studies conducted a survey of the Scarboro community to identify environmental and health
concerns of the residents. The surveyors attempted to elicit responses from the whole community
and achieved an 82 percent response rate. Additionally, with support from DOE Oak Ridge
Operations, the Joint Center has been working with the community since 1998 to help residents
articulate their environmental, health, economic, and social needs. Because Scarboro is small, the
community assessment provided new information not available through sources such as the U.S.
Census Bureau. It also identified Scarboro’s strengths and weaknesses and illustrated the relative
unimportance of environmental health issues to other community concerns. Environmental and

B-6

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

health issues are not a priority for most Scarboro residents; rather, the community is more
concerned about crime and security, children, and economic development. The Joint Center
recommended more active community involvement in city and community planning (Friday and
Turner 2001).
Summary of CDC Activities
Scarboro Community Health Investigation, July 2000. In November 1997, a Nashville
newspaper published an article about illnesses among children living near the nuclear weapons
facility at the ORR in eastern Tennessee. The article described a high rate of respiratory illness
among residents of the nearby community of Scarboro—16 children had repeated episodes of
“severe ear, nose, throat, stomach, and respiratory illnesses.” Among those respiratory illnesses
were asthma, bronchitis, sinusitis, allergic rhinitis, and otitis media. The article implied that
exposure to the ORR caused these illnesses, especially given the proximity of these children’s’
residences to ORR facilities (Thomas et al. 1997). In response, the TDOH Commissioner asked
CDC to work with the department to investigate the Scarboro situation. TDOH coordinated the
Scarboro Community Health Investigation to investigate a reported excess of respiratory illness
among children in the Scarboro community; the investigation included a community health
survey and a follow-up medical evaluation of children less than 18 years of age (Johnson et al.
2000). Both the survey and the examination components were designed to measure the rates of
common respiratory illnesses among children who reside in Scarboro, compare these rates with
national rates, and determine any unusual characteristics of these illnesses. The investigation was
not designed to find what caused the illnesses.
In 1998, a study protocol was developed and a community health survey was administered to the
members of each household in the Scarboro community. The purpose of the survey was to
determine whether the rates of certain diseases were higher in Scarboro than elsewhere in the
United States and to determine whether exposure to various factors increased residents’ risk for
health problems. In addition, information regarding occupations, occupational exposures, and
general health concerns was collected for adults. The participation/response rate of the health
investigation survey was 83 percent (220/264 households) and included 119 questionnaires about
children living in these households and 358 questionnaires about adults.
In September 1998, CDC released the preliminary results of the survey. The asthma rate was 13
percent among children in Scarboro, compared with national estimates of 7 percent among all
children aged 0–18 years and 9 percent among African American children aged 0–18 years. The
Scarboro rate was, however, within the range of rates from 6 to 16 percent reported in similar
studies throughout the United States. The wheezing rate among children in Scarboro was 35
percent, compared with international estimates ranging from 1.6 to 36.8 percent. With the
exception of unvented gas stoves, no statistically significant association was found between
exposure to common environmental asthma triggers and asthma or wheezing illness (Johnson et
al. 2000). Common environmental asthma triggers might include pests, environmental tobacco
smoke, and the presence of dogs or cats in the home. Or they might include potential
occupational exposures such as living with an adult who works at the ORR or living with an
adult who works with dust and fumes and brings exposed clothes home for laundering. In any
event, the survey found no asthma trigger/wheezing illness link.
Using the information obtained in the health investigation survey, 36 children, including those
identified in the media report, were invited to receive a physical examination. These

B-7

examinations were conducted in November and December 1998 to confirm the results of the
community survey, to establish whether children with respiratory illnesses were getting the
medical care they needed, and to determine whether the children reported in the newspaper to
have respiratory medical problems really had these problems. Children who were invited to
participate met one or more conditions:
• Severe asthma, defined as more than 3 episodes of wheezing or visiting an emergency room
because of these symptoms;
• Severe undiagnosed respiratory illness, defined as more than 3 episodes of wheezing and
visiting an emergency room because of these symptoms;
• Respiratory illness and no regular source of medical care; or
• Identified as having respiratory illness in newspaper reports.
Of the 36 children invited, 23 participated in the physical examination. Some of the eligible 36
children had moved out of Scarboro; others either were not available or decided not to
participate.
During the physical examination, nurses asked children and their parents a series of questions
about the child’s health. Volunteer pediatricians reviewed the results of the nurse interview and
examined the children. In addition to direct physical examinations, children also underwent a
blood test and a special breathing test. If the examining doctor thought the child needed an x-ray
to complete the assessment, this was done. All examinations, tests, and transportation to and
from Knoxville were provided free of charge.
Immediately after the examinations, the results were reviewed. None of the children had findings
that needed immediate intervention. A number of laboratory tests were found to be either above
or below the normal range, such as blood calcium level, blood hemoglobin level, or breathing
test abnormality. Following the initial review of results, laboratory results were communicated
by letter or telephone to the parents of the children and their doctors. If the parents did not want
the results sent to a doctor, the results were given to the parents by telephone. The parents of
children with any health concern identified as a result of the examination were sent a personal
letter from Paul Erwin, M.D., of the East Tennessee Regional Office of the TDOH, informing
them of the need for follow-up with their medical provider. If they did not have a medical
provider, they were to contact Brenda Vowell, RNC, Public Health Nurse, East Tennessee
Regional Office of the TDOH, for help in finding a provider and possible TennCare or
Children’s Special Service.
In January 1999, a team of physicians representing CDC, TDOH, the Oak Ridge medical
community, and the Morehouse School of Medicine reviewed the findings of the physical
examinations and the community survey. Of the 23 children who were examined, 22 had
evidence of some form of respiratory illness (reported during the nurse interview or discovered
during the doctor’s examination). Overall, the children appeared healthy and no problems that
needed urgent management were identified. Several children had mild respiratory illnesses at the
time of the examination; only one child had findings of an abnormality of the lungs at the time of
the examination. None of the children had wheezing. The examinations did not indicate any
unusual pattern of illness among children in Scarboro. The illnesses that were detected were not
more severe than would be expected and were typical of those that might be found in any
community. The findings of examinations essentially confirmed the results of the community
B-8

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

health survey. The results of the review were presented on January 7, 1999, at a community
meeting in Scarboro (Johnson et al. 2000).
Three months after the letters went to the parents and physicians about the findings, attempts
were made to telephone the parents of children who participated. Eight parents were successfully
contacted. Because some of the parents had more than one child who was examined, questions
addressed the health of 14 children. Parents of nine children could not be contacted despite
attempts on several days to contact them by telephone.
Of the 14 children whose parents had been contacted, seven had seen a doctor since the
examinations. In most cases, the health of the child was the about the same, although one child
had been hospitalized because of asthma, and another child’s asthma medication had been
increased to treat a worsening asthma condition. Several children had nasal allergies, and several
parents mentioned difficulties in obtaining medicines because of cost and lack of coverage by
TennCare for the particular medicines. Health department nurses subsequently have assisted
these parents in getting the needed medicines (Johnson et al. 2000).

B-9

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Appendix C. Summary Briefs and Factsheets
• ATSDR’s Health Consultation on the Y-12 Weapons Plant Chemical Releases Into East Fork
Poplar Creek
• ATSDR’s Exposure Investigation, Serum PCB and Blood Mercury Levels in Consumers of
Fish and Turtles from Watts Bar Reservoir
• TDOH’s Phase I Dose Reconstruction Feasibility Study
• TDOH’s Task 2 Study: Mercury Releases from Lithium Enrichment at the Oak Ridge Y-12
Plant—A Reconstruction of Historical Releases and Off-Site Doses and Health Risks
• FAMU’s Scarboro Environmental Study
• U.S.EPA’s September 2001 Sampling Report for the Scarboro Community

C-1

ORRHES Brief
Oak Ridge Reservation Health Effects Subcommittee

Public Health Consultation, Y-12 Weapons Plant
Chemical Releases into East Fork Poplar Creek,
Oak Ridge, Tennessee, April 5, 1993
Site: Oak Ridge Reservation
Conducted by: Agency for Toxic
Substances and Disease Registry
Time Period: Early 1990s
Location: East Fork Poplar Creek and
Floodplain Area

Purpose
The purpose of the health consultation was to evaluate
published environmental data and to assess health
risks associated with Y-12 Weapons Plant releases at
the Oak Ridge Reservation.

Background
Between 1950 and 1963, the Department of Energy
(DOE) Y-12 Weapons Plant used mercury in a lithium
separation process. DOE officials estimate that 110
metric tons of mercury were released to the East Fork
Poplar Creek (EFPC), and that an additional 750 metric
tons of mercury used during that period could not be
accounted for. Releases of mercury to the creek con­
taminated instream sediments, and periodic flooding
contaminated floodplain soils along the creek. Land
uses along the floodplain are residential, commercial,
and recreational. Furthermore, residents used the sediment to enrich private gardens, and the city of Oak
Ridge used creek sediment as fill material on sewer
belt lines. In 1983, the state of Tennessee publicly disclosed that sediment and soil in the EFPC floodplain
were contaminated with mercury. That same year, the
Oak Ridge Task Force initiated remediation of public
and private lands within the city of Oak Ridge.
In 1992, during Phase IA of the EFPC remedial investi­
gation, DOE conducted preliminary sampling of soil,
sediment, surface water, and groundwater from the
EFPC floodplain area. During 1990 and 1991, DOE
sampled for contaminants in EFPC fish through its
Biological Monitoring and Abatement Program.

Study design and method
This was a health consultation conducted by the Agency
for Toxic Substances and Disease Registry (ATSDR).
An ATSDR health consultation is a verbal or written
response from ATSDR to a specific request for information about health risks related to a specific site, chemical release, or the presence of hazardous material. In
this case, DOE requested that ATSDR comment on the
health threat posed by past and present chemical releas­
es from the Y-12 Weapons Plant to the East Fork Poplar
Creek. To conduct the consultation, ATSDR evaluated
DOE’s preliminary environmental sampling data for
metals, volatile and semivolatile organic compounds,
radionuclides, and polychlorinated biphenyls (PCBs).
Health consultations may lead to specific actions, such
as environmental sampling, restricting site access, or
removing contaminated material, or ATSDR may make
recommendations for other activities to protect the
public’s health.

Study group
ATSDR did not conduct a study.

Exposures
ATSDR estimated human exposure to contaminated
EFPC floodplain soil, sediments, surface water,
groundwater, fish, and air.

Outcome measure
ATSDR did not review health outcome data.

Results
Only mercury in soil and sediment, and PCBs and mercury in fish, are at levels of public health concern. Other
contaminants, including radionuclides found in soil,
sediment, and surface water, are not at levels of public
health concern. Data were not available on radionu­
clides in fish.
Elevated levels of mercury, up to 2,240 parts per
million (ppm), were found in a few soil and sediment
samples from all three creek areas sampled. The mer­
cury in the EFPC soil consisted primarily of some

Y-12 Chemical Releases into EFPC

relatively insoluble inorganic forms of mercury (mer­
cury salts and metallic mercury), with less than 1% of
the mercury in organic form.
Mercury Salts in Soil
The primary routes of inorganic mercury exposure for
people (particularly for children) who fish, play, or
walk along the creek and floodplain, are through
ingestion of soil from hand-to-mouth activities and
from excessive dermal exposure. Following ingestion,
absorption of inorganic mercury compounds across the
gastrointestinal tract to the blood is low in both people
and animals. Long-term exposure to the EFPC flood­
plain soil containing elevated levels of mercury may
result in body burdens of mercury that could result in
adverse health effects. The kidney is the organ most
sensitive to the effects of ingestion of inorganic mer­
cury salts. Effects on the kidney include increased
urine protein levels and, in more severe cases, a reduc­
tion in the glomerular filtration rate, which is a sign of
decreased blood-filtering capacity.
Metallic Mercury in Soil
The metallic mercury vapor levels in the ambient air
at the three creek areas sampled are not at levels of
public health concern. However, excavation of con­
taminated soil may result in mercury vapor being
released from the soil, especially as the air tempera­
ture increases. Such releases may increase ambient air
levels of mercury vapor, which could pose a health
risk to unprotected workers and the public. Once
inhaled, metallic mercury vapors are readily absorbed
across the lungs into the blood; however, metallic
mercury is poorly absorbed through dermal and oral
routes. Exposure to mercury vapor may elicit consis­
tent and pronounced neurologic effects.
Organic Mercury in Fish
Organic mercury is the primary form of mercury found
in fish. Frequent ingestion of EFPC fish over the long
term may result in neurotoxic effects. Concentrations
of mercury in EFPC fish samples ranged from 0.08
ppm to 1.31 ppm. Studies on the retention and excre­
tion of mercury have shown that approximately 95% of
an oral dose of organic mercury is absorbed across the
gastrointestinal tract. Neurodevelopmental effects have
been seen in infants following prenatal exposure via
maternal ingestion of organic mercury in fish.
PCBs in Fish
Frequent and long-term ingestion of EFPC fish could
result in a moderate increased risk of developing can­
cer. Concentrations of PCBs in EFPC fish samples
ranged from 0.01 ppm to 3.86 ppm. PCBs are widely
distributed environmental pollutants commonly found
in blood and fat tissue of the general population. PCBs

are classified as a probable human carcinogen by the
U.S. Environmental Protection Agency. PCBs have
been shown to produce liver tumors in mice and rats
following intermediate and chronic oral exposure.
Groundwater samples collected from shallow monitor­
ing wells along the EFPC floodplain were shown to
contain elevated levels of metals and volatile organic
compounds. There was no evidence, however, that
groundwater from shallow aquifers was being used for
domestic purposes. The municipal water system, which
is used by most Oak Ridge residents, receives water
from Clinch River upstream of the DOE reservation.

Conclusions
In some locations along the creek, mercury levels in
soil and sediment pose a threat to people (especially
children) who ingest, inhale, or have dermal contact
with contaminated soil, sediment, or dust while playing,
fishing, or taking part in other activities along the
creek’s floodplain.
Mercury and PCBs were found in fish fillet samples
collected from the creek. Although people who eat fish
from the creek are not at risk for acute health threats,
people who frequently ingest contaminated fish over a
prolonged period have a moderate increased risk of (1)
adverse effects to the central nervous system and kidney
and (2) developing cancer.
ATSDR did not have enough information on groundwa­
ter use along the East Fork Poplar Creek to comment
on the contamination of groundwater in shallow, private
wells along the creek. However, contamination detected
in wells along the creek does not pose a threat to people
who receive municipal water.
ATSDR made the following recommendations.
• Determine the depth and extent of mercury contam­
ination in the EFPC sediments and floodplain soil.
• As an interim measure, restrict access to the con­
taminated soil and sediment, or post advisories to
warn the public of the hazards.
• Continue the Tennessee Department of
Environment and Conservation EFPC fish advisory.
• Continue monitoring fish from the creek for the
presence of mercury and PCBs.
• Complete the survey of well water use along the
EFPC floodplain.
• Sample shallow private wells near the creek for
PCBs, volatile organic compounds, and total and
dissolved metals.

ORRHES Brief
Exposure Investigation

Oak Ridge Reservation Health Effects Subcommittee

Exposure Investigation, Serum PCB and Blood

Mercury Levels in Consumers of Fish and Turtles

from the Watts Bar Reservoir, March 5, 1998

Site: Oak Ridge Reservation
Conducted by: ATSDR
Time period: 1997
Study area: Watts Bar Reservoir

Purpose
The purpose of this exposure investigation
was to determine whether people consuming
moderate to large amounts of fish and turtles
from the Watts Bar Reservoir were being
exposed to elevated levels of polychlorinated
biphenyls (PCBs) or mercury.

Background
Previous investigations of the Watts Bar
Reservoir and Clinch River evaluated many con­
taminants, but identified only PCBs in reservoir
fish as a possible contaminant of current health
concern. The U.S. Department of Energy (DOE)
and the Tennessee Department of Environment
and Conservation (TDEC) detected PCBs at lev­
els up to approximately 8 parts per million (ppm)
in certain species of fish from the reservoir.
PCBs were detected in turtles at levels up to 3.3
ppm in muscle tissue and up to 516 ppm in adi­
pose tissue. Mercury is a historical contaminant
of concern for the reservoir due to the large
quantities released from the Oak Ridge
Reservation. However, recent studies have not
detected mercury at levels of health concern in
surface water, sediments, or fish and turtles from
the Watts Bar Reservoir.

The 1994 DOE remedial investigation for the
Lower Watts Bar Reservoir and the 1996 DOE
remedial investigation for Clinch River/Poplar
Creek concluded that the fish ingestion pathway
had the greatest potential for adverse human
health effects. The Agency for Toxic Substance
and Disease Registry’s (ATSDR’s) 1996 health
consultation of the Lower Watts Bar Reservoir
reached a similar conclusion. These investiga­
tions based their conclusions on estimated PCB
exposure doses and estimated excess cancer risk
for people consuming large amounts of fish over
an extended period of time. Fish ingestion rates,
however, provide large uncertainty to these risk
estimates. In addition, these estimated exposure
doses and cancer risks do not consider consump­
tion of reservoir turtles because of the uncertain­
ties regarding turtle consumption.
ATSDR conducted this investigation primarily
because of the uncertainties involved in estimat­
ing exposure doses and excess cancer risk from
ingestion of reservoir fish and turtles. Also, pre­
vious investigations did not confirm that people
are actually being exposed or that they have
elevated levels of PCBs or mercury. In addition,
a contractor for the Tennessee Department of
Health (TDOH) recommended that an extensive
region-wide evaluation be conducted of relevant
exposures and health effects in counties sur­
rounding the Watts Bar Reservoir. Prior to the
initiation of such evaluations, ATSDR believed
that it was important to determine whether
mercury and PCBs were actually elevated in
individuals who consumed large amounts of
fish and turtles from the reservoir. Mercury was
included in this exposure investigation because it
was a historical contaminant of concern released
from the Oak Ridge Reservation.

Exposure Investigation

Study Design and Methods
This exposure investigation was cross-sectional
in design as it evaluated exposures of the fish
and turtle consumers at the same point in time.
However, because serum PCB and mercury
blood levels are indicators of chronic exposure,
the results of this investigation provide infor­
mation on both past and current exposure for
each study participant.
Exposure investigations are one of the approach­
es that ATSDR uses to develop better characteri­
zation of past, present, or possible future human
exposure to hazardous substances in the environ­
ment. These investigations only evaluate expo­
sures and do not assess whether exposure levels
resulted in adverse health effects. Furthermore,
this investigation was not designed as a research
study (for example, participants were not ran­
domly selected for inclusion in the study and
there was no comparison group), and the results
of this investigation are only applicable to the
participants in the study and cannot be extended
to the general population.
Specific objectives of this investigation includ­
ed measuring levels of serum PCBs and blood
mercury in people consuming moderate to large
amounts of fish or turtles, identifying appropri­
ate health education activities and follow-up
health actions, and providing new information
to help evaluate the need for future region-wide
assessments.
Study Group
The target population was persons who con­
sumed moderate to high amounts of fish and
turtles from the Watts Bar Reservoir. ATSDR
recruited participants through a variety of
means, including newspaper, radio, and televi­
sion announcements, as well as posters and fly­
ers placed in bait shops and marinas. ATSDR
representatives also made an extensive, proac­
tive attempt to reach potential participants by
telephoning several hundred individuals who
had purchased fishing licenses in the area.

ATSDR interviewed more than 550 volunteers.
Of these, 116 had eaten enough fish to be
included in the investigation. To be included in
the investigation, volunteers had to report eating
one or more of the following during the past
year: 1 or more turtle meals; 6 or more meals of
catfish and striped bass; 9 or more meals of
white, hybrid, or smallmouth bass; or 18 or
more meals of largemouth bass, sauger, or carp.
Exposures
Human exposures to PCBs and mercury from
fish and turtle ingestion were evaluated.
Outcome Measure
Outcome measures included serum PCB
and total blood mercury levels. ATSDR also
collected demographic and exposure informa­
tion from each participant (for example, length
of residency near the reservoir; species eaten,
where caught, and how prepared).
Results
The 116 participants resided in eight Tennessee
counties and several other states. The mean age
was 52.5 years and 58.6% of the participants
were male and 41.4% were female. A high
school education was completed by 65%.
Eighty percent consumed Watts Bar Reservoir
fish for 6 or more years, while 65.5% ate
reservoir fish for more than 11 years. Twenty
percent ate reservoir turtles in the last year.
The average daily consumption rate for fish or
turtles was 66.5 grams per day.
Serum PCB levels above 20 parts per billion
(ppb) were considered elevated, and only five
individuals had elevated serum PCB levels. Of
the five participants with elevated PCB levels,
four had levels between 20 and 30 ppb. One
participant had a serum PCB level of 103.8
ppb, which is higher than levels found in the
general population. None of the participants
with elevated PCB levels had any known
occupational or environmental exposures that
might have contributed to the higher levels.

Exposure Investigation

Only one participant had an elevated blood
mercury level—higher than 10 ppb. The
remaining participants had mercury levels
up to 10 ppb, which is comparable to levels
found in the general population.

Conclusions
Serum PCB levels and blood mercury levels in
participants were similar to levels found in the
general population.
Based on the screening questionnaire, most
of the people who volunteered for the study
(over 550) ate little or no fish or turtles from
the Watts Bar Reservoir. Those who did eat fish
or turtles from the reservoir indicated that they
would continue to do so even though they were
aware of the fish advisory.

Dose Reconstruction Feasibility Study

ORRHES Brief

Oak Ridge Reservation Health Effects Subcommittee

Dose Reconstruction Feasibility Study

Oak Ridge Health Study Phase I Report

Site: Oak Ridge Reservation
Study area: Oak Ridge Area
Time period: 1942–1992
Conducted by: Tennessee Department
of Health and the Oak Ridge Health
Agreement Steering Panel

Purpose
The Dose Reconstruction Feasibility Study
had two purposes: first, to identify past
chemical and radionuclide releases from the
Oak Ridge Reservation (ORR) that have the
highest potential to impact the health of the
people living near the ORR; and second, to
determine whether sufficient information
existed about these releases to estimate the
exposure doses received by people living
near the ORR.

Background
In July 1991, the Tennessee Department of
Health initiated a Health Studies Agreement
with the U.S. Department of Energy (DOE).
This agreement provides funding for an
independent state evaluation of adverse health
effects that may have occurred in populations
around the ORR. The Oak Ridge Health
Agreement Steering Panel (ORHASP) was
established to direct and oversee this state
evaluation (hereafter called the Oak Ridge
Health Studies) and to facilitate interaction
and cooperation with the community.
ORHASP was an independent panel of local
citizens and nationally recognized scientists
who provided direction, recommendations,

and oversight for the Oak Ridge Health
Studies. These health studies focused on the
potential effects from off-site exposures to
chemicals and radionuclides released at the
reservation since 1942. The state conducted
the Oak Ridge Health Studies in two phases.
Phase 1 is the Dose Reconstruction Feasibility
Study described in this summary.

Methods
The Dose Reconstruction Feasibility Study
consisted of seven tasks. During Task 1, state
investigators identified historical operations at
the ORR that used and released chemicals and
radionuclides. This involved interviewing both
active and retired DOE staff members about
past operations, as well as reviewing historical
documents (such as purchase orders, laborato­
ry records, and published operational reports).
Task 1 documented past activities at each
major facility, including routine
operations, waste management practices,
special projects, and accidents and incidents.
Investigators then prioritized these activities
for further study based on the likelihood that
releases from these activities could have
resulted in off-site exposures.
During Task 2, state investigators inventoried
the available environmental sampling and
research data that could be used to estimate
the doses that local populations may have
received from chemical and radionuclide
releases from the ORR. These data, obtained
from DOE and other federal and state
agencies (such as the U.S. Environmental
Protection Agency, Tennessee Valley

Dose Reconstruction Feasibility Study

Authority, and the Tennessee Division of
Radiological Health), were summarized by
environmental media (such as surface water,
sediment, air, drinking water, groundwater,
and food items). As part of this task,
investigators developed abstracts which
summarize approximately 100 environmental
monitoring and research projects that
characterize the historical presence of
contaminants in areas outside the ORR.
Based on the results of Tasks 1 and 2, investi­
gators identified a number of historical facility
processes and activities at ORR as having a
high potential for releasing substantial quanti­
ties of contaminants to the off-site environ­
ment. These activities were recommended for
further evaluation in Tasks 3 and 4.
Tasks 3 and 4 were designed to provide an
initial, very rough evaluation of the large
quantity of information and data identified in
Tasks 1 and 2, and to determine the potential
for the contaminant releases to impact the
public's health. During Task 3, investigators
sought to answer the question: How could
contaminants released from the Oak Ridge
Reservation have reached local populations?
This involved identifying the exposure path­
ways that could have transported contaminants
from the ORR site to residents.
Task 3 began with compiling a list of contami­
nants investigated during Task 1 and Task 2.
These contaminants are listed in Table 1.
The contaminants in the list were separated
into four general groups: radionuclides,
nonradioactive metals, acids/bases, and
organic compounds. One of the first steps in
Task 3 was to eliminate any chemicals on
these lists that were judged unlikely to reach
local populations in quantities that would pose
a health concern. For example, acids and bases
were not selected for further evaluation
because these compounds rapidly dissociate in
the environment and primarily cause acute

health effects, such as irritation. Likewise,
although chlorofluorocarbons (Freon) were
used in significant quantities at each of the
ORR facilities, they were judged unlikely to
result in significant exposure because they also
rapidly disassociate. Also, some other
contaminants (see Table 2) were not selected
for further evaluation because they were used
in relatively small quantities or in processes
that are not believed to be associated with
significant releases. Investigators determined
that only a portion of contaminants identified
in Tasks 1 and 2 could have reached people in
the Oak Ridge area and potentially impacted
their health. These contaminants, listed in
Table 3, were evaluated further in Tasks 3
and 4.
The next step in Task 3 was to determine, for
each contaminant listed in Table 3, whether a
complete exposure pathway existed. A com­
plete exposure pathway means a plausible
route by which the contaminant could have
traveled from ORR to off-site populations.
Only those contaminants with complete
exposure pathways would have the potential to
cause adverse health effects. In this feasibility
study, an exposure pathway is considered
complete if it has the following three elements:
• A source that released the contaminant
into the environment;
• A transport medium (such as air, surface
water, soil, or biota) or some combination
of these media (e.g., air ➔ pasture ➔
livestock milk) that carried the contami­
nant off the site to a location where
exposure could occur; and
• An exposure route (such as inhalation,
ingestion, or—in the case of certain
radionuclides that emit gamma or beta
radiation—immersion) through which a
person could come into contact with the
contaminant.

Dose Reconstruction Feasibility Study

In examining whether complete exposure
pathways existed, investigators considered
the characteristics of each contaminant and
the environmental setting at the ORR.
Contaminants that lacked a source, transport
medium, or exposure route were eliminated
from further consideration because they lacked
a complete exposure pathway. Through this
analysis, investigators identified a number of
contaminants with complete exposure
pathways.

In Task 5, investigators described the historical
locations and activities of populations most
likely to have been affected by the releases
identified in Task 4. During Task 6,
investigators compiled a summary of the
current toxicologic knowledge and hazardous
properties of the key contaminants.
Task 7 involved collecting, categorizing,
summarizing, and indexing selected
documents relevant to the feasibility study.
Study Group

During Task 4, investigators sought to deter­
mine qualitatively which of the contaminants
with complete exposure pathways appeared to
pose the greatest potential to impact off-site
populations. They began by comparing the
pathways for each contaminant individually.
For each contaminant, they determined which
pathway appeared to have the greatest poten­
tial for exposing off-site populations, and they
compared the exposure potential of the conta­
minant's other pathways to its most significant
pathway. They then divided contaminants into
three categories—radionuclides, carcinogens,
and noncarcinogens—and compared the
contaminants within each category based on
their exposure potential and on their potential
to cause health effects. This analysis identified
facilities, processes, contaminants, media, and
exposure routes believed to have the greatest
potential to impact off-site populations. The
results are provided in Table 4.
The Task 4 analysis was intended to provide
a preliminary framework to help focus and
prioritize future quantitative studies of the
potential health impacts of off-site contamina­
tion. These analyses are intended to provide
an initial approach to studying an extremely
complex site. However, care must be taken in
attempting to make broad generalizations or
draw conclusions about the potential health
hazard posed by the releases from the ORR.

A study group was not selected.
Exposures
Seven completed exposure pathways
associated with air, six completed exposure
pathways associated with surface water, and
ten completed exposure pathways associated
with soil/sediment were evaluated for
radionuclides and chemical substances
(metals, organic compounds, and polycyclic
aromatic hydrocarbons) released at the ORR
from 1942 to 1992.
Outcome Measures
No outcome measures were studied.

Conclusions
The feasibility study indicated that past
releases of the following contaminants have
the greatest potential to impact off-site
populations.
• Radioactive iodine
The largest identified releases of radioac­
tive iodine were associated with radioac­
tive lanthanum processing from 1944
through 1956 at the X-10 facility.
• Radioactive cesium
The largest identified releases of radioac­
tive cesium were associated with various
chemical separation activities that took
place from 1943 through the 1960s.

Dose Reconstruction Feasibility Study

• Mercury
The largest identified releases of mercury
were associated with lithium separation
and enrichment operations that were
conducted at the Y-12 facility from
1955 through 1963.
• Polychlorinated biphenyls
Concentrations of polychlorinated
biphenyls (PCBs) found in fish taken from
the East Fork Poplar Creek and the Clinch
River have been high enough to warrant
further study. These releases likely
came from electrical transformers and
machining operations at the K-25 and
Y-12 plants.
State investigators determined that sufficient
information was available to reconstruct past
releases and potential off-site doses for these
contaminants. The steering panel (ORHASP)
recommended that dose reconstruction
activities proceed for the releases of radioac­
tive iodine, radioactive cesium, mercury, and
PCBs. Specifically they recommended that the
state should continue the tasks begun during

the feasibility study, and should characterize
the actual release history of these contaminants
from the reservation; identify appropriate fate
and transport models to predict historical
off-site concentrations; and identify an
exposure model to use in calculating doses
to the exposed population.
The panel also recommended that a
broader-based investigation of operations and
contaminants be conducted to study the large
number of ORR contaminants released that
have lower potentials for off-site health effects,
including the five contaminants (chromium VI;
plutonium-239, -240, and -241; tritium; arsenic;
and neptunium-237) that could not be
qualitatively evaluated during Phase 1 due to a
lack of available data. Such an investigation
would help in modifying or reinforcing the
recommendations for future health studies.
Additionally, the panel recommended that
researchers explore opportunities to conduct
epidemiologic studies investigating potential
associations between exposure doses and
adverse health effects in exposed populations.

Dose Reconstruction Feasibility Study

TABLE 1
LIST OF CONTAMINANTS INVESTIGATED DURING TASK 1 AND TASK 2
X-10

K-25

Y-12

Radionuclides
Americium-241
Argon-41
Barium-140
Berkelium
Californium-252
Carbon-14
Cerium-144
Cesium-134,-137
Cobalt-57,-60
Curium-242,-243,-244
Einsteinium
Europium-152,-154,-155
Fermium
Iodine-129, -131, -133
Krypton-85
Lanthanum-140
Niobium-95
Phosphorus-32
Plutonium-238, -239, -240, -241
Protactinium-233
Ruthenium-103, -106
Selenium-75
Strontium-89, -90
Tritium
Uranium-233,-234, -235, -238
Xenon-133
Zirconium-95

Neptunium-237
Plutonium-239
Technetium-99
Uranium-234, -235, -238

Neptunium-237
Plutonium-239, -239, -240, -241
Technetium-99
Thorium-232
Tritium
Uranium-234, -235, -238

Beryllium
Chromium (trivalent and hexavalent)
Nickel

Arsenic
Beryllium
Chromium (trivalent and hexavalent)
Lead
Lithium
Mercury

Acetic acid
Chlorine trifluoride
Fluorine and fluoride compounds
Hydrofluoric acid
Nitric acid
Potassium hydroxide
Sulfuric acid

Ammonium hydroxide
Fluorine and various fluorides
Hydrofluoric acid
Nitric acid
Phosgene

Benzene
Carbon tetrachloride
Chloroform
Chlorofluorocarbons (Freons)
Methylene chloride
Polychlorinated biphenyls
1,1,1-Trichloroethane
Trichloroethylene

Carbon tetrachloride
Chlorofluorocarbons (Freons)
Methylene chloride
Polychlorinated biphenyls
Tetrachloroethylene
1,1,1-Trichloroethane
Trichloroethylene

Nonradioactive Metals
None initially identified

Acids/Bases
Hydrochloric acid
Hydrogen peroxide
Nitric acid
Sodium hydroxide
Sulfuric acid

Organic Compounds
None initially identified

Dose Reconstruction Feasibility Study

TABLE 2
CONTAMINANTS NOT WARRANTING
FURTHER EVALUATION IN TASK 3 AND TASK 4
Radionuclides
Americium-241
Californium-252
Carbon-14
Cobalt-57
Cesium-134
Curium-242, -243, -244
Europium-152, -154, -155
Phosphorus-32
Selenium-75
Uranium-233
Berkelium
Einsteinium
Fermium
Nonradioactive Metals
Lithium
Organic Compounds
Benzene
Chlorofluorocarbons (Freons)
Chloroform
Acids/Bases
Acetic acid
Ammonium hydroxide
Chlorine trifluoride
Fluorine and various fluoride compounds
Hydrochloric acid
Hydrogen peroxide
Hydrofluoric acid
Nitric acid
Phosgene
Potassium hydroxide
Sulfuric acid
Sodium hydroxide

Dose Reconstruction Feasibility Study

TABLE 3
CONTAMINANTS FURTHER EVALUATED IN TASK 3 AND TASK 4
Radionuclides

Nonradioactive Metals

Organic Compounds

Argon-41
Barium-140
Cerium-144
Cesium-137
Cobalt-60
Iodine-129, -131, -133
Krypton-85
Lanthanum-140
Neptunium-237
Niobium-95
Plutonium-238, -239, -240, -241
Protactinium-233
Ruthenium-103, -106
Strontium-89, 90
Technetium-99
Thorium-232
Tritium
Uranium-234 -235, -238
Xenon-133
Zirconium-95

Arsenic
Beryllium
Chromium (trivalent and hexavalent)
Lead
Mercury
Nickel

Carbon tetrachloride
Methylene chloride
Polychlorinated biphenyls
Tetrachloroethylene
1,1,1-Trichloroethane
Trichloroethylene

Dose Reconstruction Feasibility Study

TABLE 4

HIGHEST PRIORITY CONTAMINANTS, SOURCES,

TRANSPORT MEDIA, AND EXPOSURE ROUTES

Contaminant

Source

Transport Medium

Exposure Route

Iodine-131, -133

X-10
Radioactive lanthanon (RaLa)
processing
(1944-1956)

Air to vegetable to dairy
cattle milk

Ingestion

Cesium-137

X-10
Various chemical
separation processes
(1944-1960s)

Surface water to fish

Ingestion

Soil/sediment

Ingestion

Soil/sediment to vegetables;
livestock/game (beef); dairy
cattle milk

Ingestion

Air

Inhalation

Air to vegetables;
Livestock/game (beef);
dairy cattle milk

Ingestion

Surface water to fish

Ingestion

Soil/sediment to
livestock/game (beef);
vegetables

Ingestion

Surface water to fish

Ingestion

Mercury

Polychlorinated
biphenyls

Y-12
Lithium separation
and enrichment operations
(1955-1963)

K-25 and Y-12
Transformers and machining

ORRHES Brief

Oak Ridge Reservation Health Effects Subcommittee

Mercury Releases from Lithium Enrichment at the Oak Ridge

Y-12 Plant—a Reconstruction of Historical Releases and Off-

Site Doses and Health Risks, Reports of the Oak Ridge Dose

Reconstruction, Vol. 2, July 1999 (Task 2 Report)

Site: Oak Ridge Reservation
Conducted by: ChemRisk/ORHASP for the
Tennessee Department of Health
Time period: 1950 to 1990

Purpose
The purpose of the Task 2 study was to conduct
a detailed investigation of potential off-site
doses and health risks from historical releases of
mercury from the Y-12 plant on the Oak Ridge
Reservation (ORR) in Oak Ridge, Tennessee.
Specifically, the study quantified past mercury
releases from the Y-12 plant, characterized
environmental concentrations from these releases,
defined potential pathways of human exposure in
neighboring communities, and estimated human
exposure doses and human health hazards between
1950 and 1990.

Background
In July 1991, the Tennessee Department of Health
in cooperation with the U.S. Department of Energy
initiated a Health Studies Agreement to evaluate the
potential for exposures to chemical and radiological
releases from past operations at the ORR. The
Oak Ridge Does Reconstruction Feasibility
Study, conducted in 1992–1993, recommended
that dose reconstructions be conducted for several
contaminants with potential negative health effects,
including mercury releases from the Y-12 plant.
The ORR is located in eastern Tennessee,
approximately 25 miles west-northwest of
Knoxville. The Y-12 plant was built in 1945 as part
of the Manhattan Project. Located at the eastern

end of Bear Creek Valley, the Y-12 plant is within
the corporate limits of the city of Oak Ridge and is
separated from the main residential areas of the city
by Pine Ridge. The East Fork Poplar Creek (EFPC)
originates from a spring beneath the Y-12 plant and
flows northeasterly through the plant and through
residential and commercial sections of the city of
Oak Ridge.
From the early 1950s to the early 1960s, the Y-12
plant released large quantities of mercury into the
environment. These releases resulted from lithium
enrichment operations using a process known as
Colex (column-based exchange process), during
which lithium isotopes are separated by transferring
them between water-based solutions of lithium
hydroxide and lithium in mercury. Between the
early 1950s, when two large-scale production
facilities were built, and 1962, when production of
enriched lithium ceased, approximately 24 million
pounds of mercury were used. During this time, the
Y-12 plant released mercury to the air and surface
water; more than 200 individual Y-12 waste water
outfalls drained into EFPC.
In response to public concern over the potential
for adverse health effects from mercury exposure,
Y-12 mercury emissions and contamination of offsite environments have been investigated. EFPC
has been routinely sampled and analyzed for
mercury since 1953, producing what might be the
longest record of mercury release from any site in
the world. Additional investigations of the off-site
environment beginning around 1970, showed high
concentrations of mercury in soils, sediments, and
fish downstream from the Y-12 plant. For example,
in 1983, members of the Mercury Task Force
conducted an analysis of Y-12 quantified mercury
releases; which acted as the foundation for the Task
2 investigation.

Mercury Releases from Lithium Enrichment at the Oak Ridge Y-12 Plant

Methods
The project team’s review of mercury releases
and environmental concentrations began with an
examination of records assembled by members
of the 1983 Mercury Task Force. However, the
Task 2 investigation differed from the 1983
Mercury Task Force in that it 1) conducted a
more thorough records review; 2) verified data
used to calculate historical mercury releases and
adjusted the variables used to estimate mercury
releases, including ventilation rates, air and water
concentrations, and water flow rates; and 3) revised
mercury release estimates.
Additionally, the Task 2 team estimated mercury
concentrations—including elemental mercury
(the dominant form in air), inorganic mercury
(the dominant form in water, soil, and food), and
organic mercury or methylmercury (the dominant
form in fish)—in different environmental media:
●

The Task 2 team estimated mercury
concentrations in the waters of EFPC at locations
downstream from the Y-12 plant between 1950
and 1990, based on independently verified
measurements of concentrations and flow rates.
These estimates accounted for downstream
reductions in concentrations due to dilution by
additional water and mercury loss to other media
(e.g., adherence to sediment and volatilization to
air).

●

The Task 2 team calculated mercury
concentrations in air based on estimates of
annual releases from Y-12 between 1953 and
1962. Estimates of mercury concentrations in
air focused on the Wolf Valley and Scarboro
communities and were based on wind direction
and proximity to the Y-12 plant, respectively.
Mercury concentrations in air were further
examined through measurements of mercury in
tree rings of red cedars growing in the EFPC
floodplain, and by modeling the volatilization
of mercury from EFPC and the dispersion of
mercury in air to neighboring communities.

●

The Task 2 team estimated concentrations of
mercury in soil and EFPC sediment for multiple
populations based on sampling conducted as part
of the EFPC Floodplain Remedial Investigation
in 1991–1992. Mercury concentration estimates
included adjustment factors to account for

higher concentrations in the past than during
more recent data collection. Additional soil
concentration data were based on limited soil
sampling conducted in Scarboro by Oak Ridge
Associated Universities in 1984.
●

The Task 2 team also estimated concentrations
of mercury in edible plants using measurements
of airborne mercury deposition to vegetation
(samples collected near the city of Oak Ridge in
the late 1980s) and transfer of mercury from soil
to below-ground vegetables and pasture grass
(measurements collected in the Oak Ridge area
in the mid-1980s and in 1993). The project also
estimated the transfer of mercury to milk and
meat after intake by cattle based on studies from
the literature.

●

Finally, the Task 2 team estimated historical
annual consumption of fish collected from
EFPC and from locations downstream, including
the Clinch River, Poplar Creek, and Watts Bar
Reservoir, from 1950 to 1990. Estimates were
based on measured mercury concentrations in
fish collected after 1970, mercury concentrations
measured in fish at other sites with comparable
mercury levels in water and sediments, studies
of possible mercury content in live fish, and data
from sediment cores collected during the mid­
1980s.

Based on historical and current environmental
measurements, Task 2 estimated mercury doses
through all applicable exposure pathways to off-site
populations who lived near the Y-12 plant between
1950 and 1990. Dose estimates were also based on
historical release information, demographic data,
and published information on rates of intake—
either deliberate or incidental—of air, water, soil,
and food. Exposure doses to mercury in fish were
evaluated based on the number of fish meals
consumed per year: >1 to 2.5 meals/week (category
1), >0.33 to 1 meal/week (category 2), and 0.04
to 0.33 meal/week (category 3). The Task 2 team
used established toxicity benchmark values for
comparison with estimated doses, including U.S.
Environmental Protection Agency (EPA) reference
doses (RfDs), Agency for Toxic Substances and
Disease Registry (ATSDR) minimal risk levels
(MRLs), and lowest or no observed adverse effects
levels (LOAELs or NOAELs).

Mercury Releases from Lithium Enrichment at the Oak Ridge Y-12 Plant

Exposures
The Task 2 team considered multiple exposure
routes that were most likely to contribute to human
exposure to mercury, including:
●

Inhalation of contaminated air due to direct
releases from the Y-12 plant and volatilization
from EFPC.

●

Dermal contact with contaminated surface water
from EFPC.

●

Incidental ingestion of contaminated surface
water from EFPC.

●

Consumption of contaminated fish found in
EFPC, the Clinch River, Poplar Creek, and the
Watts Bar Reservoir.

●

Dermal contact with contaminated sediment and
floodplain soil from EFPC.

●

Incidental ingestion of contaminated soil.

●

Consumption of homegrown fruits and vegetables
contaminated by mercury in the air and/or soil.

●

Consumption of beef tissue and/or milk due
to local cattle consumption of pasture grass
contaminated by mercury in the air, soil, and/or
surface water.

Study Subjects
Multiple populations live in proximity to the Y-12
plant, as well as along EFPC, which flows through
residential and commercial sections of the city of
Oak Ridge. The Task 2 team identified six off-site
populations who could potentially be exposed to
mercury via one or more of the exposure pathways
identified above:
●

Oak Ridge community residents who lived near
the EFPC floodplain may have been exposed to
mercury from the air or garden-grown produce.

●

Scarboro community residents, located
approximately one-third mile north of the ORR
border, may have been exposed to mercury from
various sources due to air, water, sediment, and/
or fish contamination. Scarboro has historically
been the closest residential area to the Y-12 plant.

●

Students at the Robertsville Junior High School,
located along the banks of EFPC, may have been

exposed to mercury from air, water, sediment,
and/or soil contamination.
●

Residents of the Wolf Valley area, approximately
5 miles downwind from the Y-12 plant, may
have been exposed to mercury in direct airborne
releases from the plant.

●

Residents who lived and farmed along the EFPC
floodplain may have been exposed to mercury
from contaminated air, garden-grown produce,
dairy cattle, water, sediment, and/or fish.

●

The angler population who caught and consumed
fish from waterways downstream from the Y-12
plant, including EFPC, Poplar Creek, the Clinch
River, and the Watts Bar Reservoir may have
been exposed to mercury in the fish.

The size of potentially affected populations varied
greatly. During the Task 2 period of study, the early
1950s to early 1990s, the angler fishing population
was estimated to be less than 100 individuals.
However, the population size of the Oak Ridge
community was estimated between 15,000 and
30,000 individuals.

Results
Mercury releases from the Y-12 plant to the air and
the EFPC were found to be greater than previously
estimated by the 1983 Mercury Task Force. The
Task 2 team estimated that the Y-12 plant released
approximately 73,000 pounds of mercury to the air
during the period of enriched lithium production
(1953–1962) and 280,000 pounds of mercury to the
EFPC from 1950 to 1993—an increase of 43 and 18
percent, respectively, more than the estimates of the
1983 Mercury Task Force.
The Task 2 team assessed doses based on the type
and route of mercury exposure:
●

Air (elemental mercury): The 95% upper
confidence limit (UCL) for the estimated
elemental mercury doses from inhalation
exceeded the RfD for Scarboro community
children during the mid- to late-1950s and for
EFPC floodplain families (adults and children)
during the mid-1950s to early 1960s. The farm
families along the EFPC floodplain had the
highest estimated inhalation doses. During all
years, estimated doses for Scarboro residents

Mercury Releases from Lithium Enrichment at the Oak Ridge Y-12 Plant

were between 10 and 40 percent of the inhalation
doses estimated for farm families along the
EFPC floodplain. This difference is due to the
closer proximity of EFPC floodplain residents to
the creek. Average elemental mercury doses for
all populations during all years did not exceed
the NOAEL.
●

Ingestion and contact (inorganic mercury):
Estimated 95% UCL total inorganic mercury
doses, from all pathways except inhalation and
fish consumption, exceeded the RfD during the
mid- to late- 1950s at all communities of concern
for at least one year. Average inorganic mercury
doses for all populations during all years did not
exceed the NOAEL. At five of the six locations,
excluding the Robertsville School, estimated
doses were largely contributed to by ingestion of
contaminated homegrown produce.
For residents living in the EFPC floodplain,
estimated doses also exceeded the RfD through
the mid-1960s and early-1970s, particularly
for children. Doses to these individuals were
estimated to be high because they were
assumed to live close to EFPC on the edge
of the floodplain and to be exposed through
multiple pathways, including consumption of
contaminated produce, contact with surface
water and soil, etc. Although the EFPC
floodplain farm family population was relatively
small, between 10 and 50 individuals per year, it
is likely that mercury doses to some individuals
posed a potential health risk.

●

Ingestion of fish (methylmercury):
Estimated 95% UCL methylmercury doses
from consumption of fish exceeded the
methylmercury RfD (based on in utero
exposure) at all locations. Depending on the
number of fish meals per week, estimated doses
exceeded the RfD for several years in the 1950s
and 1960s (category 3: 0.04-0.33 meal/week) to
all years of examination, 1950-1990 (category
1: >1-2.5 meals/week). At Watts Bar Reservoir,
Clinch River, and Poplar Creek, estimated
doses for category 1 fish consumers exceeded
the RfD even at the lower bound of the annual
average dose (2.5th percentile) during multiple
years. Estimated doses for fish consumption also
exceeded the NOAEL for methylmercury (based
on in utero exposure) for category 1 consumers

from the Watts Bar Reservoir (1956–1960) and
for all categories of consumers from the Clinch
River and Poplar Creek (category 1: 1950–1975,
category 2: 1950–1964, category 3: 1957).
For all exposure pathways of interest, the highest
annual average mercury doses are estimated to
have occurred during the mid- to late-1950s. These
were the years of highest releases of mercury
from the Y-12 plant to the air and EFPC. Overall,
estimated total mercury doses to farm families
who lived near the EFPC floodplain, particularly
children, are the highest of all evaluated exposure
populations due to their proximity to the creek.
The estimated doses are due predominantly
to a combination of inhalation of volatilized
mercury from EFPC and consumption of locally
grown fruits and vegetables contaminated from
airborne mercury. Estimated total doses for other
populations are lower. For example, highest
estimated doses for Wolf Valley and Scarboro
community residents are 30- to 40-times and
9-times lower, respectively, than the highest doses
estimated for farm families living near the EFPC
floodplain. Estimated methylmercury doses to
fish consumers are also relatively high. Estimated
doses for residents consuming fish from the Clinch
River and Poplar Creek were about 4-fold higher
than doses for consumers eating fish from the
Watts Bar Reservoir.

Conclusions
Estimates of mercury releases previously reported
by the 1983 Mercury Task Force were incomplete
and have been revised by the Task 2 team to
reflect larger historic releases of mercury to the air
and surface water than previously thought.
Based on dose reconstructions, multiple exposure
pathways may have resulted in exposures to
mercury at potentially harmful annual average
doses. Specifically, the Task 2 report highlights
several high exposure-risk activities:
●

Consumption of any fish from EFPC, the Clinch
River, or Poplar creek, and consumption of more
than 3–4 meals of fish per year from the Watts
Bar Reservoir, during the mid- to late-1950s.
These limits on fish consumption are based on
childhood methylmercury exposure.

Mercury Releases from Lithium Enrichment at the Oak Ridge Y-12 Plant

●

Consumption of fruits or vegetables that grow
above-ground from backyard gardens in the
Scarboro community or within several hundred
yards of the EFPC floodplain.

●

Recreational use of EFPC (e.g., fishing and

wading) for more than 10–15 hours per year.

●

Living or attending school within several
hundred yards of the EFPC floodplain or in the
Scarboro community (from inhalation of airborne
mercury). The highest estimated elemental
(airborne) mercury doses were for children living
in these communities.

While multiple exposure pathways may have
resulted in mercury intake above the RfDs, the
likelihood of this was greatest during the period of
highest mercury releases from the Y-12 plant in the
mid-1950s to early 1960s.
Furthermore, results show that the annual average
doses through some exposure pathways were

likely insignificant, given the distance from
contamination sources, small populations sizes,
and/or low ingestion rates, even during the years
of highest mercury releases from the Y-12 plant.
Based on this information, the Task 2 team
concluded that the following behaviors were not
likely to have resulted in exposure to mercury at
annual average doses above RfDs:
●

Consumption of beef from cattle that grazed in
downwind/downstream from the Y-12 plant,

●

Consumption of produce from backyard gardens
located more than one mile from the EFPC
floodplain (excluding the Scarboro community in
the 1950s and early 1960s), and

●

Living or attending school more than one
mile from the EFPC foodplain (excluding the
Scarboro community in the 1950s and early
1960s).

ORRHES Brief
Oak Ridge Reservation Health Effects Subcommittee

Scarboro Environmental Study
Site: Oak Ridge Reservation
Conducted by: Environmental Sciences
Institute at Florida Agricultural and Mechanical
University, Environmental Radioactivity
Measurement Facility at Florida State University,
Bureau of Laboratories of the Florida Department
of Environmental Protection, Jacobs Engineering,
DOE subcontractors in the Neutron Activation
Analysis Group at Oak Ridge National Laboratory
Time Period: 1998
Location: Scarboro, Tennessee

Purpose
The purpose of the study was to address com­
munity concerns about environmental monitor­
ing in the Scarboro neighborhood.

Background
This study was conducted in response to
Scarboro community residents’ concern about the
validity of measurements taken at air monitoring
station 46 located in the Scarboro community and
external radiation results from past aerial surveys.
The study was designed to incorporate commu­
nity input and meet the requirements of an EPA
investigation of this type. The analytical compo­
nent of the study was conducted by the
Environmental Sciences Institute at Florida
Agriculture and Mechanical University (FAMU)
and its contractual partners at the Environmental
Radioactivity Measurement Facility at Florida
State University and the Bureau of Laboratories
of the Florida Department of Environmental
Protection, and by DOE subcontractors in the
Neutron Activation Analysis Group at the Oak
Ridge National Laboratory.

1

Method
Soil, sediment and surface water samples were
collected in the Scarboro neighborhood and
analyzed for mercury, radionuclides, and organ­
ic and inorganic compounds. Initial radiological
walkover surveys were conducted to identify
hot spots prior to sample collection, and some
samples were collected from these areas with
the highest radiological counts.
A total of 48 samples were collected; 40 were
surface soil samples (within top 2 inches) and 8
were sediment/surface water samples. All sam­
ples were analyzed for mercury, gross alpha/beta
content, uranium, and gamma emitting radionu­
clides. Gross alpha-beta content was conducted to
screen samples for further analysis. Gamma-ray
spectroscopy measurements were made to check
for the presence of naturally occurring and man
made radionuclides. Neutron activation analysis
was used to analyze all soil and sediment samples
for uranium isotopes (U-238 and U-235).
Approximately 10% of the samples collected
(4 soil, 1 sediment and 1 surface water sample)
were tested for the presence of analytes on the
target compound list (TCL), the target analyte
list (TAL), and Strontium-90. Alpha spec­
troscopy was also used to test these samples for
isotopes of uranium, plutonium, and thorium.
To determine whether a sample measurement
was within normal background levels, the value
was compared to the 95th percentile of the dis­
tribution of results obtained in the Background
Soils Characterization Project (BSCP) study.
Scarboro data were specifically compared to
results from the Chickamauga Bethel Valley
group in the BSCP study because this geologic
formation best approximates the geologic for­
mation underlying the Scarboro community.

The 95th percentile value is the value at or below which 95% of the samples fall in a distribution. For example, if 100
soil samples were collected and tested for mercury, and the 95th percentile value was found to be 0.5 parts per billion
(ppb), 95 of the samples would have a value of 0.5 ppb or less.

Scarboro Environmental Study

Study Subjects
No groups were studied.
Exposures
Exposures studied included mercury, gammaray emitting radionuclides, TCL organics, TAL
inorganics, Strontium-90, and uranium, thori­
um, and plutonium isotopes.
Outcome Measures
Health outcomes were not studied.

Results
Mercury: Mercury values in the Scarboro soil
samples ranged from 0.021 milligrams per kilo­
gram (mg/kg) to 0.30 mg/kg, with a median
value of 0.11 mg/kg. Two samples (192 S.
Benedict Ave and Parcel 570, Wilberforce)
exceeded the 95th percentile value for mercury
for the Bethel Valley Chickamauga Group, but
were less than the 95th percentile for the K-25
Chickamauga Group.
Mercury was not detected in surface water
samples. Mercury values in Scarboro sediment
ranged from 0.018 mg/kg to 0.12 mg/kg.
Comparison of sediment values to BSCP data
was not possible.
Gamma-ray spectroscopy measurements: Most
gamma-ray emitting radionuclides fell within the
range of expected values. In a few cases the
radioisotopes U-238 (Th-234) and U-235
exceeded the 95th percentile values for the
BSCP formations; however, the mean values for
U-235 and U-238 were within one standard
deviation of the BSCP medians. This means that,
on average, it is unlikely that uranium was pres­
ent in Scarboro soil at elevated concentrations.
Uranium Isotopic Analysis by Neutron
Activation Analysis: The average Uranium-238
value (1.39 PicoCurie per microgram (pCi/μg) for
the Scarboro samples fell within the range of val­
ues determined by both alpha spectroscopy and
gamma-ray spectroscopy in the BSCP study. The
mean ratio of uranium-235 to uranium-238 was

0.0093 + 0.0021. Five soil samples (4 in Parcel
570, and 117/119 Spellman Ave) contained U­
235/U-238 weight ratios greater than might be
expected, suggesting enrichment in uranium-235.
10% samples: Antimony, selenium, silver, sodi­
um and thallium were rarely detected in any of
the samples. Lead and zinc concentrations in
one soil sample (117/119 Spellman Avenue)
exceeded the 95th percentile for all BSCP geo­
logic formations.
The pesticides alpha-chlordane (1700 ppb),
gamma-chlordane (2800 ppb), heptachlor (190
ppb), and heptachlor epoxide (970 ppb) were
detected in one soil sample (117/119 Spellman
Avenue). No other organic contaminants were
detected in Scarboro samples.
The maximum Strontium-90 value fell within
the 95th percentile from the BSCP study.
Using alpha-spectroscopy analysis, most of the
concentrations and ratio values for uranium,
thorium, and plutonium isotopes were within
expected ranges when compared to results from
the BSCP study. However, one soil sample
(117/119 Spellman Avenue) showed enrichment
of both U-234 and U-235 relative to U-238.

Conclusions
Mercury concentrations measured in this study
ranged from 0.021 mg/kg to 0.30 mg/kg. These
values are generally within the range of values
given in the BSCP report.
Radionuclide results including total uranium
concentrations were within expected ranges.
However, approximately 10% of soil samples
showed evidence of enrichment in uranium-235.
One of 6 samples contained organic compounds
on the TCL (alpha- and gamma-chlordane, hep­
tachlor and heptachlor epoxide) above detection
limits. In this same sample, lead and zinc con­
centrations exceeded typical values obtained in
the BSCP study by a factor of two.

EPA Sampling Report for the Scarboro Community

ORRHES Brief

Oak Ridge Reservation Health Effects Subcommittee

September 2001 Sampling Report for the Scarboro

Community, Oak Ridge, Tennessee, April 2003

Site: Oak Ridge Reservation
Conducted by: U.S. EPA
Time Period: 2001
Location: Scarboro, Tennessee

Purpose
The purpose of the U.S. Environmental
Protection Agency (EPA) sampling event was to
re-sample 20% of the sampling locations investi­
gated by the Environmental Sciences Institute at
Florida Agricultural and Mechanical University
(FAMU) for the U.S. Department of Energy
(DOE) in 1998. The results of these samples
were to be compared to those collected by
FAMU. By comparing the results, EPA would:
• Verify the 1998 chemical, metal, and radio­
logical data collected and analyzed by DOE,
• Identify any substance(s) not analyzed by
DOE and evaluate those analytical data gaps,
• Determine the source(s) of uranium and
other radionuclides, and
• Evaluate whether unreasonable risk to
human health may be present.

Background
Beginning in 1997, the Scarboro Chapter of
the National Association for the Advancement
of Colored People (NAACP) contacted EPA
with concerns that the Scarboro community
was possibly being exposed to emissions from
the Y-12 plant located at DOE’s Oak Ridge
Reservation (ORR). They were concerned that
the community could be experiencing negative
health impacts.

In May 1998, DOE responded to the concerns
of the citizens by contracting with FAMU to
conduct the Scarboro Community Environmental
Study. FAMU and its contractual partners at the
Environmental Radioactivity Measurement
Facility at Florida State University, the Bureau
of Laboratories of the Florida Department of
Environmental Protection, and the Neutron
Activation Analysis Group at the Oak Ridge
National Laboratory collected and analyzed sam­
ples from 48 locations in the Scarboro communi­
ty. Forty soil and eight sediment and/or surface
water samples were collected. The results of the
Scarboro Community Environmental Study were
released in September 1998. However, EPA
states they did not receive the DOE sampling
and analysis plan for review prior to its imple­
mentation nor was EPA able to participate in or
observe the FAMU and DOE field sampling.
Therefore, to verify the FAMU and DOE’s sam­
pling, EPA developed a draft sampling plan, EPA
Proposed Sampling and Analysis Plan for the
Scarboro Community, in July 1999, and present­
ed it to the Oak Ridge Site Specific Advisory
Board at its September 1, 1999, meeting. The
EPA solicited and received comments from the
Oak Ridge community-at-large.

Methods
On September 25, 2001, representatives of
the EPA (specifically, Region 4, Science and
Ecosystem Division (SESD), Enforcement
Investigation Branch (EIB) personnel) collected
a total of 10 environmental samples from eight
separate properties within the Scarboro commu­
nity. Six surface soil samples (6 inch interval),
two sediment samples, and two surface water
samples were collected from nine separate
locations (two samples were collected at one

EPA Sampling Report for the Scarboro Community

of the eight properties). Additionally, at the
request of local residents, core soil samples (12
inch interval) were taken from two locations to
determine the depth at which uranium is pres­
ent. Sample sites were selected based on:
• The May 1998 DOE study,
• Reconnaissance performed in February 23,
1999, by SESD-EIB personnel,
• Information gathered during the February
1999 and September 2001 public meetings
held in Oak Ridge, and
• Professional judgment regarding where an
unreasonable risk to human health might be
found, if such were to exist.
All samples were collected and handled in
accordance with the EPA, Region 4, SESD’s
Environmental Investigations Standard
Operating Procedures and Quality Assurance
Manual, May 1, 1996. Surface soil was collected
using a pre-cleaned 3-inch diameter stainless
steel hand auger from the interval of 0-6 inches.
Core samples were taken at a depth of 0-12 inch­
es to determine the presence of uranium. Samples
for volatile organic compounds (VOCs) were not
homogenized prior to being placed in the sample
container. Because wading was possible in each
surface water body, surface water samples were
collected directly into the sample container, prior
to taking sediment samples. Surface water sam­
ples were not filtered in the field. Sediment sam­
ples were collected with a stainless steel scoop or
spoon and were homogenized.
The samples were analyzed by the EPA National
Air and Radiation Environmental Laboratory
(NAREL) located in Montgomery, Alabama, for
the following contaminants: radionuclides, met­
als (including mercury), VOCs, semi-volatile
organic compounds (SVOCs), pesticides, and
polychlorinated biphenyls (PCBs). In order to
evaluate the presence of lithium in the samples,
the laboratory Lithium Internal Standard for
trace metal analysis was used as evidence that
there is little, if any, lithium present in the sam­
ples collected by EPA.

In addition, personnel from the EPA, Region 4,
Office of Technical Services conducted a radia­
tion walkover (a qualitative screening) of the
areas selected for sampling to determine
whether radiation existed above background
levels. The survey was performed using a sodi­
um iodide detector and GM Pancake probe to
identify the presence of uranium isotopes and
other gamma-emitting isotopes.
Study Subjects: No groups were studied.
Exposures: No exposures were studied.
Outcome Measures: Health outcomes were not
studied.
Results: To evaluate the results of the analyti­
cal sampling EPA used the following guidance
and standards:
• Under the Safe Drinking Water Act (SDWA)
standards were created to control the level
of contaminants that are in drinking water.
EPA used this guidance for the surface
water samples that were collected.
Maximum contaminant limits (MCLs) are
legally enforceable health protective stan­
dards (National Primary Drinking Water
Standards). National Secondary Drinking
Water Standards (NSDWS) are non­
enforceable standards that provide guidance
on cosmetic effects a contaminant might
have on the quality of the water.
• Preliminary Remediation Goals (PRGs) are
risk-based values used for screening soil and
sediment samples at contaminated sites. The
PRG is a number that represents the lowest
risk level of the Comprehensive
Environmental Response, Compensation, and
Liability Act (CERCLA) protective risk range
(1×10-6 to 1×10-4) for cancer effects. For noncancer effects the PRG represents the Hazard
Index (HI) value of 1.0 (see next bullet).
• The Hazard Quotient/Hazard Index (HQ/HI)
is a ratio of the exposure level for a single
toxic substance to the reference dose of that
substance over the same exposure period.

EPA Sampling Report for the Scarboro Community

The HI is the sum of all HQ values from all
toxic substances that a person is exposed to
from a common source. A HQ or HI less
than 1.0 indicates that the exposure is not
sufficient to yield a health concern for a life­
time (70 years) of daily exposure.
• Gamma Spectroscopy was used as a screen
to analyze gamma-emitting isotopes which
indicate radioactive decay.
• Gross Alpha/Gross Beta levels were used as
a screen to determine if individual radionu­
clides should be sampled.
Radionuclides
The qualitative walkover screening did not
detect radiation above background levels. None
of the radionuclide analytical values exceeded
normal background levels, MCLs, or PRGs.
The two core samples collected from 0 to 12
inches below the ground surface indicate that
uranium levels are below the PRG or back­
ground levels within the U.S.
The uranium results indicated that there was
uncertainty associated with uranium enrichment
due to the uranium isotope levels being at
either background levels and/or detection lim­
its. However, even if there is potentially some
uranium enrichment in the uranium isotopes in
the Scarboro soil and sediment, the actual lev­
els of uranium isotopes are still within the U.S.
and Oak Ridge background ranges.
Lithium. The laboratory results could not support
a positive presence of lithium in the samples col­
lected by EPA. The evidence indicates there is lit­
tle, if any, lithium present in the samples.
Metals
All metals, including mercury, in the surface
water, sediment, and soil samples were unde­
tected or below MCLs, NSDWS, or PRGs with
the following exceptions:
• Aluminum. The NSDWS of 50-200 μg/L for
aluminum was exceeded in both surface
water samples (1,030 μg/L and 1,640 μg/L).

• Arsenic. The PRG of 0.39 mg/kg for
arsenic was exceeded in both sediment
samples (1.62 mg/kg and 5.17 mg/kg) and
four soil samples (5.64 mg/kg, 3.66 mg/kg,
4.68 mg/kg, and 6.39 mg/kg).
• Iron. The NSDWS of 300 μg/L for iron was
exceeded in both surface water samples
(769 μg/L and 1,160 μg/L). The PRG of
23,000 mg/kg for iron was exceeded in
three soil samples (23,100 mg/kg, 25,400
mg/kg, and 25,400 mg/kg).
• Manganese. The NSDWS of 50 μg/L for
manganese was exceeded in one of the sur­
face water samples (65.5 μg/L). The PRG
of 1,800 mg/kg for manganese was exceed­
ed in one soil sample (1,930 mg/kg).
VOCs and SVOCs
No VOCs were detected in the surface water
samples. The following VOCs were detected in
the soil and/or sediment samples: cyclote­
trasiloxane, benzoic acid, acetic acid, 1R-alpha­
pinene, and dodecane. The following SVOCs
were detected in the surface water, soil, or sedi­
ment samples: butyl benzyl phthalate, di-n­
butyl phthalate, and dibutyl phthalate. These
VOCs and SVOCs are generally attributed to
sampling and/or laboratory activities and are
not considered to be related to the ORR or the
Scarboro area.
Pesticides and PCBs
All pesticides and PCBs in the surface water,
sediment, and soil samples were undetected or
below MCLs, NSDWS, or PRGs with the fol­
lowing exceptions:
Alpha-chlordane and gamma-chlordane were
detected in one sediment sample (0.50 J μg/kg
and 0.75 J μg/kg, respectively). Alpha-chlor­
dane was detected in two soil samples (11
μg/kg and 14 μg/kg). Gamma-chlordane was
also detected in two soil samples (12 μg/kg and
30 μg/kg). Heptachlor was detected in one soil
sample (13 μg/kg). Heptachlor epoxide was
detected in one soil sample (11 μg/kg).

EPA Sampling Report for the Scarboro Community

Conclusions
EPA stated that the results of the analysis did not
reveal any chemicals or radionuclides at levels
that warrant a health or environmental concern.
• The level of radiation was below back­
ground levels and the radionuclide analyti­
cal values did not indicate a level of health
concern. Uranium levels in the core soil
samples were also below background lev­
els. There is no indication that lithium was
present in the analyzed samples at levels
that would warrant health concern.
• Aluminum, iron, and manganese are natu­
rally occurring in the geologic formations
of the Oak Ridge area, indicating that these
are not related to releases from DOE opera­
tions. Regardless, they are not present at
levels of health hazard.
• Arsenic has both carcinogenic and noncar­
cinogenic health effects. The HI value for
arsenic indicates that an assumed exposure
level could be above the protective level for
noncarcinogenic effects. However, the value
did not exceed the CERCLA protective risk
range (1×10-4) for its carcinogenic effects.
• The detected VOCs and SVOCs are plasti­
cizers, solvents, softening agents, and/or
column artifacts and their presence is gener­
ally attributed to sampling and/or laboratory
activities. Therefore, they are not consid­
ered to be site related and no further evalua­
tion was conducted.
• The presence of pesticides indicates possi­
ble past use by the homeowner/resident.
They are not considered to be site related
and no further evaluation was conducted.
The results of both the EPA and DOE sampling
effort are consistent in their findings. These
results confirm that existing soil, sediment, and
surface water quality pose no risk to human
health within the Scarboro community. There is
not an elevation of chemical, metal, or radionu­
clides above a regulatory health level of con­

cern. The Scarboro community is not currently
being exposed to substances from the Y-12
facility in quantities that pose an unreasonable
risk to health or the environment. The EPA does
not propose to conduct any further environmen­
tal sampling in the Scarboro community.
If additional environmental information
becomes available, EPA proposes that the fol­
lowing recommendations be implemented:
1. DOE should develop a written procedure to
receive citizen and community complaints
regarding discharges, emissions, or other
releases originating from the ORR. The proce­
dure should identify and provide for a timely
response and follow-up action. Additionally,
DOE should develop a communication strate­
gy to inform the residents and other communi­
ty members or stakeholders of its findings.
2. If additional environmental information
becomes available regarding Scarboro that
warrants an investigation by DOE, the sam­
pling plan, if developed, should be reviewed
and approved by the EPA and the Tennessee
Department of Environment and Conservation
(TDEC), as regulatory oversight agencies to
the Federal Facility Agreement (FFA).
3. Any future health investigations conducted by
DOE of the impacts of its operations on the
Scarboro or the greater Oak Ridge communi­
ty should be coordinated with the Oak Ridge
Reservation Health Effects Subcommittee
(ORRHES) of the Agency for Toxic
Substances and Disease Registry (ATSDR).
4. Upon the release of recommendations by the
ORRHES to the ATSDR, DOE, EPA, and
TDEC with stakeholder involvement will
scope the off-site (off DOE reservation)
operable unit. The results of this activity will
be the preparation of a Preliminary
Assessment/Site Inspection, which is cur­
rently planned for September 30, 2005. This
commitment is a DOE FFA milestone.

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Appendix D. Toxicologic Implications of Mercury Exposure
ATSDR’s toxicological profiles (ToxProfiles) identify and review the key peer-reviewed
literature that describes the toxicologic properties of particular hazardous substances ToxProfiles
also present other pertinent literature, but describe it in less detail than do the key studies.
ToxProfiles are not intended as exhaustive documents, but they do reference more
comprehensive sources of specialty information.
In 1999, ATSDR published an updated ToxProfile for mercury (ATSDR 1999). This document,
like all such profiles, characterizes the toxicologic and adverse health effects information for the
hazardous substance it describes. The discussion below is drawn from the updated profile for
mercury, except where otherwise noted.
What is mercury?
Mercury occurs naturally in the environment. It is found in three forms: metallic mercury (also
known as elemental mercury), inorganic mercury, and organic mercury. Metallic mercury is a
shiny, silver-white metal that is a liquid at room temperature. Metallic mercury is the elemental
or pure form of mercury—it is not combined with other elements. Metallic mercury metal is the
familiar liquid metal used in thermometers and some electrical switches. At room temperature,
some of the metallic mercury will evaporate and form mercury vapors. Mercury vapors are
colorless and odorless.
Inorganic mercury compounds occur when mercury combines with elements such as chlorine,
sulfur, or oxygen. These mercury compounds are also called mercury salts. Most inorganic
mercury compounds are white powders or crystals, except for mercuric sulfide (also known as
cinnabar), which is red and turns black after exposure to light.
When mercury combines with carbon, the compounds formed are called “organic” mercury
compounds or organomercurials. The environment contains a potentially large number of organic
mercury compounds; however, by far the most common organic mercury compound in the
environment is methylmercury. Like the inorganic mercury compounds, methylmercury is a
“salt” (for example, methylmercuric chloride). When pure, most forms of methylmercury are
white crystalline solids.
Several forms of mercury occur naturally in the environment. The most common natural forms
are metallic mercury, mercuric sulfide (cinnabar ore), mercuric chloride, and methylmercury.
Some microorganisms (bacteria and fungi) and natural processes can change the mercury in the
environment from one form to another. The most common organic mercury compound that
microorganisms and natural processes generate from other forms is methylmercury.
How can mercury enter and leave my body?
A person can be exposed to mercury from breathing in contaminated air, from swallowing or
eating contaminated water or food, or from having skin contact with mercury. Not all forms of
mercury easily enter your body, even if they come in contact with it. To know which form of
mercury you have been exposed to is important, as is by which route (air, food, or skin).
When you swallow small amounts of metallic mercury, for example from a broken oral
thermometer, virtually none (less than 0.01 percent) of the mercury will enter your body through
the stomach or intestines, unless they are diseased. When you breathe in mercury vapors,
however, most (about 80 percent) of the mercury enters your bloodstream directly from your

D-1

lungs, and then rapidly goes to other parts of your body, including the brain and kidneys. Once in
your body, metallic mercury can stay for weeks or months. When metallic mercury enters the
brain, it is readily converted to an inorganic form and is “trapped” for a long time. Metallic
mercury in the blood of a pregnant woman can enter her developing child. Most of the metallic
mercury will accumulate in your kidneys, but some metallic mercury can also accumulate in the
brain. Most of the metallic mercury absorbed into the body eventually leaves in the urine and
feces, while smaller amounts leave the body in the exhaled breath.
Inorganic mercury compounds do not generally vaporize at room temperatures as will elemental
mercury. And if inorganic mercury compounds are inhaled, they are not expected to enter your
body as easily as inhaled metallic mercury vapor. When inorganic mercury compounds are
swallowed, generally less than 10 percent is absorbed through the intestinal tract; however, up to
40 percent may enter the body through the stomach and intestines in some instances. Some
inorganic mercury can enter your body through the skin, but only a small amount will pass
through your skin compared with the amount that gets into your body from swallowing inorganic
mercury. Once inorganic mercury enters the body and gets into the bloodstream, it moves to
many different tissues. Inorganic mercury leaves your body in the urine or feces over a period of
several weeks or months. A small amount of the inorganic mercury can be changed in your body
to metallic mercury and leave in the breath as a mercury vapor. Inorganic mercury accumulates
mostly in the kidneys and does not enter the brain as easily as metallic mercury. Inorganic
mercury compounds also do not move as easily from the blood of a pregnant woman to her
developing child. In a nursing woman, some of the inorganic mercury in her body will pass into
her breast milk.
Methylmercury is the form of mercury most easily absorbed through the gastrointestinal tract
(about 95 percent absorbed). After you eat fish or other foods contaminated with methylmercury,
it enters your bloodstream easily and goes rapidly to other parts of your body. Only small
amounts of methylmercury enter the bloodstream directly through the skin. Organic mercury
compounds may evaporate slowly at room temperature and may enter your body easily if you
breathe in the vapors. Once organic mercury is in the bloodstream, it moves easily to most
tissues and readily enters the brain. Methylmercury in the blood of a pregnant woman will easily
move into the blood of the developing child and then into the child’s brain and other tissues. Like
metallic mercury, methylmercury can be changed by your body to inorganic mercury. When this
happens in the brain, the mercury can remain there for a long time. When methylmercury does
leave your body after you have been exposed, it leaves slowly over a period of several months,
mostly as inorganic mercury in the feces. As with inorganic mercury, some of the methylmercury
in a nursing woman’s body will pass into her breast milk.
How can mercury affect my health?
The nervous system is very sensitive to mercury. In poisoning incidents that occurred in other
countries, some people who ate fish contaminated with large amounts of methylmercury or seed
grains treated with methylmercury or other organic mercury compounds developed permanent
damage to the brain and kidneys. Permanent damage to the brain has also been shown to occur
from exposure to sufficiently high levels of metallic mercury. Whether exposure to inorganic
mercury results in brain or nerve damage is not as certain, given that it does not easily pass from
the blood into the brain.

D-2

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Metallic mercury vapors or organic mercury may affect many different areas of the brain and
their associated functions, resulting in a variety of symptoms. These include personality changes
(irritability, shyness, nervousness), tremors, changes in vision (constriction (or narrowing) of the
visual field), deafness, loss of muscle coordination, loss of sensation, and difficulties with
memory.
Because different forms of mercury do not all move through the body in the same way, they have
different effects on the nervous system. When metallic mercury vapors are inhaled, they readily
enter the bloodstream and are carried throughout the body and can move into the brain.
Breathing in or swallowing large amounts of methylmercury also results in some of the mercury
moving into the brain and affecting the nervous system. Inorganic mercury salts, such as
mercuric chloride, do not enter the brain as readily as does methylmercury or metallic mercury
vapor.
The kidneys are also sensitive to the effects of mercury. It accumulates in the kidneys and causes
higher exposures to these tissues, and thus more damage. If large enough amounts enter the
body, all mercury forms can cause kidney damage. If the damage caused by the mercury is not
too great, the kidneys are likely to recover once the body clears itself of the contamination.
Short-term exposure (hours) to high levels of metallic mercury vapor in the air can damage the
lining of the mouth and irritate the lungs and airways. This can cause tightness of the chest, a
burning sensation in the lungs, and coughing. Other effects from exposure to mercury vapor
include nausea, vomiting, diarrhea, increases in blood pressure or heart rate, skin rashes, and eye
irritation. Damage to the lining of the mouth and lungs can also occur from exposure to lower
levels of mercury vapor over longer periods (for example, in some occupations where workers
were exposed to mercury for many years). Most studies of humans who breathed metallic
mercury for a long time indicate that mercury from this type of exposure does not affect the
ability to have children. Studies in workers exposed to metallic mercury vapors have also not
shown any mercury-related increase in cancer. Skin contact with metallic mercury has been
shown to cause an allergic reaction (skin rashes) in some people.
In addition to kidney effects, inorganic mercury can damage the stomach and intestines. If
swallowed in large amounts, inorganic mercury can produce symptoms of nausea, diarrhea, or
severe ulcers. Effects on the heart have also been observed in children after accidentally
swallowing mercuric chloride. Symptoms included rapid heart rate and increased blood pressure.
Little information is available on the effects in humans from long-term, low-level exposure to
inorganic mercury.
Animal studies provide limited information about whether mercury causes cancer in humans
(ATSDR 1999). U.S.EPA has determined that mercuric chloride and methylmercury are possible
human carcinogens (EPA 2012a, 2012b). International Agency for Research on Cancer (IARC)
has determined that methylmercury compounds are possibly carcinogenic to humans (Group 2B),
and metallic mercury and inorganic mercury compounds are not classifiable as to their
carcinogenicity to humans (Group 3) (IARC 1997).
How can mercury affect children?
Methylmercury eaten or swallowed by a pregnant woman or metallic mercury that enters her
body from breathing contaminated air can also pass into the developing child. Inorganic mercury
and methylmercury can also pass from a mother’s body into breast milk and into a nursing

D-3

infant. Methylmercury can also accumulate in an unborn baby’s blood to a concentration higher
than the concentration in the mother.
For similar exposure routes and forms of mercury, the harmful health effects seen in children are
similar to the effects seen in adults. High exposure to mercury vapor causes lung, stomach, and
intestinal damage, and, in severe cases, death due to respiratory failure. These effects are similar
to those seen in adult groups who inhale metallic mercury vapors at work.
Children who breathe metallic/elemental mercury vapors, eat foods or other substances
containing phenylmercury or inorganic mercury salts, or use mercury-containing skin ointments
for an extended period may develop a disorder known as acrodynia, or pink disease. Acrodynia
can result in severe leg cramps; irritability; and abnormal redness of the skin, followed by
peeling of the hands, nose, and soles of the feet. Itching, swelling, fever, fast heart rate, elevated
blood pressure, excessive salivation or sweating, rashes, fretfulness, sleeplessness, weakness, or
a combination of these symptoms, may also be present. This syndrome was once thought to
occur only in children, but recent reported cases in teenagers and adults have shown that they too
can develop acrodynia.
In critical periods of development before children and fetuses are born, and in the early months
after birth, they are particularly sensitive to the harmful effects of metallic mercury and
methylmercury on the nervous system. Harmful developmental effects may occur when a
pregnant woman is exposed to metallic mercury and some of the mercury is transferred into her
developing child.
As with mercury vapors, exposure to methylmercury is more dangerous for young children than
for adults, because more methylmercury easily passes into the developing brain of young
children and may interfere with the development process. The effects on the infant may be subtle
or more pronounced, depending on the amount to which the fetus or young child was exposed.
Is there a medical test to determine whether I have been exposed to mercury?
Reliable and accurate ways to measure mercury levels in the body are available. These tests
involve taking blood, urine, or hair samples, and must be performed in a doctor’s office or in a
health clinic. Nursing women may have their breast milk tested for mercury levels, if any of the
other samples tested are found to contain significant amounts of mercury. Most of these tests,
however, do not determine the form of mercury to which you were exposed. Mercury levels
found in blood, urine, breast milk, or hair may be used to determine whether adverse health
effects are likely to occur. Mercury in urine is used to test for exposure to metallic mercury vapor
and to inorganic mercury forms. Measurement of mercury in whole blood or scalp hair is used to
monitor exposure to methylmercury. Urine is not useful for determining methylmercury
exposure. Levels found in blood, urine, and hair may be used together to predict health effects
possibly caused by the different forms of mercury.
What recommendations has the federal government made to protect human health?
The U.S. Environmental Protection Agency (U.S.EPA) and the U.S. Food and Drug
Administration (FDA) have set a limit of 2 parts inorganic mercury per billion (ppb) parts of
water in drinking water. U.S.EPA has determined that a daily exposure (for an adult of average
weight) to inorganic mercury in drinking water at a level up to 2 ppb is not likely to cause any
significant adverse health effects. FDA has set a maximum permissible level of 1 part of

D-4

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

methylmercury in a million parts (ppm) of seafood products sold through interstate commerce (1
ppm is a thousand times more than 1 ppb) (FDA 2011).
Occupational Safety and Health Administration (OSHA) regulates levels of mercury in the
workplace. It has set limits of 0.1 milligrams of mercury per cubic meter of air (mg/m3) for
organic mercury and 0.05 mg/m3 for metallic mercury vapor in workplace air to protect workers
during an 8-hour shift and a 40-hour work week. National Institute for Occupational Safety and
Health (NIOSH) recommends limiting that the amount of metallic mercury vapor in workplace
air be to an average level of 0.05 mg/m3 during a 8-hour work shift (DHHS and DOL 1978).

D-5

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Appendix E. Task 2 Pathway Discussions
The Task 2 Air Mercury Concentration Models
The earliest off-site ambient air mercury concentrations were measured in 1986, but the highest
Y-12 mercury releases to air occurred during the period from 1953 through 1962. 35 Therefore,
the Task 2 team used models to estimate historic off-site air mercury concentrations. Different
models were used to estimate air mercury concentrations for receptor populations in Wolf
Valley, Scarboro, and people living near the East Fork Poplar Creek (EFPC) floodplain.
Wolf Valley residents were chosen as an affected population. Historically, they were the closest
population to the Y-12 plant in the predominant downwind direction in the chain of valleys—
Bear Creek Valley, Union Valley, and Wolf Valley—that includes the Y-12 plant. Scarboro is
the closest residential population to the Y-12 plant, but it is separated from the Y-12 buildings by
Pine Ridge. Still, air emissions from the Y-12 plant windows, vents, and roof stacks could have
migrated over Pine Ridge.
Studies of mercury in trees growing in or near the EFPC floodplain conducted during the 1990s
suggested that EFPC was a source of significant mercury releases to the air. The Task 2 team
modeled air mercury concentrations resulting from the volatilization of mercury from the EFPC
floodplain to the following receptor locations and “near-floodplain” resident populations:
•
•
•
•
•

Scarboro community
Robertsville School
Oak Ridge community population #1
Oak Ridge community population #2
EFPC floodplain farm family

The Task 2 team considered the Scarboro community as the only receptor population whose air
was affected by both direct mercury releases to the atmosphere from the Y-12 plant and
volatilization of mercury to the air from EFPC. The Task 2 team used three models (or
combinations of models)—the U.S. Environmental Protection Agency (U.S.EPA) ISCST3
dispersion model, the x/Q model, and the EFPC volatilization model—to estimate mercury
concentrations in air at each potentially exposed community.

Wolf Valley Residents
The Task 2 team modeled air concentrations of mercury for the years from 1953 through 1962
for Wolf Valley residents using the U.S.EPA ISCST3 (EPA 1995b). This model uses a Gaussian
dispersion equation to calculate air concentrations at a remote location from the releases. It is an
appropriate model to use in relatively flat terrain.
A separate source term (mass per unit time) was estimated for each of 114 Y-12 building
emission points (windows, stacks, and vents) for each year that the buildings were known to
have been in operation. The U.S.EPA model predicted mercury concentrations in Wolf Valley
for each year from each source term. The sum of contributions from each point source resulted in
the total annual mercury air concentrations (in units of milligrams of mercury per cubic meter of
air, mg/m3) in Wolf Valley. The estimated air mercury concentrations in Wolf Valley for 1953
35

Lithium separation at the Y-12 plant using the Colex process ended in June 1963. The Task 2 team estimated air
source terms from 1953 through 1962.
E-1

through 1962 ranged from 0.0000008 to 0.000014 mg/m3 (ChemRisk 1999a). The peak value
(0.000014 mg/m3) was in 1955.
Task 2 estimated that the total uncertainty in the estimated annual average mercury
concentrations in Wolf Valley was ± 44 percent (ChemRisk 1999a). This figure included
uncertainties in the source buildings’ air mercury concentrations, emission rates from the
building sources, and in the air dispersion model.
The selection of the U.S.EPA model for this application appears to be appropriate. ATSDR
considers that Task 2 team’s reported estimates of air mercury concentrations in Wolf Valley
resulting from this model are reasonable.

Scarboro Community: Emissions from Y-12 Buildings
The Task 2 team recognized that the U.S.EPA ISCST3 dispersion model was not appropriate for
the Scarboro community—the terrain is not flat between the Y-12 plant and Scarboro. The Task
2 team considered other dispersion models but did not find any suitable models that could
adequately predict air concentrations over Pine Ridge. Consequently, Task 2 used a different
kind of model based on uranium data to estimate air mercury concentrations in Scarboro.
The model is based on the assumption that the relationship between air mercury concentrations
in Scarboro and mercury release quantities from the Y-12 plant is the same as the relationship
between air uranium concentrations in Scarboro and uranium release quantities from Y-12. If the
assumption is correct, then annual average air mercury concentrations in Scarboro can be
calculated by multiplying annual mercury release quantities times the ratio of uranium
concentrations in Scarboro divided by uranium
Model Equation
releases from the Y-12 plant. 36
C = Raa X Empirical (X/Q) (s/m3)
Task 2 designed a “custom” distribution from 20
C = Concentration of mercury at Scarboro (mg/m3)
discrete x/Q values using uranium data from 1986
Raa = Annual average release rate of mercury from Ythrough 1995 (ten x/Q values for uranium-238 and
12 (mg/s)

ten values for uranium-234/235). 37 The
Empirical X/Q (s/m3) = Annual average concentration

of uranium in Scarboro (pCi/m3)

consistency of the ratios is good for uranium2
234/235 (linear regression analysis, r = 0.97) and
Annual average release rate of uranium (pCi/s)

not as good for uranium-238 (r2 = 0.64). The data
The mathematical quantity, “empirical chi over Q” (or
are only from years with relatively low uranium
X/Q) is based on two physical quantities: Greek letter
chi (X) represents the measured air uranium
releases because we do not have data from years
concentrations in Scarboro and Q represents annual
with high releases. Among the data, the highest
uranium release rates from Y-12 to the air.
estimated annual uranium release (210 kg in
1986) was nearly 30 times smaller than the estimated amount of uranium released in 1959—the
year with the highest estimated annual air uranium release (6,200 kg). The linearity and
predictive value of the model is unknown for the years with high uranium releases (1953 through
the middle 1960s). The validity of the model is also unknown for the years when mercury
releases to air were highest (10,260 kg in 1955).

36

The xlQ model was developed for the Task 6 (Y-12 uranium) report. Additional information is provided in the
Task 6 report (ChemRisk 1999b).
37
The Task 2 report does not describe how it designed the “custom” distribution from the uranium data or what is
“custom” about the distribution.

E-2

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

The primary assumption of this model is that mercury releases to the air from the Y-12 plant
behave the same as uranium air releases from Y-12. The Task 2 report provides the following
discussion points:
• Both uranium and mercury were released to the air from a variety of locations spread over
the Y-12 site, in many cases from the same buildings. Uranium was released from short
stacks on top of buildings more often than mercury, which in turn was released more often
from windows and other ventilation sources. Therefore, uranium was generally released from
greater heights than mercury. This might have resulted in more uranium crossing over Pine
Ridge than mercury. However, uranium was released as solid particles and would likely have
experienced a higher wet and dry deposition rate than mercury. All the mercury releases were
assumed to be elemental vapor and would be expected to travel higher and further than
uranium. Therefore, more mercury could have traveled over Pine Ridge than uranium. The
Task 2 report suggests that the differences between the physical behavior of uranium and
mercury were not likely large enough to have had a significant impact on relative
atmospheric mercury concentrations in Scarboro, but there are no data that support or refute
this assumption.
• Mercury released into the air from the Y-12 plant might behave like uranium from Y-12 if
the particle sizes of mercury and uranium released were similar. Data do not support this
presumption. As a vapor, the average mercury droplet size (i.e., the geometric mean
aerodynamic diameter) would be in the vicinity of 1 micrometer (μm) or smaller. In a 1975
study of uranium operations at the Y-12 plant, the measured median airborne uranium
particle diameters, for different types of uranium operations, were between 1.1 μm and 3.3
μm (mean = 2.3 μm) (Sanders 1975). If mercury quickly became attached to other particulate
matter in the air, the similarity between the behavior of mercury and uranium in air might be
stronger. However, ATSDR found no studies that described the immediate fate and transport
of mercury releases coming from the Y-12 facilities.
Task 2 applied the custom xlQ distribution to the annual estimated airborne mercury release rates
from the Y-12 plant for the years 1953 through 1962. Annual uranium and mercury release
estimates from Y-12 were assumed to be evenly distributed over the years in question. This
calculation produced the estimated annual average mercury concentrations in air from 1953
through 1962 for the Scarboro community (see Table E-1).
The Task 2 team included the estimated air mercury concentrations for Scarboro, however the
data were not presented numerically (ChemRisk 1999a). 38 Therefore, ATSDR calculated annual
average air mercury concentrations using the minimum, mean, and maximum xlQ values from
the Task 2 report. Uncertainties in the estimated mean air mercury concentrations are bounded
by the estimated minimum and maximum concentrations (see Table E-1).

38

.Data were presented in a difficult-to-read bar chart (ChemRisk 1999a; Figure 7-2).

E-3

Table E-1. Estimated Annual Average Air Mercury Concentrations in Scarboro

Mercury Concentrations

Y-12 Mercury
Release Rates
lbs y-1

mg/sec

Minimum x/Q
(3.50E-08 sec/m3)
mg/m3

1953

1496

2.15E+01

7.53E-07

4.73E-06

1.46E-05

1954

3438

4.94E+01

1.73E-06

1.09E-05

3.36E-05

1955

22606

3.25E+02

1.14E-05

7.15E-05

2.21E-04

1956

13831

1.99E+02

6.96E-06

4.37E-05

1.35E-04

1957

5902

8.48E+01

2.97E-06

1.87E-05

5.77E-05

1958

9243

1.33E+02

4.65E-06

2.92E-05

9.03E-05

1959

7803

1.12E+02

3.93E-06

2.47E-05

7.63E-05

1960

3714

5.34E+01

1.87E-06

1.17E-05

3.63E-05

1961

2475

3.56E+01

1.25E-06

7.83E-06

2.42E-05

1962

2456

3.53E+01

1.24E-06

7.77E-06

2.40E-05

Year

Mean x/Q
(2.20E-07 sec/m3)
mg/m3

Maximum x/Q
(6.80E-07 sec/m3)
mg/m3

Source: ChemRisk 1999a

Values in mg/m3 are calculated from lbs/y.

Bold indicates the year with the highest annual average mercury concentrations in Scarboro.

The highest Y-12 air mercury releases, and therefore the highest annual average mercury
concentrations in Scarboro, were in 1955. But the annual average air mercury concentrations in
Scarboro include mercury from both the Y-12 releases and from EFPC.
We do not know whether air releases of mercury behaved like those of uranium. We do not know
whether the xlQ “custom distribution” is an accurate depiction of the relationship between the
mercury quantities released and the air mercury concentrations in Scarboro. ATSDR has no basis
for reliably evaluating the air mercury concentrations generated from this model.

Mercury Concentrations in Air Due to Volatilization from EFPC
The Task 2 team recognized that Pine Ridge partially limits the air exchange between the Y-12
plant and Oak Ridge communities, including Scarboro. Still, analyses of mercury in red cedar
core samples collected near East Tulsa Road in the EFPC floodplain in 1993 showed that air
mercury concentrations had been elevated in neighborhoods beyond Scarboro during the years of
peak mercury releases from Y-12 (see Table E-2).
Table E-2. Mercury Concentrations Detected in Tree Rings from the EFPC Floodplain
Year
1950
1951
1952
1953
1954
1955
1956

Y-12 E1

Y-12 E2

Y-12 W

EFPC-2

EFPC-3

EFPC-4

EFPC-5

EFPC-6

0.47
0.40
0.36
0.36
0.36
0.25
0.16

0.20001
0.34
0.34
0.52
0.47
0.46
0.46

0.48
0.45
0.66
0.75
1.1
0.67
0.98

5.3
5.3
5.3
7.2
7.2
7.2
7.2

1.8
1.8
1.8
1.8
2.7
2.7
2.7

ND
ND
ND
ND
ND
ND
ND

ND
ND
ND
ND
4.6
4.6
4.6

1.2
0.61
0.37
0.31
0.29
0.33
0.25

E-4

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Year

Y-12 E1

Y-12 E2

1957
0.16
0.32
1958
0.11
0.14
1959
0.11
0.10
1960
0.077
0.10
1961
0.077
0.068
1962
0.077
0.068
1963
0.042
0.043
1964
0.042
0.043
1965
0.042
0.043
1966
0.035
0.043
1967
0.033
0.043
1968
0.029
0.043
1969
0.030
0.032
1970
0.021
0.032
1971
0.019
0.032
1972
0.016
0.018
1973
0.016
0.018
1974
0.016
0.018
1975
0.016
0.018
1976
0.016
0.018
1977
0.016
0.0097
1978
0.014
0.0097
1979
0.014
0.0097
1980
0.014
0.0097
1981
0.014
0.0097
1982
0.014
0.0012
1983
0.015
0.0012
1984
0.016
0.0012
1985
0.016
0.0012
1986
0.0078
0.0012
1987
0.0067
0.0012
1988
0.0039
0.0082
1989
0.0035
0.0049
1990
0.0044
0.0043
1991
0.0022
0.0043
1992
0.0020
0.0027
1993
0.0020
0.0027
Source: ChemRisk 1999a

Units are in parts per million (ppm)

Y-12 W

EFPC-2

EFPC-3

EFPC-4

EFPC-5

EFPC-6

1.1
1.2
1.2
0.76
0.76
0.95
0.95
1.5
1.5
1.6
1.6
1.0
1.0
0.47
0.47
0.23
0.23
0.13
0.13
0.085
0.085
0.058
0.058
0.048
0.048
0.058
0.058
0.060
0.031
0.019
0.023
0.030
0.050
0.018
0.016
0.010
0.012

7.2
1.5
1.5
1.5
1.5
1.5
1.5
1.5
0.14
0.14
0.14
0.14
0.14
0.14
0.14
0.050
0.050
0.050
0.050
0.050
0.050
0.050
0.343
0.343
0.343
0.343
0.343
0.343
0.343
0.070
0.070
0.070
0.070
0.070
0.070
0.070
0.070

2.7
2.7
3.0
3.0
3.0
3.0
3.0
0.49
0.49
0.49
0.49
0.49
1.7
1.7
1.7
1.7
1.7
0.632
0.63
0.63
0.63
0.63
0.093
0.093
0.093
0.093
0.093
0.059
0.059
0.059
0.059
0.059
0.12
0.12
0.12
0.12
0.12

ND
ND
0.22
0.22
0.22
0.22
0.22
0.050
0.050
0.050
0.050
0.050
0.016
0.016
0.016
0.016
0.016
0.058
0.058
0.058
0.058
0.058
0.0040
0.0040
0.0040
0.0040
0.0040
0.0057
0.0057
0.0057
0.0057
0.0057
0.0074
0.0074
0.0074
0.0074
0.0074

5.1
5.1
0.63
0.63
0.63
0.63
0.63
0.29
0.29
0.29
0.29
0.29
0.32
0.32
0.32
0.32
0.32
0.16
0.16
0.16
0.16
0.16
0.092
0.092
0.092
0.092
0.092
0.13
0.13
0.13
0.13
0.13
0.074
0.074
0.074
0.074
0.074

0.29
0.26
0.17
0.17
0.17
0.17
0.17
0.098
0.098
0.098
0.098
0.098
0.036
0.036
0.036
0.036
0.036
0.014
0.014
0.014
0.014
0.014
0.011
0.011
0.011
0.011
0.011
0.0055
0.0055
0.0055
0.0055
0.0055
0.0014
0.0014
0.0014
0.0014
0.0014

E-5

Plants take up and release mercury through their leaves and stems—uptake of mercury through
plant roots is minimal. The Task 2 team studied mercury in tree rings in hopes of using the
quantity of mercury found in tree rings to estimate annual average air mercury concentrations for
the years represented by each ring. The Task 2 team, however, determined that the tree ring data
could not reliably predict air mercury concentrations for several reasons:
• Mercury concentrations in rings did not correlate well with mercury release quantities in
different years.
• Mercury concentrations in specific rings, corresponding to particular years, were not similar
in trees that were close together.
• Analyses of the ratios of tree ring concentrations were not consistent between different trees.
• Mercury concentrations in rings in some trees corresponding to years before the lithium
separation process was in full production were higher in some cases than in subsequent
years. 39
The Task 2 report suggested that the mercury did not remain in individual rings; it may have
migrated across rings inside the tree. Therefore, the Task 2 team could not reliably assign the
measured mercury concentrations to specific years. As a result, the Task 2 team abandoned its
effort to estimate historic air mercury concentrations from tree core samples. Therefore, the Task
2 report modeled air mercury concentrations from the volatilization of mercury from the
floodplain.
The Task 2 team looked at EFPC floodplain soil emissions. Data collected in 1993 during a
study of the EFPC floodplain indicated that mercury concentrations in the air directly over
mercury-contaminated soil were 340 times lower than air mercury concentrations directly over
EFPC water (Lindberg et al. 1995). 40 Task 2 also reviewed studies in the scientific literature and
concluded that mercury emissions from EFPC soils were insignificant compared with mercury
emissions from EFPC water. Therefore, the Task 2 team modeled mercury in air originating from
EFPC surface water only.
The Task 2 team modeled air mercury concentrations from the volatilization of mercury from
EFPC to the following five potentially exposed communities:
•
•
•
•
•

Scarboro community
Robertsville School
Oak Ridge community population #1
Oak Ridge community population #2
EFPC floodplain farm family

The Task 2 team estimated the amount of mercury that volatilized from EFPC by dividing the
entire length of EFPC into 403 theoretical rectangular segments, each with a width of 15 meters
and a length between 15 and 140 meters (see Figure E-1) (ChemRisk 1999a). The Task 2 team
assumed the volatilization rate was constant throughout EFPC. But the starting mass of mercury
39

The Task 2 team indicated that mercury concentrations in the tree ring corresponding to 1938—before the
Manhattan Project began—was higher than in subsequent years in a tree on the west end of Y-12 property
(ChemRisk 1999a).
40
Concentrations of mercury in air over water were modeled; concentrations in air over soil were measured. These
data were from separate studies.

E-6

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

at each segment was the amount released from the Y-12 plant less the amount of mercury lost
from the water from each of the preceding upstream segments. Therefore, the amount of mercury
that volatilized from each segment was a function of its distance from Y-12. No adjustments in
the volatilization fraction were made for the variations in the creek flow.
The estimated mass of mercury lost from each water segment (in grams of mercury discharged to
air per year [g/y]) was used as a line source term in version 96113 of the U.S.EPA’s ISCST3
dispersion model (EPA 1995b). The dispersion model calculated air mercury concentrations at
the various potentially exposed communities. In the dispersion model the Task 2 team used 1987
meteorological data from the Y-12 East Meteorological station. The Task 2 team included an
uncertainty factor to account for uncertainty in the air dispersion model, but did not include a
factor for the uncertainty or variability in the meteorological data. The amount of mercury that
was released into EFPC at the Y-12 plant is provided by the annual source terms for Y-12
mercury releases to water.
The fraction of mercury that will volatilize from EFPC depends on the amount of dissolved
gaseous mercury (DGM) in the water, as well as the physical conditions of the water and the
adjacent air. DGM is dissolved elemental mercury; it is the only mercury species in water that
will significantly volatilize from water. Elemental mercury is only slightly soluble in water (56
μg/L at 25° C), but supersaturation (the build-up of DGM beyond its equilibrium concentration)
has often been documented in environmental water systems. Conditions in the water, such as the
water temperature, pH, stream flow, and mixing of the water column may favor either the loss of,
or the formation of, DGM. Higher temperatures and higher wind currents at the water surface,
for example, will increase the volatilization of DGM from the water to air. Water agitation and
air flow at the water’s surface may significantly affect the propensity of DGM to overcome
surface energy barriers to volatilization (Saouter et al. 1995). Higher pH will favor the reduction
(chemical conversion) of mercuric forms of mercury to elemental mercury, while lower pH will
favor the oxidation (chemical conversion) of elemental mercury to mercurous and mercuric
species. The presence of minerals and organic matter in the water favor the oxidation of
elemental mercury and the removal of DGM from the water. Finally, DGM may be formed either
biotically (mediated by microscopic organisms) or abiotically (occurring chemically without
microscopic organisms) in the water.
Measurements of DGM in EFPC during the 1950s are not available. The only data that
characterize stream conditions, available from the 1950s, are some pH and flow measurements.
The pH values and the flow volumes during the 1950s, as well as the many curves in the EFPC
bed, would generally favor the formation and volatilization of mercury. But these data are
insufficient to estimate with any precision or known accuracy the amount of mercury that
volatilized. The magnitude of their effects or those of competing processes occurring in the creek
are not known.
For the volatilization fraction, the Task 2 team assumed a distribution of values: a minimum, a
best estimate, and a maximum value equal to 1, 5, and 30 percent, respectively, of the total
mercury mass released annually to the creek. The Task 2 team derived these percentages from a
1995 published study of Reality Lake—a settling pond within EFPC on the Y-12 property. But
the Task 2 report did not present the derivation of these numbers, and the study does not clearly
support the range of values the Task 2 team selected (Saouter et al. 1995).

E-7

Source: ChemRisk 1999a

E-8

Figure E-1. Conceptual Model for Mercury Releases from EFPC

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Although the Task 2 team also assumed the minimum, best estimate, and maximum fractions
from a logtriangle distribution, it provided no justification for that choice. A logtriangle
distribution provides greater weight to the lower concentration estimates and less weight to the
higher ones. ATSDR has seen no evidence to favor one portion of the distribution of the
volatilization fractions over any other portion. For example, we do not know the time-average
distribution of wind patterns at the water surface, or the pattern of variability of DGM
concentrations in EFPC during a typical year during the 1950s. Mercury may have volatilized
less frequently in the high-lying volatilization fractions than in the low-lying fractions, but no
evidence supports such an assumption.
George R. Southworth, affiliated with the Oak Ridge National Laboratory, estimated that the
EFPC mercury evasion rate may be about 3 percent of the total mercury flux over the length of
EFPC (Southworth GR, personal communication, February 14, 2005). Southworth based his
calculation on the amount of DGM
EFPC Mercury Evasion Rate Calculations
measured in EFPC in 1997 as well as
Mean dissolved gaseous mercury in EFPC (summer, 1997) = 1.1 ng/L =
estimates of the total mercury in the
0.0011 ng/cm3.
water and the surface area of EFPC.
Mass transfer coefficient = 10 cm/h.
Southworth’s calculations appear in
Evasion flux = 0.0011 ng/cm3 × 10 cm/h1 = 0.011 ng/cm2/h = 110
the text box to the right:
2
ng/m /h.

Southworth emphasized that the
volatilization fraction he calculated
(3 percent) is imprecise. It depends
on many variables that can vary
widely and are not well determined.

EFPC surface area = length × width = 25,000 m × 10 m = 250,000 m2.
Total surface flux = creek area × evasion flux = 27.5 mg/h = 660 mg/d.
Hg flux through the creek = 22 g/d.
Therefore, 660 mg/d - 22 g/d = 0.03 or 3 percent.

The Task 2 team’s best estimate value of 5 percent is similar to Southworth’s estimate of 3
percent. Still, they are both based on 1990s data. Between the 1950s and 1990s, many changes
occurred at the Y-12 facilities that affected what was released into EFPC. To determine whether
either value accurately predicts mercury volatilization from EFPC during the 1950s is
impossible. Similarly, no evidence supports the assumption that the fraction of total mercury in
the creek that volatilized was similar in both decades.
The minimum best estimate and maximum volatilization fractions generated three source terms
for each segment of EFPC for each year and produced three air mercury concentrations at each
potentially exposed community for each year.
The highest estimated mercury releases from the Y-12 plant to EFPC, and consequently the
highest air mercury emissions from EFPC, occurred in 1957. 41 The Task 2 team estimated air
mercury concentrations for each of the five potentially exposed communities using the 5 percent
mercury volatilization fraction (see Table E-3; ChemRisk 1999a). The mercury concentrations in
Table E-3 for the Scarboro community do not include the contribution from the xlQ model. Table
E-4 presents the combined air mercury concentrations for the Scarboro community.

41

The highest air mercury concentration in the Scarboro community occurred in 1955, due to a significant
component from the xlQ model for that year.

E-9

Table E-3. Estimated Air Mercury Concentrations (mg/m3)1

1

Year

EFPC
Floodplain
Farm Family

Scarboro
Community

Robertsville
School

Oak Ridge
Location 1

Oak Ridge
Location 2

1953

6.4E-05

6.5E-06

4.3E-06

2.2E-06

1.1E-06

1954

3.8E-05

3.9E-06

2.6E-06

1.3E-06

6.3E-07

1955

1.9E-04

2.0E-05

1.3E-05

6.6E-06

3.2E-06

1956

1.6E-04

1.6E-05

1.1E-05

5.4E-06

2.6E-06

1957

3.9E-04

4.0E-05

2.6E-05

1.3E-05

6.5E-06

1958

3.5E-04

3.5E-05

2.3E-05

1.2E-05

5.8E-06

1959

1.0E-04

1.0E-05

6.9E-06

3.5E-06

1.7E-06

1960

3.8E-05

3.8E-06

2.5E-06

1.3E-06

6.3E-07

1961

3.6E-05

3.6E-06

2.4E-06

1.2E-06

5.9E-07

1962

2.5E-05

2.5E-06

1.7E-06

8.4E-07

4.1E-07

1963

1.7E-05

1.7E-06

1.1E-06

5.6E-07

2.8E-07

Estimates are based on the volatilization of mercury at five receptor locations from EFPC.

Table E-4. Combined Estimated Air Mercury Concentrations for Scarboro (mg/m3)1

1

Year

EFPC (5% vf)

x/Q (mean)

Sum

% due to x/Q

1953

6.5E-06

4.7E-06

1.1E-05

42%

1954

3.9E-06

1.1E-05

1.5E-05

74%

1955

2.0E-05

7.2E-05

9.1E-05

78%

1956

1.6E-05

4.4E-05

6.0E-05

73%

1957

4.0E-05

1.9E-05

5.8E-05

32%

1958

3.5E-05

2.9E-05

6.4E-05

45%

1959

1.0E-05

2.5E-05

3.5E-05

70%

1960

3.8E-06

1.2E-05

1.6E-05

75%

1961

3.6E-06

7.8E-06

1.1E-05

68%

1962

2.5E-06

7.8E-06

1.0E-05

76%

1963

1.7E-06

1.7E-06

Estimated concentrations are from both the xlQ model and the volatilization of mercury from EFPC.

EFPC =
xlQ
Sum
%

the Task 2 air mercury concentration from the volatilization of EFPC using a volatilization fraction
of 5 percent
= the Task 2 air mercury concentration from Y- 12 air mercury releases using the Task 2 xlQ model
and the mean xlQ value
= EFPC

+ x/Q columns
= the percentage which the xlQ-derived concentration is of the whole (sum)

E-10

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

The Task 2 Water Model for Mercury Concentrations in EFPC
The Task 2 team developed a model to estimate water mercury concentrations at different
locations along EFPC from 1950 through 1990. The Task 2 team did not estimate exposures to
mercury in surface water downstream from EFPC—the water mercury concentrations below
EFPC were considered insignificant.
Task 2 selected the following potentially exposed communities for exposures to surface water:
•
•
•

Scarboro community
Robertsville School students
EFPC floodplain farm family42

The Task 2 team estimated mercury concentrations in water for each of the three populations by
selecting areas along EFPC corresponding to the closest populations. The Task 2 EFPC mile
marker locations corresponding to the Scarboro community, Robertsville School students, and
the EFPC floodplain farm family are EFPC Mile 14, Mile 12, and Mile 10, respectively 43 (see
Figure 15).
The basis of the Task 2 model was the average annual water mercury release estimates and
additional specific water mercury concentration data generated or compiled by the Task 2 team.
The annual release estimates were calculated from data available in ORR weekly, monthly, and
quarterly environmental reports. The reported data include mercury concentrations in weekly
composite water samples
Task 2 Equations for Calculating EFPC Mercury Concentrations at each
collected in EFPC on the
Reference Location
Y-12 property and weekly
Cref (mg/L) = CY-12 (mg/L) x Water Concentration Ratio
average flow volumes of
Where:
EFPC on Y-12 property.
The reported monthly and
Cref = Mercury concentration in water at a population reference location
quarterly data are
CY-12 = Mercury concentration in water at Y-12
averages calculated from
Water Concentration Ratio = Dilution Ratio × (1 - fraction lost to other compartments)
the weekly data. Not all
The Dilution Ratio, estimated from the size of the drainage basin at Cref, is:
the data are available for
Dilution Ratio = Y-12 discharge volume (in cubic feet per second, cfs)
all the time periods. In
Y-12 discharge volume (cfs) + EFPC inflow volume (cfs)
addition to measurements
Larger volumes of runoff to EFPC result in a smaller dilution ratio. The smaller the
at Y-12, Oak Ridge
dilution ratio, the smaller the water concentration ratio and the more the water mercury
personnel collected water
concentration is reduced downstream at reference locations (Cref) compared with the
samples on or close to a
concentration at Y-12 (CY-12).
weekly basis between
1955 and 1961, just upstream of the confluence of EFPC with Poplar Creek. The samples from
EFPC near the Poplar Creek confluence contained between 1 and 60 percent (average = 11
percent) of the estimated mercury concentrations in EFPC directly below the discharge point at
the Y-12 plant during the same time period.

42

Despite that EFPC does not run through the Scarboro community, the Task 2 team thought children from Scarboro
might have played in or near EFPC.
43
Mile marker numbers increase from the juncture of EFPC with Poplar Creek (EFPC Mile 0) up to the source of
EFPC at the Y-12 plant (EFPC Mile 14.4).

E-11

The Task 2 team assumed some of the difference in the mercury concentrations in water at each
end of EFPC was due to dilution, and some was due to the loss of mercury to soil, sediment, and
air. Task 2 first estimated the portion of the difference that was due to dilution and attributed the
remainder of the difference to the loss of mercury to soil, sediment, and air.
The Task 2 team obtained information about the area of the drainage basin from a 1985 study by
the Tennessee Valley Authority (TVA 1985a) and about the percent of precipitation runoff to
EFPC from a 1967 U.S. Geological Survey (USGS) report. TVA (1985a) divided the drainage
basin into sections along EFPC according to the location of tributaries that feed surface water
runoff into the creek. The Task 2 team calculated drainage basin areas for each potentially
exposed community by interpolating between the nearest drainage areas in the TVA study for
each potentially exposed community along EFPC. With annual precipitation data obtained from
USGS (1967), Task 2 calculated inflow volumes at each of the three potentially exposed
communities for each year from 1950 to 1990. Task 2 used these data to estimate the effect of
dilution on mercury concentrations at the potentially exposed communities along EFPC and for
the creek as a whole.
The Task 2 team used average Y-12 release volumes 44 for the 24 calendar quarters from 1956
through 1961, the drainage basin data, and the precipitation data. The Task 2 team estimated that
the volume flow at the EFPC-Poplar Creek junction increased approximately 3.6 times over the
volume flow at the Y-12 plant. This is an average dilution ratio of 0.26 (range: 0.15–0.42) over
the expanse of EFPC.
The Task 2 team estimated that on average, EFPC lost about 58 percent (range: -160 to 97) of
mercury from water to sediment and air for each of the 24 calendar quarters. The -160 percent and
two other negative values occurred during 1956 and the first quarter of 1957. Negative values
indicate no losses of mercury to sediment and air, and surface water runoff had less effect than
proposed. Or that less surface water runoff occurred than estimated. But this is counterintuitive—
it indicates that the validity and, therefore, the results of the model are in question for those
quarters. The average mercury loss estimates for the remainder of 1957 through 1961 (ignoring
the earlier, inconsistent data), over the expanse of the creek, was 79 percent.
In 1984, TVA collected 141 soil core samples from 30 transects across EFPC (TVA 1985b).
From the core data, TVA estimated that the total mass of mercury in the EFPC sediment and
EFPC floodplain was 157,000 pounds. This mass is approximately 57 percent of the estimated
275,000 pounds of mercury that the Task 2 team estimated the Y-12 plant had released to EFPC
from 1953 through mid-1984. This result is roughly the same as the 79 percent mercury mass the
Task 2 team estimated using the water model above. Both estimates suggest a large fraction of
the mass of mercury released from the Y-12 plant was lost to sediments, with only a small
fraction of mercury lost to air. The Task 2 team also referenced a study that showed more than
99 percent of mercury transported in surface water was associated with the solid phase
(particulate matter or sediment) (Lindberg et al. 1991).
From these analyses, Task 2 assumed that EFPC water lost 70 ± 30 percent of its mercury mass
to other environmental compartments (soil, sediment, and air) over the full length of the creek.
This number is not an exact numerical derivation—it includes a relatively large degree of
uncertainty.
44

Water released to EFPC in cubic feet per second (cfs).

E-12

Y-12 Plant
EFPC Mile 14.7
(mg/L)

Scarboro
EFPC Mile 14
(mg/L)

Robertsville School
EFPC Mile 12
(mg/L)

EFPC Floodplain
EFPC Mile 10
(mg/L)

1958

2.330

2.037

1.505

1.092

1959

0.680

0.601

0.418

0.304

1960

0.240

0.213

0.139

0.101

1961

0.200

0.175

0.122

0.086

1962

0.120

0.107

0.075

0.055

1963

0.086

0.078

0.057

0.044

1964

0.044

0.039

0.026

0.019

1965

0.095

0.083

0.057

0.041

1966

0.043

0.039

0.028

0.020

1967

0.031

0.026

0.017

0.012

1968

0.005

0.005

0.003

0.002

1969

0.006

0.005

0.004

0.003

1970

0.026

0.022

0.016

0.011

1971

0.006

0.005

0.004

0.003

1972

0.001

0.001

0.001

0.000

1973

0.065

0.054

0.033

0.023

1974

0.015

0.013

0.075

0.005

1975

0.001

0.001

0.001

0.000

1976

0.001

0.001

0.001

0.000

1977

0.002

0.002

0.001

0.001

1978

0.001

0.001

0.001

0.000

1979

0.002

0.002

0.001

0.001

1980

0.002

0.002

0.001

0.001

1981

0.002

0.002

0.001

0.001

1982

0.002

0.003

0.002

0.001

1983

0.002

0.002

0.001

0.001

1984

0.002

0.001

0.001

0.001

1985

0.003

0.002

0.001

0.001

1986

0.000

0.002

0.001

0.001

1987

0.008

0.003

0.002

0.001

1988

0.002

0.002

0.001

0.001

1989

0.002

0.001

0.001

0.001

1990

0.002

0.001

0.001

0.001

Year

1

Concentrations for 1950, 1951, and 1952 were calculated using the percentages in Table E-5. Task 2 did not
calculate “dilution only” concentrations for those years.

E-14

Discussion of the Task 2 Water Model
The Task 2 team developed the water model as a method of estimating average annual water
mercury concentrations. The estimated water mercury concentrations were then used to calculate
average annual mercury exposure doses. These dose estimates, however, should be used with
caution: predicted concentrations of mercury in water are not always reliable and the model is
not sufficiently precise to evaluate the more important short-term exposures.
The Task 2 model includes three assumptions: 1) Over the length of EFPC, mercury
concentrations decrease due to dilution and due to mercury loss from water to soil, sediment, and
air; 2) Between 40 and 90 percent of the mercury mass released from the Y-12 plant to EFPC
was lost from the water to soil, sediment, and air over the full length of EFPC, and 3) the loss of
mercury to other environmental compartments is linear with the distance from the Y-12 plant.
But the data suggest more is going on than just dilution and linear loss of mercury mass.
Task 2 derived the mercury mass partition value (70 ± 30 percent of the mass of mercury lost to
sediment and air over the length of EFPC) using mercury concentration data from both ends of
EFPC. This partition range is very broad and has limited interpretive value. The two water
mercury concentration data sets are minimally correlated, even when they are adjusted for
changes in water volume (correlation coefficient [r] = 0.37). The mercury concentrations at the
Poplar Creek end of EFPC are inconsistent relative to the concentrations at the Y-12 plant,
probably because of many significant chemical and physical processes affecting the dissolved
mercury mass during its transport through the creek. The exchange of mercury between water
and other compartments (sediment and air, for example) is complex and may depend on many
variables such as water temperature, flow rates, turbulence, amount of precipitation, surface
runoff, amount and types of mercury in “storage depots” in the floodplain soils and sediments,
and the quantity and physical properties of organic and particulate matter present. These
processes are not quantitatively characterized in the scientific literature. The low correlation of
the data means the model is not predictive. The lack of accuracy of the model was demonstrated
by its failure to predict sizeable mercury losses for three calendar quarters in 1956 and 1957.
The Task 2 model-estimated mercury concentrations are also limited—they are annual averages.
ATSDR notes that the longer the duration over which periodic data are averaged, the lower the
peak values. Thus, the average annual water mercury concentrations are lower than some of the
quarterly concentrations for the same period. The average quarterly concentrations are lower
than some of the monthly concentrations, and the average monthly concentrations are lower than
some of the weekly concentrations.
ATSDR believes that some of the assumptions used by the Task 2 team may not be
representative of actual exposure conditions. Many of the exposures to EFPC water occurred
over periods of time shorter than 1 year. Children did not typically play in EFPC over the winter
months, and if they did, they were not likely to have ingested much water. And notwithstanding
Task 2’s assumption, a child 3 years of age and younger playing in the creek is unlikely. Older
children may have played in the creek over several (or many) years, but each year they likely
took time off from playing in the creek. In any event, the Task 2 average annual mercury doses
provide only an estimate of exposures averaged over a full year—an exposure that is least likely
to be a public health concern.
To estimate the short-term reduction of mercury mass in EFPC, ATSDR considered comparing
on a weekly basis (rather than quarterly) the concentration data from the water samples collected

E-16

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

from each end of EFPC in the 1950s. But no evidence supports the assumption that the
predictability or linearity of the Task 2 model increases with shorter periods. Any quantitative
evaluation based on such an exercise would thus suffer from a lack of confidence.
The Task 2 Model for Mercury Concentrations in Soil and Sediment
The Task 2 team estimated doses and risks associated with direct exposures to contaminated soil
and sediment for the following populations:
•
•
•

EFPC floodplain farm family
Robertsville School students
Scarboro community

The direct exposure pathways are 1) ingestion of mercury-contaminated soil or sediment, and 2)
dermal absorption of mercury from skin contact with mercury-contaminated soil or sediment.
For each selected group, the Task 2 team identified samples collected from areas of the
floodplain or creek likely to have been contacted by people.
The Task 2 team used soil samples from two studies to estimate past mercury concentrations in
both soil and sediment in the EFPC floodplain and Scarboro (see Table E-8). The two studies are
the Science Applications International Corporation (SAIC) EFPC Floodplain Remedial
Investigation (RI) from 1990–1992 (SAIC 1994a) and the Oak Ridge Associated University
(ORAU) study in 1984 (Hibbitts 1984, 1986; TDHE 1983). The EFPC RI study included more
than 2,800 core (16-inch long) soil samples, with many of the samples from the EFPC
floodplain, but it did not include any soil samples from Scarboro. The ORAU study included
more than 3,000 soil samples from the EFPC floodplain and properties throughout Oak Ridge
(including Scarboro), but they were only surface samples (0 to 3 inches below the surface)
(ChemRisk 1999a).
Table E-8. Data Sources for Past Soil and Sediment Mercury Concentrations
Environmental
Pathway

EFPC Farm
Family

Robertsville
School

Scarboro
Community

Soil

EFPC RI

EFPC RI

ORAU

Sediment

EFPC RI

EFPC RI

EFPC RI

The EFPC RI included soil samples from throughout the EFPC floodplain. The samples were
plotted on transects, imaginary lines that cross the EFPC floodplain at right angles to the creek.
The RI included 159 transects that crossed the full length (23.2 kilometers or 14.4 miles) of the
creek. Each was separated by approximately 100-meter (330-foot) intervals. Samples were taken
at the edge of the water and every 20 meters (65 feet) away from the creek, up to the elevation of
the 100-year floodplain (see Figure 18).
The RI core samples had already been collected, mixed together (i.e., composited), and analyzed
before the dose reconstruction project began. Thus the mercury concentrations at various depths
in those samples could not be determined. But other studies could provide data that allowed the
Task 2 team to estimate the possible vertical mercury distribution. A 1993 study indicated that
most of the mercury in the EFPC floodplain was contained within the first 16 inches of soil
(Henke et al. 1993). This was attributed to the tendency for elemental and mercuric mercury to

E-17

stay bound to soil and to the fact that elemental mercury is not very soluble in water. With time,
cleaner soil and sediment accumulates on top of the more highly contaminated soil and sediment.
In 1992, SAIC conducted a study called the Vertical Integration Study (VIS). SAIC took five 16­
inch EFPC soil cores and analyzed each 1-inch depth separately. The cores were taken at four
locations:
•
•
•
•

EFPC confluence with Poplar Creek
Grand Cove Subdivision
Bruner’s Center site (two core samples)
National Oceanic and Atmospheric Administration (NOAA) property

Key findings included the observation that the highest mercury concentrations were deep in the
core, and the lowest concentrations were found near the top of the core sample. And when
composited, the mercury concentration of the top 16 inches of soil was approximately equal to
the average mercury concentration from the individual 16 inches analyzed separately. Task 2
used this observation and the average stratification of mercury in the VIS core samples to
construct a table of soil concentration adjustment factors (see Table E-9).
Table E-9. Task 2 Soil Concentration Adjustment Factors
Year

Adjustment Factor (%)

1950–1954
1955–1958
1959–1962
1963–1966
1967–1970
1971–1974
1975–1978
1979–1982
1983–1986
1987–1990
1991–1994
Source: ChemRisk 1999a

100–400
200–500
50–300
50–300
40–200
10–100
5–100
3–50
1–50
2–50
1–30

The Task 2 team assumed that the highest mercury concentrations in the VIS core samples were
attributable to the period of the highest mercury releases (from 1955–1959), and that the rate of
soil deposition in all samples was a constant ¼ inch per year. The Task 2 team assigned specific
years to the vertical distribution of mercury concentrations in the VIS samples. The
concentrations for different years were then converted to percentages of the average composited
concentration. These percentages are the concentration adjustment factors. They are presented as
ranges to account for the uncertainty in the actual value of the soil or sediment mercury
concentration. The adjustment factors were multiplied by the average soil mercury
concentrations in the composited core samples (the top 16 inches of soil) from the EFPC RI to
estimate annual average soil and sediment mercury concentrations for the years 1950–1990.
Historic soil and sediment mercury concentrations for the EFPC floodplain farm family and the
Robertsville School students were calculated from the RI samples collected near Mile 10 (± 0.5
mile) and near Mile 12 (± 0.5 mile) in the EFPC floodplain, respectively. The historic sediment
samples for Scarboro residents were calculated from EFPC RI samples collected near the
floodplain’s Mile 14 (± 1 mile).
E-18

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

During the RI, samples were defined as either soil or sediment on the basis of where they were
collected in proximity to EFPC. Samples collected at the edge of EFPC were considered
sediment. But samples used for evaluating soil exposure pathways (the EFPC floodplain farm
family and Robertsville School children, for example) included the full set of samples collected
from just beyond the edge of the creek, to the elevation of the 100-year floodplain.
For the Scarboro community, mercury concentrations in sediment were calculated from the RI
core samples, and mercury concentrations in soil were calculated from the 1984 ORAU study
data. The ORAU study included a total of 57 samples from Scarboro—16 samples from
Hampton Road and 41 samples near the intersection of Tulsa and Tuskegee Roads. All of the
samples were surface samples (0 to 3 inches deep). The Task 2 report does not indicate how
historic soil mercury concentrations in the Scarboro community were estimated.
Task 2 used the VIS samples to calculate annual mercury concentrations from the many
composited RI core samples. The range of mercury concentrations as a percentage of the average
concentration within some of the VIS core samples is wide, varying from less than 1 to 380
percent of the average mercury concentration in the composite sample. The location where the
minimum and maximum mercury concentrations are found in the VIS samples often varies
between the samples (the overall pattern of mercury concentrations measured at different depths
in the two samples collected from the same location [Bruner’s Center] for example, are not
similar). To compensate, the Task 2 team extended the ranges of the adjustment factors beyond
the measured range for the years with the highest mercury releases to EFPC. Thus in the
composite samples, Task 2 increased the upper range of the adjustment factors for the years of
the highest mercury releases from 380 percent to 500 percent of the mercury concentration.

Task 2 Results
Task 2 used the soil and sediment mercury concentrations—estimated from its model—to
calculate average annual mercury doses for the three potentially
Doses exceeding the
exposed communities. None of the Task 2 estimated doses from soil or
RfD or MRL do not
sediment ingestion for 1950 through 1990 exceeded U.S.EPA’s RfD or
necessarily presuppose
ATSDR’s minimal risk level (MRL) for inorganic mercury. For 1950
adverse health effects.
through 1966 (except 1962), however, Task 2 estimated upper-end
doses to EFPC floodplain farm children could have exceeded the inorganic mercury RfD (though
not the MRL) from dermal contact with soil. 46 Also, for 1958 only, Task 2 estimated upper-end
doses Robertsville School children could have exceeded the inorganic mercury RfD (though not
the MRL) from dermal contact with soil. Still, none of the dermal mercury doses calculated at
the 50 percentile exceeded either agency’s health guideline value, and none of the calculated
doses from sediment exposures exceeded either agency’s health guideline value.

46

The “upper-end” doses are the 97.5 percentile doses, which, according to the Task 2 report, are the 97.5 percentile
confidence levels of the probability density functions (PDFs). The PDFs, which characterize the distribution of
doses for each specific pathway, were calculated by Task 2 using Monte Carlo simulations. The 97.5 percentile
doses are less likely to occur than doses at lower probability levels; they are calculated with the most extreme
exposure assumptions. Task 2, however, considers the highest doses are possible because the full range of
assumptions used in its calculations was considered possible.

E-19

Discussion
ATSDR reviewed ORAU soil data. ATSDR identified 43 surface (0–3 inches below surface) soil
samples collected in the Scarboro area in 1984. 47 The highest soil mercury concentration among
the 43 samples was 3.8 ppm, below ATSDR’s comparison value of 20 ppm. ATSDR does not,
however, consider the mercury concentrations in ORAU samples collected in the top 3 inches of
soil in 1984 as representative of past mercury concentrations in Scarboro soils. Core data from a
1992 study indicate that the floodplain soil layers with the highest mercury concentrations are
buried beneath as much as 10 inches of soil and sediment (ChemRisk 1999a). The near-surface
soil data collected in Scarboro would not likely reflect historical mercury concentrations in soil
and sediment.
The overall weighted-average adjustment factor for the years 1950 through 1990 is nearly 130
percent. This ensures that overall, none of the mercury measured in the top 16-inch cores is
“lost” through modeling. The Task 2 team assumed, however, that mercury deposition occurred
at a constant rate over the floodplain and that mercury does not migrate significantly in the soil.
No studies demonstrate how well these assumptions hold. The model might increase the mercury
levels for some years and decrease them in other years, relative to the true concentration values.
This averaging effect could underestimate exposures in years with high mercury releases or in
areas with high mercury deposits, even considering the wide range of adjustment factors that the
Task 2 team adopted for those years. The very small number of samples in the VIS and the poor
consistency between mercury concentrations at similar depths suggest that the model is not
reliable. Given the small number of samples on which the adjustment factors are based and given
the nonuniformity of concentrations within each of the vertical layers, considerable uncertainty
surrounds whether the extended adjustment factors adequately reflect the true pattern of mercury
distribution in the core samples since 1950.
Additionally, the Task 2 model may not sufficiently account for the mass of mercury in EFPC.
The Task 2 team only applied the adjustment factors to the uppermost core data. In some areas of
the floodplain, multiple core samples were collected from a single location (maximum of five
core samples deep). Historical soil or sediment mercury concentrations could be underestimated
if significant mercury were present below 16 inches. SAIC estimated that 18 percent of the soil
volume contaminated with mercury at levels greater than or equal to 50 ppm lay in the second
core “horizon” (16–32 inches below ground surface), and 29 percent of the soil volume
contaminated with mercury at levels greater than or equal to 200 ppm lay in the second core
horizon. These analyses indicate that for the highest contaminated regions of the floodplain, the
Task 2 efforts to assign soil mercury concentrations to individual years are not reliable.
Given the uncertainties described above, in this public health assessment ATSDR decided to
evaluate the soil data without considering the Task 2 team’s method of assigning an estimated
timeframe of mercury deposition.
The Task 2 Model for Mercury Concentrations in Fish
Before 1970, fish downstream from the Y-12 plant were not collected and analyzed for mercury.
But the largest releases of mercury from the Y-12 plant to EFPC occurred during the 1950s and
early 1960s. For the years 1950–1990, the Task 2 team estimated average annual mercury
concentrations in fish from three bodies of water:
47

Some additional samples may have been collected in Scarboro, but ATSDR only identified 43 samples.

E-20

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

• EFPC
• Poplar Creek (downstream of EFPC) and the Clinch River (downstream of Poplar Creek)
• Tennessee River/Watts Bar Reservoir (downstream of the Clinch River)
The Task 2 team estimated mercury doses from eating fish from EFPC for residents of the
Scarboro community and the EFPC floodplain farm family. The Task 2 team also estimated
mercury doses for people who ate fish from Poplar Creek/Clinch River and the Tennessee
River/Watts Bar Reservoir. Where the latter fish-eating populations lived was not identified;
people who fish in these waters come from all around the area.
The Task 2 team considered that if mercury concentrations in fish were proportional to mercury
concentrations reported in sediments, historic sediment data could be used to estimate past
mercury concentrations in fish. Task 2 therefore studied the relationship between mercury
concentrations in fish and mercury concentrations in surface sediment samples collected during
the 1970s and 1980s in EFPC, Poplar Creek, the Clinch River, and the Tennessee River (to Watts
Bar Dam). Fish data were compared with sediment data from samples collected near one another
in the water. Linear, semi-log, and log-log regression analyses were conducted of mercury
concentrations in bluegill sunfish and largemouth bass and compared to mercury concentrations
in sediment. The database for other fish species was too small to analyze and both bluegill
sunfish, and largemouth bass are resident sport species anglers commonly catch for eating.
Mercury concentrations in bluegill sunfish and largemouth bass correlated well with surface
sediment mercury concentrations using linear regression analysis. 48 The mercury concentrations
in sediments that were co-located with fish samples ranged from 0.18 to 99 ppm (bluegill
sunfish) and 0.18 to 46 ppm (largemouth bass).
The general approach was to apply the regression equations for bluegill sunfish and largemouth
bass (developed from 1970s and 1980s fish and sediment data) to mercury concentrations in
sediment for the years 1950–1990, to estimate fish mercury concentrations for those years. Some
characteristics of the model the Task 2 team used to estimate mercury in fish are described
below.
• The sediment mercury concentrations used for these calculations were estimated from six
sediment core samples taken in the 1980s from EFPC, Poplar Creek, the Clinch River, and
the Tennessee River. The team assigned different years to different core depths based on an
analysis of mercury and cesium-137 in the sediment samples and estimates of the annual
quantities of mercury and cesium-137 released from the Y-12 plant. Concentrations of both
mercury and cesium-137 in sediment layers were assumed to be proportional to the annual
quantities of mercury and cesium-137 released from the Y-12 plant.
• To estimate the fish mercury concentrations, the Task 2 team used one core sample and one
surface sediment sample for fish from EFPC, three core samples from Poplar Creek and the
Clinch River, and two core samples from the Watts Bar Reservoir. The six sediment cores
analyzed to estimate past mercury concentrations in fish were collected from the following
six locations:
• New Hope Pond, in EFPC immediately downstream from the Y-12 plant,

48

The squared correlation coefficients (r2) for bluegill sunfish and largemouth bass were 0.69 and 0.66, respectively,
indicating a good correlation.

E-21

• Poplar Creek near the confluence with EFPC,
• The Clinch River approximately midway between the confluence of Poplar Creek and the
confluence of the Clinch River with the Tennessee River,
• One mile up from the confluence of the Clinch River,
• Just past the confluence of the Clinch River in the Tennessee River, and
• Eight miles upstream of Watts Bar Dam in the Tennessee River (Watts Bar Reservoir).
Having generated the regression model, Task 2
During 1969, a chloralkali plant on the St. Clair
dispensed with it when the sediment mercury
River discharged approximately 30 pounds of
concentrations in core samples exceeded the regression
elemental mercury per day to the river.
limits. Task 2 did not assume that the correlation
In 1970, sediment mercury concentrations of
between fish and sediment mercury concentrations was
up to 1,700 ppm were measured in the river. In
linear beyond the range of the data used in the
1971, mercury was analyzed in fish collected
regression analysis. The Task 2 team, then, did not
in the river and further downstream in Lake St.
Clair.
apply the regression equations to sediment mercury
From this study, the Task 2 team selected
concentrations above 99 ppm. For years corresponding
mercury concentrations for past years (during
to sediment layers whose mercury concentrations
years of peak releases from the Y-12 plant)
exceeded those in the linear regression model (99
from fish that were comparable species and
ppm), Task 2 used default fish mercury concentrations
sizes to those in Poplar Creek and the Clinch
from a fish study in 1971 from the St. Clair River and
River (Wren 1996).
Lake St. Clair in the Great Lakes region. This was the
case for some layers of sediment in EFPC and Poplar Creek.

Task 2 Evaluation of EFPC Fish Concentrations
For EFPC, the Task 2 team examined a 1982 sediment core sample collected from the upper end
of EFPC in New Hope Pond, downstream of Y-12 buildings. For the lower end of EFPC, before
EFPC feeds into Poplar Creek, no core samples were taken, but a surface sediment sample was
collected in 1982. The Task 2 report noted that the surface sediment mercury concentration at the
lower end of EFPC was approximately 20 percent of the surface sediment mercury concentration
at New Hope Pond. The Task 2 team assumed that the historic sediment mercury concentrations
at the lower end of EFPC were 20 percent of those for the same years at New Hope Pond.
But the New Hope pond was dredged in 1973. The sediment core only included sediment as old
as 1973; any preexisting sediment was removed at that time. All the New Hope Pond mercury
concentrations in sediment between 1973 and 1982 exceeded the upper end of the sediment
concentrations used to generate the regression equations. And all of the fish concentrations at
New Hope Pond, including those before 1973, were default values from the St. Clair River/Lake
St. Clair study. The lower limit, mean, and upper limit fish concentrations that the Task 2 team
selected for fish in EFPC from the St. Clair River/Lake St. Clair study were 1, 1.7, and 4 ppm for
bluegill sunfish, and 2, 3.2, and 4.5 ppm for largemouth bass.
At the lower end of EFPC, the same default fish mercury concentrations from the St. Clair
River/Lake St. Clair study were used for the years between 1950 and 1964. The Task 2 team
assumed that sediment mercury concentrations exceeded the sediment regression limit values for
those years. Beginning in 1965, the Task 2 team reported that it applied the regression equations
to the estimated sediment concentrations to calculate fish mercury concentrations for the lower
end of EFPC. The Task 2 report, however, does not present the sediment mercury concentrations

E-22

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

it used with the regression equations for those calculations. Information gaps mean data gaps in
certain periods. 49 After 1982 (after the date of the New Hope Pond core sample, for example),
the Task 2 team presumably used analytical data from fish collected in EFPC. But the report
does not describe what data it used or how it calculated fish mercury concentrations for those
later years.
The highest Task 2 estimated annual fish mercury concentrations in EFPC were for the years
from 1950 to 1964. The minimum, mean, and maximum average annual fish mercury
concentrations for those years were 1.5, 2.5, and 4.3 ppm, respectively. These values are the
averages of the default fish mercury concentrations from the St. Clair River/Lake St. Clair study
for bluegill sunfish and largemouth bass; they were not calculated using the regression equations.

Task 2 Evaluation of Poplar Creek/Clinch River and Tennessee River/Watts Bar Reservoir
Fish Concentrations
For the sediment core sample collected at the Poplar Creek location below the confluence of
EFPC, mercury concentrations in core sample layers corresponding to the years from 1956 to
1961 exceeded the maximum surface sediment mercury concentrations used to generate the
correlation equations. 50 Again, Task 2 took default mean and maximum fish mercury
concentrations (3.3 and 7 ppm, respectively) from the St. Clair study. The same values were used
for both bluegill sunfish and largemouth bass. 51 In other years and at sediment core sample
locations farther downstream, Task 2 used the regression equations to calculate fish mercury
concentrations.
The Task 2 team averaged together the estimated fish mercury concentrations at the locations of
the sediment core samples in each water segment. It also averaged together the estimated
mercury concentrations of the bluegill sunfish and largemouth bass.
The Task 2 team calculated 95 percent confidence intervals around the predicted mean fish
concentrations associated with sediment core mercury concentrations, using the regression
model’s estimated standard error. The averaging of mercury concentrations in fish from different
locations in a water segment, from two fish species, and the use of confidence intervals based on
the regression model resulted in three mercury concentrations (a minimum, a mean, and a
maximum) for each year (1950–1990) for each water segment.
Generally, the sediment mercury concentrations (and correlated fish mercury concentrations) the
Task 2 team used were higher closer to the Y-12 plant and decreased with distance downstream.
Table E-10 contains the mean fish mercury concentrations for each surface water segment for the
years 1950–1970.

49

For example, if the sediment mercury concentrations at the lower end of EFPC were assumed to be 20 percent of
those in New Hope Pond, but the oldest sediment in New Hope Pond was from 1973, what sediment data were
used between 1965 and 1972?
50
The upper Poplar Creek sediment mercury concentrations from 1956–1961 ranged between 156 and 460 ppm.
51
The Task 2 team used different values than those used for fish from EFPC.

E-23

Table E-10. Estimated Annual Average Mercury Concentrations in Fish (1950–1970)

EFPC
(ppm)

Poplar Creek/Clinch
River (ppm)

Watts Bar Reservoir
(ppm)

1950

2.5

1.1

0.13

1951

2.5

1.1

0.13

1952

2.5

1.1

0.16

1953

2.5

1.4

0.17

1954

2.5

1.2

0.19

1955

2.5

0.9

0.34

1956

2.5

2.2

0.52

1957

2.5

2.6

0.66

1958

2.5

2.5

0.74

1959

2.5

2.4

0.74

1960

2.5

2.2

0.52

1961

2.5

2.0

0.29

1962

2.5

1.9

0.29

1963

2.5

1.2

0.27

1964

2.5

0.97

0.25

1965

2.5

0.82

0.25

1966

2.5

0.73

0.23

1967

2.5

0.63

0.22

1968

2.4

0.52

0.22

1969

2.4

0.55

0.20

1970

2.4

0.58

0.19

Year

Concentrations are based on fresh weight samples.

ppm:
parts per million

Task 2 Mercury Doses to Humans
The Task 2 team used the estimated fish mercury concentrations to calculate mercury doses for
past fish consumption. 52 Table E-11 contains the mean fish ingestion rates that the Task 2 team
used in its dose calculations. The Task 2 team generated unspecified “custom” distributions of
childhood ingestion rates from the adult rates.
The Task 2 team calculated doses using a Monte Carlo simulation. This produces a central dose
value and lower and upper bound values corresponding to the 95 percent confidence interval
around the central value. The Task 2 team compared its estimated mercury doses to U.S.EPA
RfDs for ingestion of methylmercury.

52

Most of the mercury found in fish is methylmercury.

E-24

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Table E-11. Task 2 Average Fish Mercury Dose Ingestion Rates

Population

Location

Body Size

Ingestion Rate
(g/d)

Ingestion Rate
(m/y)

Scarboro

EFPC

adult

1.2

2.6

Scarboro

EFPC

child

0.27

0.58

EFPC Floodplain Farm Family

EFPC

adult

1.2

2.6

EFPC Floodplain Farm Family

EFPC

child

0.27

0.58

Commercial Angler

Poplar Creek/Clinch River

adult

2.2

4.7

Commercial Angler

Poplar Creek/Clinch River

child

0.49

1.1

Recreational Angler

Poplar Creek/Clinch River

adult

18

39

Recreational Angler

Poplar Creek/Clinch River

child

4.0

8.6

Commercial Angler

Watts Bar Reservoir

adult

24

52

Commercial Angler

Watts Bar Reservoir

child

5.4

12

Recreational Angler

Watts Bar Reservoir

adult

30

64

Recreational Angler

Watts Bar Reservoir

child

6.7

14

Source: ChemRisk 1999a

g/d:
grams per day

m/y:
meals per year

The adult ingestion rates are arithmetic means of lognormal distributions.

A fish meal is assumed to be approximately 6 ounces or 170 grams.

Using the ingestion rates presented in Table E-11, the Task 2 team determined that none of the
estimated methylmercury central dose values for Scarboro residents and the EFPC floodplain
farm family (adults or children) who ate fish from EFPC exceeded the RfD for methylmercury.
But at the upper bound end of the estimated dose range, all of the estimated doses for the same
two populations exceeded the RfD for all the years from 1950 through 1990.
For people who fished in Poplar Creek or the Clinch River, the central doses of recreational
fishers exceeded the RfD for methylmercury for the years from 1950 through 1964. For Watts
Bar Reservoir (Tennessee River) fishers, the central dose value for methylmercury exceeded the
RfD for 1957, 1958, and 1959 only.
At the high end of the dose range (the 97.5 percentile doses), all the Task 2 report estimated
doses to recreational anglers who fished in Poplar Creek/Clinch River and both recreational and
commercial anglers who fished in Watts Bar Reservoir exceeded the RfD. The upper bound
estimated doses to commercial anglers who fished in Poplar Creek/Clinch River exceeded the
RfD from 1950 through 1967.
Discussion
The sediment core samples were used to estimate mercury concentrations in fish. These values
were generally spread out across the upper and lower ends of each water segment between EFPC
and Watts Bar Dam. The small sample size, however, may not adequately represent the past
sediment mercury concentrations (and correlated fish tissue concentrations) in the surface water
segments downstream from the Y-12 plant: only three core samples were used in the Task 2

E-25

model to represent nearly 12 miles of Poplar Creek and the Clinch River. Only two core samples
were used to represent approximately 30 miles of the Tennessee River.
Mercury in the EFPC floodplain soil is not distributed evenly. Nor is it simply deposited in
quantities inversely proportional to distance from the Y-12 plant. Sediment is often mobile in
these surface stream beds. Numerous regions of the stream beds have no apparent sediment
accumulation at all. Even in places where sediment accumulates, it may be subject to significant
agitation and dispersion. No core sediment samples were found to be co-located with surface
sediment samples; thus, evaluating the consistency in mercury measurements between the two
types of samples was not possible. In addition, no independent means were available to judge
how representative the core sample layers are of surface sediment mercury concentrations across
the miles of creeks and rivers in past years.
The Task 2 team used default mercury concentrations for fish in EFPC and Poplar Creek in some
years; published studies suggested limits to the amounts of mercury that fish can bioaccumulate.
The Task 2 team listed three laboratory studies that indicate mercury body burdens ranging
between 10 and 20 ppm are lethal to rainbow trout. Yet the relevance of those studies to fish in
EFPC in the 1950s is questionable. 53
The St. Clair River and Lake St. Clair studies of mercury in fish suggest limits to the amounts of
mercury that bluegill sunfish and largemouth bass accumulate. The water environments in those
studies, however, may have been dissimilar to EFPC in ways that affected available fish diets,
methylmercury production, and the fish accumulation of mercury. Many hazardous substances
(such as industrial cleaning chemicals) were released in large quantities to EFPC during the
earlier decades of the Y-12 plant operations. These releases likely contributed to poor aquatic
health and to smaller numbers of fish and smaller sized fish in EFPC than in later years. The St.
Clair studies included similar sizes and species of fish (bluegill sunfish and largemouth bass) as
those analyzed in EFPC. But different conditions may have obtained (more aquatic tropic layers,
for example) in the St. Clair studies that affected the bioaccumulation of mercury differently than
in EFPC. Moreover, the maximum mercury concentrations in sediments reported from the St.
Clair studies (up to 1,700 ppm) were about one-half the maximum mercury concentrations
measured in the EFPC floodplain soils (3,420 ppm). ATSDR does not have sufficient
information to determine whether the St. Clair mercury concentrations in fish are good
surrogates for those in EFPC and Poplar Creek during the 1950s and 1960s. Consequently,
ATSDR thinks the fish mercury concentrations, which the Task 2 team adopted for the mercury
dose reconstruction, do not reflect adequately the level of uncertainty associated with these data.
In summary, ATSDR believes the Task 2 team relied on fewer sediment core samples than
needed to estimate adequately past mercury concentrations in sediment, And consequently, to
provide reliable estimates of fish tissue concentrations from these water bodies. The applicability
of the St. Clair data is unknown and need to be explored further before data from this study can
be used with confidence.

53

For example, in the trout studies, mercuric chloride was put into the water, whereas elemental mercury and
mercuric nitrate were released from the Y-12 plant into EFPC.

E-26

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Task 2 Vegetation Model for Mercury Concentrations
The Task 2 team calculated deposition of mercury from the air to above-ground vegetation. Total
deposition was calculated by adding the amount deposited during dry conditions to the amount
deposited during wet conditions.
• The dry deposition component of the equation takes the total dry deposition velocity and
accounts for the amount retained by the vegetation in relation to the mass of the vegetation.
• The wet deposition component of the equation takes into account climatological conditions.
This component requires additional parameters to calculate wet deposition velocity,
specifically the washout ratio and the average annual precipitation rate.
The dry and wet deposition components are then added together to calculate the total deposition
from the air onto vegetation.

ATSDR’s Technical Review
ATSDR’s technical reviewers commented that the Task 2 report’s assumptions in estimating air
to plant mercury transfer appeared reasonable. One reviewer, however, criticized the report for
combining the distinct issues of mercury deposition on plants and mercury absorption by plants.
Another reviewer commented that the report had probably slightly overestimated the deposition
of mercury on fruits and fruiting vegetables. He pointed out the following:
• The analysis treats mercury deposition as a function of mass, rather than surface area.
Because fruits and fruiting vegetables (peppers, tomatoes, squash, for example) have lower
surface area-to-mass ratios, the report’s analysis probably exaggerated the degree of mercury
accumulation.
• Estimating mercury in plant fruits and stems based on deposition is likely an overestimate;
mercury is unlikely to be translocated within the plant.
• The analysis assumes that airborne Hg° deposited on plant surfaces is completely oxidized to
Hg+2. Because this process is gradual, however, a portion of the Hg° deposited onto plant
surfaces is lost due to revolatilization.
• The analysis assumes that the mercury ingested in aboveground fruits and vegetables is Hg+2,
however a portion of this is Hg°, which has a low absorption rate in the gastrointestinal tract.
• The use of mass interception factors for small aerosols, mists, and gases may overestimate
the accumulation of mercury in vegetation depending on the aerosols/mists/gases used to
determine the factors. Hg° is relatively insoluble, and will likely stay near the air (that is, the
surface).
A third technical reviewer commented on the huge uncertainty in the calculations of mercury
transfer to vegetation. Still, he noted that the estimates are probably adequate to assess their
contribution to the overall exposure of persons. The fourth technical reviewer noted several
uncertain components in the calculation of air concentrations and deposition to vegetation. To
remove some of the uncertainty, he suggested the required data could be obtained by a field
study or a wind-tunnel (environmental chamber) study.

E-27

Discussion
Task 2’s approach seems reasonable. It might even be the best estimate available. But how
accurately this model represents actual exposures from 50 years ago is unclear. The key
parameters with the greatest apparent influence on the estimated concentration (air concentration,
weathering rate of vegetables, fraction of mercury remaining after washing, and the
bioavailability factor, for example) are either 1) highly uncertain, 2) taken from literature relating
to radionuclides in plants, or 3) based on professional judgment. Given these observations, to
determine what the estimated numbers truly mean is difficult. Using past ATSDR modeling
experience, estimating historical air concentrations is a challenge. And estimating plant tissue
concentrations that result from air concentrations adds an entire level of complexity, as well as
uncertainty.

E-28

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Appendix F. Evaluation of Mercury Emissions from Selected Electricity
Generating Facilities

-MEMORANDUM­
DATE:

June 15, 2005

TO:

Jack Hanley and Bill Taylor, ATSDR

FROM:

John Wilhelmi, ERG

RE:

Oak Ridge Reservation: Evaluation of Mercury Emissions from Selected
Electricity Generating Facilities

This memo presents ERG’s evaluation of past air emissions of mercury from electricity
generating facilities near the Oak Ridge Reservation, such that ATSDR has context for
evaluating past inhalation exposures to mercury in the vicinity of Oak Ridge. ERG used two
different analyses to comment on this matter. First, for qualitative insights on air quality impacts
from electricity generating facilities, this memo presents a brief review of EPA’s 1997 “Mercury
Study Report to Congress” (EPA 1997). Second, the memo presents quantitative estimates of air
quality impacts from an electricity generating facility operated by the Tennessee Valley
Authority (TVA). The memo concludes with summary statements based on the two different
types of analyses. Citations for all references are presented at the end of the memo.
Review of EPA’s 1997 “Mercury Study Report to Congress.” For general insights into
potential mercury air quality impacts from power plants, ERG first reviewed EPA’s 1997
“Mercury Study Report to Congress” (EPA 1997)—an extensive overview of the environmental
and health impacts associated with environmental releases of mercury. The following paragraphs
summarize key statements from this report, specifically those that pertain to coal-fired power
plants. No references are provided in this section, as all information was taken from the EPA
report (EPA 1997).
• Emissions. The EPA report includes a detailed inventory of anthropogenic emissions sources
of mercury for a 1994-1995 baseline. The report acknowledges that significant amounts of
mercury are also released from non-anthropogenic sources, including natural sources (e.g.,
volcanoes) and sources that “re-emit” mercury to the environment after it deposits from the air
(e.g., volatilization from oceans, soils, and other media).
The inventory of anthropogenic sources considers more than 30 different source categories,
including electricity generating facilities, incinerators, chlor-alkali facilities, mobile sources,
and numerous others. Emissions from coal-fired boilers, which ranked highest of all these
source categories, were estimated to account for 33 percent of the total nationwide mercury air
emissions from anthropogenic sources. Emissions estimates for these power plants were
computed from multiple input parameters, including coal throughput, average concentration of
mercury in coal, and mercury reductions attributed to coal cleaning and air pollution controls.

F-1

Mercury emitted from these sources can be found in different chemical forms (elemental and
compounds) and different physical forms (vapor phase and particle-bound), and the speciation
of mercury emissions significantly affects fate and transport properties, as described below.
Mercury species emitted from coal-fired power plants reportedly vary with coal type, boiler
design, and operating conditions. The EPA report presents limited data on speciation for these
sources, but suggests the following mercury speciation for air emissions from coal-fired power
plants: 50 percent as elemental mercury vapor, 30 percent as divalent mercury vapor, and 20
percent as particle-bound mercury.
• Fate and transport. Though the EPA report includes an extensive multi-media fate and
transport analysis of local, regional, and global mercury cycling, this memo focuses on
conclusions that pertain to atmospheric transport on local scales (i.e., less than 50 km from the
emissions source). On these local scales, the report repeatedly emphasizes that fate and
transport behavior of mercury depends largely on its chemical and physical state.
On the one hand, elemental mercury vapor can remain airborne for roughly 1 year and
transport thousands of miles from emissions sources. The primary removal mechanisms for
the mercury vapor are deposition, chemical conversion to mercury compounds, and uptake
and retention by plants. However, such mechanisms appear to have fairly slow kinetics, as
EPA modeling results suggest that only a small percentage (<5 percent) of mercury vapor
emissions deposits to the surface within 50 km of a coal-fired plant. Because of this, elemental
mercury vapor typically accounts for the majority of total airborne mercury (see next section).
On the other hand, airborne mercury compounds (divalent mercury) and particle-bound
mercury have estimated residence times in the atmosphere of a few days or less. These forms
of mercury are more readily removed from the atmosphere by both dry and wet deposition
processes. Therefore, these forms of mercury account for smaller percentages of total airborne
mercury.
• Ambient air concentrations. According to several environmental monitoring studies, elevated
mercury concentrations in multiple environmental media have been measured around large
mercury emissions sources. However, no comprehensive monitoring data are available to
quantify the exact extent to which various emissions sources contribute to measured air
concentrations. Qualitatively, ambient air concentrations of mercury at any given location will
depend on the locations of nearby sources, the amounts and species of mercury emitted, and
local meteorological conditions.
EPA’s report includes a brief review of several ambient air monitoring studies published in
the 1990s. In all studies and monitoring locations considered, average concentrations of total
airborne mercury were less than 50 ng/m3—EPA’s Reference Concentration (RfC) for
mercury. Moreover, the monitoring results clearly showed that most airborne mercury is in the
form of mercury vapor: average air concentrations of mercury vapor were consistently at least
20 times greater than corresponding average concentrations of particulate-bound mercury.
EPA’s report also presents monitoring data from a single study designed to characterize
mercury air quality impacts from a coal-fired power plant. That study reported no significant
differences between particulate-bound mercury concentrations measured 5 km upwind and 5
km downwind from the source of concern; no information was provided on whether the study
considered vapor phase concentrations.

F-2

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

In addition to summarizing measured concentrations, EPA’s report presents estimated
concentrations based on dispersion modeling analysis. Of particular interest, EPA evaluated
air quality impacts from a generic “large coal-fired power plant” (i.e., a plant with 975
Megawatt capacity that emits 230 kg of mercury to the air per year). Using typical stack
parameters and mercury speciation data, the modeling predicted that ground-level ambient air
concentrations of mercury at distances 2.5 km, 10 km, and 25 km from the generic power
plant would be less than 1.7 ng/m3—the background concentration attributed to natural
sources and re-emitted mercury. Thus, the incremental air quality impacts from large coalfired power plants were estimated to be essentially negligible in comparison to EPA’s RfC.
• Exposure and risk. The EPA report repeatedly emphasizes that, nationwide, exposure to
mercury is dominated by the fish ingestion pathway. This conclusion was based on estimated
exposures for numerous scenarios, including evaluations of exposures in the vicinity of coalfired power plants. Chlor-alkali plants were the only industrial source category predicted to
cause notable exposures via the inhalation pathway. Although EPA’s report does not provide
quantitative risk or hazard estimates, the modeling results clearly show that the estimated air
quality impacts from the generic coal-fired power plant were below appropriate health
benchmarks.
Screening Modeling Analysis. To supplement the general information available from EPA’s
“Mercury Study Report to Congress,” ERG conducted a screening dispersion modeling analysis
to examine potential air quality impacts from the Tennessee Valley Authority (TVA) Kingston
Fossil Plant. 54 Construction of this facility was completed in 1955 and operations continue today.
The facility currently consumes approximately 14,000 tons of coal per day and has a winter net
generating capacity of 1,456 Megawatts (TVA 2005). Thus, current operations appear to be
slightly larger than those considered in EPA’s modeling efforts of a “large coal-fired power
plant.” Information on coal usage data for earlier years is not available.
The purpose of the screening analysis was to estimate coal usage rates at the Kingston Fossil
Plant that might be expected to cause elevated air quality impacts in the immediate vicinity of the
Y-12 Plant, located more than 25 km away. ERG used a screening model (SCREEN3) to
estimate air quality impacts based on the following release parameters:
• Stack height = 100 feet (30.5 meters)
• Stack diameter = 15 feet (4.6 meters)
• Stack exit velocity = 70 feet/second (21.3 meters/second)
• Stack exit temperature = 270 degrees Fahrenheit (405 degrees Kelvin)
With one exception, these release parameters were estimated from recent data that the
Department of Energy compiled on electricity generating facilities across the country. 55 As the
exception, the stack height was set artificially low to reflect the approximate stack heights at the

54

ERG did not evaluate air quality impacts from the Bull Run Plant, because construction of that facility was not
completed until 1967, which is several years after the time frame of interest for ATSDR’s evaluation of mercury
issues.
55
ERG ran sensitivity analyses on the model to assess the impacts of uncertainty in the input parameters. Lower
stack heights, lower exit velocities, and lower exit temperatures would all lead to higher estimates of air quality
impacts, but the modeling analysis was not unusually sensitive to any of these parameters. For instance, a 10%
decrease in stack height resulted in only a 5% increase in estimated air concentrations at the receptors of interest.

F-3

Kingston Fossil Plant during the time when the Y-12 facility released considerable quantities of
mercury. Several additional assumptions were programmed into the model:
• ERG assumed that all mercury in the coal burned at the Kingston Fossil Plant became
airborne, with none collected by pollution controls, removed in coal cleaning processes, or
sequestered in ash. This assumption should serve to overstate actual air quality impacts.
• ERG assumed that all mercury is released as elemental vapor and remains airborne throughout
the modeling domain. By not considering deposition, this assumption causes the model to
overstate the amounts of mercury in air and available for human exposure.
• ERG assumed that annual average concentrations of mercury near Y-12 are 8 percent of the
maximum hourly average concentrations. This factor is documented in EPA guidance for
screening analyses (EPA 1992) and is used to extrapolate the 1-hour maximum levels in the
SCREEN3 outputs to longer averaging times. According to EPA, “a degree of conservatism is
incorporated in the factor to provide reasonable assurance that maximum concentrations…will
not be underestimated” (EPA 1992). ERG further notes that the factor will tend to overstate
long-term air quality impacts with increased distance from the emissions source. Thus, ERG
has reason to believe that using this factor could considerably overstate air quality impacts.
• ERG assumed no complex terrain separates the Kingston Fossil Plant and the Y-12 Plant. In
reality, several small ridges separate these two areas, and these ridges would likely inhibit
atmospheric transport of the Kingston Fossil Plant’s emissions toward the Y-12 area. By not
considering these terrain features, the screening analysis likely overstates the potential air
quality impacts in the vicinity of Y-12.
• ERG used data from a recent EPA guidance document on estimating air emissions from
electricity generating facilities (EPA 2000) for a default concentration of mercury in coal.
That document lists typical mercury concentrations for coal mined in different states across
the country. ERG used the highest mercury composition in the entire document (0.42 ppm by
weight) in the calculations of air quality impacts. While using the highest mercury
composition figure is likely another conservative assumption, ERG acknowledges that the
mercury content of coal in specific mining areas might exceed the highest statewide average
used in this analysis. The screening analysis can be further refined if TVA were to provide
composition data for the coal that was previously used at the Kingston Fossil Plant.
Based on the aforementioned input parameters and assumptions, the SCREEN3 model outputs
predict that ambient air concentrations of mercury near Y-12 likely would not have exceeded the
RfC (0.05 jglm3) unless the Kingston Fossil Plant was burning nearly 275,000 tons of coal per
day. For reference, this coal throughput is approximately 20 times greater than the current coal
usage rates and almost undoubtedly exceeds the processing capacity of the facility. In other
words, even when considering the combination of multiple assumptions that likely overstate air
quality impacts, it seems exceedingly unlikely that air emissions from the Kingston Fossil Plant
could have caused ambient air concentrations near the Y-12 Plant to approach health
benchmarks.
ERG acknowledges that this screening analysis has inherent limitations and uncertainties. Most
notably, the analysis only estimates air quality impacts, which may not adequately represent
actual conditions. However, the approach of including multiple conservative assumptions (i.e.,
assigning highly uncertain inputs values that are known to overstate air quality impacts) provides

F-4

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

some confidence that this analysis does not underestimate actual air concentrations. Additionally,
the sensitivity analysis provides further confidence that the modeling outputs are not strongly
dependent on the stack parameters that were chosen as model inputs. There are several
opportunities for reducing model uncertainty. These include, but are not limited to, obtaining
site-specific data on actual coal usage for the time frame of interest, obtaining data on the typical
mercury content of the coal that was burned, or using a refined dispersion model. However, the
results of this screening analysis suggest that additional modeling for this issue might not be
necessary.
Conclusions and Recommendations. The following summary statements are supported by the
analyses presented earlier in this memo:
• EPA’s “Mercury Study Report to Congress” suggests that emissions from coal-fired power
plants have extremely limited incremental effects on ground-level air quality. The modeling
analyses EPA conducted on a hypothetical coal-fired power plant found essentially no
ground-level impacts at locations 2.5 km, 10 km, and 25 km downwind.
• Consistent with these general findings, ERG’s screening modeling analysis showed that past
mercury emissions from the TVA Kingston Fossil Plant almost certainly did not have
substantial air quality impacts (i.e., concentrations approaching the RfC) near the Y-12 Plant,
even when considering a series of health-protective assumptions.
References
[EPA 1992] US Environmental Protection Agency. 1992. Screening Procedures for Estimating
the Air Quality Impact of Stationary Sources, Revised. Office of Air and Radiation and Office of
Air Quality Planning and Standards. EPA-454/R-92-019. October 1992.
[EPA 1997] US Environmental Protection Agency. 1997. Mercury Study Report to Congress.
Office of Air Quality Planning and Standards and Office of Research and Development. EPA­
452/R-97-003. December 1997.
[EPA 2000] US Environmental Protection Agency. 2000. EPCRA Section 313 Industry
Guidance: Electricity Generating Facilities. Office of Pollution Prevention and Toxics. EPA 745­
B-00-004. February 2000.
Turner RR, Bogle MA, Heidel LL, et al. 1992. Mercury in ambient air at the Oak Ridge Y-12
Plant, July 1986 through December 1990. Govt. Reports Announcements and Index (GRA&I)
Issue 02.
Turner RR, Bogle MA. 1993. Ambient air monitoring for mercury around an industrial complex.
In: Chow W, Connor KK, eds. Managing hazardous air pollutants state of the art. Boca Raton,
Florida: Lewis Publishers, 162-172.
[TVA 2005] Tennessee Valley Authority. 2005. Information accessed from TVA’s website
(www.tva.gov). Site last accessed June 14, 2005.

F-5

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Appendix G. Past Exposure Pathway Parameters
Surface Water Ingestion
As far as ATSDR has been able to determine, East Fork Poplar Creek (EFPC) water has not been
used as a primary source of drinking water since the time the Y-12 plant was built in the early
1940s. ATSDR’s exposure pathway evaluation of mercury in EFPC water thus includes only
incidental ingestion and dermal contact with the water.
ATSDR has sufficient anecdotal information that children played and swam in EFPC. For
example, several people told ATSDR that they did so as children, or they knew children who did.
ATSDR also knows of adults who waded through the creek for various reasons and on occasion
possibly fell into the creek. Some people washed their horses in EFPC. ATSDR knows, with less
certainty, how many children and adults played or swam in the creek, how often they did, how
much the children weighed, who played in the creek, and how much water they swallowed when
they played in the creek. These exposure parameters are based on a series of assumptions, as
described below.
• Body Weight (BW): The mean weight of an 8-year-old child is 28.1 kg (EPA 1997). The body
weight could have been lower, but ATSDR thought the chances were less likely that such a
small child would be playing in EFPC.
• Intake Rate (IR): ATSDR knows that children get water in their mouths when they swim.
ATSDR assumed that children who swam inadvertently swallowed 0.15 liters of water each
day they were in the creek (EPA 1997). ATSDR surmises that children old enough to play in
the creek knew not to swallow the water intentionally, but that children inadvertently do
swallow water is well known.
• Exposure Frequency (EF): ATSDR assumed a child could have played in the water for up to
2 weeks (for acute exposures) or intermittently for 75 days over
For the longer-duration
the course of a year (for intermediate-duration exposures).
exposures, the dose is
ATSDR selected 75 days for intermittent exposures as follows:
calculated from an average
first, Oak Ridge receives an average of 60 inches of rain or snow
of mercury concentrations
over a calendar quarter. For
(combined) per year. Therefore, ATSDR estimated that children
acute exposures, the dose is
did not play outside for approximately 3 months during the year
calculated from (higher)
because of wet weather. ATSDR also assumed that another 3
average weekly water
months were too cold to play outside in the creek. In the
mercury concentrations.
remaining 6 months, during 3 of those months children might
have played outside 15 days per month, and for the remaining 3 months they might have
played outside 10 days per month. In this estimate, the total number of days a child played
outside, and in EFPC, was 75 days. This means that a child played in EFPC 20 percent of the
days of the year (75 days - 365 days = 0.2), which ATSDR considers a conservative
estimate.

G-1

Summary of Assumptions Implemented for Analysis of the Water Exposure Pathway
• To determine how much of the mercury released to
EFPC was elemental mercury was not possible. No
reliable information provides the dates or quantities
of elemental mercury disposed of in EFPC. One
suggestion is that the amount of elemental mercury
increased after mercury spills occurred. Elemental
mercury is measurable in water, but it has a very
low solubility (0.056 mg/L at 25° C). Elemental
mercury is also very bio-unavailable. Thus
ATSDR’s calculations assumed that 100 percent of
the inorganic mercury in water behaved like the
ionic forms of inorganic mercury, such as mercuric
nitrate. This was a conservative assumption—
mercuric nitrate is one of the most bioavailable
forms of mercury.

ATSDR’s Human and Environmental
Exposure Assumptions for the Surface
Water Ingestion Pathway
� A child weighing 28.1 kg swam or played in
EFPC for as many as 75 days a year, and
accidentally swallowed 0.15 liters of water
from the creek each day he or she played in
the creek.
� The child played in the creek daily, for up to
two weeks, on some occasions (for acute
exposures); and intermittently for 75 days
during a year at other times (for longerduration exposures).
� The mercury in the water was 100 percent
inorganic mercury when inorganic mercury
doses were calculated.
� 100 percent of the methylmercury in the
water is bioavailable and 60 percent of the
inorganic mercury in the water is
bioavailable.
� Weekly water mercury concentrations were
used to evaluate acute exposures.
� Quarterly water mercury concentrations
were used to evaluate longer-duration
exposures.

• The portion of methylmercury in EFPC during the
1950s and 1960s was less than 1 percent of the total
mercury. For the purposes of calculating doses to
methylmercury, ATSDR assumed that the portion
of methylmercury was equal to 0.3 percent of the
total mercury concentration. This percent was the
highest measured concentration of methylmercury
found in the scientific literature. For the purposes
of calculating exposure doses to inorganic mercury in water, ATSDR assumed that 100
percent of the mercury in the water samples was inorganic mercury. These were conservative
assumptions; the methylmercury portion was likely less than 0.3 percent.
• ATSDR hypothesized that there was a loss of mercury to sediment and air between its source
at the Y-12 plant and the nearest property off site where children could have played. How
much mercury was lost to sediment and air is not known, but because that distance is
relatively short, we assumed the amount of mercury lost was insignificant. The values of the
reported mercury concentrations due to loss of mercury from the water thus were not
reduced. This was a conservative assumption—some mercury was in fact lost to sediment
and air.
• ATSDR surmised that some mercury in the water remained dissolved. And that some
mercury precipitated and was bound to other inorganic or organic species. Mercuric sulfide
for example was present in the soils. To assume that mercuric sulfide formed in the water
was reasonable. But how much inorganic mercury was fully dissolved and how much was
not dissolved was not known. Thus we made no specific assumption concerning the
proportion of dissolved and undissolved inorganic mercury in water; it doesn’t help to
identify the amount of mercury that was bioavailable. We did not suggest that the distribution
of bioavailable and biounavailable inorganic mercury necessarily was in the same proportion
as the distribution of dissolved and undissolved mercury in water. We knew that fully

G-2

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

dissolved mercuric chloride was not 100 percent bioavailable. At the same time, precipitated
or bound mercury could become dissolved and bioavailable in the stomach.
• ATSDR assumed that the relative bioavailability of inorganic mercury in EFPC in the 1950s
was 60 percent. This value was calculated from the reported bioavailability of mercuric
nitrate (15 percent) divided by the reported upper range of the
No sufficient data are
bioavailability of mercuric chloride in adult mice (25 percent). This
available to estimate
assumption is equivalent to assuming that the mercury in the water in
the bioavailability of
EFPC is absorbed into the bloodstream to the same extent as mercuric
inorganic mercury in
EFPC during the
nitrate. This is likely a conservative assumption—it does not consider
that some of the mercury was lost from the water and some might have

1950s.
been less bioavailable than was mercuric nitrate.

Although the relative bioavailability factor is highly uncertain and variable, ATSDR’s
conclusion is not strongly dependent on the choice of bioavailability factors. A higher relative
bioavailability factor means more data (more weeks) are available
Mercury Water Exposure
when the mercury concentrations exceeded the acute oral inorganic
Pathway Data Assessment

mercury MRL; a lower relative bioavailability factor means fewer

Limitations
weeks when the data exceeded the MRL. Using a relative
� Missing data prior to 1956
bioavailability factor of 60 percent (ATSDR’s choice), weekly
� Analytical methods for

concentrations that exceed the MRL are available during the years

measuring mercury were no
better than ± 40 percent
1956, 1957, and 1958. If the relative bioavailability factor is lowered
� Not known how much
to 40 percent, only weekly data during 1957 and 1958 exceed the

mercury was lost to

MRL. Only at a relative bioavailability below 11 percent would all
sediment and air
of the weekly mercury concentrations fall below the acute oral
� No real good sense of the
MRL. But no compelling evidence suggests reducing the relative
relative bioavailability of
bioavailability below 11 percent, which is an absolute bioavailability
mercury in EFPC water
for inorganic mercury of less than 3 percent.

Results
Using all of the above-mentioned assumptions, ATSDR calculated mercury doses and made the
following observations:
• The calculated short-term inorganic mercury doses from ingestion of water from EFPC
between May and September were above the ATSDR acute oral inorganic mercury MRL in
1956, 1957, and 1958, but not in other years.
• The calculated longer-duration inorganic mercury doses were below the ATSDR intermediate
oral inorganic mercury MRL for all years.
• The calculated methylmercury doses were below the ATSDR chronic oral methylmercury
MRL for all years.
Soil-Sediment Ingestion
ATSDR considered two types of mercury in the soil—inorganic mercury and methylmercury.
The mercury in EFPC floodplain soil and sediment is primarily inorganic mercury, but a small
amount is methylmercury. Methylmercury is slowly formed in sediment and soils by bacteria or
fungi which attach methyl groups to inorganic mercury. Conditions which favor the conversion
of inorganic mercury to methylmercury are not well understood. Measurements of

G-3

methylmercury in soil from the EFPC floodplain range from 0.0008 to 0.0044 percent of the total
mercury in the soil (SAIC 1994c). When considering the inorganic mercury exposures, ATSDR
assumed that the data (representing total mercury) is 100 percent inorganic mercury; and when
considering methylmercury exposures, ATSDR assumed that 0.0044 percent of the total mercury
is methylmercury.
The EFPC RI data are presented as composite samples using the average mercury concentrations
in 12-inch, 16-inch, or 24-inch cores. If the mercury in a 16-inch core sample (for example) is
entirely localized in a 3-inch layer and the remainder of the core soil is clean, then the average
mercury concentration in that 3-inch layer before being composited (i.e., mixed and blended
together) will theoretically be 5.3 times higher than the mercury concentration in the entire core
after being mixed and reported as a composite (i.e., 16 inches - 3 inches = 5.3). Multiplying the
average core concentration times the multiplier (5.3) results in the theoretical maximum mercury
concentration for a 3-inch layer. ATSDR calculated the theoretical maximum mercury
concentrations for 3-inch layers for all the EFPC RI soil core data in this way. 56 The results of
these calculations are referred to as the “adjusted” RI data.
Specific mercury concentrations that ATSDR used in the calculations are discussed below:
• Intake Rate (IR): Experimental studies have
reported soil intake rates for children range
from approximately 40 to 270 milligrams per
day (mg/day) with 100 mg/day representing
the best estimate of the average intake rate.
There are very few data on soil ingestion by
adults, but limited experimental studies
suggest a soil intake rate in adults of up to 100
mg/day, with an average intake of 50 mg/day
(EPA 1997). ATSDR used soil intake rates of
100 mg/day for adults and 200 mg/day for
children.

Soil Ingestion Exposure Dose Equation
D = (C x IR x AF x EF x CF) / BW
Where,
D = exposure dose (mg/kg/day)

C = mercury concentration (mg/kg/day)

IR = intake rate of contaminated soil (mg/day)

AF = bioavailability factor (unitless)

EF = exposure factor (unitless)

CF = conversion factor (10-6 kg/mg)

BW = body weight (kg)

Dermal Contact with Soil Exposure Dose Equation
D = (C x A x AF x EF x CF) / BW

Where,
Young children (6 years old and younger)
D = exposure dose (mg/kg/day)

occasionally exhibit soil-pica behavior which
C = mercury concentration (mg/kg)

is typically characterized by soil intake rates
A = soil adhered (mg/day)

between 1,000 and 5,000 mg/day. These
AF = bioavailability factor (unitless)

children intentionally eat soil and ingestion in
EF = exposure factor (unitless)

these cases is not accidental. Occurrence of
CF = conversion factor (10-6 kg/mg)

BW = body weight (kg)

soil-pica behavior is rare (less than 1 percent
of young children in the U.S. population) but
rates vary widely. Soil-pica behavior is influenced by the child’s nutritional status and the
quality of child care and supervision. ATSDR does not know whether soil-pica behavior
occurred among children living near the EFPC floodplain. However, if it did occur, it
represents a worst-case intake rate. Pica behavior is considered under acute exposures.
ATSDR assumed an intake rate of 5,000 mg/day for children who exhibit soil-pica behavior.
This rate is 25 times higher than our default intake rate for children and may lead to adverse
56

Different core lengths have different multipliers: 3.3 for 10-inch cores, 4 for 1-foot cores, 8 for 2-foot cores, and
5.3 for 16-inch cores.

G-4

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

health effects at soil mercury concentrations 25 times lower than those concentrations which
cause effects in children who ingest soil incidentally. ATSDR did not consider pica-soil
behavior further in this report.
• Soil Adhered (A): There are few studies available which provide consistent and reliable
information regarding the amount of soil that adheres to the skin. ATSDR used U.S.EPA
default values for the total amount of soil that adheres to the skin. These values are based on
estimates of the exposed body surface area for people in different age groups. For children
the value is 525 milligrams (mg) and for adults the value is 326 mg of soil (ATSDR 2005;
EPA 1997, 2001).
• Bioavailability (AF): When a person swallows mercury-contaminated soil or gets it on his or
her skin, not all of the mercury is absorbed into the body. Some mercury remains with the
soil and passes through the gastrointestinal tract and is eliminated in the feces. Similarly,
when mercury-contaminated soil adheres to the skin, not all the mercury in the soil is
absorbed through the skin. The fraction or percent of mercury in the soil absorbed into the
blood is called the mercury bioavailability.
Revis et al. (1989) reported that EFPC floodplain soils contain 84–98 percent mercuric
sulfide, an insoluble salt. Although mercuric sulfide is very insoluble in water, 57 studies
comparing it with mercuric chloride show that its bioavailability is greater than is predicted
from water solubility alone. In one mouse study, the kidney deposition of mercury was
approximately 30–60 times lower in mice exposed to mercuric sulfide as compared with
mice exposed to mercuric chloride. This study does not provide a measure of bioavailability,
but it does show that mercuric sulfide is absorbed from the gastrointestinal tract at a
measurable extent (Schoof and Nielsen 1997). From this and other studies, the bioavailability
of mercuric sulfide is known to be considerably lower than mercuric chloride, although
studies to measure its specific bioavailability have not been identified in the scientific
literature (ATSDR 1999).
In the early 1990s, Sheppard et al. (1995) studied heavy metals in soils and reported that the
bioavailability of mercury in soil-amended diets in laboratory mice was 44 percent of that in
diets consisting of feed alone. This means that independent of other factors, the soil matrix
by itself will decrease the bioavailability of ingested inorganic mercury. ATSDR used this
figure (0.4) in the dose estimates to reflect the fact that we are considering ingestion of
mercury-contaminated soil and not mercury dissolved in water, as given to laboratory
animals in the studies used to derive the MRLs, for example.
The highest oral bioavailability reported in the scientific literature for any inorganic mercury
species is 38 percent for mercuric chloride administered in water to week-old suckling
laboratory mice; for adult mice the figure is 25 percent (Kostial et al. 1978). Due to the soil
matrix, ATSDR assumed the oral bioavailability of inorganic mercury in soil was 40 percent
of these figures, or 15 percent and 10 percent for children and adults, respectively. ATSDR
recognizes that these oral bioavailability factors are very likely conservative because they do
not necessarily account for the diversity of mercury species in EFPC floodplain soil, most of
which may be less bioavailable than mercuric chloride. But no sufficient evidence establishes
a lower bioavailability factor.
57

The solubility product constant (Ksp) for HgS at 25°C is 2E-53.

G-5

In contrast to inorganic mercury, methylmercury seems to be nearly completely absorbed (95
percent) following ingestion (Miettinen 1973). There is no evidence that a soil matrix inhibits
the absorption of methylmercury from the gut; therefore, ATSDR assumed that
methylmercury is 95 percent bioavailable, if ingested.
In contrast to oral bioavailability of mercury in soil, no quantitative data describe the dermal
bioavailability of mercury in soil. ATSDR thus assumed the dermal bioavailabilities of
inorganic mercury and organic mercury in soil are the same as the oral bioavailabilities.
• Exposure Factor (EF): ATSDR considered both acute exposure (1–14 days) and
intermediate duration exposure (15–364 days in a year). For acute exposure, only exposures
to soil or sediment with very high mercury concentrations were considered because humans
can eliminate mercury before harm occurs, if the exposures are not too high or too frequent.
Exposures of intermediate duration may involve soil from a variety of locations and with a
range of mercury concentrations. ATSDR calculations for intermediate exposures included
average soil mercury concentrations from multiple groupings of data.
The exposure factor expresses how often or how long a person is exposed to a contaminated
medium. For a short-term or acute exposure, the exposure factor is 1. This indicates that for
the duration of the exposure, a person is exposed continuously or daily. For intermediate- and
long-term exposures, however, ATSDR calculates an average exposure over the duration that
exposures occur. In the case of ingesting soil or sediment, exposures might have occurred
over several years, but not necessarily in consecutive days. ATSDR assumed that exposure to
soil or sediment does not occur every day of the year, but rather is largely dependent on
season and weather conditions. An exposure factor of 90 days a year (or one-quarter year)
was used as the maximum number of days in a year a person was exposed to mercurycontaminated soil from the EFPC floodplain.
• Body Weight (BW): ATSDR assumed a body weight of 70 kg (154 pounds) for adults and
28.1 kg (62 pounds) for children. Sometimes, ATSDR and U.S.EPA assume higher weights
than these values, but these are more conservative. Smaller body weights in the exposure
dose equations result in higher mercury doses when all other parameters are the same.

Results
Table G-1 contains the soil and sediment dose calculations for acute, intermediate, and chronic
exposure. The lowest concentration that results in doses above ATSDR’s oral mercury MRL is
2,400 ppm.

G-6

2.8 × 10-4
1.4 × 10-1
3.3

8.4 × 10-1
2.6

ratio dose to MRL
ratio dermal dose to oral dose

2,400
0.00033
0.1
0.25
70

1.7 × 10-3

2,400
0.00053
0.15
0.25
28.1

Intermediate
Child
Adult

D = exposure dose

C = contaminant concentration
A = soil adhered
AF = bioavailability factor
EF = exposure factor
BW = body weight

D = C x A x AF x EF / BW

3.7 × 10-1

2.6 × 10-3

2,400
0.0002
0.15
1
28.1

9.6 × 10-1
2.6

6.7 × 10-3

2,400
0.00053
0.15
1
28.1

Child

Inorganic

4.3 × 10-2

3.2 × 10-1

ratio dose to MRL

Dermal Exposure Route

8.6 × 10-5

2,400
0.0001
0.1
0.25
70

6.4 × 10-4

2,400
0.0002
0.15
0.25
28.1

Adult

Adult

G-7

1.6 × 10-1
3.3

1.1 × 10-3

2,400
0.00033
0.1
1
70

Acute

4.9 × 10-2

3.4 × 10-4

2,400
0.0001
0.1
1
70

Acute

Intermediate
Child
Adult
Child

0.007

Inorganic
0.002

D = exposure dose

C = contaminant concentration
IR = intake rate
AF = bioavailability factor
EF = exposure factor
BW = body weight

D = C x IR x AF x EF / BW

Oral Exposure Route
MRL

1.6 × 10-3
2.6

4.7 × 10-7

0.11
0.00053
0.95
0.25
28.1

Child

6.0 × 10-4

1.8 × 10-7

0.11
0.0004
0.95
0.25
28.1

Child

3.9 × 10-4
3.3

1.2 × 10-7

0.11
0.00033
0.95
0.25
70

Adult

6.2 × 10-3
2.6

1.9 × 10-6

0.11
0.00053
0.95
1
28.1

Chronic
Child

2.4 × 10-3

7.1 × 10-7

Organic

1.2 × 10-4

Child
0.11
0.0002
0.95
1
28.1

Chronic

3.6 × 10-8

0.11
0.0001
0.95
0.25
70

Adult

Organic
0.0003

Table G-1. Soil and Sediment Exposure Dose Calculations

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

1.6 × 10-3
3.3

4.7 × 10-7

0.11
0.00033
0.95
1
70

Adult

4.8 × 10-4

1.4 × 10-7

0.11
0.0001
0.95
1
70

Adult

unitless
unitless

mg/kg/day

mg/kg
kg/day
unitless
unitless
kg

Units

unitless

mg/kg/day

mg/kg
kg/day
unitless
unitless
kg

Units
mg/kg/day

Consumption of Fish
The only significant human exposure pathway to methylmercury in fish is ingestion of fish.
Estimates of mercury exposure are based on a series of assumptions that account for how much
mercury is in the fish, how much fish people eat, and how
Fish Ingestion Exposure Dose Equation
much mercury that is swallowed is absorbed into the
D = (C x IR x AF x CF) / BW
bloodstream.
• Mercury Concentrations (C): ATSDR calculated
human methylmercury doses from the fish data
presented in Table 12. For chronic exposures,
ATSDR considered the highest average (i.e., mean)
mercury concentrations in fish samples collected from
each sampling location (EFPC, Poplar Creek, Clinch
River, and Watts Bar Reservoir). For acute exposures,
ATSDR considered the maximum reported mercury
concentration in fish collected from each sampling
location.

Where,
D = exposure dose (mg/kg/d)
C = mercury concentration (mg/kg)
IR = intake rate of contaminated fish (mg/d)
AF = bioavailability factor (unitless)
CF = conversion factor (10-6 kg/mg)
BW = body weight (kg)
The exposure factor (EF) which is used in
several other exposure dose equations is
figured into the intake rate and does not
appear separately in the equation.

• Intake Rate (IR): Intake rates vary widely between individuals and are highly uncertain for
each population group. For chronic exposures, ATSDR used the mean and maximum adult
fish intake rates developed by Task 2 (except the maximum intake rate ATSDR used for
EFPC was the U.S.EPA rate for average daily fish consumption for recreational anglers in
small ponds or streams 58). Each of the child intake rates are one-half of the adult rates rather
than the 20 percent that Task 2 used because, in our model, ATSDR used an older child who
would eat more fish than Task 2 used in its model. For acute exposures, ATSDR assumed a
person would eat one or two whole fish meals consisting of 170 grams (6 ounces) or 340
grams (12 ounces) of fish, respectively. Refer to Table G-2 (ingestion—kg/day) and Table G­
3 (ingestion—meals/year) for location-specific chronic consumption rates.
The Task 2 fish consumption rates were discussed in meetings with the Oak Ridge Health
Agreement Steering Panel (ORHASP), which oversaw the Oak Ridge Dose Reconstruction
efforts. ORHASP members expressed limited confidence concerning the consumption rates.
However, the maximum Task 2 rates for the Watts Bar Reservoir are only slightly higher
than the highest fish consumption rates that ATSDR staff recorded during interviews with
anglers around the Watts Bar Reservoir during the 1997 exposure investigation (ATSDR
1998).

58

Task 2 did not present a maximum fish ingestion rate for EFPC fishers.

G-8

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Table G-2. Chronic Fish Intake Rates (kg/day)

Average Consumption Rates
(kg/day)
Recreational
child
adult

Maximum Consumption Rates
(kg/day)
Recreational
child
adult

EFPC
0.0006
0.0012
0.002
0.004
0.009
0.018
0.033
0.065
Poplar Creek/Clinch River1
Watts Bar Reservoir
0.015
0.03
0.055
0.11
EFPC: East Fork Poplar Creek
kg/day: kilograms per day
1
Poplar Creek and Clinch River are presented together because the Task 2 investigation does not separate these
two locations, and therefore, intake rates can only be calculated as one combined location.

Table G-3. Chronic Fish Intake Rates (meals/year)
Average Consumption Rates
(meals/year)
Recreational
child
adult

Maximum Consumption Rates
(meals/year)
Recreational
adult
child

EFPC
1.3
2.6
4.3
8.6
19
39
70
140
Poplar Creek/Clinch River1
Watts Bar Reservoir
32
64
120
240
EFPC: East Fork Poplar Creek
One fish meal = 6 ounces
1
Poplar Creek and Clinch River are presented together because the Task 2 investigation does not separate these
two locations, and therefore, intake rates can only be calculated as one combined location.

• Bioavailability (AF): ATSDR assumed that the mercury measured in fish is 100 percent
methylmercury and that the methylmercury is completely bioavailable (i.e., bioavailability =
1) for both children and adults.
• Body Weight (BW): ATSDR assumed a body weight of 70 kg (154 pounds) for adults and
28.1 kg (62 pounds) for children.
Consumption of Fruits and Vegetables
The only significant exposure pathway to mercury in garden vegetation is ingestion of fruits and
vegetables. Estimates of mercury exposure are based on a series of assumptions that account for
how much mercury is in the produce, how much produce people eat, and how much ingested
mercury is absorbed into the bloodstream (ATSDR 2005):
• Mercury Concentration (C): ATSDR assumed that the total mercury measured in fruits and
vegetables is inorganic mercury. Mercury speciation studies of plants grown in soil with
inorganic mercury contamination indicate that the mercury taken into plants is taken up as
inorganic mercury (i.e., mercuric ions) (ChemRisk 1999a).
• Intake Rate (IR): ATSDR used an intake rate from the U.S.EPA Exposure Factors Handbook
(EPA 1997) for people living in the South. Adults and children were reported to eat 2.27

G-9

grams of homegrown vegetables per kilogram of body weight per day (g/kg/day) (EPA
1997). Note that a body weight factor is already incorporated into the intake rate.
• Bioavailability (AF): In contrast to oral bioavailability of mercury in soil, there is limited
quantitative data describing the oral bioavailability of mercury in produce. Therefore,
ATSDR assumed that the oral bioavailability of inorganic mercury in produce is the same as
the oral bioavailability in soil. ATSDR assumed the oral bioavailabilities of inorganic
mercury in produce are 15 percent and 10 percent for children and adults, respectively.
• Exposure Factor (EF): ATSDR assumed the same
exposure factor as for soil exposures; that is, people will
eat home-grown fruits and vegetables during 25 percent
of the days in a year for intermediate exposures (EF =
0.25) and everyday for acute exposures (EF = 1).

Edible Vegetation Ingestion Exposure
Dose Equation

Results

D = exposure dose (mg/kg/d)

D = (C x IR x AF x EF x CF)
Where,
C = mercury concentration (mg/kg)

Using the average mercury concentration of 1.6 ppm from
IR = intake rate of contaminated produce
leafy vegetables from the ORAU and SAIC data sets, the
(g/kg/day)
intermediate exposure doses to both children and adults are
AF = bioavailability factor (unitless)
well below the ATSDR inorganic mercury intermediate oral
EF = exposure factor (unitless)
MRL (0.002 mg/kg/day). Using the highest mercury
CF
= conversion factor (10-3 g/kg)
concentration measured in edible fruits and vegetables
among the ORAU and SAIC data sets (3.2 ppm for kale
leaf), the resulting acute exposure doses (for children and adults) are below the ATSDR
inorganic mercury acute oral MRL (0.007 mg/kg/day). Table G-4 presents the exposure dose
calculations for acute and intermediate ingestion of fruits and vegetables.
Table G-4. Fruit and Vegetable Exposure Dose Calculations
Oral Exposure Route
MRL

Inorganic

D = C x IR x AF x EF

0.002

0.007

intermediate

Acute

Units
mg/kg/day

child

Adult

child

adult

C = contaminant concentration

1.6

1.6

3.2

3.2

mg/kg

IR = intake rate

2.27

2.27

2.27

2.27

g/kg/day

AF = bioavailability factor

0.15

0.1

0.15

0.1

unitless

EF = exposure factor

0.25

0.25

1

1

unitless

CF = conversion factor

10-3

10-3

10-3

10-3

0.0001

0.00009

0.001

0.0007

0.07

0.04

0.14

0.1

D = exposure dose
ratio dose to MRL

G-10

g/kg
mg/kg/day
unitless

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Appendix H.

What You Need to Know About Mercury in Fish and Shellfish

H-1

However, nearly all fish and shellfish contain
traces of mercury. For most people, the risk
from mercury by eating fish and shellfish is
not a health concern. Yet, some fish and
shellfish contain higher levels of mercury that
may harm an unborn baby or young child’s
developing nervous system. The risks from
mercury in fish and shellfish depend on the
amount of fish and shellfish eaten and the
levels of mercury in the fish and shellfish.
Therefore, the Food and Drug Administration
(FDA) and the Environmental Protection
Agency (EPA) are advising women who may
become pregnant, pregnant women, nursing
mothers, and young children to avoid some
types of fish and eat fish and shellfish that are
lower in mercury.

healthy diet. Fish and shellfish contain highquality protein and other essential nutrients,
are low in saturated fat, and contain omega-3
fatty acids. A well-balanced diet that includes
a variety of fish and shellfish can contribute
to heart health and children’s proper growth
and development. So, women and young
children in particular should include fish or
shellfish in their diets due to the many
nutritional benefits.

Fish and shellfish are an important part of a

The Facts

EPA-823-F-04-009

For further information about the safety of locally
caught fish and shellfish, visit the Environmental
Protection Agency’s Fish Advisory website
www.epa.gov/ost/fish or contact your State or Local
Health Department. A list of state or local health
department contacts is available at
www.epa.gov/ost/fish. Click on Federal, State, and
Tribal Contacts. For information on EPA’s actions to
control mercury, visit EPA's mercury website at
www.epa.gov/mercury.

For further information about the risks of mercury
in fish and shellfish call the U.S. Food and Drug
Administration’s food information line toll-free at
1-888-SAFEFOOD or visit FDA’s Food Safety
website www.cfsan.fda.gov/seafood1.html.

from the
U.S. Food and Drug Administration
U.S. Environmental Protection Agency

Women Who Might Become Pregnant
Women Who are Pregnant
Nursing Mothers
Young Children

Advice for

w
Know
Aboutt
Mercury
n Fish
h
in
d
and
h
Shellfish

hatt You
u
W Need
d to
o

3 Safety Tips
1.

Do not eat:
• Shark
• Swordfish
• King Mackerel
• Tilefish
They contain high
levels of mercury.

B

y following these 3 recommendations for selecting and eating fish or shellfish, women and young
children will receive the benefits of eating fish and shellfish and be confident that they have reduced
their exposure to the harmful effects of mercury.

2.

Eat up to 12 ounces (2 average meals) a week of a

variety of fish and shellfish that are lower in mercury.
• Five of the most commonly eaten fish that are low
in mercury are shrimp, canned light tuna, salmon,
pollock, and catfish.
• Another commonly eaten fish, albacore (“white”) tuna
has more mercury than canned light tuna. So, when
choosing your two meals of fish and shellfish, you may
eat up to 6 ounces (one average meal) of albacore tuna
per week.

3.

Check local advisories about
the safety of fish caught by family
and friends in your local lakes,
rivers, and coastal areas.
If no advice is available, eat up to
6 ounces (one average meal) per week
of fish you catch from local waters,
but don’t consume any other fish
during that week.

Follow these same recommendations when feeding fish and shellfish to your young child, but serve smaller portions.
v or
Visit the Food and Drug Administration’s Food Safety Website www.cfsan.fda.gov
h
the Environmental Protection Agency’s Fish Advisory Website www.epa.gov/ost/fish
for a listing of mercury levels in fish.

Frequently Asked Questions about Mercury in Fish and Shellfish:

what is mercury?
Mercury occurs naturally in the environment
and can also be released into the air through
industrial pollution. Mercury falls from the air
and can accumulate in streams and oceans and
is turned into methylmercury in the water. It is
this type of mercury that can be harmful to your
unborn baby and young child. Fish absorb the
methylmercury as they feed in these waters and
so it builds up in them. It builds up more in
some types of fish and shellfish than others,
depending on what the fish eat, which is why
the levels vary.

I’m a woman who could have children
but I’m not pregnant - so why should I be
concerned about methylmercury?
If you regularly eat types of fish that are high in
methylmercury, it can accumulate in your blood
stream over time. Methylmercury is removed
from the body naturally, but it may take over a
year for the levels to drop significantly. Thus, it
may be present in a woman even before she
becomes pregnant. This is the reason why
women who are trying to become pregnant
should also avoid eating certain types of fish.

Ishellfish?
s there methylmercury in all fish and
Note:
If you have questions or
think you've been exposed
to large amounts of
methylmercury, see your
doctor or health care
provider immediately.

Nearly all fish and shellfish contain traces of
methylmercury. However, larger fish that have
lived longer have the highest levels of
methylmercury because they’ve had more time
to accumulate it. These large fish (swordfish,
shark, king mackerel and tilefish) pose the
greatest risk. Other types of fish and shellfish
may be eaten in the amounts recommended by
FDA and EPA.

I don’t see the fish I eat in the advisory.
What should I do?
If you want more information about the levels in the
various types of fish you eat, see the FDA food safety
website www.cfsan.fda.gov/~frf/sea-mehg.html or the
EPA website at www.epa.gov/ost/fish.

w
hat about fish sticks and fast food
sandwiches?
Fish sticks and “fast-food” sandwiches are commonly
made from fish that are low in mercury.

The advice about canned tuna is in the advisory,
but what's the advice about tuna steaks?
Because tuna steak generally contains higher levels of
mercury than canned light tuna, when choosing your
two meals of fish and shellfish, you may eat up to
6 ounces (one average meal) of tuna steak per week.

w
hat if I eat more than the recommended
amount of fish and shellfish in a week?
One week’s consumption of fish does not change the
level of methylmercury in the body much at all. If you
eat a lot of fish one week, you can cut back for the
next week or two. Just make sure you average the
recommended amount per week.

where do I get information about the safety of
fish caught recreationally by family or friends?

Before you go fishing, check your Fishing Regulations
Booklet for information about recreationally caught
fish. You can also contact your local health department
for information about local advisories. You need to
check local advisories because some kinds of fish and
shellfish caught in your local waters may have higher or
much lower than average levels of mercury. This
depends on the levels of mercury in the water in which
the fish are caught. Those fish with much lower levels
may be eaten more frequently and in larger amounts.

Reviewer Comment

Overarching Figure. The mercury cycle is complex. It is recommended that a
figure be used to outline key source-fate-exposure pathways, and
integrate/connect the various studied components (e.g., air, water, soil, fish,
people). Such a figure may help readers better understand the nature and extent
of contamination. Many mercury cycle figures exist and here are some examples
that may be tailored to this particular case:

3

I-1

Sediments and Methylation Potential. Sediments are a critical component of the
mercury cycle. While the reviewer appreciates that little bottom sediment exists in
the East Fork Poplar Creek (pg 81), there should be some mention of the critical
role that sediments play in the biomethylation of inorganic mercury into
methylmercury. Have studies characterized the biomethylation potential of the
sediment in the study region?

http://www.oar.noaa.gov/spotlite/2008/spot_mercury.html
http://people.uwec.edu/piercech/Hg/mercury_water/cycling.htm
http://sofia.usgs.gov/projects/index.php?project_url=evergl_merc

The contamination appears to be adequately described and the information which
is needed in order to draw conclusions about health risk, or the absence of health
risk, is summarized in a readily usable form. In general, the site maps are useful
in understanding the extent of contamination although the occasional figure can
be improved. For example, in Figure 17 it is most difficult to discern marked Hg
concentrations from the rest of the Figure, even upon higher resolution. I suggest
a different color scheme, or perhaps some other way to get this information
across to the reader.

2

Does the public health assessment adequately describe the nature and extent of contamination?
1
Yes, to the extent possible given the available data. In many cases, data are not
sufficient to draw conclusions about the nature and extent of contamination in the
early periods of site use (e.g., in the 1950s and 1960s). Please see the attached
detailed comments concerning specific issues.

Reviewer

ATSDR added information about biomethylation of inorganic mercury into
methylmercury to the background section of the public health assessment.

ATSDR added information about the mercury cycle to the background section
of the public health assessment.

Thank you for the comment. Two additional figures (Figure 20 and Figure 21)
were created and added to the public health assessment to present the
information more clearly to the reader.

Thank you for the comment. The detailed comments are addressed below.

ATSDR Response

ATSDR received the following comments from independent peer reviewers on the Oak Ridge Reservation: Evaluation of Y-12
Mercury Releases public health assessment. For comments that questioned the validity of statements made in the document, ATSDR
verified or corrected the statements.

Appendix I. Peer Reviewer Comments and ATSDR Responses

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Other Mercury Sources. While the focus of the Report is on the Y-12 facility, in
order to understand the totality of mercury contamination in the study region, it is
necessary to explore (or at least mention) other pertinent sources of mercury. In
the general population, most are exposed to elemental mercury via personal
dental amalgam and to methylmercury via store-bought fish. This should be
emphasized in the report. Given all the concerns in the area about mercury
exposure and the perception that local contamination is the principal source, it is
highly likely that the principal sources of mercury to residents are personal
amalgams and store-bought fish such as canned tuna. There is also mention of
negligible mercury inputs from a nearby electrical generating facility (Appendix F)
but what about other facilities (e.g., coal-fired plants, chlor-alkali, etc.) in the
region, both current and historic. A listing of these seems relevant. With very
recent advances in isotopic speciation of mercury (e.g., Bergquist & Blum,
Science 318(5849): 417-420), future research may be able to tease apart sourcefate-exposure pathways with greater precision.

ATSDR evaluated probable exposure scenarios relevant to human health in
this public health assessment. Hunting for terrestrial animals in the floodplain
is not common. No one is eating terrestrial animals from the floodplain on a
regular basis. Therefore, we do not believe this is a realistic pathway of
concern.

Terrestrial Risk. A relatively new study by Prof. Dan Cristol published in Science
(320(5874):335) documents for the first time that methylmercury contamination
can move up the terrestrial ecosystem to levels of health concern to songbirds
and other organisms that feed on invertebrates in riparian habitats. While most
studies focus on methylmercury risks from the aquatic ecosystem, new attention
needs to be paid to the terrestrial system. I am not sure if this is of clear human
health risk in this particular study region, but is something to be investigated given
the immense point source and worthy of mention in the report.

ATSDR added information about other sources of mercury to the background
section of the public health assessment.

The DOE conducted an ecological baseline risk assessment of the EFPC
floodplain as part of the remedial investigation and feasibility study (RI/FS).
Aquatic species (variety of fish, crayfish, algae, and heterotrophic microbes),
terrestrial vegetation, and terrestrial species (e.g., shrews, mice, wren, blue
heron, earthworms, and insects) were targeted for population surveys and
measured for body burden analyses. The ecological baseline risk assessment
concluded that the terrestrial resources generally exhibited less body burden
of mercury and less risk to the terrestrial species than the aquatic species.
Mercury in terrestrial species was not at levels of human health concern.

ATSDR Response

Reviewer Comment

I-2

Does the public health assessment adequately describe the existence of potential pathways of human exposure?
1
Largely yes. ATSDR has adequately described the applicable exposure
ATSDR added “E” to the eliminated pathways and added a definition for
pathways. It would be helpful if ATSDR provided a discussion as to why certain
eliminated.
pathways are not relevant and were not evaluated. This could be easily done in a
table.

Reviewer

I-3

Biomarkers were used in the following Y-12 mercury exposure-based studies.
A table has not been added to the public health assessment because the
appropriate mercury biomarkers and analytical analyses are easily located in
the literature and may change over time.

Mercury Biomarkers. Hair largely reflects exposure to methylmercury, urine
largely reflects exposure to elemental/inorganic mercury, and blood represents
exposure to both organic and inorganic sources of mercury. None of the exposure
assessments discussed in the report utilized all three biomarkers, which is a
limitation. Given the importance of characterizing exposures in the mercury risk
assessment, it is recommended that a table be included which articulates the
three main mercury biomarkers, what form of mercury they assess, thresholds for
each, etc. so that community members are better informed when planning future
exposure-based studies.

A physician at the Emory University School of Public Health requested that
ATSDR and the Centers for Disease Control and Prevention (CDC) National
Center for Environmental Health (NCEH) facilitate clinical laboratory support
by the NCEH Environmental Health Laboratory for patients referred to Emory
by the Oak Ridge physician in 1992 and 1993. Because of patient-to­
physician and physician-to-physician confidentiality, results of the clinical
analysis have not been released to public health agencies. However, the
Emory physician did not recommend that ATSDR conduct a follow-up

At the request of an Oak Ridge physician, ATSDR evaluated clinical
information in 1992, on 45 patients tested for heavy metals to determine
whether exposure to metals caused these patients' illnesses. ATSDR
concluded that this case series did not provide sufficient evidence to associate
low levels of metals with these diseases.

In June and July 1984, TDOH and CDC conducted a pilot survey to document
human body levels of inorganic mercury (see Section II.F.5). Additionally, to
follow up on the findings of previous studies and investigations of the Watts
Bar Reservoir, ATSDR conducted an exposure investigation in 1996, to
measure actual PCB levels in serum and mercury levels in blood of people
consuming moderate to large amounts of fish and turtles from the Watts Bar
Reservoir (see Section II.F.4).

Other agencies have addressed occupational exposures to mercury from the
Y-12 plant. This assessment focuses on the general community/non-worker
population. The degree of exposure is very different between the two
populations. ATSDR's summary of the studies of mercury workers, including
the Albers et al. 1988 study, can be found at
http://www.atsdr.cdc.gov/HAC/oakridge/phact/c_3.html#314.

Occupational Exposures. The report is focused on community exposures but
what about workers (past, present), many of whom may live in the neighboring
communities? Is there any occupational exposure data that may increase
understanding of risks to the broader community? For example, a study by Albers
et al. (Ann Neurol 1988; 24: 651-659) documented tremendous urinary mercury
levels among 502 Y-12 plant workers, and associated these exposures with a
range of peripheral neuropathies.

3

Thank you for the comment.

ATSDR Response

ATSDR does a very nice job of laying out potential pathways of exposure, and in
explaining that conservative (or safety-oriented) decisions were made in
comparing these pathways with available health information. The biggest
exposure of any mercury form is from fish consumption and ATSDR expanded
this area appropriately.

Reviewer Comment

2

Reviewer

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Reviewer Comment

Further, ATSDR and ORRHES jointly created a fact sheet on available
environmental and occupational medical resources
(http://www.atsdr.cdc.gov/HAC/oakridge/factsheets/env_med_res.html ).

On August 27, 2002, ORRHES "determined that discussion of public health
activities related to the establishment of a clinic, clinical evaluations, medical
monitoring, health surveillance, health studies, and/or biological monitoring is
premature to ATSDR’s Public Health Assessment (PHA) process. Thus, the
ORRHES recommends that formal consideration of these issues be
postponed until the ATSDR PHA process identifies and characterizes an
exposure of an off-site population at levels of health concern."

From 2001 to August 2002, ATSDR staff worked with ORRHES members and
interested individuals to review the community concerns and issues related to
the need for an environmental health clinic in Oak Ridge. The ORRHES
reviewed the public comments/concerns from members of the community
expressing the need for a clinic in the Oak Ridge area to serve exposed/ill
persons and conducted a program review of state and federal environmental
and clinical programs and mandates. This program review included a review
of Congressional mandates of federal agencies with regards to environmental
clinical assessments, and the limits placed by the Congressional mandates
that specifies what they can do within their programs. The clinical program
review included a comparison that highlights target populations, types of
assessments, and criteria for screening /medical evaluation, and follow-up
actions/benefits for each agency. The minutes from the ORRHES meetings on
December 4, 2001, March 26, 2002, and August 27, 2002, are particularly
relevant to this topic and are posted on ATSDR's Oak Ridge Reservation
Public Health Web site at
http://www.atsdr.cdc.gov/HAC/oakridge/meet/orrhes.html.

investigation concerning any exposure and recommended to the Tennessee
Department of Health (TDOH) that they not conduct a study of these
individuals as a group.

ATSDR Response

I-4

Are all relevant environmental, toxicological, and radiological data (i.e., hazard identification, exposure assessment) being appropriately used)?
Yes, the available data are appropriately used. The lack of data, as noted above,
1
Thank you for the comment. The detailed comments are addressed below.
is a problem in terms of the conclusions that could be drawn. Please see the
detailed comments.

Reviewer

3

2

Reviewer

I-5

See comment above about biomarkers.

ATSDR clarified in the text that Table 7 is focused on the health guidelines for
the different forms of mercury referenced in this public health assessment.
Additional toxicological studies can be found in ATSDR's Toxicological Profile
for Mercury and EPA's Integrated Risk Information System (IRIS).

The focus of Table 7 is to provide ATSDR's and EPA's health guidelines and
the basis for them. It is not meant to be a comprehensive literature review.
The studies cited are the basis for the development of the health guidelines.
Other more recent studies are not included because they were not used to
develop the health guidelines.

Public health assessments are written primarily for the public. The document
discusses each guideline and their associated doses and endpoints, which is
sufficient information for the target audience. To provide the additional source
of information for the more scientific reader, ATSDR added a footnote that
references Dourson et al. 2001.

One toxicological detail that I suggest adding, however, is the reasoning behind
the differences among FDA, ATSDR and EPA on the methylmercury “safe” dose,
whether it is called ADI, MRL, or RfD. For example, on Page 93, lines 18-28,
ATSDR avoids discussing the well established controversy on the choice of the
study for the basis of the ATSDR’s MRL and FDA’s ADI on one side, and EPA’s
RfD on the other. The fact remains that PCB contamination of Faroese breast
milk is the likely source of neurological effects in this population, when compared
to the absence of both PCBs in breast milk fed to, and of neurological effects in,
Seychelles infants who actually end up with more methylmercury exposure than
the Faroese infants. I encourage ATSDR to add some text on this controversy,
perhaps as a footnote. (See for example Dourson et al., 2001. Uncertainties in
the reference dose of methylmercury. Neuor Toxicology, 22:677-689.)

Mercury Health Risks – Tables. Table 7 provides a good overview of mercury
risks but it should be emphasized that this table does not contain all studies but
rather selected studies. Not all studies have been reviewed here. I am also
surprised that most studies cited are rather dated. This table may be a good place
to indicate what key biomarkers are used (hair, blood, urine) and associated
thresholds.

Thank you for the comment.

ATSDR Response

Overall, the information cited was appropriate for the task at hand. ATSDR did not
overly write the toxicology section, for example, but rather kept it to a level that
balanced the need to know, with the EGO effect (eyes-glazed-over) that often
comes with too much technical detail.

Reviewer Comment

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Reviewer

For methylmercury exposure in this public health assessment, ATSDR
focused on the most sensitive and best-documented health endpoint, which is
neurodevelopmental effects from prenatal methylmercury exposure. ATSDR
acknowledges recent studies have found limited evidence of association
between exposure to methylmercury and cardiovascular effects. However, at
this time more research is needed in epidemiological studies, dose-response
assessment, mechanisms, exposure assessment, and cardiovascular end
points to develop a dose-response relationship between methylmercury
exposure and cardiovascular effects (NAS 2000; Roman 2011).

Cardiovascular Impacts. There are an increasing number of recent studies
documenting cardiovascular effects associated with methylmercury exposure.
While there exists inconsistencies across studies (similar to neurodevelopmental
outcomes that vary across widely referenced cohorts), the possibility of mercuryassociated cardiovascular risk exists. Such a possibility should be discussed in
this ATSDR science-based report rather than revealed to community members
from other media sources.

Most of the environmental data (especially current data) evaluated for this
public health assessment were collected during the RCRA Superfund
process. Documents prepared for the RCRA programs must meet specific
standards for adequate quality assurance and control measures for chain-of­
custody procedures, laboratory procedures, and data reporting. The historical
data come from a variety of studies that used the best technologies available
at the time.
The detection limits for the surface water data ranged from 0.03-0.067 ppb
from 2005 to 2009. Detection limits prior to 2005 are not available, except the
detection limits from the RI/FS ranged from 0.05 to 0.2 ppb.

Analytical Uncertainty. An important discussion concerning analytical uncertainty
and data quality is offered in the sections focused on mercury levels in air and
water, but what about in soil, fish, and food (Sect IV – Public Health Evaluation)?
At least a sentence or two in each of those sections concerning QA/QC
parameters would help the reader appreciate the quality of the data presented.
Were accredited methods used? Were appropriate blanks included in all batch
runs? Using standard reference materials and replicates, what was the accuracy
and precision? Etc.

Detection Limits. There are several tables (e.g., Tables 18, 19, 20) that show
values to be below detection limits. For this information to be properly
understood, the analytical detection limits need to be provided. Perhaps as a
footnote? In only one case I could find (pg 120; blood mercury) were detection
limits provided, and this made it easier to put the data into perspective.

I-6

The fact that the general U.S. population on average consumes lower
amounts of seafood per capita is not relevant to deciding whether the Faroe
Islands and Seychelles studies apply to the U.S. population. Consumption of
seafood is popular in the United States, has increased over the years, and
occurs with greater frequency in coastal U.S. populations. Tomasallo et al. did
not evaluate neurocognitive performance in children from in utero exposure to
methylmercury; therefore, this paper has no bearing on the issues of mercury
toxicity. In addition, the National Academy of Sciences (NAS) recommends
that the Faroe Islands study be used to derive the health guideline for
mercury.

Projecting Faroe/Seychelles Data to Current Study (marine vs. freshwater).
Neurodevelopmental risks of methylmercury are largely calculated from
longitudinal birth cohort studies based in the Faroes and Seychelles. These are
populations that mainly consume marine fish, and do not necessarily reflect the
general U.S. population. Additionally, there exists the possibility that mercury
derived from marine-based fish do not necessarily predict risks associated with
mercury derived from freshwater-based fish. For example, see the paper by
Tomasallo et al. (Environ Res, 2010; 110: 62-69) which shows divergent results
between consumers of Great Lakes fish vs. marine fish. The caveats of using the
Faroe/Seychelles data to the current situation need to be clear.

Mozaffarian D. Fish, mercury, selenium and cardiovascular risk: current evidence
and unanswered questions. Int J Environ Res Public Health 2009;6:1894-1916

Mozaffarian D, Rimm EB. Fish intake, contaminants, and human health:
evaluating the risks and the benefits. JAMA 2006;296:1885-1899

Here are two review papers:

ATSDR Response

Reviewer Comment

Reviewer Comment

ATSDR Response

2

ATSDR added an explanation to each summary conclusion statement.

ATSDR clarified the text.

ATSDR clarified the text.

Page 4. ATSDR’s conclusion statements here and elsewhere should state briefly
the “why” part of the answer. For example, the second bullet under conclusion for
past mercury exposure could be extended with the phrase “because the releases
were below established safe concentrations.” In fact, page 5 of the text gives an
example to which each of these conclusionary statements should strive.
Specifically in regards to past exposure to mercury from EFPC fish, the
conclusion states that the estimated methylmercury exposure doses are below
ATSDR’’s and USEPA’s health guidelines. [Note here that I would clarify this
statement to say that the exposures are lower than ATSDR’’s and USEPA’s
“safe” doses, and thus thought to be without any risk.]

Page 6, second paragraph of second bullet. I do not understand the reference to
“most… doses”. Do you mean doses less than the safe dose because fish
consumption was less than that stated above? The juxtaposition is confusing as it
now reads.

Page 9, first bullet and elsewhere. What is a “comparison” value? Do you mean
the MRL? RfC? Safe concentration in air? I have no problem with referring MRLs,
RfDs and ADIs as “safe” doses, as long as this language is appropriately
caveated (see Barnes and Dourson, 1988, Reference Dose (RfD): Description
and use in health risk assessments. Reg. Tox Pharmaco. 8:471-486, page 481
with some language in this regard).

I-7

Thank you for the comment.

I believe that it does, but offer some suggestions to the conclusionary language at
front. For example,

Does the public health assessment accurately and clearly communicate the health threat posed by the site?
1
In general, yes. I do have concerns regarding the use of the term “risk” to discuss ATSDR used the NAS health effect level for the basis of comparison because
it was considered to have the most sensitive endpoint as well as the lowest
methylmercury and neurodevelopmental effects (of which I approve) versus
statements about causation of specific effects (i.e., renal effects) with inorganic
dose.
mercury exposure. See the specific comments for details. Another overarching
Specific comments are addressed below.
issue is that the effect levels from the Faroe Islands study are frequently used as
a comparison value for the estimated exposures but the Seychelles Islands data
are not used. As I noted in the specific comments, it would also be clearer if the
findings were also presented in tabular form.

Reviewer

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

3

Reviewer

ATSDR added information about the benefits of eating fish to the public health
assessment.

To put the estimated exposure doses into perspective, ATSDR added six
graphs to show the past and current doses from eating fish compared to the
health guidelines and health effect levels for methylmercury.

Nutrient-Toxicant Interactions. An important component of any assessment
concerning methylmercury risk from consuming fish requires a discussion of the
benefits posed by nutrients in fish, such as selenium and omega-3 fatty acids.
While this information is present in Appendix H (EPA/FDA – “What you need to
know about mercury in fish and shellfish”) and on pg 95, the information is not
tailored for this particular study. In the analysis of fish from local water sources,
were beneficial nutrients assessed? Do local residents believe that fish
consumption is good for health?

Risk Thresholds and Relative Risk. Figures 10, 11, and 12 outline risk thresholds
for the different mercury forms. Why not overlay onto this chart the range of
mercury exposures calculated for area residents? This would put the exposure
risk data into perspective, and visually document that exposures are indeed
relatively low.

I-8

Unfortunately, current science does not adequately support a robust analysis
of multiple chemical exposures and their interactions. Debate continues in the
scientific community about how best to evaluate exposure to a chemical
mixture both from a single pathway and from multiple, combined pathways. In
addition, estimating combined doses from multiple pathways is hampered by a
lack of knowledge of the levels of chemicals people are exposed to through
various pathways.

No, the health effect levels refer to inorganic mercury LOAELs (specifically,
autoimmune effects observed in rats). Figure 12, Figure 13, and Figure 14
show the health effect levels as well as the health guidelines for elemental,
inorganic, and organic mercury, respectively.

Page 10, last line and elsewhere. Is the phrase “health effect levels” in reference
to the NAS subtle neurological effects in Faroese children? ATSDR may wish to
show a graph of ranges of safe dose, intermediate dose and doses associated
with effect levels. An example is attached of this in reference to interpretation of
biological equivalents.

Multiple Stressors. It is clear that several chemical and non-chemical stressors
exist in the study region. However, there is very little mention or discussion about
the possibility of cumulative health impacts that may arise under this highly
relevant scenario. Mercury is only one of several environmental threats in the
region, and a complete public health assessment needs to discuss the
interactions of mercury with other stressors such as PCBs and socioeconomic
status. It is recommended that some text be devoted to multiple stressors and
ensuing public health risks.

ATSDR Response

Reviewer Comment

Reviewer

I-9

Community Concerns - Perception versus Reality. This is reflected in Sect VI
(Community Health Concerns). ATSDR responses are well versed, though I am
surprised that more studies concerning ‘knowledge, attitude, and beliefs’ (e.g.,
Appendix B-5) were not conducted in the region. The perception of living in a
polluted landscape may in fact be more damaging than the toxic chemicals
themselves. Here is a relevant paper that could be referred to - Kroll-Smith, J. S.,
and Couch, S. R. (1991). As if exposure to toxins were not enough: the social and
cultural system as a secondary stressor. Environmental Health Perspectives 95,
61-66.

Cumulative risks are difficult to accurately assess, given the uncertainty
involved with the historical doses. The population exposed the most from
multiple pathways are EFPC floodplain farm families. They may have been
exposed to harmful levels of elemental mercury in the air prior to 1963. They
may have been exposed to harmful levels of inorganic mercury in the surface
water and sediment prior to cleanup in 1996 and 1997. They may currently be
exposed to harmful levels of methylmercury in the fish, if warning signs are
ignored.

Multiple Mercury Exposure Routes. Similar to a comment earlier, all the risk
calculations are focused on single pathways of mercury exposure. However, area
residents are exposed to mercury via several different routes, in a simultaneous
manner, and it needs to be wondered what the cumulative risks are to all these
mercury sources. In addition, a nice diagram outlining the pertinent source-fate­
exposure routes would enable the reader to better appreciate the situation faced
by community residents.

ATSDR. 2003. Assessing the Health Education Needs of Residents in the
Area of the Oak Ridge, Tennessee. Association of Environmental and
Occupational Clinics, George Washington University. May 2003.

Jennifer Friday and Robin Turner. 2001. Scarboro Community Assessment
Report. Joint Center for Political and Economic Studies, Washington, DC.
August 2001.

Michael Benson and William Lyons. 1994. Report of Knowledge, Attitudes,
and Beliefs Survey of Residents of an Eight-County Area Surrounding Oak
Ridge, Tennessee. Tennessee Department of Health, Oak Ridge Health
Agreement Steering Panel, and Oak Ridge Reservation Local Oversight
Committee. August 12, 1994.

In addition, ATSDR used the information in the following studies:

Responding to community health concerns is an essential part of ATSDR’s
overall mission and commitment to public health. ATSDR actively gathered
comments and other information from those who live or work near the ORR,
and has addressed over 500 specific community health concerns in the public
health assessments (see Section VI for community health concerns related to
mercury releases from the Y-12 plant).

As discussed, exposure to the different forms of mercury result in different
health endpoints. The surface water and sediment doses from exposure to
inorganic mercury would have a cumulative effect on the same endpoint
(kidneys). ATSDR has already concluded that past exposure to both pathways
increased their risk of harmful renal effects. Exposure to elemental mercury in
the air prior to 1963 may also increase the risk of harmful renal effects,
however, ATSDR does not have adequate information to determine whether
this past exposure could have caused harmful renal effects.

ATSDR Response

Reviewer Comment

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Reviewer Comment

ATSDR Response

2

I-10

Page 58, line 14. Please add the words “at or” before the word “less” in the
phrase “Estimated doses less than these values…” These words also have to be
added elsewhere in the text.

Are there any other comments about the public health assessment that you would like to make?
1
Please see the attached list of detailed comments.

ATSDR added several figures and simplified the take home message to better
translate the results from the public health assessment to the public.

A Typical Exposure Scenario. In order to better translate the Report to the public,
why not offer a few typical case scenarios? Mercury risks to a child that grew up
in the region? Mercury risks to a resident that moved here in the 1980s? Etc.

ATSDR made the editorial change.

Thank you for the comments.

Site-specific exposure scenarios were evaluated by the ORHASP. Their final
report, Releases of Contaminants from Oak Ridge Facilities and Risks to
Public Health, is available at
http://health.state.tn.us/ceds/oakridge/ORHASP.pdf.

ATSDR added an Overall Conclusions section to the summary.

Take Home Message. The take home message is not obvious. The bulleted
conclusions are lengthy, spanning several pages. Given the importance of
straightforward public health messaging, is it not possible to offer a clearer
conclusion? I do realize the complexity of the case may not permit a simple
conclusion, but there may be other more effective means of communication such
as a summary table/figure.

3

Thank you for the comment. ATSDR added six graphs to show the estimated
past and current doses from eating fish compared to the health guidelines and
health effect levels for methylmercury.

I support ATSDR’s conclusions, with minor improvements as mentioned. The
overall intent of the text has been met and the further response to individual
public comments was helpful. It may be that occasional exposures to individuals
caused effects from this pollution (e.g., skin rashes from exposure to silver­
appearing-riparian-zone muck), but the population risk from acute and chronic
exposures was ATSDR’s concern and the text more than adequately addressed
these concerns. In fact, ATSDR was very good about explaining the risks
associated with exposures above the “safe” health level (e.g., MRL), and the fact
that it investigates more fully such risks and more exacting exposures when the
screening level exposure assessment exceeds these health levels. I suggest a
graphic to more easily show this, similar to what Sean Hayes of Summit
Toxicology has done in his Biological Equivalents work.

2

Are the conclusions and recommendations appropriate in view of the site’s condition as described in the public health assessment?
1
Yes. It is unfortunate that the limited data prevent most conclusions regarding
Thank you for the comment.
exposure in the 1950s and 1960s and that the data to describe current conditions
date from the late 1990s. Within the constraints of these limitations, the
conclusions and recommendations are appropriate.

Reviewer

Thank you for the comment.
Thank you for the comment.

Page 111, Table 21 and elsewhere. The precision of these wrought risk values
should be changed to appropriate levels.

Figures 10, 11 and 12 are so very nicely done. Good work!

None provided

3

I-11

Are there any other comments?
2
None, but thanks for the opportunity to help. The text was very nicely done and I
am proud be colleagues with several ATSDR scientists that no doubt assisted
with this work.

In general, you have not allowed sufficient resource to support individual
reviewers times, especially if they do the review as a member of another
organization. While many of us do not mind pro bono activity, it will likely limit
ATSDR’s ability to obtain qualified reviewers.

2

Thank you for the comment.

Thank you for the comment.

Thank you for the comment.

Thank you for the comment.

ATSDR made the editorial change.

Page 104, text box. I would add the phrase “including sensitive subgroups” to this
paraphrase of EPA’s definition of RfD.

Significant Digits. There is inconsistent use in the number of significant digits
(e.g., Tables 17, 20, 21, etc.).

The reference to a child's water intake is only a note explaining that ATSDR
used a comparison value for surface water that is based on ingesting 1 liter of
water per day, which is an overestimate for recreational exposure.

Page 102, line 11. I am having trouble visualizing lifetime consumption at a child’s
water intake of 1 liter per day. However, this note, starting on line 7 is exceedingly
important and should be place up front in the executive summary section (I
apologize if it is already there and I missed it).

Thank you for the comment.

ATSDR did not use the ITER database to access the MRLs or RfDs. Instead,
ATSDR referenced the primary sources for this information.

Page 58, line 25. You may also wish to reference the International Toxicity
Estimates for Risk (ITER) database found on the National Library of Medicine’s
Toxnet for international risk values. Incidentally, this data base also includes
ATSDR MRLs and all of EPA’s IRIS information.

Overall. The Report is nicely written and organized. The overview of mercury, it’s
various chemical forms and associated risks are clear and factual (e.g., Table on
page 1 is nice). Basic public health concepts are well explained (e.g., “what is
meant by exposure” on page 52; Figure 9 – ATSDR Chemical Screening
Process). In most cases caveats and limitations are offered (e.g., pg 77).

ATSDR Response

Reviewer Comment

Are there any comments on ATSDR's peer review process?
1
No. The amount of time allocated for peer review is adequate.

3

Reviewer

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

1

Reviewer

There are few studies in humans regarding absorption of ingested inorganic
mercury. Reports of up to 40% absorption are from animal studies, so are not
specific to any age group.
ATSDR made the editorial change.

ATSDR added an Overall Conclusions section to the Summary.
ATSDR made the editorial change.
ATSDR added an explanatory statement after each conclusion bullet.

Page 1. Inorganic mercury. 40% absorption. Is this specific to any age group?

Page 1. Organic mercury. I would suggest "eating contaminated fish". I would
also suggest "to various tissues including the brain" and "Effects on the
developing nervous system in children are the primary health concern".
(Developmental effects is rather vague)

Pages 4-14 generally. ATSDR should consider using a matrix-type table to
summarize these findings. The repeated statements are very hard to follow.

Page 4. 2nd bullet. "elemental mercury released from the Y-12 plant"

Page 4, 4th bullet. Is it possible to say why ATSDR cannot draw a conclusion?
For example:

ATSDR made the editorial change.
ATSDR made the editorial change.

Page 4, 11th bullet. There is an extra period at the end.

Page 5, 1st bullet. Consider: "mercury in soil that increased their risk of adverse
renal effects." Again the discussion should be about risk, not disease causation.

I-12

ATSDR made the editorial changes.

Page 4, 6th and 7th bullets. "Children who swallowed water containing inorganic
Hg while playing in EPFC…". I also have a problem with the conclusion here
regarding causation of specific health effects. This occurs at multiple places in the
document. I would recommend "…and 1958 may have received doses that have
been associated with renal (kidney) effects in experimental animals." or
alternatively "may have increased risk of developing renal (kidney) effects based
on comparison with data derived in animal studies." In many places the document
talks (appropriately) about an increased level of risk. I don't see why for this
particular set of comparisons, the document discusses disease causation. This is
a risk assessment and the conclusions should really discuss risk.

"Due to a lack of adequate data, ATSDR cannot conclude…". The lack of a stated
reason for the inability to draw a conclusion might read strangely to someone
reading only this section.

No, the point is to highlight the primary exposure pathway for each form of
mercury.

ATSDR Response

Page 1. Elemental mercury. Is there any reason to mention why ingestion is not
an important pathway? For the general public, this might seem a bit strange.

Reviewer Comment

Reviewer
ATSDR added footnotes in the body of the public health assessment to
explain that the neurodevelopmental effects discussed by the NAS were
observed at a population level; not on an individual basis.

Page 5, 5th bullet. Here and elsewhere I think ATSDR should include
reference/comparison to the Seychelles study. The Seychelles study is the basis
for ATSDR's own MRL yet, aside from mentioning that, all comparisons are to the
Faroes and NAS values. It should also be noted here (and elsewhere) that the
effects observed in the Faroes were population level effects and not observable
on an individual basis.

ATSDR added a text box to explain the conclusions used for exposure to
methylmercury in fish.

See response above.
See response above.
In cases where there is no harm/risk, ATSDR prefers to use the more
common language, especially in the summary for the general public.
ATSDR made the editorial change.

Current is defined as 1990 to present to be consistent with the other public
health assessments already released for the Oak Ridge Reservation.

Page 6, 2nd bullet. It is a bit unclear that in the first paragraph ATSDR says there
is a risk of neurodevelopmental effects but then in the next paragraph states that
no doses exceed the NAS health effect level. Please clarify. Perhaps explaining
that the increased risk is due to being above the MRL and RfD but that the risk is
only slight (and theoretical) because the exposures are below the NAS level
where effects have actually been observed.

Page 6 4th and 5th bullets. Same comment as above.

Page 7, 3 and 4th bullet. Same comment as above.

Page 7, 5th bullet. "were not harmed from exposure to methylmercury"? Again,
please keep this in the context of risk. "were not at increased risk due to their
estimated methylmercury exposure"

Page 9, 4th section (and elsewhere). Please consider italicizing "avoid contact
with water (bacterial advisory)" to make it clear this is the sign’s message.
Otherwise the sentence reads very awkwardly.

Page 10, 3rd section. "Children who played in the EFPC floodplain at the NOAA
and Bruner sites before soil removal activities in 1996 and 1997" The section is
titled current exposure yet ATSDR is talking about activities that occurred more
than a decade ago. Please clarify.

I-13

Whether children are as sensitive to the neurotoxic effects of mercury as the
fetus is uncertain because studies were not done on children not exposed in
utero. ATSDR added this uncertainty to the discussion in the body of the
public health assessment.

Page 6, 2nd bullet. Throughout the document, ATSDR lumps all individuals 0-18?
years of age together as "children". Yet it is likely only young children who are at
risk of MeHg exposure. The literature studies (Seychelles, Faroes, New Zealand)
have looked at children exposed in utero. I am not aware of any studies that
looked at children exposed only in later age (e.g., 7 or 10 years) without having
been exposed in utero. ATSDR might want to clarify what it meant by children in
the discussions.

ATSDR also added a text box to explain that any health effects discussed in
the public health assessment are observed at a population level and are not
observable on an individual basis.

ATSDR Response

Reviewer Comment

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Reviewer

Thank you for the comment.
ATSDR’s conclusion for past exposures to mercury are based on the State of
Tennessee’s Oak Ridge Health Studies – Dose Reconstruction Project Task 2
estimates of mercury known to be lost to the air and surface water, not the
Task 2 estimates of not accounted for mercury. See the discussion on pages
57 and 58 of the public health assessment on the Oak Ridge Dose
Reconstruction Project, ATSDR’s technical reviewer’s comments on Task 2,
and ATSDR’s decision to use the Task 2 estimates of mercury known to be
lost.

I reviewed Section II but have no comments. The information presented is
straightforward and easy to understand.

Page 50, Table 5. Does ATSDR discuss the potential implications of the lost
mercury on the overall conclusions? What if 10% of the unaccounted for Hg
actually got into the environment? That could be important. As regards the
footnote, it is not clear how "accounting purposes" could provide estimates in the
pound range. Is it a percentage allocation method? Can ATSDR provide a bit
more explanation? Also what is the basis for the 2,437,752 pounds of lost and not
accounted for mercury in the 1977 inventory report? Was it derived using the
censored numbers? Maybe this could be further explained in footnote "*".

I-14

ATSDR noted the uncertainty about exposures occurring after birth being
based on exposures occurring in utero to the body of the public health
assessment.

Page 12, bottom section. "Children who ignore the posted warning signs". What
age range is relevant to this point? Children who are fishing are probably at least
6 or 8 years old. Are the RfD or MRL relevant for them? Some statement about
the uncertainty in using an RfD based on risks of in utero exposure to assess
risks of exposure later in childhood should be included somewhere in the
document.

Page 52. ATSDR states what the PHAWG was but never says what its
conclusions were.

The advisory for freshwater fish does not specifically include shellfish,
however, ATSDR believes that most people in the area assume shellfish is
included in the advisory.

Page 12, bottom section. The text box is a nice tool for highlighting an important
point. Does the fish consumption advisory address shellfish? If the signs just say
fish, will people assume that means shellfish as well?

ATSDR clarified that the PHAWG understood and recognized the limitations
and recommendations of the Task 2 report, and agreed with the use of the
Task 2 report in the public health assessment.

In Section III.B of the public health assessment, ATSDR explains the findings
of the three previous studies (1977 Mercury Inventory Report, 1983 Mercury
Task Force, and Task 2) that investigated the past mercury inventories,
estimated mercury releases to the air and water, and estimated mercury not
accounted for. This section of the public health assessment also explains the
basis of the estimates for the known lost mercury, estimates of the not
accounted for mercury, and theories of why the mercury inventories have not
been accounted for. ATSDR also acknowledges the fact that the
discrepancies in these estimates will never be confidently accounted for and
that more mercury might have been released to the environment. Since the
public health assessment narrative includes the explanation for the estimates
in Table 5, the footnote is not necessary and has been removed.

ATSDR Response

Reviewer Comment

Reviewer
ATSDR clarified that if one of the elements is missing, the exposure pathway
is considered incomplete.
Thank you for the comment.
The redundant statement could not be found.

ATSDR made the editorial change.

The RMEG is based on animal studies using mercuric chloride, but is used for
screening all forms of inorganic mercury in soil.

Page 52. In the text box, please consider adding the statement that an exposure
pathway without any of the five elements is incomplete and presents no risk.

Page 53. I like figure 9 (with the screening pans). It is very easy to understand.

Page 54. "concentrations detected at or below ATSDR's comparison values do
not warrant health concern". This is a bit redundant since the same statement is
made on Page 53.

Page 54. Text box. “Overestimated values are considered conservative”. The
statement by itself seems a bit odd. To the public, "overestimated" will just mean
"wrong". Maybe consider "Conservative values are developed with assumptions
that are more likely to overestimate than underestimate actual risks".

Page 56. Table 6. LTHA/MCLG/MCL? Does the RMEG really pertain only to
HgCl2 or to other forms of inorganic mercury? Should ATSDR note in the text that
the FDA limit for methylmercury in fish is 1 ppm as a means of putting the RSL in
context?

ATSDR clarified the reviews.

ATSDR moved the text boxes.
ATSDR references the latest sources of information for the ATSDR MRLs,
EPA RfDs, and NAS BMDL. ATDSR is working on completing a new
toxicological profile on alkyl and dialkyl mercury, and in the process will
update the chronic oral MRL based on post-parturition neurobehavioral
effects. However, until the toxicological profile is revised by the agency, the
1999 version is THE reference for ATSDR mercury MRLs.

Page 57. How are reviews by the Health Effects/MRL workgroup different from
Agency wide MRL workgroup reviews? Please clarify. Also "Expert panel of
external peer reviews" should it be reviewers?

Page 57. The text boxes describing the NOAEL/LOAEL should be on the next
page where actual study data are discussed.

Page 58, middle of page. The ATSDR toxicological profile on mercury is from
1999, or more than a decade old. Can a more recent reference be provided?

I-15

ATSDR made the editorial change.

Page 57. footnote §. The rates and factors are different not only for different
media but also for receptors of different ages.

The FDA limit is not appropriate for this public health assessment. This health
assessment evaluates local fish consumption. The FDA limit is based on a
market basket approach, which uses samples collected from supermarkets,
grocery stores, and fast food restaurants. ATSDR fears that adding another
comparison value will only confuse the general public.

ATSDR Response

Reviewer Comment

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Reviewer
ATSDR clarified that basing the health assessment on the most sensitive
endpoint (neurodevelopmental effects) is likely protective of carcinogenic
effects.

ATSDR listed mercuric chloride and mercuric nitrate, instead of mercuric
sulfide, as the examples in Table 7.
The case studies are not used to develop health guideline values. The health
guidelines are based on the LOAELs and NOAELs, as noted in Figure 12.

Page 59. Is it worth stating that the neurodevelopmental effects of methylmercury
provide a more stringent basis for evaluating exposures than the potential
carcinogenicity? If mercury is a carcinogen it must be a fairly weak one. I believe
the general thinking is that regulatory values and guidelines based on the
neurodevelopmental effects would likely be protective of carcinogenic effects.

Page 60. Table 7. Is mercuric sulfide a relevant example compound for the site?
What about mercuric nitrate, the form that was actually released?

Page 61 Figure 10. The "case studies' are the studies used to develop health
guideline values? This is unclear in the Figure. Can the authors of the studies be
noted? Also, the figure references an acute exposure case study but does not
provide an acute exposure guideline value. This should be explained in the text.

The appropriate studies are cited in Table 7.The BMDL is a range that
encompasses the NOAEL.
No sampling was conducted at that time. It has been almost 50 years, and
there is no way to know which homes or buildings might have been
contaminated. Further, many potentially affected homes have since been torn
down.
ATSDR clarified the intent of the statement.

Page 63, Figure 12. Would it be possible to indicate the studies involved in some
way? Without reading the text it appears strange that the chronic NOAEL is
above the BMDL.

Page 69. Two locations where a space is needed between "from" and "1953".
Later in the page there is a discussion of workers bringing mercury home into
their houses. Was any sampling data available at all? Is there any reason for
doing sampling at the current time?

Page 70, bottom of page. ATSDR indicates it has no concerns regarding the EPA
dispersion modeling. Do you mean concerns about the model that was used and
its assumptions or concerns about the results of the model (or both)?

I-16

ATSDR agrees. This difference could be due to study design or endpoints.
Since the chronic study is not used to develop the health guidelines, no further
explanation is necessary.

Page 62 Figure 11 is a nice way to show the various effect levels. A few
questions do arise. The intermediate and acute NOAELs are below the chronic
exposure NOAEL and LOAEL. This could be due to differences in study design or
endpoints but it is unexpected. Perhaps an explanation is necessary.

Table 7 explains that an acute MRL is not available for elemental mercury.

Many of the study details (such as author and year) are provided in Table 7.
As noted in the narrative that refers to the figures, more detailed information
about the studies can be found in ATSDR's Toxicological Profile for Mercury
(ATSDR 1999) and U.S.EPA’s Integrated Risk Information System (IRIS).

ATSDR Response

Reviewer Comment

Reviewer
ATSDR added the reasons noted in Appendix E to the discussion.

The pivotal feature of the EFPC Volatilization Model is the volatilization
fraction, which is the fraction of metallic mercury mass in EFPC that volatilized
from the water. Task 2 assumed a log triangular distribution of values, with a
minimum value a “best estimate,” and a maximum value equal to 1, 5, and 30
percent, respectively, of the total mercury mass released annually to the
creek. Task 2 apparently selected these values from data collected in the
1990s. ATSDR suggests that conditions in EFPC were too different in the
1990s compared with the 1950s to warrant unqualified application of those
values. Task 2 did not explain how it derived the volatilization fractions it used,
and ATSDR believes this key variable needs to be justified. There is not
enough support for the model upon which to draw public health conclusions.
The Task 2 models are further explained in Appendix E. For clarification,
ATSDR added language from the appendix to the text here.
ATSDR agrees. Thank you for the comment.

ATSDR made the editorial change.
The 11 million gallon per day was based on the average of flow rates
measured from 1955–1957 (ChemRisk 1999a).
ATSDR clarified the intent of the paragraph.

Due to the nitric acid in the liquid waste, loss of mercury would have been
minimal because the acidity would have favored dissolved ionic mercury.

Page 71. I agree that the Chi over Q model, if based on uranium, would not be
expected to provide a good estimate of mercury air concentrations. Presumably
the uranium would be in the form of particulate whereas mercury would largely be
in the form of a vapor. That being said, ATSDR should list the reasons why they
do not agree with the model. The current text seems dismissive without really
explaining why.

Page 72. A comma is needed after "a minimum value". Did the researchers use
sensitivity analysis to explore the impact of the volatilization fraction decision on
the final model results? Is this something ATSDR could do? For example, would a
normal or lognormal distribution with the same parameters yield very different
results? If the top value were 60 percent versus 30 percent, would that change
the results significantly? While lack of support for the distribution chosen is a
problem, it seems problematic to simply throw the results away. Doing so keeps
ATSDR from making any conclusions at all regarding these locations.

Page 73, second bullet. Why is the EPA ISCST3 model appropriate? ATSDR is
throwing out the volatilization model for lack of explanation; it should not fail to
justify it's own basis for retaining the model that is accepted.

Page 73, fifth bullet. While it is clear that elemental mercury in a home could pose
problems if it was not kept in a sealed container, ATSDR has no measurement
data indicating there was exposure. The whole argument is based on purely
anecdotal information.

Page 73, second from last bullet. Add to the end "where harm is not expected."

Page 75. Is there any basis for the 11 million gallon per day assumption? Was
that the average flow rate?

Page 77. Is there a typo? Lack of acidification would have resulted in samples
being more basic which would favor elemental over ionic mercury. That should
have maximized not minimized volatilization.

Page 77, line 33. The sample was composited over a week? Would the
preservation issues noted above have been a problem over that time frame?

I-17

ATSDR Response

Reviewer Comment

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Reviewer
ATSDR clarified the intent of the paragraph.

ATSDR made the editorial change.

Because exposure to the different forms of mercury result in different health
effects, and the primary exposure to the different media are to different forms
of mercury, it makes sense to discuss the relevant health effects for each
exposure pathway.
ATSDR made the editorial change.

For this paragraph, inorganic mercury refers to both elemental and ionic.
ATSDR clarified the intent of the paragraph.
The mercuric nitrate was in an aqueous solution (Rahola et al. 1973). ATSDR
added the citation to the text.
ATSDR made the editorial change.
ATSDR made the editorial change.

ATSDR made the editorial change.
Methylmercury concentration in canned tuna is generally less than 0.2 ppm
(ATDSR 1999).

Page 79, lines 4-7. The Southworth data indicate that the *percent* of total
mercury that is organic decreases with increasing total mercury concentration.
This suggests saturation of the methylation capacity of the ecosystem. However,
ATSDR seems to be implying that methylmercury concentrations in the 1950s
would not be higher in an absolute sense than those measured in the 1990s
when much higher total mercury levels were seen in the 1950s. I am not sure I
would take the results of the Southworth et al. data quite that far. Higher total
mercury concentrations would result in more methylmercury, just not
proportionally more.

Page 79, line 18. Harm to the lining of the GI tract would occur at levels far above
those reported in the EFPC. Please clarify the substantial difference in dose. It
might be actually good to give the value.

General comment. The health effects information is scattered across different
sections of the document. It might be good to put it all together in one place and
then refer back to that section as appropriate.

Page 79, line 26. "mercury absorbs poorly into the blood from the GI tract" Just to
be clear where in the body you mean, mercury is absorbed into the blood very
efficiently in the alveoli.

Page 79, line 28. "inorganic mercury" means elemental or ionic?

Page 80, line 11. What was the medium in the mercuric nitrate bioavailability
studies, water, soil? This could be important to exposures at the site.

Page 80, line 24. As noted earlier, I would prefer if the report discussed increased
risks rather than making statements about actual health effects.

Page 80, bottom. I think it would be better to say "water from EFPC containing
xxx mercury" than "with mercury". "With" is more often used in the context of a
voluntary consumption.

Page 89. Same comment regarding causation of specific health effects.

Page 90. Just an observation. So mercury concentrations in fish tissues
downstream of the Y-12 plant are not that different, only slight above, than the
range currently seen in canned tuna?

I-18

ATSDR Response

Reviewer Comment

Reviewer
Doses are compared to the Seychelles studies, when appropriate (i.e., when
methylmercury in fish is being evaluated).
ATSDR made the editorial changes.

The Seychelles studies are used as a point of comparison, when appropriate.
ATSDR added “Current conditions are not likely to be different than those in
the late 1990s, because there have been no significant mercury releases and
remediation activities involving mercury at Y-12 are being monitored.”

ATSDR changed the text to read “Adults and children are unlikely to
participate in recreational activities that would result in exposure to EFPC
surface water, especially since signs are posted to warn the public to avoid
contact with the water because of the bacterial contamination.”
No, the detection limits varied and are well below the comparison value of 2
ppb. The detection limits ranged from 0.03-0.067 ppb from 2005 to 2009.
Detection limits prior to 2005 are not available, except the detection limits from
the RI/FS ranged from 0.05 to 0.2 ppb.
ATSDR changed the text to read “ATDSR assumed that adults weighed 70 kg
and were exposed to the maximum concentration for 30 years, and children
weighed 28.1 kg and were exposed to the maximum concentration for 6
years.”
ATSDR made the editorial change.

Page 93. The Seychelles studies are used as a basis for comparison to the
predicted exposure levels. That is reasonable and balanced. This should be done
elsewhere in the document.

Page 93, generally. Again, the report authors discuss neurodevelopmental effects
in terms of increased risk and don’t state that children will have
neurodevelopmental problems. The risk statement is made on the basis of human
data. Why was the same approach not taken for renal effects from inorganic
mercury where the data come from rodents (thus making extrapolation even more
uncertain)?

Page 97. The Seychelles studies should be used as a point of comparison
particularly in this concluding section.

Page 100. The title of this section is "current EFPC air” but the most recent data
are over a decade old. The report should have a statement indicating that current
conditions are not likely to be different than those in the late 1990s. This certainly
is reasonable (assuming no new releases or no significant change in buried
sediments) but should be explicitly stated so that data from the late 1990s can be
used to make statements about current conditions.

Page 102. line 13. "recreational activities that would involved drinking EFPC
surface water"

Page 106, Table 19. Was the detection limit 2 ppb in all cases? The range in the
detection limit should be noted in the footnote.

Page 111, line 11. ATSDR should mention that implicit here is that the individual
is exposed at the location with 2,240 ppm mercury, and only that location,
repeatedly for the full 6 or 30 years.

Page 111, line 29. "ingest 100 mg/day of sediment for 18 days/yr"

I-19

ATSDR Response

Reviewer Comment

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

Reviewer
Current conditions are not likely to be different than those in 1991 and 1996,
because there have been no significant mercury releases and the deep
channel sediments have not been disturbed.
ATSDR made the editorial changes.

Maximum concentrations are first used for the initial screening of several
exposure scenarios and when appropriate (e.g., for acute exposures).
ATSDR made the editorial changes.

Page 113. Again, current shallow near shore sediment is being characterized by
data collected in 1991 and 1996, which is a bit of a stretch.

Page 118, line 13. "unlikely for pregnant adults and young children to eat one
meal a month...". Again the RfD and MRL are most relevant for comparing
exposures in these two groups. Comparisons for other types of individuals will
likely substantially overestimate risk.

Page 119, line 7. Why is the maximum concentration first used? It seems that
average concentrations are typically used elsewhere.

Page 124, line 28. ATSDR states that even children (no age specified) who are
not exposed in utero are "at risk" if they eat the same fish as their mothers. I'm
not sure this kind of vague statement is helpful. What does "at risk" mean in this
context? Non-pregnant women and others are also "at risk" if by “at risk” you
mean some increased level of risk, however small. I am not aware of any
particular study that indicates that children exposed post utero, particular at an
age where they eat locally caught fish (certainly after several years of age) are at
particular risk of mercury exposure. Please provide a citation.

As stated, ATSDR used health-protective assumptions that likely
overestimated soil/sediment consumption. Doses were calculated for EFPC,
Oak Ridge, Scarboro, and LWBR (see Table 31). As explained, all of the
estimated exposure doses for potential pica child exposures are below the
health effect levels available in the toxicological and epidemiological literature.
Correct.

Page 154. Table 31. I agree that pica children are unlikely to ingest EFPC soil or
sediment (certainly not on a regular basis) due to advisories or public knowledge
about the contamination but ATSDR’s dismissal of the results of the calculation is
not particularly convincing. Why are the maximum concentrations used? Does
this make sense from a chronic exposure? Did the calculation assume that every
day of outdoor pica activity (52 days per year) involved EFPC soils/sediment?

Starting at Section IV. I did not provide comments on general public comments or
previously published documents contained in the appendices because these are
presumably not open to revision.

I-20

ATSDR added a text box to explain that any health effects discussed in the
public health assessment are observed at a population level and are not
observable on an individual basis.

Page 151. Please add some language discussing the nature of the effects at
issue with methylmercury exposure. These are subtle test performance effects
that cannot be observed at an individual level. They can only be observed when
large numbers of children are tested under similar conditions. A bit of discussion
about the lack of concordance between the Seychelles and Faroes studies would
also be helpful here.

ATSDR believes it is prudent public health practice to advise parents of young
children about the potential health effects from mercury to young children. The
brain of very young children is still developing and at increased risk of adverse
health effects from methylmercury exposure. Also, FDA and EPA recommend
that young children avoid some types of fish and eat fish that are lower in
mercury to receive the benefits of eating fish and be confident that young
children reduce their exposure to the harmful effects of mercury.

ATSDR Response

Reviewer Comment

Reviewer

I-21

I reviewed Appendices E, F and G but have no comments to offer on these
sections.

The detailed toxicological information on methylmercury from the Faroe Island
and Seychelles cohorts is discussed in the body of the public health
assessment in the appropriate sections on past and current exposure to
methylmercury from eating fish. Appendix D is prepared for the general public
and contains general information on mercury, exposure to mercury, and
health effects from exposure. The information in this section is from the
ATSDR Public Health Statement which is the summary chapter from the
Toxicological Profile for Mercury (ATSDR 1999). The general information in
this appendix is consistent with the more recent findings of the Faroe Island
and Seychelles cohorts.

Page D-1. Much of the discussion of mercury toxicity comes from the ATSDR
toxicological profile for mercury. It would be useful to supplement that material
with updated information on the Faroe Island and Seychelles cohorts (the findings
are consistent with earlier results but are confirmed at older ages), discuss
mercury and autism which has been a subject of some interest to the general
public, and discuss the potential role of methylmercury exposure in
cardiovascular disease (emerging as a potentially significant health concern for
methylmercury exposure).

Thank you for your review.

“…children with any ASD conditions and those without ASD had similar
ethylmercury exposures at the end of each exposure period from pregnancy
to 20 months of age. Exposure to ethylmercury from thimerosal-containing
immunizations during pregnancy (prenatally), or as a young child, was not
associated with any of the ASD outcomes. The researchers found that the
results were similar between boys and girls—thimerosal-containing
immunizations did not increase the risk of any of the ASD outcomes.”

A discussion of exposure to ethylmercury from thimerosal in vaccines and
autism spectrum disorder (ASD) is not necessary in a public health
assessment on mercury released from the Y-12 pant. First, the forms of
mercury and the routes of exposure are different for these two mercury
exposures. Second, CDC published a study titled “Prenatal and Infant
Exposure to Thimerosal from Vaccines and Immunoglobins and Risk of
Autism" in October 2010. This study concluded the following:

ATSDR Response

Reviewer Comment

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

In general, the document is well written and the risk assumptions (e.g., exposure
factors used in does equations) do not differ significantly from those that EPA
uses and recommends.

It is amazing how many of my questions were addressed by this very thoughtful
report. It is a nice benchmark for the pre-coal fly ash spill era. There are several
sections that are especially helpful: understanding exposure and comparison
values; and speciation and bioavailability. I also like the occasional reminders
about how to interpret exposure values threaded throughout the report. These
explanations and reminders should be very helpful to the uninitiated reader.

Page 98, Section IV.A.6. Mercury in Local Produce. It is wonderful to see such a
variety of fruits and vegetables being tested. I hope food analyses will be done
for heavy metals and other potential contaminants in the LWBR area sometime.
If there is no need, forgive my curiosity.

Timely information about local dietary patterns and actual food consumption is
very helpful in understanding these findings. The determination of fish
consumption and estimation of mercury intake by local anglers is a welcome
step forward in methodology.

3

4

5

6

J-1

We hope this will help the public better understand that many recreational
activities downstream of the Y-12 plant carry no risk. Additionally, identification
of the current exposure scenarios that continue to pose a health risk will help
focus cleanup efforts and warnings where most needed.

2

General comments on document
1
The CAP is pleased to review this comprehensive evaluation of possible health
effects from past and current mercury releases. Our review found that the
document addresses all the concerns we have heard expressed regarding past
and present exposure scenarios.

Public Comment

Thank you for your comment.

Thank you for your comment.

Thank you for your comment.

Thank you for your comment.

Thank you for your comment.

Thank you for your comment.

ATSDR Response

ATSDR received the following comments from the public during the public comment period (October 4, 2011 to November 11, 2011)
for the Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases public health assessment. For comments that questioned the
validity of statements made in the document, ATSDR verified or corrected the statements.

Appendix J. Responses to Public Comments

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

It is important that a document of this nature be both user-friendly and
understandable by the audience which it addresses. Given the nature of the
information and analysis provided, this document does as well as can be expected
in being understandable by its intended audience.

Some improvement might be gained by making the document more user-friendly
by providing more information to the reader upfront and not making them search
for the information somewhere else in the document. [These specific comments
are under the editorial comment section.]

For those reading the document for a more in-depth understanding of the issues
involved, a more careful job of providing and citing references might be done.
Not all readers will be aware of the various studies that have been performed
and the documents generated as a result of the Oak Ridge studies. [These
specific comments are under the editorial comment section.]

I just saw this in the news clips. Does this relate in any way to the ATSDR report
from earlier this year?

8

9

10

11

Y-12 is required under a state permit to do toxicity testing. The complex
maintains a nuclear arsenal and provides nuclear fuel to the Navy and to
research reactors worldwide.

J-2

Officials will compare recently collected biological monitoring data from the creek
with data from other local streams.

The creek, which originates at the complex, had mercury and other contaminants
discharged into its shallow waters years ago.

OAK RIDGE, Tenn. (AP) - Officials at the Y-12 National Security Complex in Oak
Ridge on Tuesday will update the status of East Fork Poplar Creek.

Posted: Dec 06, 2011 4:05 AM EST Updated: Dec 06, 2011 4:05 AM EST

Y-12 complex to update conditions at creek
(AP)

I have three comments, all related to the presentation of the results and
conclusions. In most of the results, they make a very clear connection between
the concern, basis for the analysis, and conclusion. However, I offer comments
on those that appear to be exceptions (included in appropriate categories
herein).

Public Comment

7

This news clip is not related to the ATSDR PHA, which evaluates past and
current human exposures to mercury released from the Y-12 plant. Rather, this
newspaper article refers to the recent findings from the Oak Ridge National
Laboratory’s biological monitoring of East Fork Poplar Creek. The findings were
published in several journal articles in January 2011 with Mark J. Peterson of
ORNL as the lead author. The titles of associated articles can be searched
through the DOE’s online Energy Citations Database at
http://www.osti.gov/energycitations/.

The recommended changes were incorporated (as appropriate).

The recommended changes were incorporated (as appropriate).

Thank you for your comment.

Thank you for your comment.

ATSDR Response

ATSDR Response

According to the U.S. Geological Survey’s National Hydrography Dataset (2000),
“A reach is a continuous, unbroken stretch or expanse of surface water” (see
http://nhd.usgs.gov/chapter1/chp1_data_users_guide.pdf). The uses of “stretch”
referenced by the commenter were replaced with “reach” except for on p. 23, line
8, where the term “stretch” is used to define the distance rather than the water
body itself. To avoid confusion, this use of “stretches” was changed to “Lower
Watts Bar Reservoir (LWBR) extends from the confluence of the Tennessee
River and the Clinch River downstream to the Watts Bar Dam.”

What is the difference between a reach and a stretch? [reach (p. 28, lines 2-6);
stretch (p. 23, line 8; p. 81, line 19; p. 84, line 8).]

14

J-3

Ingestion rates vary widely between each population group. A detailed
description of the methodology used to derive the different intake rates is
presented in the “Consumption of Fish” section in Appendix G of the PHA.
Footnote text was added to the Summary in Section I to refer the reader to
Appendix G for detailed information on how the intake rates were derived.
ATSDR also clarified the language describing each ingestion rate in a
subsequent version of the PHA.

The description of “high ingestion rates” in the Summary (Section I) are widely
divergent and potentially confusing. In the “Fish EFPC” discussion, a high
ingestion rate is described as up to 9 meals from EFPC per year. This does not
appear to be an especially high ingestion rate and may need additional
information to support this definition. Alternatively in the “Fish from Poplar Creek
and Clinch River” section, high ingestion rates are defined as “three meals a
week or higher.” This definition is significantly higher than the one provided for
the EFPC. The gap between these definitions of high ingestion rates may be

confusing. Additional context is needed if both definitions are correct for their

respective water bodies.

13

Clarification of verbiage
We agree. Text similar to that presented in Appendix D of the PHA has been
12
Numerous names are used to refer to mercury contamination in the PHA.
added to the first page of the document in Section I.A. Background, under the
Sometimes the generic “mercury” is used interchangeably with more specific
species names or “total mercury.” In other cases, organic and inorganic mercury heading “Mercury in the environment.”

are used interchangeably with methyl mercury or mercuric chloride, respectively.

It may be confusing to the lay reader (and occasionally to the technical review) to

understand and follow the various terms used, especially since the fate of

elemental mercury in the environment is not described until more than 50 pages

into the PHA. It would be useful to add a brief description of the different species

of mercury (such as that presented Appendix D, Section D-1, Lines 29-34)

should be presented in the [sic]

Public Comment

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

PROPOSED RESOLUTION: I offer that they may consider revising to state that
eating fish or crayfish from the EFPC has the potential to harm people’s health, if
that is the conclusion – which it appears to be. The verbiage with the qualifier
could follow that statement, including the additional concern that fish advisories
should stay in place due to the PCB concern.

This verbiage appears to be permission to eat fish; however, expecting the
general public to keep account of the number of fish meals per month is
inherently challenging.

CONCERN: I assume that this aspect has the potential to be controversial,
especially since the previous evaluation found that eating fish from EFPC
“…could have harmed fetuses of pregnant women and babies of nursing
women.” – page 3. It goes on to state that children and adults who ate fish
periodically are not expected to have been harmed.

Page 151, line 10-12, and also page 7. The report states, “ATSDR concludes
that based on the level of mercury, eating fish and crayfish from EFPC once a
month is not expected to harm people’s health if the warning signs are removed
in the future. All of the concentrations of mercury in EFPC fish and crayfish were
higher than the comparison value. Currently, it is unlikely that anyone is actually
eating fish from EFPC because of the fish consumption advisory.”

Public Comment

J-4

PROPOSED RESOLUTION: While I agree with the conclusion, the report will be
used to address public concerns and I offer that consideration be given (if it is
not too onerous) to collecting and analyzing samples to provide a clear,
compelling, unqualified conclusion that there is no concern in a way that can be
justified to a non-scientific public skeptic.

CONCERN: In the absence of samples, they appear to make the circumstantial
case that, due to the flow of the water channel and sediment layering, mercury in
the sediment deposits would not be a concern.

Mercury sampling
16
Page 150, lines 14-18 and also page 6. The report states, “ATSDR concludes
that coming in contact with mercury in the soil near the LWBR is not expected to
harm people’s health. Even though no soil samples have been collected from the
LWBR, the occurrence of harmful health effects from exposure to mercury in soil
along the LWBR shoreline is not a concern.”

15

As mandated by Congress, ATSDR does not perform environmental sampling
for its PHAs. The agency can recommend sampling if the PHA’s conclusions
suggest this need to fill data gaps. However, this is not the case with respect to
the LWBR soil. Even though soil samples have not been collected near the
LWBR, ATSDR does not believe mercury from ORR operations has
contaminated the nearby soil based on the agency’s past evaluation of the
LWBR and because mercury detected in the near-shore LWBR sediment levels
was found at levels less than 1 ppm—much lower than the comparison value of
20 ppm for mercury in soil. Specifically, ATSDR previously evaluated this
potential exposure scenario in its 1996 Health Consultation: US DOE Oak Ridge
Reservation (Lower Watts Bar Reservoir Operable Unit) (available at
http://www.atsdr.cdc.gov/hac/pha/pha.asp?docid=1361&pg=0). ATSDR
evaluated surface sediments in shallow areas of the reservoir using maximum
concentrations of contaminants and worst case scenarios, including if surface
sediments were dredged and used as surface soil at residential properties.
ATSDR concluded that the maximum chemical contaminant concentrations

Children who ignore the posted warning signs and eat one meal of EFPC fish a
month have a small increased risk of subtle neurodevelopmental effects. Eating
one or more crayfish meals a month from EFPC increases that risk.

Children born to or nursing from women who ignore the posted warning signs
and eat one meal of fish caught from EFPC a month are not at risk of being
harmed from exposure to methylmercury. However, eating one or more crayfish
meals a month from the EFPC floodplain increases the risk of subtle
neurodevelopmental effects.

We agree with the commenter wholeheartedly, and had changed this conclusion
before receiving this comment. With respect to current exposure to mercury from
EFPC fish and shellfish, the document now concludes the following:

ATSDR Response

J-5

Mercury bioaccumulation
17
Page 126, lines 38-41. Does bioaccumulation of mercury over time affect this
conclusion?

Public Comment

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

If mercury is ingested repeatedly, it will accumulate. However, based on
ATSDR’s evaluation conducted in this PHA and its review of several past
investigations that examined consumption of LWBR fish and turtles (i.e., ATSDR
1998, ATSDR et al. 2000, Rowley et al. 1985, TDEC 1997), bioaccumulation of
mercury would not affect the agency’s conclusion that no further analysis of
health outcome data is appropriate for mercury exposure via ingestion of
moderate to high amounts of LWBR fish in the mid-1990s. Moreover, TDEC has
a public health advisory which advises the public to prevent and/or limit its
consumption of certain fish from the LWBR due to PCB contamination, which in
turn, would limit the public’s consumption of LWBR fish in general (and therefore,
limit the consumption of mercury-contaminated fish) (see the advisories at
http://www.tn.gov/twra/fish/contaminants.html).

ATSDR believes that the levels of mercury in these deep channel sediments
would be much higher than those in soil along the shoreline. We have added text
to the PHA that further explains our rationale behind this finding that exposure to
mercury in soil along the LWBR shoreline is not a concern based on the worstcase scenarios evaluated in the agency’s 1996 health consultation.

ORR-related mercury that has accumulated in the LWBR river channel
sediments (where little, if any, exposure would occur) is buried under as much as
80 cm of cleaner sediment and several meters of water. In its health
consultation, ATSDR also evaluated the potential exposure (ingestion,
inhalation, and dermal contact) if these subsurface sediments were removed and
used as surface soil on residential properties. ATSDR concluded that the
potential exposure to mercury would not pose a health concern, even if these
deep sediments were dredged and used as residential soil.

(including mercury) would not present a public health hazard.

ATSDR Response

Is anyone currently assessing the bioaccumulation of mercury in the food web at
the LWBR? It is by no means clear to me that significant bioaccumulation of
mercury is occurring, but it seems appropriate to inquire. It would appear to be
necessary for someone to monitor the mercury levels in fish and turtles for as
long as it is considered to be a contaminant of concern by health agencies-­
ATSDR and TDH. Phase III of the coal fly ash spill cleanup includes an EPA
“health risk assessment,” and plans are being made to determine what
“constituents of concern” will be monitored over time. It would seem prudent to
look into these questions.

Public Comment

20

J-6

Page 131, ATSDR's response to comment #6. The aim in dredging in Phase I of
the coal fly ash spill cleanup was to disturb the sediment as little as possible.
The interagency working group was consulted. I don't know if anyone monitored
mercury in environmental media in real time.

Mercury sampling
19
Page 130, ATSDR's response to comment #2. Does this statement pertain to
mercury alone or are you clearing food grown in this area with respect to all
potential contaminants?

18

Thank you for your comment. ATSDR’s response to comment #6 refers
specifically to mercury released from the ORR that traveled to the LWBR river
channel and deposited in deep and shallow reservoir sediment. It does not;
however, pertain to the cleanup of the coal fly ash spill related to the TVA’s
Kingston Fossil Fuel Plant that occurred on December 22, 2008. The TVA’s plant
spill is not addressed in or covered under the scope of this PHA, which deals
only with mercury releases from the Y-12 plant.

This statement refers only to locally grown produce associated with the subject
of this PHA: mercury released from the Y-12 plant. However, ATSDR has
evaluated several other potential exposures associated with ORR-related
contaminants and locally grown foods. For more information on these
evaluations, see ATSDR’s PHAs on uranium releases from the Y-12 plant
(ATSDR 2004), radionuclide releases from White Oak Creek (ATSDR 2006a),
iodine 131 releases from the X-10 site (ATSDR 2008), ORR-wide
polychlorinated biphenyl (PCB) releases (ATSDR 2009), uranium and fluoride
releases from the K-25 site (ATSDR 2010), contaminant releases from the Toxic
Substances Control Act (TSCA) incinerator (ATSDR 2005a), and screening of
current (1990 to 2003) environmental data. All of these PHAs are available on
ATSDR's ORR website at http://www.atsdr.cdc.gov/sites/oakridge/.

The commenter is correct that mercury can bioaccumulate through the aquatic
food chain. TDEC’s Division of Water Pollution Control
(http://www.tn.gov/environment/wpc/), the Tennessee Wildlife Resources Agency
(TWRA) (http://www.tn.gov/twra/fish/Reservoir/fishcomsurv/fishcomsurv.html),
and other agencies, such as the Tennessee Valley Authority
(http://www.tva.gov/environment/ecohealth/wattsbar.htm), routinely perform
biological monitoring and analyze data obtained from lakes and streams across
the state for contaminants of concern (e.g., mercury in fish tissue), including the
Lower Watts Bar Reservoir, to determine any areas with elevated levels of
contaminants over time. In addition, DOE is required by the LWBR ROD (DOE
1995c) to monitor and evaluate any changes in contaminant levels in fish. DOE’s
2010 monitoring results indicated the concentrations of mercury in LWBR fish
were low, with average concentrations below EPA’s recommended water quality
criterion for fish tissue of 0.3 mg/kg. Overall, DOE concluded that monitoring of
LWBR fish shows mercury levels are below federal advisory levels (Bechtel
Jacobs 2011).

ATSDR Response

24

On page 151, in the callout box, the report states that “it would be prudent public
health practice to avoid eating turtle organs.”

23

J-7

The ORAU/Vanderbilt health screening program for Roane County volunteers
did not screen people for mercury and some other heavy metals. It is likely these
elements were omitted because they were looking specifically for health effects
that correspond to the composition of environmental media (e.g., coal fly ash).
The question remains whether these volunteers experienced chronic mercury
exposure from other sources over the years including the possibility of mercury
exposure through consumption of locally caught fish.

PROPOSED RESOLUTION: Recommend that the fish advisory be expanded to

included eating turtles.

CONCERN: While the ATSDR acknowledges that turtle consumption is

uncertain for the area, and that they likely overestimate it by assuming it is equal

to fish consumption, this statement appears in the report but there is no

corresponding recommendation to expand the fish advisories (listed on page

152) to turtles.

Page 132, ATSDR's response to comment #10. Really? I have been somewhat
confused by admonitions to follow the fish advisories with respect to mercury
while at the same time being told that there is no problem because the mercury
is buried in deep sediment. Do you mean that if you follow the fish advisories,
there will be no problem with mercury exposure? If you do not follow the fish
advisories and eat locally caught fish, does this still apply?

22

A map showing the locations of these properties might be helpful.

Potential exposure to mercury
21
Page 31, Lines 1-6:

Public Comment

Oak Ridge Reservation: Evaluation of Y-12 Mercury Releases
Public Health Assessment

The potential past and current exposures evaluated by ATSDR in this PHA for
locally caught fish apply to these individuals as they would anyone else who
consumes fish from the water bodies evaluated. For more information, see
Section IV.A.5 for the past evaluation and Section IV.B.7 for the current
evaluation.

The Oak Ridge Associated Universities (ORAU’s) and Vanderbilt University
Medical Center’s medical screenings of citizens living near where the TVA’s
Kingston Fly Ash Plant spill occurred are not under the purview of this document,
nor are they associated with ATSDR. Though, these medical screenings of over
200 people revealed no adverse physical health effects associated with fly-ash
components (ORAU 2012).

This callout box was removed during a subsequent revision of the document.

ATSDR’s response to comment #10 refers only to potential current exposures to
EFPC and LWBR surface water and sediment because these are potential
media that would be associated with new homes near EFPC. ATSDR’s public
health evaluation in this PHA determined that exposure to the current levels of
mercury in these particular media—surface water and sediment—are not at
levels expected to cause harmful health effects. This is completely separate from
ATSDR’s conclusion on current exposure to mercury in fish and shellfish from
the EFPC, which the agency determined in this PHA, can harm human health
under certain scenarios. For detailed information on ATSDR’s current evaluation,
refer to Section IV.B.7. ATSDR recommends people heed the consumption
advisories for EFPC fish.

ATSDR added Figure 6 to the PHA to show the locations of the properties
mentioned.

ATSDR Response

25

Public Comment

J-8

It is unlikely that it would be too hot outside to play in the creek. If anything, the
heat would likely increase the likelihood of playing in the water.

Page G-1, Lines 30-31:

ATSDR Response
This change was made.