Volume 1 Text Only

REPORT ON
EVALUATION OF NC DEQ PRIVATE WELL DATA
VOLUME 1 - REPORT
DUKE ENERGY
by Haley & Aldrich, Inc.
Boston, Massachusetts
for Duke Energy
www.haleyaldrich.com

Table of Contents
Page
List of Tables
List of Figures
iv
List
of Acronyms
v
1.
Introduction
1
1.1 APPROACH TO EVALUATION OF PRIVATE WELL DATA
2
2.
Sources of Data
4
2.1 DEQ PRIVATE WELL DATA
4
2.2 DEQ REFERENCE OR BACKGROUND WELL DATA
5
2.3 DUKE BACKGROUND WELL DATA
5
3.
Screening Methodology
6
3.1 SCREENING LEVELS
6
3.2 SCREENING METHODOLOGY
6
3.3 CCR RULE CONSTITUENTS
7
4.
Summary of Screening Results
9
4.1 DEQ PRIVATE WELL RESULTS
9
4.1.1 Screening Results
9
4.1.2 "Do Not Drink" Letters
10
4.2 DEQ BACKGROUND PRIVATE WELL RESULTS
10
4.2.1 Screening Results
10
4.2.2 "Do Not Drink" Letters
11
4.3 DUKE BACKGROUND PRIVATE WELL RESULTS
11
4.3.1 Screening Results
12
5.
Background
13
5.1 BACKGROUND THRESHOLD VALUES
13
5.2 LITERATURE SOURCES OF BACKGROUND
14
5.3 COMPARISON TO BACKGROUND
14
5.3.1 Boron
14
5.3.2 Vanadium
15
5.3.3 Hexavalent Chromium
15
6.
Evaluation of Potential Correlation
16
6.1 CORRELATION EVALUATION APPROACH
16
6.2 ALLEN STEAM STATION
17

Table of Contents
Page
6.3 ASHEVILLE STEAM ELECTRIC PLANT
17
6.4 BELEWS CREEK STEAM STATION
18
6.5 BUCK STEAM STATION
19
6.6 CAPE FEAR STEAM ELECTRIC PLANT
19
6.7 CLIFFSIDE STEAM STATION
20
6.8 HF LEE ENERGY COMPLEX
20
6.9 MARSHALL STEAM STATION
21
6.10 MAYO STEAM ELECTRIC PLANT
21
6.11 ROXBORO STEAM ELECTRIC PLANT
22
6.12 LV SUTTON ENERGY COMPLEX
22
6.13 WH WEATHERSPOON POWER PLANT
23
6.14 CORRELATION SUMMARY
23
7. Screening Level Considerations
25
7.1 VANADIUM
25
7.2 HEXAVALENT CHROMIUM
26
8. Summary
28
9. References 29
Tables
Figures
Appendix A - Correlation Charts

List of Tables
Table No. Title
1
Summary of NC DEQ and Duke Private Well and Background Sample and Well Counts
2
Summary of Screening Values
3
Statistical Summary of DEQ Private Well Sampling Data
4
Summary of DEQ Private Well Sampling Data Screening
5
Statistical Summary of DEQ Background Private Well Sampling Data
6
Summary of DEQ Background Private Well Sampling Data Screening
7
Statistical Summary of Duke Energy Background Private Well Sampling Data -
Excluding Sutton Plant
8
Summary of Duke Energy Background Private Well Sampling Data Screening -
Excluding Sutton Plant
9
Comparison of DEQ Private Well Sampling Data Detected Results to Background
10
Summary of Available Background Data
11
Summary of Background Levels for Boron and Hexavalent Chromium from AWWA
12
USEPA UCMR3 Results for Hexavalent Chromium in Public Water Systems in the U.S.
13
Summary Statistics of DEQ Private Well Sampling Data by Station
14
Coefficients of Determination (R2) for Constituent Correlation Charts
iii %UICH

List of Figures
Figure No.
1
2
3
4
5
6
7
8a
8b
9
10
11
12
13
14a
14b
14c
15
Title
Duke Energy Coal -Fueled Fleet
Allen Steam Station - Receptor and Background Wells Sampled by NCDEQ
Asheville Steam Electric Plant - Receptor Wells Sampled by NCDEQ
Belews Creek Steam Station - Receptor Wells Sampled by NCDEQ
Buck Steam Station - Receptor and Background Wells Sampled by NCDEQ
Cape Fear Steam Electric Plant - Receptor Wells Sampled by NCDEQ
Cliffside Steam Station - Receptor Wells Sampled by NCDEQ
HF Lee Energy Complex - Inactive Basin Receptor Wells Sampled by NCDEQ
HF Lee Energy Complex - Active Basin Receptor Wells Sampled by NCDEQ
Marshall Steam Station - Receptor and Background Wells Sampled by NCDEQ
Mayo Steam Electric Plant - Receptor Wells Sampled by NCDEQ
Roxboro Steam Electric Plant - Receptor Wells Sampled by NCDEQ
LV Sutton Energy Complex - Receptor Wells Sampled by NCDEQ
WH Weatherspoon Power Plant - Receptor Wells Sampled by NCDEQ
Boron — Background Data Comparison
Vanadium — Background Data Comparison
Hexavalent Chromium — Background Data Comparison
Map of Vanadium in Groundwater in North Carolina
iv

List of Acronyms
ATSDR Agency for Toxic Substances and Disease Registry
AWWA American Water Works Association
BTV Background Threshold Value
CaIEPA
California Environmental Protection Agency
CAMA
Coal Ash Management Act
CAP
Corrective Action Plan
CCR
Coal Combustion Residual
CSA
Comprehensive Site Assessment
CSF
Cancer Slope Factor
DENR
Department of Environment and Natural Resources
DEQ
Department of Environmental Quality
DHHS
Department of Health and Human Services
HRE Health Risk Evaluation
HSSR Hydrogeochemical and Stream Sediment Reconnaissance
ID Identification
IOM Institute of Medicine of the National Academy of Sciences
IRIS Integrated Risk Information System
MCL Maximum Contaminant Levels
mg/kg milligrams per kilogram
mg/L milligrams per liter
NCPH
North Carolina Public Health Department
NJDEP
New Jersey Department of Environmental Protection
NTP
National Toxicology Program
NURE
National Uranium Resource Evaluation
PWS Public Water System
RfC
Reference Concentration
RfD
Reference Dose
RSL
Risk -Based Screening Level
SAB
Science Advisory Board
SMCL
Secondary Maximum Contaminant Levels

List of Acronyms (continued)
UCL
Upper Confidence Limit
UCMR3
Unregulated Contaminant Monitoring Rule
ug/day
micrograms per day
ug/L
micrograms per liter
UL
Tolerable Upper Intake Level
UPL
Upper Prediction Limit
USEPA
U.S. Environmental Protection Agency
USGS
U.S. Geological Survey
USL
Upper Simultaneous Limit
UTL
Upper Tolerance Limit

1. Introduction
North Carolina passed the Coal Ash Management Act (CAMA) of 2014 (CAMA, 2014), which is primarily
administered by the North Carolina Department of Environment and Natural Resources (DENR), now the
North Carolina Department of Environmental Quality (DEQ).
CAMA requires the DENR (now DEQ) to, as soon as practicable, but no later than December 31, 2015,
prioritize for the purpose of closure and remediation coal combustion residuals surface impoundments,
including active and retired sites, based on these sites' risks to public health, safety, and welfare, the
environment, and natural resources.
To this end, CAMA includes the following requirements for coal -fueled facilities that manage coal ash or
coal combustion residuals, the material that results from the combustion of coal for the creation of
electric energy:
An assessment of groundwater at coal combustion residuals surface impoundments
Corrective action for the restoration of groundwater quality at coal combustion residuals
surface impoundments
Duke Energy (Duke) owns and operates, or has operated 14 coal -fueled electric generating facilities in
the state of North Carolina. Figure 1 shows the locations of the Duke coal -fueled fleet in North Carolina.
The facilities are color -coded to indicate those facilities (seven) that have operating coal -fueled units,
and facilities where all coal -fueled units been retired (seven).
Each of these facilities is subject to investigation of groundwater under CAMA. Per CAMA, the
investigation reporting milestones for each facility include:
• A Groundwater Assessment Work Plan
• A Groundwater Assessment Report, referred to as a Comprehensive Site Assessment (CSA)
Once the investigation is completed, a Corrective Action Plan (CAP) is prepared.
Duke has submitted the Groundwater Assessment Work Plans and CSAs as required by the CAMA
schedule. Duke is in the process of developing the CAP reports. As agreed with DEQ the CAP reports
are being prepared in two parts. CAP -1 reports have been prepared and submitted for all facilities (as
of December 8, 2015). CAP -2 reports are in progress. The CAP -2 reports will also include a site-specific
human health and ecological risk assessment that will be used to inform the remedial decision making
for each facility.
CAMA also requires a survey of drinking water supply wells and replacement of water supplies when
DEQ makes a determination that they are contaminated. DEQ has yet to make such a determination
under CAMA. Duke has voluntarily provided bottled water to residents who received a "Do Not Drink"
notice from DEQ and North Carolina Department of Health and Human Services (DHHS) (see Section 4
for a discussion of the private well sampling results).

As required under CAMA, Duke provided to DEQ a survey of drinking water wells in the vicinity of each
of its facilities. DEQ then selected wells for sampling, and coordinated the sampling.
DEQ released data in August 2015 from the sampling of public and private wells located in the vicinity of
Duke ash ponds (the August 20, 2015 announcement with links to the data is here:
http://porta1.ncdenr.org/web/guest/coaIashnews/-/blogs/updated-well-water-testing-results-and-
information-posted-3? 33 red irect=%2Fweb%2Fguest%2Fcoa lash news). Eligibility for testing was
broadly defined as wells that were located within a half -mile radius of each facility.
The objective of this report is to provide a review of the DEQ private well data with respect to regulatory
and risk-based screening levels, as well as to available data on background levels of many of the
constituents in groundwater in the vicinity of the Duke facilities, in the state of North Carolina, and in
the U.S. It is hoped that the results of this risk-based review will help inform DEQ as they review the
drinking water results.
This report has been prepared in two volumes. Volume 1 provides the text, figures, tables, and
appendices. Volume 2 provides the detailed results of the facility -by -facility, well -by -well, sample -by -
sample screening of the private wells conducted for this evaluation, as described in Section 2.
1.1 APPROACH TO EVALUATION OF PRIVATE WELL DATA
There is not a single metric that can be used to identify if a well has been impacted by a release from a
coal ash management unit. This is due in large part to the fact that all of the constituents that are
present in coal ash and that could be released to groundwater are naturally occurring. The challenge is
to understand these background conditions, and in that context evaluate whether there has been an
impact from a release from coal ash.
Based on our understanding of the behavior of constituents that can be released from coal ash into
groundwater, USEPA has identified those constituents that are considered together to be indicators of a
potential release from coal ash; these are the CCR Rule Appendix III constituents. Of these, boron and
sulfate are the most common constituents used to evaluate the potential for an impact in groundwater.
Constituent concentrations alone are not sufficient to identify whether impact has occurred. There
must also be a transport pathway from the coal ash management unit of interest to the specific well or
wells of interest. For an exposure pathway to be complete, the following conditions must exist (as
defined by USEPA (1989)):
1. A source and mechanism of chemical release to the environment;
2. An environmental transport medium (e.g., air, water, soil);
3. A point of potential contact with the receiving medium by a receptor; and
4. A receptor exposure route at the contact point (e.g., inhalation, ingestion, dermal contact).
Thus, to understand if a particular well or wells have been impacted by a release from a coal ash
management unit, the following are needed:
• An evaluation of the magnitude of concentrations of the constituents in the well,

An evaluation of those detected constituents in relation to background concentrations in
groundwater,
• An evaluation of the potential correlation between the co -presence and concentration of
constituents considered to be indicators of a release from a coal ash management unit, and
• Consideration of the information available on the potential for there to be a complete transport
pathway between a coal ash management unit and a well.
The evaluation of private well data is further complicated by a lack of information on construction,
depth, and well screen for these wells. Each aquifer or water -bearing unit has its own constituent
characteristics, and these can differ greatly between different water -bearing unit layers. These water -
bearing units can be separated by confining layers that prevent communication between water -bearing
units. Thus, evaluation of analytical data from private wells in a given area provides information about
the areal extent of constituents in groundwater, but these data do not necessarily provide information
on vertical extent. Because coal ash management units are present on the land surface, the depth to
which groundwater may be impacted by a release depends on the setting of the coal ash management
unit, and the groundwater regime in the area. Coal ash constituents in groundwater are generally
limited to the surficial or upper most aquifers where ash may be placed.
To further our understanding of the DEQ private well data, an evaluation has been conducted using
several lines of evidence in a weight of evidence approach. This has included:
• Constituent presence/absence in private wells, and the concentration,
• Comparison to sources of information on background concentrations of constituents in
groundwater locally, regionally, and nationally,
• Evaluation of potential correlations between constituent presence/absence and concentration,
and
• Evaluation in the context of groundwater flow at each of the facilities.
3 1�%UICH

2. Sources of Data
Private well analytical data are available from three sources, which are discussed in the following sub-
sections:
DEQ Private Well Data: Available at: http://portal.ncdenr.org/web/guest/coal-ash-news-blog/-
/blogs/updated-well-water-testing-results-and-information-posted-
3? 33 red irect=%2Fweb%2Fguest%2Fcoal-ash-news-blog. NC DEQ provided Duke with an Excel
version of the PDF table that is posted to the website. Note that the Excel file contains blank
results rows and some location identifications (IDs) have a single strike -though line. This
formatting has not been changed here.
DEQ Reference or Background Private Well Data: Available at:
http://Portal.ncdenr.org/web/guest/coal-ash-news-blog/-/blogs/groundwater-reconnaissance-
well-water-sampling-study-results-posted? 33 red irect=%2Fweb%2Fguest%2Fcoal-ash-news-
blog. NC DEQ provided Duke with Excel versions of 16 of the 24 analytical results posted on the
website. The Excel versions were used in this evaluation and supplemented with the missing
results that are available in PDF on the website.
• Duke Background Private Well Data: Data provided by Duke.
2.1 DEQ PRIVATE WELL DATA
DEQ collected 500 samples of water from 333 private wells; some wells were resampled more than one
time to address technical or analytical issues. The information is provided by facility in Table 1. In
addition to the PDF table of the data posted on the DEQ website, DEQ provided Duke personnel with an
Excel version of the data table. The Excel data have been used as provided; there are some locations
where results are not reported, and there are some locations where the ID has a strike though. As the
reasons for these edits are not known, the Excel format was used as provided.
DEQ provided a "Summary of Well Testing at Coal Ash Ponds" on their website (see:
http://Portal.ncdenr.org/c/document library/get file?uuid=09b93b4b-039f-4986-bc50-
7d1227b422bb&groupld=14). The well count provided in the document is similar to the well count
developed from the spreadsheet provided by DEQ to Duke, and summarized in Table 1. The reasons for
this discrepancy have not been determined.
The DEQ private well data are provided by facility in Volume 2 of this report. These facilities are:
• Allen Steam Station
• Asheville Steam Electric Plant
• Belews Creek Steam Station
• Buck Steam Station
• Cape Fear Steam Electric Plant
• Cliffside Steam Station
• HF Lee Energy Complex
• Marshall Steam Station
• Mayo Steam Electric Plant
4 L1LY1'SRN7

• Roxboro Steam Electric Plant
• LV Sutton Energy Complex
• WH Weatherspoon Power Plant
Note that no private well samples were collected in the vicinity of the Dan River Steam Station or the
Riverbend Steam Station.
2.2 DEQ REFERENCE OR BACKGROUND WELL DATA
In what it describes as a reconnaissance study, DEQ collected 24 water samples from 24 locations in the
vicinity of three of the Duke facilities: Allen Steam Station (seven [7] locations), Buck Steam Station
(seven [7] locations), and Marshall Steam Station (ten [10] locations).
DEQ provides the following description: "The study provides a limited evaluation of the distribution of
metals and other parameters that may be naturally occurring in the groundwater, and provides data for
staff to develop a better understanding of background concentrations of metals and other parameters in
areas that are not hydraulically connected to groundwater beneath Duke Energy's coal-fired power
plant facilities."
The DEQ background data are provided by facility in Appendix N of Volume 2 of this report.
Figure 2 though Figure 13 provide the approximate locations of the private wells and background private
wells sampled by DEQ for each facility, based on available information.
2.3 DUKE BACKGROUND WELL DATA
Duke developed a background private well data set by offering to sample private drinking water wells
for facility employees, contractors, and others associated with Duke Energy, if their well was located
generally between 2 miles and 10 miles of each facility.
As shown on Table 1, 198 samples were collected and analyzed by Duke. The Duke background data are
provided by facility in Appendix O of Volume 2 of this report.
5 �UICH

3. Screening Methodology
3.1 SCREENING LEVELS
The screening levels used in this evaluation are provided on Table 2. They are from both State and U.S.
Environmental Protection Agency (USEPA) sources, as follows:
• 2L Standards: NCAC. 2013. 15A NCAC 02L.0202. Groundwater Standard (2L), Classifications
and Water Quality Standards Applicable to Groundwaters of North Carolina. North Carolina
Administrative Code. April 1, 2013. Available at:
http://Portal.ncdenr.org/c/document library/get file?uuid=laa3fa13-2cOf-45b7-ae96-
5427fb1d25b4&groupld=38364
Federal Drinking Water Standards: Maximum Contaminant Levels (MCLS) and Secondary
Maximum Contaminant Levels (SMCLs). USEPA. 2012. 2012 Edition of the Drinking Water
Standards and Health Advisories. Spring 2012. Available at:
http://water.epa.gov/drink/contaminants/index.cfm. Note that the MCLS are enforceable
standards for public drinking water supplies, and that the SMCLs are not enforceable standards.
DHHS Screening Levels: NC DHHS. 2015. DHHS Screening Levels. Division of Public Health,
Epidemiology Section, Occupational and Environmental Epidemiology Branch. April 24, 2015.
Available at:
http://Portal.ncdenr.org/c/document library/get file?p I id=1169848&folderld=24814087&na
me=DLFE-112704.PDF. Note that these screening levels have been developed for water supply
well sampling near coal ash facilities, and do not apply to other areas of the state, or to drinking
water supplies.
USEPA Risk -Based Screening Levels (RSLs): USEPA. 2015a. USEPA Risk -Based Screening
Levels. November 2015. Available at: http://www2.epa.gov/risk/risk-based-screening-table-
generic-tables. The RSLs are purely risk-based levels, derived using standard default exposure
parameters, that do not take into consideration treatment technologies or regulatory issues.
3.2 SCREENING METHODOLOGY
The analytical results for each well for all three data sets described in Section 2 are compared to each of
these four types of screening levels.
SOURCES OF DATA
SCREENING LEVELS
DEQ Private Well Data
2L Standards
DEQ collected 500 samples of water from 333 private
wells
DEQ Reference or Background Well Data
Federal Drinking Water Standards
DEQ collected 24 water samples from 24 locations in
the vicinity of three of the Duke facilities
Duke Background Well Data
DHHS Screening Levels
Duke collected 198 samples from private wells located
between 2 miles and 10 miles of the stations
USEPA Risk -Based Screening Levels (RSLs)
6 1�%UICH

The screening comparisons are presented in Volume 2 of this report. The DEQ private well data are
organized by facility, in alphabetical order. The results for each facility are separated by tabs in
Volume 2. For example, the Appendix A tab provides the results for the Allen Steam Station. A Table of
Contents is also provided for Volume 2.
Each table provides the analytical data by facility, by well location, and by constituent. Each row
presents the analytical data from one well sample. The constituents have been organized in columns
left to right based on the constituents required for groundwater monitoring under the federal Coal
Combustion Residual (CCR) Rule (USEPA, 2015b) (see Section 3.3, below). Thus, Appendix III (Detection
Monitoring) constituents are listed first, followed by Appendix IV (Assessment Monitoring) constituents,
followed by all others. In the "All Others" category, vanadium has been listed first — the remaining are
presented by alphabetical order.
It is important to note that all of these constituents are naturally present in our environment, as shown
by work conducted by the U.S. Geological Survey (USGS, 2014), and they are also naturally occurring in
groundwater (USGS, 2011).
The four sets of screening levels are listed at the top of each table. There are four screening tables for
each set of private well data for each facility; each of the tables provides the results of screening using
one set of screening levels. For example, for Allen Steam Station:
• The first table presents the results of the comparison to 2L groundwater standards,
• The second table presents the results of the comparison to federal MCLS/SMCLs,
• The third table presents the results of the comparison to DHHS screening levels, and
• The fourth and last table presents the results of the comparison to RSLs.
Yellow highlighting is used to indicate results that are above the applicable screening level; gray
highlighting is used to indicate that the detection limit for a non -detect result is above the applicable
screening level.
3.3 CCR RULE CONSTITUENTS
The constituents identified for Detection Monitoring and Assessment Monitoring under the CCR Rule
were identified by USEPA as part of the six-year long rule-making process that culminated in the 2015
Final Rule (USEPA, 2015b).
The constituents included on the DEQ private well sampling analyte list that are also identified for
detection monitoring in the CCR Rule are: boron, calcium, chloride, pH, sulfate, and total dissolved
solids. These are constituents that are considered to be indicators of potential releases from a coal ash
management unit. The presence of these constituents in groundwater is not enough to conclude that
there has been a release from a coal ash management unit; it is the magnitude of the concentrations,
the potential correlations between the constituent concentrations, and importantly the information on
the hydrogeology of an area that are used to make such a determination.
The constituents identified by USEPA in Appendix IV for Assessment Monitoring are those that USEPA
has identified as constituents most likely to be present in groundwater at levels that may present a risk
7 L1LY1'SRN7

to human health or the environment if there has been a release from a coal ash management unit. The
Appendix IV constituents included in the DEQ private well sampling are:
Antimony
Cadmium
Mercury
Arsenic
Chromium
Molybdenum
Barium
Cobalt
Selenium
Beryllium
Lead
Thallium
The Appendix IV constituents were identified by USEPA using a very conservative national risk
assessment evaluating potential impacts of ash management units on groundwater (USEPA, 2015c).
This is not to say that all of these constituents would be present at concentrations above screening
levels in all locations if there has been a release from an ash management unit — only that there is the
potential for release of these constituents that could result in concentrations in groundwater above
screening levels. This potential is the basis for the development of the Appendix IV list. The USEPA risk
assessment considered many other constituents, as shown in the table from the Executive Summary of
the risk assessment:
Table E5-1. List of Chettllcal Cot7Stitti tits Evaltlated it] the CCR Risk Assesstrtent
■ Aluminurn
■ Cadmium
■ Iron
■ Molybdenum
Strontium
■ Ammonia
■ Calcium
■ Lanthanum
■ Mickel
Sulfate
■ Antimony
■ Chloride
■ Lead
■ Nitrate f Nitrite
Sulfide
■ Arsenic
i Chromium
■ Lithium
• Selenlum
Thallium
• Barium
a Cobalt
® Magnesium
■ Silicon
Uranium
■ Beryllium
Copper
■ Manganese
■ Silver
Vanadium
■ Baron
Fluoride
Mercury
■ Sodium
Zinc
Most notably, vanadium was evaluated quantitatively but was not included in Appendix IV based on the
national risk assessment results.

4. Summary of Screening Results
This section provides a summary of the private well screening results. Two of the available data sets are
for background locations within the vicinity of the Duke facilities; these results are also compared to the
screening levels. (See Section 3 for descriptions of the data sets and screening levels used). Summary
statistics are calculated for the background data sets, and a background threshold value (BTV) is
calculated for each constituent in the background data sets. The private well data are then compared to
the BTVs for the two background data sets.
To put the results in context, other sources of data on background levels of some of the constituents are
provided for comparison. Specific results from the screening are also presented.
4.1 DEQ PRIVATE WELL RESULTS
Table 3 provides a statistical summary of the 500 DEQ private well sample results. The table provides
the frequency of detection both numerically and by percent, the arithmetic mean of the detected
concentrations, and the percentiles of all of the results (i.e., the percentiles include consideration of the
detection limits for the results reported as non -detect). The table also identifies the type of distribution
of the data (e.g., normal, nonparametric, etc.).
Of the inorganic constituents analyzed, the least frequently detected constituents (less than 10%) are:
antimony, beryllium, cadmium, mercury, selenium, and thallium. The most frequently detected
constituents (>90%) are: calcium, chloride, barium, magnesium, potassium, sodium, and strontium.
4.1.1 Screening Results
Volume 2 provides the screening tables for all of the private well data provided publicly by DEQ. The
results are summarized in Table 4. This table provides the frequency of detection and the range of
detected concentrations for all private wells sampled by DEQ combined. The summary provides the
number of results above each of the four screening levels for both detected results and the results
reported as not detected. In addition, for each constituent, the frequency of a detected result at a
concentration below all four screening levels is provided.
The following summarizes the results with respect to each screening level:
• There are 15 constituents with at least one detected result at a concentration above a 2L
standard. The most frequent of these are for pH (outside of the target pH range), and
vanadium.
• There are 18 constituents with at least one detected result at a concentration above a DHHS
standard. The most frequent of these are for vanadium and hexavalent chromium.
• There are 11 constituents with at least one detected result at a concentration above a federal
MCL. The most frequent of these is for pH (outside of the target pH range).
9 1�%UICH

• There are 6 constituents with at least one detected result at a concentration above a USEPA RSL.
The most frequent of these is arsenic. Note that the RSL for arsenic is so low that all detections
of arsenic are above the RSL, as are all detection limits.
The following constituents are not detected above any screening level: boron, barium, beryllium,
cadmium, mercury, nickel, and selenium.
There are a total of 19,263 analytical results for the DEQ private well sampling. Of these, 805 results or
4% are above a 2L standard, 733 results or 4% are above a DHHS screening level, 476 results or 2% are
above an MCL, and 100 results or 0.5% are above a USEPA RSL. There are 5,670 analytical results or
29.4% that are not above any screening level.
From this summary, pH, vanadium, and hexavalent chromium have the highest frequency of results
above a screening level. From almost 500 samples, there are 256 results outside of the regulatory pH
range, there are 344 results above the 2L and DHHS screening levels for vanadium, and there are 239
results above the DHHS screening level for hexavalent chromium.
4.1.2 "Do Not Drink" Letters
Based on these results, the DHHS reviewed the sampling results for each well and conducted a health
risk evaluation (HRE). The HRE was sent with the analytical results to each well owner. According to the
DEQ (NC DEQ, 2015), of the 327 HREs prepared and issued, 288 or 88% included a "Do Not Drink"
warning. Only 15 well owners were provided "okay recommendations." The majority of the "Do Not
Drink" advisories were for vanadium and hexavalent chromium.
4.2 DEQ BACKGROUND PRIVATE WELL RESULTS
Table 5 provides a statistical summary of the DEQ background well sample results. The table provides
the frequency of detection both numerically and by percent, the arithmetic mean of the detected
concentrations, and the percentiles of all of the results (i.e., the percentiles include consideration of the
detection limits for the results reported as non -detect).
This is a smaller data set (n=24) than the private well data set (n=500), and the results are from the
vicinity of only three Duke facilities: Allen, Buck, and Marshall. Of the inorganic constituents analyzed,
the following were not detected: antimony, beryllium, cadmium, cobalt, mercury, and thallium. The
least frequently detected constituents (less than 10%) are: arsenic. The most frequently detected
constituents (>90%) are: calcium, chloride, barium, magnesium, potassium, sodium, and strontium.
In addition, because this is a background data set, a BTV was calculated for each constituent using
USEPA's ProUCL (Version5.0.00) program (USEPA, 2013). See Section 5.1 for additional detail on the
derivation of the BTVs.
4.2.1 Screening Results
Appendix N of Volume 2 provides the screening tables for all of the background private well data
provided publicly by DEQ. The results are summarized in Table 6.
10 1�%UICH

There are a total of 936 analytical results for the DEQ background well sampling. Of these, 32 results or
3% are above a 2L standard, 35 results or 4% are above a DHHS screening level, 16 results or 2% are
above an MCL, and 2 results or 0.2% are above a USEPA RSL. There are 249 analytical results or 26.6%
that are not above any screening level.
From this summary, pH, vanadium, and hexavalent chromium have the highest frequency of results
above a screening level. For vanadium, all 19 detected results are above the 2L and DHHS screening
levels for vanadium. All 12 detected results for hexavalent chromium are above the DHHS screening
level.
4.2.2 "Do Not Drink" Letters
HREs were also conducted by the DHHS for the DEQ background wells. Of the HREs prepared, 20 of the
24 well owners received a "Do Not Drink" recommendation from the DHHS. There were 19 "Do Not
Drink" recommendations for vanadium, and 12 "Do Not Drink" recommendations for hexavalent
chromium with additional recommendations for retesting.
As a reminder, these wells are located in areas described by DEQ as "not hydraulically connected to
groundwater beneath Duke Energy's coal-fired power plant facilities."
Thus, in this background data set, vanadium and hexavalent chromium are present in groundwater at
concentrations above the DHHS screening levels, at similar frequencies to the results for the DEQ private
well data for wells located in the vicinity of a Duke facility.
4.3 DUKE BACKGROUND PRIVATE WELL RESULTS
Table 7 provides a statistical summary of the Duke background well sample results; these are 198
samples collected from wells with a distance of 2-10 miles from a Duke facility. The table provides the
frequency of detection both numerically and by percent, the arithmetic mean of the detected
concentrations, and the percentiles of all of the results (i.e., the percentiles include consideration of the
detection limits for the results reported as non -detect).
This is a larger data set (n=198) than the DEQ background data set (n=24). Review of the data (see
Appendix O of Volume 2) indicates that the results for the background wells near the Sutton facility are
higher than for other locations. The boron results for all other locations in the Duke background data
set range from 5.1 to 113 micrograms per liter (ug/L); there are six results for Sutton that at 212 to
928 ug/L are above the background range for the remaining data set. The results for Sutton, therefore,
are not included in the calculation of the background summary statistics, as their inclusion would tend
to bias the results high. The reason for these higher boron levels are not currently known, but could be
associated with the local geology and/or the proximity to the coast. The concentration of boron in
seawater ranges from 4 to 5 milligrams per liter (mg/L) (Health Canada, 1991).
Of the inorganic constituents analyzed, the least frequently detected constituents (less than 10%) are:
beryllium, cadmium, mercury, selenium, and thallium. The most frequently detected constituents
(>90%) are: calcium, chloride, sulfate, magnesium, potassium, sodium, and strontium.

In addition, because this is a background data set, a BTV was calculated for each constituent using
USEPA's ProUCL (Version5.0.00) program (USEPA, 2013). See Section 5.1 for additional detail on the
derivation of the BTVs.
4.3.1 Screening Results
Appendix O of Volume 2 provides the screening tables for the background private well data from
samples collected by Duke. The results are summarized in Table 8.
There are a total of 4,995 analytical results for the Duke background well sampling. Of these, 271 results
or 5.4% are above a 2L standard, 293 results or 5.8% are above a DHHS screening level, 123 results or
2.4% are above an MCL, and 48 results or 0.1% are above a USEPA RSL.
Vanadium and hexavalent chromium have the most instances of a result above a screening level in the
Duke background data set.
DHHS did not review these data, but using their criteria, "Do Not Drink" letters could be issued to the
owners of wells where vanadium and hexavalent chromium were detected above the 2L and/or DHHS
screening levels.

S. Background
The DEQ private well results are compared to background levels derived statistically, or available in the
literature.
5.1 BACKGROUND THRESHOLD VALUES
USEPA provides a statistical calculation software package, ProUCL (USEPA, 2013). Data can be
summarized statistically using the program; it is most commonly used in risk assessment to calculate the
95% upper confidence limit (UCL) on the arithmetic mean, which is the value used to define an exposure
point concentration. ProUCL can also be used to develop BTVs for use in understanding background
conditions for a particular medium.
USEPA guidance discusses the use of upper percentile, upper prediction limit (UPL), upper tolerance
limit (UTL), and upper simultaneous limit (USL) as possible BTVs (USEPA, 2013). Upper percentile and
UPL values are associated with a high false positive rate (i.e., site data or even other background data,
when compared to the values, has a high probability of being identified as 'higher' than background and
thus indicative of potential contamination, when in practice such a finding is simply an artifact of the
limited statistical power of the upper percentile and UPL values). Therefore, USEPA recommends
against using those values as BTVs. Rather, USEPA recommends the use of the USL, followed by the UTL,
as BTV values. The following excerpted from USEPA (2013) provides an overview of the UTL and USL
values.
"Based upon an established background data set free of outliers and representing a single
statistical population, a USL95 represents that statistic such that all observations from the
"established" background data set are less than or equal to the USL95 [95% USL] with a CC
[confidence coefficient] of 0.95. A parametric USL takes the data variability into account. It is
expected that all current or future observations coming from the background population
(comparable to background population, unimpacted site locations) will be less than or equal to
the USL95 with CC, 0.95 (Singh and Nocerino, 2002). The use of a USL as a BTV estimate is
suggested when a large number of onsite observations (current or future) need to be compared
with a BTV.
It is noted that by definition, USL95 does not discard any observation. The false positive error
rate does not change with the number of comparisons, as the USL95 is designed to perform
many comparisons simultaneously. Furthermore, the USL95 also has a built in outlier test
(Wilks, 1963), and if an observation (current or future) exceeds USL95, then that value definitely
represents an outlier and may not come from the background population. The false negative
error rate is controlled by making sure that the background data set represents a single
background population free of outliers. Typically, the use of a USL95 tends to result in a smaller
number of false positives than a UTL95-95 [95% Upper Tolerance Limit with 95% coverage],
especially when the size of the background data set is greater than 15."
Tables 5 and 7 provide the BTVs calculated for the DEQ background data set and the Duke background
data set, respectively. For each analyte, the BTV is provided, the underlying data distribution that the
13 1�%UICH

BTV is based on is identified, and the type of USL identified by ProUCL as appropriate for the data set is
provided.
Table 9 compares the frequency and range of detected results from the DEQ private well results to the
BTVs calculated for both the Duke background data set and the DEQ background data set.
5.2 LITERATURE SOURCES OF BACKGROUND
To provide additional context, available information on background levels of the analyzed constituents is
presented in Table 10. The data are from the U.S. Geological Survey (USGS, 2011) and the North
Carolina Public Health Department (NCPH, 2010).
USGS also provides groundwater data as part of the National Uranium Resource Evaluation (NURE)
Hydrogeochemical and Stream Sediment Reconnaissance (HSSR) Program (USGS, 2006). Review of the
supporting documentation for the USGS 2011 groundwater data set from the "Trace Elements and
Radon in Groundwater Across the United States, 1992-2003" report does not indicate whether or not
the NURE data were included in the analysis. The USGS 2011 report was used here as it provides readily
available statistical summaries of the background data. The NURE data is provided as an interactive
website; time considerations did not allow for extraction and evaluation of data from the website for
inclusion in this evaluation.
Background data specifically for boron and hexavalent chromium are available from an American Water
Works report (AWWA, 2004), and are presented in Table 11.
Table 12 provides additional background data for hexavalent chromium. To understand the occurrence
of hexavalent chromium and other constituents in U.S. drinking water that are currently not regulated
under the Safe Drinking Water Act, the USEPA is conducting a nationwide survey of public water systems
(PWS) under the Third Unregulated Contaminant Monitoring Rule (UCMR3) (USEPA, 2015d,e). Table 12
provides a summary of the results for all states.
5.3 COMPARISON TO BACKGROUND
The project -specific BTVs and the literature data have been used to compare to the DEQ private well
data. The comparison focuses on the constituents that are consistently above screening levels,
specifically vanadium and hexavalent chromium. Boron is also evaluated as an indicator constituent for
the CCR Rule.
5.3.1 Boron
The 90th percentile concentration for boron of 100 ug/L in the DEQ private well data set is below the
Duke background BTV of 113 ug/L; the DEQ background BTV is much lower (39.87 ug/L), though this is
based on a much smaller data set. The 75th percentile of the DEQ private well data (43.38 ug/L) is close
to the DEQ background BTV.
The average concentration of boron in drinking water supplies in the U.S., 167.9 ug/L, as reported by the
AWWA in Table 11, is well above the BTVs for the Duke and DEQ private well drinking water results as
presented in Table 9 and listed above, and is well above the DEQ private well 90th percentile
14 1�%UICH

concentration of 100 ug/L. The DEQ private well concentrations are within the USGS background range
for boron. These comparisons are shown graphically on Figure 14a.
Thus, the levels of boron reported in the DEQ private well data are consistent with the information on
background levels of boron in North Carolina and in the U.S.
5.3.2 Vanadium
The 901h percentile value for vanadium of 13.3 ug/L for the DEQ private well data is below the DEQ
background BTV (24.07 ug/L) and well below the Duke background BTV (127.8 ug/L).
The concentrations of vanadium reported in the DEQ private well data set are consistent with both the
background data collected by Duke and DEQ, and consistent with the background levels nationally, as
reported by the USGS (USGS, 2011). This is shown graphically in Figure 14b. Background concentrations
of vanadium in groundwater of North Carolina are provided on the map in Figure 15. As can be seen,
there are areas in North Carolina where background levels are elevated. The Duke facilities of Allen,
Buck, Marshall, and Belews Creek are located within these areas. Thus, the results of the DEQ private
well sampling, as well as the DEQ background well sampling, and the Duke background well sampling,
are not unexpected.
5.3.3 Hexavalent Chromium
The 901h percentile value for hexavalent chromium of 20 ug/L is below the Duke background BTV of
73.5 ug/L; the DEQ background BTV is much lower (3.168 ug/L), though this is based on a much smaller
data set. The 75th percentile of the DEQ private well data (5 ug/L) is close to the DEQ background BTV.
The average concentration of hexavalent chromium in drinking water supplies in the U.S., 1.1 ug/L, as
reported by the AWWA in Table 11, is consistent with the 901h percentile values from the Duke and DEQ
background private well data (1.5 ug/L and 1.56 ug/L, respectively).
The average concentration of hexavalent chromium in PWS in North Carolina is 0.13 ug/L, and ranges
from 0.015 ug/L to 9.1 ug/L (USEPA, 2015d,e); see Table 12. As noted on Table 2, the DHHS screening
level for hexavalent chromium is 0.035 ug/L. Thus the average concentration of hexavalent chromium in
PWS in North Carolina is well above the DHHS screening level. These comparisons are shown graphically
on Figure 14c.
Thus, the levels of hexavalent chromium reported in the DEQ private well data are consistent with the
information on background levels of boron in North Carolina and in the U.S.
15 1�%UICH

6. Evaluation of Potential Correlation
To further our understanding of the DEQ private well data, an evaluation has been conducted using
several lines of evidence in a weight of evidence approach. This has included:
• Constituent presence/absence in private wells, and the concentration,
• Comparison to sources of information on background concentrations of constituents in
groundwater locally, regionally, and nationally,
• Evaluation of potential correlations between constituent presence/absence and concentration,
and
• Evaluation in the context of groundwater flow at each of the facilities.
6.1 CORRELATION EVALUATION APPROACH
The results in the DEQ private well data set do not numerically suggest that constituents originating
from coal ash are present in any of the private wells; there are few results above screening levels (only
4% of the 19,263 analytical results are above a North Carolina screening level; see Section 4.1), and the
results are consistent with background (see Section 5.3). To further evaluate these data, the DEQ
private well results for all facilities were plotted using correlation plots to determine if the data display
the correlations that would be expected if constituents were present due to a release from a coal ash
source.
The plots focus mainly on boron, sulfate, and calcium as indicators of a potential release to groundwater
from coal ash. Boron and sulfate detection frequencies were low in the DEQ private well results (23%
and 61%, respectively), so correlations were also conducted using calcium, which is detected in 99% of
the samples (calcium was detected in 100% of both background data sets).
It must be kept in mind that these constituents are naturally occurring, as evidence by their presence in
the background wells sampled by both DEQ and Duke, and as discussed at length in Section 5. If there
was a release from a coal ash management unit that was affecting near -by private wells, one would
expect to see a pattern of correlation (consistent increasing trend) between each combination of these
three constituents.
All of the correlation plots are provided in Appendix A of Volume 1 of this report. For each facility the
following are plotted in this order:
• Boron (on the x-axis) v. Sulfate (on the y-axis)
• Boron v. Arsenic
• Boron v. Vanadium
• Boron v. Hexavalent Chromium
• Calcium v. Sulfate
• Calcium v. Arsenic
• Calcium v. Vanadium
16 1�%UICH

• Calcium v. Hexavalent Chromium
• Boron v. Calcium
Table 13 provides the minimum, maximum and mean detected concentrations by facility for each
constituent analyzed in the DEQ private well samples.
6.2 ALLEN STEAM STATION
There are 182 sample results in the DEQ private well data set for 119 wells in the vicinity of the Allen
Steam Station. The screening tables for Allen are provided in Volume 2, Appendix A. Table 13 provides
the summary statistics for the constituents evaluated in the DEQ private well data. The correlation
plots for Allen are presented in Figure 1 of Appendix A. As can be seen, there are no straight line or
near straight line correlations between any of the constituents and boron or calcium.
Only one boron result in the DEQ private well data set is above the Duke background BTV of 113 ug/L,
and only one sulfate result is above the Duke background BTV of 275 mg/L (see Section 5.1), and these
results are not co -located. Boron and calcium, and sulfate and calcium also do not show a correlation.
Thus, the three indicators of potential release from a coal ash management unit do not show a
concentration correlation.
Arsenic does not show a pattern of correlation with boron or with calcium, and the arsenic
concentrations at Allen are below the BTVs derived from both the Duke (3.968 ug/L) and DEQ (4.4 ug/L)
background data sets.
There is no correlation between vanadium and boron or calcium concentrations for Allen. There are
two vanadium results that are above the BTV for the DEQ background data (24.07 ug/L), but the results
are well below the Duke background BTV of 127.8 ug/L.
There is no correlation between boron or calcium and hexavalent chromium concentrations for Allen.
The maximum detected hexavalent chromium concentration is 8.4 ug/L which is well below the Duke
background BTV of 73.5 ug/L. There are several results above the DEQ background BTV of 3.17 ug/L,
but boron is not detected in any of these samples.
Thus, it is concluded that the private well data for the Allen facility do not show impacts from a release
from a coal ash management unit, and the results with few exceptions are consistent with background.
The results of the CSA groundwater flow evaluation for Allen are shown on Figure 2, and support the
analytical evaluation. The blue arrow flow paths indicate that groundwater is flowing away from the
residential areas towards Lake Wylie/the Catawba River.
6.3 ASHEVILLE STEAM ELECTRIC PLANT
There are nine (9) sample results in the DEQ private well data set for eight (8) wells in the vicinity of the
Asheville Steam Electric Plant. The screening tables for Asheville are provided in Volume 2, Appendix B.
Table 13 provides the summary statistics for the constituents evaluated in the DEQ private well data.
The correlation plots for Asheville are presented in Figure 2 of Appendix A. As can be seen, there are no
straight line or near straight line correlations between any of the constituents and boron or calcium.

There is one boron result (325 ug/L) that is above both the Duke (113 ug/L) and DEQ (39.87 ug/L) BTVs
(but well below all of the screening levels) (see Section 5.1). However, that sample was collected
downstream of a water treatment device, so cannot be interpreted as being representative of
groundwater conditions. That said, the sulfate concentration associated with this result is low and well
below the Duke and DEQ BTVs for sulfate, and the calcium concentration associated with this result is
117,000 ug/L, which is below the Duke BTV for calcium (150,923 ug/L) but above the DEQ BTV of
100,163 ug/L. Thus, the results do not suggest that the boron result is associated with coal ash impacts.
Arsenic, vanadium, and hexavalent chromium were not detected in any of the DEQ private well samples.
The results of the CSA groundwater flow evaluation for Asheville are shown on Figure 3, and support the
analytical evaluation. The blue arrow flow paths indicate that groundwater is flowing from the facility to
the French Broad River, and that all but one well are located across the river from the facility. Taking
into account the direction of flow in the river, most of the private wells are upstream of the facility as
well. As discussed in the CSA, streams and rivers, especially large rivers as here consistently serve as a
groundwater divide. Only under very uncommon and unusual situations would groundwater flow under
a river of this size. Thus the French Broad River is a discharge zone for groundwater. The issues are
more fully discussed in the CAP report.
Thus, it is concluded that the results of the DEQ private well testing do not show evidence of the impact
of a release from a coal ash management unit.
6.4 BELEWS CREEK STEAM STATION
There are 36 sample results in the DEQ private well data set for 24 wells in the vicinity of the Belews
Creek Steam Station. The screening tables for Belews Creek are provided in Volume 2, Appendix C.
Table 13 provides the summary statistics for the constituents evaluated in the DEQ private well data.
The correlation plots for Belews Creek are presented in Figure 3 of Appendix A. As can be seen, there
are no straight line or near straight line correlations between any of the constituents and boron or
calcium.
There are only three (3) detected results for boron, and all are well below the Duke and DEQ BTVs (see
Section 5.1). Sulfate was detected in all but three (3) samples, but the maximum detected
concentration (20.2 mg/L) is well below the Duke and DEQ BTVs of 275.3 mg/L and 148.4 mg/L,
respectively. Calcium was detected in all samples, but the maximum concentration of 43,800 ug/L is
well below the Duke and DEQ BTVs of 150,923 ug/L and 100,163 ug/L, respectively.
There are five (5) results for arsenic above the drinking water standard of 10 ug/L, and seven (7) results
above the Duke and DEQ BTVs. As the results are not correlated with the coal ash indicators boron,
sulfate and calcium, it is not known if the arsenic is naturally occurring, or if there is a local source.
The maximum detected concentration of vanadium of 23.5 ug/L as shown on Table 13 is below both the
DEQ background BTV (24 ug/L), and the Duke background BTV (127.8 ug/L).
The maximum detected concentration of hexavalent chromium of 2.1 ug/L as shown on Table 13 is
below the Duke BTV of 73.5 ug/L and the DEQ BTV of 3.17 ug/L.
18 1�%UICH

Thus, it is concluded that the results of the DEQ private well testing do not show evidence of the impact
of a release from a coal ash management unit. The results of the CSA groundwater flow evaluation for
Belews Creek are shown on Figure 4, and support the analytical evaluation. The blue arrow flow paths
indicate that groundwater is flowing away from the residential areas.
6.5 BUCK STEAM STATION
There are 114 sample results in the DEQ private well data set for 84 wells in the vicinity of the Buck
Steam Station. The screening tables for Buck are provided in Volume 2, Appendix D. Table 13 provides
the summary statistics for the constituents evaluated in the DEQ private well data. The correlation plots
for Buck are presented in Figure 4 of Appendix A. As can be seen, there are no straight line or near
straight line correlations between any of the constituents and boron or calcium.
All of the detections of boron are well below the screening levels as shown in Volume 2 Appendix D, and
below both the DEQ and Duke BTVs for boron (see Section 5.1).
There is one well that has two sets of reported results in the DEQ data set (B61 and B61R). The results
for calcium (226,000 and 246,000 ug/L) are above the Duke and DEQ BTVs of 150,923 ug/L and 100,163
ug/L, respectively, and are associated with sulfate concentrations (566 mg/L and 711 mg/L, respectively)
above the screening levels and above the Duke and DEQ BTVs for sulfate (275.3 mg/L and 148.4 mg/L,
respectively). Total dissolved solids are above screening levels and BTVs in this well, but the boron and
chloride concentrations are low. Vanadium and hexavalent chromium were not detected in this well.
The reason for these results is unclear at this time.
All concentrations of arsenic are below the drinking water standard of 10 ug/L. The maximum detected
concentration of vanadium (25.6 ug/L) shown on Table 13 is slightly above the DEQ BTV of 24.07 ug/L,
but below the Duke BTV of 127.8 ug/L (see Section 5.1). All other detected concentrations of vanadium
are below both BTVs.
All concentrations of hexavalent chromium (maximum concentration 22.3 ug/L) are below the Duke BTV
of 73.5 ug/L, though most are above the DEQ BTV of 3.168 ug/L (again, the DEQ BTV is based on a much
smaller data set; see Section 5.1).
Thus, it is concluded that the results of the DEQ private well testing do not show evidence of the impact
of a release from a coal ash management unit, however, the results at well B61/B61R bear further
scrutiny. The results of the CSA groundwater flow evaluation for Buck are shown on Figure 5, and
support the analytical evaluation. The blue arrow flow paths indicate that groundwater is flowing away
from the residential areas.
6.6 CAPE FEAR STEAM ELECTRIC PLANT
As shown in Volume 2 Appendix E, there is only one private well sample result for the Cape Fear Steam
Electric Plant. For the sake of completeness, correlation plots for Cape Fear are presented in Figure 5 of
Appendix A. The results for this well are below all screening levels, with the exception of sodium, which
is above the DHHS screening level, and arsenic which is below the drinking water standard of 10 ug/L
but above the tap water RSL. This well does not show any evidence of impact from constituents from a
19 1�%UICH

coal ash management unit. The results of the CSA groundwater flow evaluation for Cape Fear are shown
on Figure 6; the well is located across the Haw River from the facility.
6.7 CLIFFSIDE STEAM STATION
There are 29 sample results in the DEQ private well data set for 17 wells in the vicinity of the Cliffside
Steam Station. The screening tables for Cliffside are provided in Volume 2, Appendix F. Table 13
provides the summary statistics for the constituents evaluated in the DEQ private well data. The
correlation plots for Cliffside are presented in Figure 6 of Appendix A.
As shown in Volume 2, Appendix F, all concentrations of boron and sulfate are below screening levels.
The maximum detected concentrations of boron (27 ug/L), calcium (71,000 ug/L), and sulfate (13 mg/L)
are below the Duke and DEQ BTVs, as shown on Table 9. Thus, while there is some potential indication
of correlation between these constituents on some of the graphs, the results are all within the range of
background.
Arsenic concentrations (maximum concentration 2.6 ug/L) are below drinking water standards (10 ug/L
and below Duke and DEQ background (3.968 ug/L and 4.4 ug/L, respectively).
The maximum detected concentration for vanadium (22 ug/L) as shown on Table 13 is below the Duke
and DEQ BTVs of 127.8 ug/L and 24.07 ug/L, respectively (see Section 5.1).
The maximum detected concentration for hexavalent chromium (2.2 ug/L) as shown on Table 13 is
below the Duke and DEQ BTVs of 73.5 ug/L and 3.168 ug/L, respectively (see Section 5.1).
These results are consistent with the CSA results for Cliffside; as can be seen in Figure 7, groundwater
flow paths are away from residential areas. The DEQ private well results for Cliffside are all below
regional background levels.
6.8 HF LEE ENERGY COMPLEX
There are 29 sample results in the DEQ private well data set for 16 wells in the vicinity of the HF Lee
Energy Complex. The screening tables for Lee are provided in Volume 2, Appendix G. Table 13 provides
the summary statistics for the constituents evaluated in the DEQ private well data. The correlation plots
for Lee are presented in Figure 7 of Appendix A.
As shown in Volume 2, Appendix G, all boron and sulfate results are below screening levels. The
maximum detected concentration of boron is 130 ug/L, and this result is associated with the maximum
detected sulfate concentration of 38 mg/L. All sulfate concentrations are below the Duke and DEQ BTVs
for sulfate (275.3 mg/L and 148.4 mg/L, respectively) (see Section 5.1). There are two results for boron
at 42.8 ug/L and 40.3 ug/L, which are slightly above the DEQ BTV of 39.87 ug/L; these results are well
below the Duke BTV of 113 ug/L. The calcium concentration associated with the sample having the
maximum boron and sulfate concentrations is 47,600 ug/L, which is well below the Duke and DEQ BTVs
for calcium of 150,923 ug/L and 100,163 ug/L, respectively. Thus, this sample is unique compared to the
remaining samples for Lee.
zo W�LEj(
DR
6LICH

Arsenic was detected in only 1 sample, at a concentration of 1 ug/L that is well below the drinking water
standard of 10 ug/L, and below the Duke and DEQ BTVs (3.968 ug/L and 4.4 ug/L, respectively).
There are very few detections of vanadium and hexavalent chromium. The maximum detected
concentration of vanadium of 2.4 ug/L is well below the Duke and DEQ BTVs of 127.8 ug/L and 24.07
ug/L, respectively (see Section 5.1). The maximum detected concentration of hexavalent chromium of
0.32 ug/L is essentially equal to the DEQ BTV of 3.168 ug/L, and is well below the Duke BTV of 73.5 ug/L.
These results are consistent with the CSA results for Lee; as can be seen in Figure 8a and Figure 8b,
groundwater flow paths are towards the Neuse River. Other than the single boron result, sulfate,
calcium, vanadium, and hexavalent chromium results are consistent with background.
6.9 MARSHALL STEAM STATION
There are 57 sample results in the DEQ private well data set for 38 wells in the vicinity of the Marshall
Steam Station. The screening tables for Marshall are provided in Volume 2, Appendix H. Table 13
provides the summary statistics for the constituents evaluated in the DEQ private well data. The
correlation plots for Marshall are presented in Figure 8 of Appendix A.
Boron was detected in very few wells at Marshall, as shown in Figure 8 of Appendix A. Thus, there are
no relationships that can be identified between boron and the other constituents. The maximum
detected concentration of boron of 77 ug/L is below the Duke BTV of 113 ug/L and is above the DEQ BTV
for boron of 39.87 ug/L. Sulfate was detected in roughly half of the samples, and calcium was detected
in all but one of the samples. Sulfate concentrations (maximum detect is 18.7 mg/L) are all below the
Duke and DEQ BTVs for sulfate (275.3 mg/L and 148.4 mg/L, respectively) (see Section 5.1). Calcium
concentrations (maximum detect is 41,000 ug/L) are all below the Duke and DEQ BTVs for calcium of
150,923 ug/L and 100,163 ug/L, respectively.
The maximum detected concentration of arsenic of 3.2 ug/L is below the Duke and DEQ BTVs (3.968
ug/L and 4.4 ug/L, respectively). The maximum detected concentration of vanadium (14 ug/L) is below
the Duke and DEQ BTVs of 127.8 ug/L and 24.07 ug/L, respectively. The maximum detected
concentration of hexavalent chromium (2.74 ug/L) is below the Duke and DEQ BTVs of 73.5 ug/L and
3.168 ug/L, respectively. There is no discernable relationship between concentrations of calcium
compared to sulfate, arsenic, vanadium, or hexavalent chromium.
These results are consistent with the CSA results for Marshall; as can be seen in Figure 9, groundwater
flow paths are away from the residential areas. Thus, the constituents detected in the private wells in
the vicinity of Marshall are consistent with regional background, and do not indicate impact from
constituents derived from coal ash.
6.10 MAYO STEAM ELECTRIC PLANT
There are 5 sample results in the DEQ private well data set for 3 wells in the vicinity of the Mayo Steam
Electric Plant. The screening tables for Mayo are provided in Volume 2, Appendix I. Table 13 provides
the summary statistics for the constituents evaluated in the DEQ private well data. The correlation plots
for Mayo are presented in Figure 9 of Appendix A.
zi H6L/jLEj(
DRICH

Boron and arsenic were not detected in the wells at Mayo. Calcium results (maximum detect is 84,800
ug/L; see Table 13) are below the Duke and DEQ BTVs for calcium of 150,923 ug/L and 100,163 ug/L,
respectively. Sulfate results (maximum detect is 66 mg/L; see Table 13) are below the Duke and DEQ
BTVs for sulfate (275.3 mg/L and 148.4 mg/L, respectively).
Vanadium results (maximum detect is 3.03 ug/L; see Table 13) are below the Duke and DEQ BTVs of
127.8 ug/L and 24.07 ug/L, respectively. Hexavalent chromium results (maximum detect is 0.63 ug/L;
see Table 13) are below the Duke and DEQ BTVs of 73.5 ug/L and 3.168 ug/L, respectively.
These results are consistent with the CSA results for Mayo; as can be seen in Figure 10, groundwater
flow paths are away from the residential areas. Thus, the constituents detected in the private wells in
the vicinity of Mayo are consistent with regional background, and do not indicate impact from
constituents derived from coal ash.
6.11 ROXBORO STEAM ELECTRIC PLANT
There are 18 sample results in the DEQ private well data set for 11 wells in the vicinity of the Roxboro
Steam Electric Plant. The screening tables for Roxboro are provided in Volume 2, Appendix K. Table 13
provides the summary statistics for the constituents evaluated in the DEQ private well data. The
correlation plots for Roxboro are presented in Figure 11 of Appendix A.
The maximum detected concentration of boron (86 ug/L) is above the DEQ BTV for boron of 39.87 ug/L,
all of the remaining boron results are below the DEQ BTV for boron, and all boron results are below the
Duke BTV of 113 ug/L. There are two calcium results (150,000 ug/L and 145,000 ug/L) that are above
the DEQ BTV of 100,163 ug/L, but all calcium results are below the Duke BTV for calcium of
150,923 ug/L. The maximum boron and calcium results are not co -located. Sulfate results (maximum
detect is 77.6 mg/L; see Table 13) are below the Duke and DEQ BTVs for sulfate (275.3 mg/L and
148.4 mg/L, respectively).
Vanadium results (maximum detect is 10.6 ug/L; see Table 13) are below the Duke and DEQ BTVs of
127.8 ug/L and 24.07 ug/L, respectively. Hexavalent chromium results (maximum detect is 2.69 ug/L;
see Table 13) are below the Duke and DEQ BTVs of 73.5 ug/L and 3.168 ug/L, respectively.
The correlation charts do not show any specific relationships between these constituents. These results
are consistent with the CSA results for Roxboro; as can be seen in Figure 11, groundwater flow paths are
away from the residential areas. Thus, the constituents detected in the private wells in the vicinity of
Roxboro are consistent with regional background, and do not indicate impact from constituents derived
from coal ash.
6.12 LV SUTTON ENERGY COMPLEX
There are 17 sample results in the DEQ private well data set for 10 wells in the vicinity of the Sutton
Steam Electric Plant. The screening tables for Sutton are provided in Volume 2, Appendix L. Table 13
provides the summary statistics for the constituents evaluated in the DEQ private well data. The
correlation plots for Sutton are presented in Figure 12 of Appendix A.
zz H6L/jLEj(
DRICH

Approximately half of the boron results in the DEQ private well data set for Sutton are above the Duke
BTV of 113 ug/L, and above the DEQ BTV of 39.87 ug/L; the maximum detected concentration for boron
is 690 ug/L. The maximum detected concentration for sulfate, 160 ug/L, is above the DEQ BTV of
148.4 ug/L. All other results for sulfate are below both the DEQ BTV, and below the Duke BTV of
275.3 ug/L. The maximum detected concentration for calcium of 60,700 ug/L is below the Duke and
DEQ BTVs for calcium of 150,923 ug/L and 100,163 ug/L, respectively.
Vanadium results (maximum detect is 1.2 ug/L; see Table 13) are below the Duke and DEQ BTVs of
127.8 ug/L and 24.07 ug/L, respectively. Hexavalent chromium was not detected.
The relationships shown in the correlation charts for boron and sulfate and for calcium and sulfate are
indicative of a relationship between these constituents. Interestingly that relationship is not as strong
when looking at the boron and calcium chart. Boron is clearly present above background
concentrations, though below screening levels. The CSA results for Sutton are shown in Figure 12, and
there are groundwater flow paths that are in the direction of the off-site wells. Thus, it is likely that the
wells with boron concentrations above background are impacted by constituents from the Sutton ash
basins. That said, hexavalent chromium was not detected in these wells. The concentrations of
vanadium in the wells are all consistent with background. Thus, the presence of vanadium in these wells
is likely a background condition and not related to the ash basins.
It is interesting to note that in the Duke private well background data set, there are six results for Sutton
that at 212 to 928 ug/L are above the background range for the remaining data set. However, these
results may represent a localized background level, which is similar to the DEQ private well results. The
reason for these higher boron levels are not currently known, but could be associated with the local
geology and/or the proximity to the coast. The concentration of boron in seawater ranges from 4 to 5
milligrams per liter (mg/L) (Health Canada, 1991). That said, there is a relationship between boron and
sulfate and calcium for some of the DEQ private wells, and the CSA groundwater evaluations indicate
flow in the direction of the private wells, thus indicating a potential impact from coal ash on these wells.
6.13 WH WEATHERSPOON POWER PLANT
There are 3 sample results in the DEQ private well data set for 2 wells in the vicinity of the
Weatherspoon Power Plant. The screening tables for Weatherspoon are provided in Volume 2,
Appendix M. Table 13 provides the summary statistics for the constituents evaluated in the DEQ private
well data. The correlation plots for Weatherspoon are presented in Figure 13 of Appendix A. These
charts are not informative due to the small number of results.
The concentrations of boron, calcium, sulfate, and vanadium are all below the Duke and DEQ BTVs.
Arsenic and hexavalent chromium were not detected. The CSA results for Weatherspoon shown on
Figure 13 indicate that groundwater is flowing away from the two private wells. Thus it can be
concluded that these well results do not indicate impact from constituents derived from coal ash.
6.14 CORRELATION SUMMARY
The evaluations of the DEQ private well data for areas in the vicinity of these 12 Duke facilities indicate
that the majority of the results are consistent with regional background. The only correlation seen was
at Sutton, where boron concentrations are above background both the Duke and DEQ background, and
23 1�%UICH

the CSA groundwater evaluation indicated that groundwater is flowing from the site in the direction of
the off-site wells.
By virtue of the scattered nature of the correlation plots, numerical calculation of correlation is not
necessarily appropriate. However, RZ values have been calculated for each correlation plot. These are
shown in Table 14. Only two of the values are above a level of 0.8, which is generally used as a point to
indicate if further evaluation is warranted. The calcium v. sulfate R value is 0.8637; as already
discussed, both the analytical data and the CSA results indicate that some wells in the vicinity of the
Sutton facility may be impacted by coal ash constituents. The calcium v. sulfate R value for
Weatherspoon is 0.9124; as this "relationship" is based on the results of three samples from 2 wells, it is
not significant. The concentrations of boron, calcium, sulfate, and vanadium are all below the Duke and
DEQ BTVs, and the CSA results indicate that groundwater is flowing away from the two private wells.
Thus, the weight of evidence does not suggest that this is a statistically significant result.

7. Screening Level Considerations
The two DHHS screening levels with the greatest degree of uncertainty are also the screening levels
upon which the majority of the "Do Not Drink" letters were based; these are vanadium and hexavalent
chromium.
7.1 VANADIUM
The 2L and DHHS screening level for vanadium is 0.3 ug/L. Derivation of the screening level was not
available for review for this report. The screening level is in contrast to the USEPA RSL for tap water for
vanadium of 86 ug/L. There is no Federal MCL for vanadium. To put these values in context, it is
worthwhile to consider:
• Based on the DEQ/DHHS screening level for drinking water of 0.3 ug/L and assuming that an
adult drinks 2 L of water per day, the resulting daily dose of vanadium from drinking water at
the DEQ/DHHS screening level is 0.6 ug/day.
• Based on the USEPA tap water RSL for vanadium of 86 ug/L, and assuming an adult drinks 2 L of
water per day, the resulting daily dose of vanadium from drinking water at the RSL screening
level is 172 ug/day.
• The adult daily diet (Agency for Toxic Substances and Disease Registry [ATSDR], 2012) provides
approximately 20 ug/day of vanadium.
• Many adult vitamins (Centrum, 2015) provide vanadium at 10 ug/day.
• The Institute of Medicine of the National Academy of Sciences (IOM, 2001) defines the Tolerable
Upper Intake Level (UL) for vanadium as 1,800 ug/day.
As defined by the IOM:
"The Tolerable Upper Intake Level (UL) is the highest level of daily nutrient intake that is likely to
pose no risk of adverse health effect for almost all individuals" (emphasis added).
Note that the UL is 3,000 times higher than the dose of 0.6 micrograms per day (ug/day) that DHHS is
using as the basis for its "Do Not Drink" recommendations to well owners. The amount of vanadium we
get in our diet is 33 -fold higher than the DHHS "Do Not Drink" level. As noted by ATSDR:
"Vanadium is a naturally occurring element that is widely distributed in the environment. It is
found in many foods, typically in small amounts. You cannot avoid exposure to vanadium.
Exposure to the levels of vanadium that are naturally present in food and water are not
considered to be harmful."
This information is of value when evaluating the DEQ private well results for vanadium.
25 1�%UICH

7.2 HEXAVALENT CHROMIUM
Many metals can exist in different oxidation states; for some metals, the oxidation state can have
different toxicities. This is the case for chromium. Chromium exists in two common oxidation states:
trivalent chromium (chromium -3, Cr[III], Cr3 or Cr+3), and hexavalent chromium (chromium -6, Cr[VI],
Cr6 or Cr+6). Trivalent chromium is essentially nontoxic, as evidenced by its RSL for residential soil of
120,000 milligrams per kilogram (mg/kg), and the tapwater RSL of 22,000 ug/L. It can be bought over-
the-counter as a supplement, and is included in most vitamins. Hexavalent chromium has been
concluded to be a human carcinogen by the inhalation route of exposure, as identified on USEPA's
Integrated Risk Information System (IRIS) (USEPA, 2015f), the source of final USEPA-approved toxicity
values.
Currently on IRIS (USEPA, 2015f), an oral noncancer toxicity value or Reference Dose (RfD) is available
for trivalent chromium. For hexavalent chromium, IRIS provides an inhalation cancer toxicity value for
potential inhalation carcinogenic effects, and an oral RfD and inhalation noncancer toxicity value or
Reference Concentration (RfC). Note that the oral noncancer dose response value (RfD) for hexavalent
chromium is based on a study where no adverse effects were reported.
Recent studies by the National Toxicology Program (NTP) have shown that when present in high
concentrations in drinking water, hexavalent chromium can cause gastrointestinal tract tumors in mice
(NTP, 2008). Note that the drinking water concentrations in the NTP study were very high. The lowest
exposure level in the test was 5,000 ug/L — this is 50 -fold higher than the MCL of 100 ug/L for total
chromium, and 125,000 -fold higher than the DHHS screening level of 0.07 ug/L. Note that the tap water
RSL used in the screening tables, as shown on Table 2, is a value (44 ug/L) calculated using the final
USEPA toxicity values available on IRIS. The tapwater screening level that appears on the RSL table
(0.035 ug/L, USEPA, 2015a) is derived from other sources as described below.
IRIS does not present an oral cancer toxicity value for hexavalent chromium; a value developed by the
New Jersey Department of Environmental Protection (NJDEP, 2009) (and similarly developed by the
California Environmental Protection Agency [CaIEPA]) was used in the development of the RSLs. USEPA
developed a draft oral cancer dose -response value for hexavalent chromium, based on the same study
and was the same as the NJDEP value. However, it should be noted that USEPA's Science Advisory
Board (SAB) provided comments in July 2011 on the draft USEPA derivation of the oral cancer slope
factor (CSF) for hexavalent chromium and indicated many reservations with the assumptions of mode of
action, and in the derivation itself. The SAB review can be accessed at
http://cfpub.epa.gov/ncea/iris drafts/recordisplay.cfm?deid=221433. Thus, the value used to develop
the RSLs for hexavalent chromium, and likely the DHHS screening level, has been called into question by
USEPA's peer review panel.
Currently there is much scientific debate about whether the mode of action of hexavalent chromium in
very high concentrations in drinking water (as used in the NTP study) is relevant to the low
concentrations most likely to be encountered in environmental situations (Proctor, et al., 2012). Based
on detailed studies of mechanism of action, an RfD has been developed in the literature that is
protective of both noncancer and cancer effects of hexavalent chromium in drinking water, and
corresponds to a safe drinking water equivalent level of 210 ug/L (Thompson, et al., 2014), which is
above the current MCL for total chromium. Thus, the current MCL is protective. This is in contrast to
26 1�%UICH

the proposed CaIEPA drinking water level of 10 ug/L (CalEPA, 2014), and the NC DHHS screening level of
0.07 ug/L, which are orders of magnitude below this risk-based level.
Data from the American Water Works Association (AWWA, 2004) indicates that the average background
level of hexavalent chromium in water supplies in the US served by groundwater is 1.1 ug/L, though
many of the results were below the limit of detection in the study of 0.2 ug/L. The majority of the
detected results in groundwater in the AWWA study ranged from 1-10 ug/L, with a maximum of
52.6 ug/L.
The results from the AWWA study, the site-specific Duke BTV value of 73.5 ug/L, and the comparisons in
Figure 14c indicate that the detected results in the DEQ private well study (0.033 ug/L to 22.3 ug/L) are
within the range of background concentrations in North Carolina and the U.S.
27
�UICH

8. Summary
There is not a single metric that can be used to identify if a well has been impacted by a release from a
coal ash management unit. This is due in large part to the fact that all of the constituents that are
present in coal ash and that could be released to groundwater are naturally occurring. The challenge is
to understand these background conditions, and in that context evaluate whether there has been an
impact from a release from coal ash. Based on our understanding of the behavior of constituents that
can be released from coal ash into groundwater, USEPA has identified those constituents that are
considered together to be indicators of a potential release from coal ash; these are the CCR Rule
Appendix III constituents. Of these, boron and sulfate are the most common constituents used to
evaluate the potential for an impact in groundwater.
Thus, to understand if a particular well or wells have been impacted by a release from a coal ash
management unit, the following are needed:
• An evaluation of the magnitude of concentrations of the constituents in the well,
• An evaluation of those detected constituents in relation to background concentrations in
groundwater,
• An evaluation of the potential correlation between the co -presence and concentration of
constituents considered to be indicators of a release from a coal ash management unit, and
• Consideration of the information available on the potential for there to be a complete transport
pathway between a coal ash management unit and a well.
This report has evaluated the DEQ private well results:
• By constituent and by facility,
• In the context of four sets of screening levels for drinking water,
• In the context of background, both regional and national,
• In the context of the CCR Rule Appendix III constituents that are considered to be indicators of
potential release from coal ash management units to groundwater, and
• In the context of the groundwater assessments conducted as part of the CSA for each facility.
The results for boron, sulfate and calcium are not suggestive of a release, or of a significant release, of
constituents from coal ash management units to groundwater in these private well areas. There are few
instances of the USEPA CCR Rule Appendix IV and other constituents above screening levels, and in most
cases, the private well results are within the range of background, as discussed in Section 5. There are
two notable exceptions:
• Vanadium — Almost all results are above the 2L and DHHS screening levels, however, all results
are below the tapwater RSL and all results are below the background levels discussed in
Section 5.3.2.
• Hexavalent Chromium —The majority of the results are above the DHHS screening level,
however, the majority are within the range of background as discussed in Section 5.3.3 and
shown on Figure 14c.
The CSAs for each of the facilities have demonstrated that groundwater is flowing away from residential
areas in the vicinity of the facilities. Thus taken together, and with the exception of Sutton, the results
do not indicate an impact of coal ash management units at Duke facilities on private wells.
za W�LEj(
DR
6LICH

9. References
1. ATSDR. 2012. Toxicological Profile for Vanadium. Agency for Toxic Substances and Disease
Registry. Public Health Service. U.S. Department Of Health And Human Services. Available at:
http://www.atsdr.cdc.gov/substances/toxsubstance.asp?toxid=50
2. AWWA. 2004. Occurrence Survey of Boron and Hexavalent Chromium. American Water Works
Association. Available at: http://www.waterrf.org/publicreportlibrary/91044f.PDF
3. CaIEPA. 2014. Chromium -6 Drinking Water MCL. California Environmental Protection Agency.
State Water Resources Control Board. Available at:
http://www.waterboards.ca.gov/drinking water/certlic/drinkingwater/Chromium6.shtml
4. CAMA. 2014. North Carolina Coal Ash Management Act. Senate Bill S729v7. Available at:
http://www.ncleg.net/Sessions/2013/Bills/Senate/PDF/S729v7.PDF
5. Centrum. 2015. Centrum Adult Vitamins. Follow link to "Product Labeling":
http://www.centrum.com/centrum-adults
6. Health Canada. 1991. Guidelines for Canadian Drinking Water Quality: Guideline Technical
Document— Boron. Health Canada. Available at:
http://healthycanadians.gc.ca/publications/healthy-living-vie-saine/water-boron-bore-
eau/index-eng.php
7. IOM. 2001. Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium,
Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc. Institute of
Medicine. National Academy of Sciences. Available at:
http://www.nap.edu/catalog/10026/dietary-reference-intakes-for-vitamin-a-vitamin-k-arsenic-
boron-chromium-copper-iodine-iron-manganese-molybdenum-nickel-silicon-vanadium-and-zinc
8. NCAC. 2013. 15A NCAC 02L.0202. Groundwater Standard (2L), Classifications and Water
Quality Standards Applicable to Groundwaters of North Carolina. North Carolina Administrative
Code. April 1, 2013. Available at:
http://portal.ncdenr.org/c/document library/get file?uuid=laa3fa13-2cOf-45b7-ae96-
5427fb1d25b4&groupld=38364
9. NC DEQ. 2015. News Release — Well water testing results and information posted. Available at:
http://Porta1.ncdenr.org/web/guest/coalashnews
10. NC DEQ. 2015. Summary of well testing near coal ash ponds, as of August 18, 2015. Available
at: http://Portal.ncdenr.org/c/document library/get file?uuid=09b93b4b-039f-4986-bc50-
7d1227b422bb&groupld=14.
11. NC DHHS. 2015. DHHS Screening Levels. Department of Health and Human Services, Division of
Public Health, Epidemiology Section, Occupational and Environmental Epidemiology
Branch. April 24, 2015. Available at:
29 1�%UICH

http://Portal.ncdenr.org/c/document library/get file?p I id=1169848&folderld=24814087&na
me=DLFE-112704.PDF
12. NCPH. 2010. North Carolina Public Health - Epidemiology - Well Water & Health - Maps by
Contaminant Name. Available at:
http://epi.publichealth.nc.gov/oee/wellwater/by contaminant.html
13. NJDEP. 2009. Derivation of Ingestion -Based Soil Remediation Criterion for Cr+6 Based on the
NTP Chronic Bioassay Data for Sodium Dichromate Dihydrate. Division of Science, Research and
Technology New Jersey Department of Environmental Protection. Risk Assessment Subgroup of
the NJDEP Chromium Workgroup. April 8, 2009. Available at:
http://www.state.ne.us/dep/dsr/chromium/ingestion-cr.PDF
14. NTP. 2008. NTP Technical Report on the toxicology and Carcinogenesis Studies of Sodium
Dichromate Dihydrate (CAS No. 7789-12-0) in F344/N Rats and B6C3F1 Mice (Drinking Water
Studies), NTP TR 546. NIH Publication No. 08-5887. National Toxicology Program. Available at:
http://ntp.niehs.nih.gov/ntp/htdocs/It rpts/tr546.PDF
15. Proctor, DM, M Suh, LL Aylward, CR Kirman, MA Harris, CM Thompson, H Gurleyuk, R Gerads, LC
Haws, SM Hays. 2012. Hexavalent chromium reduction kinetics in rodent stomach contents.
Chemosphere. On-line pre-print of article in press. Available at:
http://www.sciencedi rect.com/science/article/pii/50045653512005978
16. Thompson, CM, CR Kirman, DM Proctor, LC Haws, M Suh, SM Hays, JG Hixond, and MA Harris.
2014. A chronic oral reference dose for hexavalent chromium -induced intestinal cancer. J. Appl.
Toxicol. 2014; 34: 525-536. Available at:
http://onIinelibrary.wiley.com/doi/10.1002/Iat.2907/ePDF
17. USEPA. 1989. Risk Assessment Guidance for Superfund, Volume 1: Human Health Evaluation
Manual, Part A. EPA/540/1-89/002. Office of Emergency and Remedial Response, Washington,
DC. December.
18. USEPA. 2012. 2012 Edition of the Drinking Water Standards and Health Advisories. Spring
2012. Available at: http://water.epa.gov/drink/contaminants/index.cfm
19. USEPA. 2013. Statistical Software ProUCL 5.0.00 for Environmental Applications for Data Sets
with and without Nondetect Observations. Software: http://www2.epa.gov/land-
research/proucl-software, and User's Guide: http://www2.epa.gov/sites/production/files/2015-
03/documents/proucl v5.0 user.PDF
20. USEPA. 2015a. USEPA Regional Screening Levels (RSLs). November 2015. Available at:
http://www.epa.gov/reg3hwmd/risk/human/rb-concentration table/Generic Tables/index.htm
21. USEPA. 2015b. Coal Combustion Residual (CCR) Rule (Hazardous and Solid Waste Management
System; Disposal of Coal Combustion Residuals From Electric Utilities; FR 80(74): 21302-
21501, April 19, 2015. Available at: http://www.gpo.gov/fdsys/pkg/FR-2015-04-17/PDF/2015-
onm1;7 Pr)F

22. USEPA. 2015c. Human and Ecological Risk Assessment of Coal Combustion Residuals. U.S.
Environmental Protection Agency, Office of Solid Waste and Emergency Response, Office of
Resource Conservation and Recovery. Final. December 2014. EPA-HQ-RCRA-2009-0640-11993.
Available at: http://www.regulations.gov/#!documentDetail;D=EPA-HQ-RCRA-2009-0640-11993
23. USEPA. 2015d. The Third Unregulated Contaminant Monitoring Rule (UCMR3): Data Summary.
Last updated January 2015.
http://water.epa.gov/lawsregs/rulesregs/sdwa/ucmr/upload/epa815s15001.PDF
24. USEPA. 2015e. The Third Unregulated Contaminant Monitoring Rule (UCMR3): Occurrence
Data. Last updated January 2015.
http://water.epa.gov/lawsregs/rulesregs/sdwa/ucmr/data.cfm#ucmr2013
25. USEPA. 2015f. Integrated Risk Information System (IRIS). Environmental Criteria and
Assessment Office. U.S. Environmental Protection Agency, Cincinnati, OH. Available at:
http://cfpub.epa.gov/ncea/iris/index.cfm
26. USGS. 2006. National Uranium Resource Evaluation (NURE) Hydrogeochemical and Stream
Sediment Reconnaissance (HSSR) Program. By Steven M. Smith. Version 1.40 (2006).
Interactive website available at: http://pubs.usgs.gov/of/1997/ofr-97-0492/index.html
27. USGS. 2011. Trace Elements and Radon in Groundwater Across the United States, 1992-2003.
Scientific Investigations Report 2011-5059. Available at:
http://water.usgs.gov/nawqa/trace/pubs/sir20ll-5059
28. USGS. 2014. Smith, D.B., Cannon, W.F., Woodruff, L.G., Solano, Federico, and Ellefsen, K.J.,
Geochemical and mineralogical maps for soils of the conterminous United States. U.S.
Geological Survey. Open -File Report 2014-1082. Available at:
http://dx.doi.org/10.3133/ofr20141082