C. Andrews - Rebuttal Report

Rebuttal Report of
Charles B. Andrews
Allen Steam Station
Buck Steam Station
Belews Creek Steam Station
Marshall Steam Station
S.S. PAPADOPULOS & ASSOCIATES, INC.
Environmental & Water -Resource Consultants
September 30, 2016
7944 Wisconsin Avenue, Bethesda, Maryland 20814-3620 9 [phone]

Rebuttal Report of
Charles B. Andrews
Allen Steam Station
Buck Steam Station
Belews Creek Steam Station
Marshall Steam Station
Prepared for:
Duke Energy Carolinas, LLC
Prepared by:
Charles B. Andrews, PhD
S.S. PAPADOPULOS & ASSOCIATES, INC.
Environmental & Water -Resource Consultants
September 30, 2016
7944 Wisconsin Avenue, Bethesda, Maryland 20814-3620 9 [phone]

Table of Contents
2'2' bV6VD06nr oe 8F b220CIVIE2' INC'
Page
Section1
Introduction................................................................................................................ 2
Section 2
Ash Removal in South Carolina................................................................................
3
Santee Cooper Grainger Generating Station..............................................................
3
WatereeStation..........................................................................................................
4
Summary....................................................................................................................
5
Section 3
Boron as a Tracer.......................................................................................................
6
Section 4
Groundwater Discharge to Surface Water.................................................................
9
Section 5
Monitoring Results from Private Wells...................................................................
12
Section6
Other Opinions.........................................................................................................
14
SlugTests.................................................................................................................
14
Monitored Natural Attenuation................................................................................
15
Flux..........................................................................................................................
15
Allen Station — May 2006 Water Levels..................................................................
16
Groundwater Flow Velocity.....................................................................................
16
RCRA Statistical Guidelines....................................................................................
17
Marshall Station — Monitoring Well AL -1 S ............................................................
17
Section7
References................................................................................................................
18
1

REPORT

2.2- bVbVDObnr02 9� V220CIV1E2' JWC'
Section 1
Introduction
I was retained by Duke Energy Carolinas, LLC (Duke) to evaluate the nature, extent, and
appropriate remedial actions for groundwater contamination from ash basins at the Allen Steam
Station, Belmont, North Carolina; the Buck Steam Station, Salisbury, North Carolina; the Belews
Creek Steam Station, Belews Creek, North Carolina; and the Marshall Steam Station, Terrell,
North Carolina. The work was conducted to assist in litigation regarding alleged violations of state
laws related to discharges to groundwater and surface water systems. I submitted experts reports
for the Allen and Buck stations on June 30, 2016 (Andrews, 2016a,b) and for the Belews Creek
and Marshall stations on August 1, 2016 (Andrews, 2016c,d).
This report is a supplement to the previously submitted expert reports. This report describes
my opinions regarding issues raised by experts for the Plaintiffs in reports issued since the
preparation of my expert reports. These reports include a supplemental expert report by Dr. Cosler
(Cosier, 2016), addendums to expert reports prepared by Dr. Campbell and Dr. Spruill (Campbell
and Spruill, 2016a,b), supplemental expert reports prepared by Dr. Parette (Parette, 2016a,b), and
a supplemental expert report prepared by Mr. Hutson (Hutson, 2016a). In addition, since my
expert report was prepared, supplements to the Comprehensive Site Assessment reports have been
issued for the Allen, Buck, Marshall and Belews Creek sites (HDR, 2016a,b,c,d).
2

2.2- bVbVDObnr02 9� V220CIV1E2' JWC'
Section 2
Ash Removal in South Carolina
Ash is currently being removed from ash basins at the Santee Cooper Grainger Generating
Station and the Wateree Station in South Carolina. Four monitoring wells at the Grainger Station
and two monitoring wells at the Wateree Station, which are adjacent to ash basins, have historically
had arsenic concentrations above the Maximum Contaminant Level (MCL) of 10 ug/L. In all of
these wells, the arsenic concentrations have declined since ash removal began. Based on the
observed trends in arsenic concentrations in these wells, experts for the Plaintiffs have opined as
follows:
"Arsenic concentrations in groundwater samples collected from monitoring wells near the
ash basins have decreased coincident with the onset and progressive removal of coal ash.
There is an empirical relationship between coal ash removal and improvement in
groundwater quality at those sites, which is what one would expect to happen when a
contaminant source is being actively eliminated. " (Campbell and Spruill, 2016a, p.17
and 2016b, p. 17).
"Groundwater monitoring data recently submitted for the Santee Cooper Grainger
Generating Station and the South Carolina Electric and Gas (SCE&G) Wateree Station
provide a clear example of the beneficial impact that ash removal can have on groundwater
quality. Arsenic is the primary ash constituent of concern to groundwater quality at both
of these locations. " (Hutson, 2016a, p. 16).
My opinion is that the decreases observed in arsenic concentrations at the Grainger and
Wateree stations are primarily the result of the recent cessation of wet ash disposal to ash basins
and the subsequent dewatering of the ash that has occurred at both of these stations, and the lining
of the berm at the Wateree station. At each of these stations, significant saturated ash remains in
the ash basins. Completion of ash removal at both stations is not scheduled to occur for several
more years.
The arsenic concentration trends observed at the South Carolina stations demonstrate the
positive benefits of cessation of wet disposal of ash and dewatering of the ash basins on arsenic
concentrations in groundwater. In addition, the arsenic concentration trends demonstrate that
significant reductions in arsenic concentrations have occurred even though saturated ash remains
in place at both of the stations. Similar positive benefits are anticipated for constituents of interest
if the cap -in-place remedies are implemented at the stations I have evaluated in detail in North
Carolina (Allen, Buck, Marshall, Belews Creek, and Marshall). I note that at the stations I have
evaluated in North Carolina, arsenic concentrations currently do not exceed the groundwater
standard in monitoring wells located at and/or beyond the compliance boundaries. In the
remainder of this section, I describe conditions at the Grainger and Wateree stations.
Santee Cooper Grainger Generating Station
The Santee Cooper Grainger Generating Station is located in Horry County, South
Carolina. The station was retired in 2012 and the structures have been removed (Santee Cooper,
2016a). There are two ash basins at the station located adjacent to the west bank of the Waccamaw

2.2- bVbVDObnroa 9� V220CIV1E2' JWC'
�S
River. Each ash basin is approximately 40 acres. The ash in the basins at the site currently is being
dewatered, excavated and reused under a November 2013 settlement between the Southern
Environmental Law Center, the Waccamaw Riverkeeper, the South Carolina Coastal Conservation
League, the Southern Alliance for Clean Energy and Santee Cooper (Southern Environmental Law
Center, 2016). A total of 1,400,000 tons of ash was estimated to be in the two ash basins prior to
commencement of excavation in 2014 (Geosyntec, 2014a, b). As of May 2016, over 550,000 tons
of ash had been removed for beneficial use, more than one third of the ash in the basins (Santee
Cooper, 2016b). The settlement requires complete removal by 2023. Based on current removal
rates, complete removal could be achieved by the end of 2019 (Southern Environmental Law
Center, 2016).
The measured arsenic
concentrations in four monitoring
wells adjacent to the ash basins from
late 2013 through April 2016 are 3303
shown the figure to the right (this —
figure was prepared by Campbell and
Spruill, 2016a,b). These wells are all Y ssab
located along the eastern basin berms. 0 an
Arsenic Detected in Groundwater Monitoring Well
Samples, Grainger Generating Station, SC
-AWS
Arsenic concentrations in the
V O �#=MWIl
four wells declined during the period o to°o
September 2013 to April 2016. These
declines are primarily the result of the
cessation of ash disposal and o
dewateringof the ponds. Significant Apia C tion 5 a«3s .� 6
p g Sample Collection Date
saturated ash remains in the ash basins
as of June 2016 (Santee Cooper, 2016b). Note that a significant reduction in the arsenic
concentration in MW -9 occurred almost immediately following the start of ash removal from the
basin. It is not probable that this concentration reduction was due primarily to the limited ash
removal that had occurred as of April 2014.
Wateree Station
The Wateree Station is located just west of Wateree River near Eastover in Richland
County, South Carolina. The Wateree Station is operated by the South Carolina Electric & Gas
Company. The station produced about 184,000 tons of ash per year from 1999 through 2010 (South
Carolina Electric and Gas, 2013a,b). Prior to 2013, ash was placed in an 80 -acre ash basin that was
constructed in a reworked borrow pit, and a second 80 -acre ash basin served as a polishing basin.
Dry handling of ash became operational in December 2012 and emergency ash sluice pipes were
removed from Unit 1 in December 2015 and from Unit 2 in March 2016. An approximately 700 -
foot long section of the eastern berm of the ash basin near MW -11 was lined in late 2011 -early
2012 to reduce seepage through the berm.
Coal ash is currently being removed from the ash basins pursuant to an August 2012
settlement between the Catawba Riverkeeper Foundation, Inc. and South Carolina Electric & Gas
Company. The ash that is removed is sold, recycled or placed in a landfill. The settlement provides
for complete removal of the ash by the end of 2020 (Catawba Riverkeeper, 2016). Approximately
2

2.2- bVbVDObnr02 9� V220CIV1E2' JWC'
496,011 tons of ash had been removed as of June 2014 and approximately 1,087,447 tons had been
removed as of June 2016.
Measured arsenic concentration in Arsenic in Monitoring wells at Wateree Station
the two monitoring wells adjacent to the LM
ash basin at the Wateree Station (located �nnw-JJ
between the ash basin and the river) from Jo°o BMW -03
early 2010 through April 2016 are shown Z
in the graph to the right. Arsenic
concentrations in MW -11 declined w
significantly once dry handling of ash < -0N
became operational in 2012 and following
lining of the berm near MW -11. This
monitoringwell is located along the °
b Jan -10 JarFll Jan -12 Jan -13 Jan -14 Jan -15 ,arr15 Jan -17
eastern berm of the primary ash basin. As
of April 2016, significant saturated ash remained in this ash basin. Arsenic concentration in
MW -03 declined markedly in 2010 and again in 2015. This monitoring well is located along the
southwestern berm of the polishing basin where no ash removal has occurred.
Summary
My evaluation of data from the Grainger and Wateree stations do not support the Plaintiffs'
experts in their claim that the removal of ash has been the primary cause of reductions in arsenic
concentrations in monitoring wells at the stations. My evaluations indicate that the decreases
observed in arsenic concentrations at the Grainger and Wateree stations are primarily the result of
the recent cessation of wet ash disposal to ash basins, the subsequent dewatering of the ash that
has occurred at both of these stations, and the lining of the berm at the Wateree station.
5

2.2- bVbVDObnr02 9� V220CIV1E2' JWC'
Section 3
Boron as a Tracer
I have opined, and continue to opine, that boron is an excellent tracer of groundwater
migration from ash basins at the Allen, Buck, Marshall and Belews Creek stations (Andrews,
2016a,b,c,d). Experts for the Plaintiffs have opined that boron is not an appropriate tracer of
groundwater migration from the ash basins. Experts have opined as follows:
`In our opinion, Duke's F&T modeling indicates that boron does not serve as a universal
indicator of coal -ash derived groundwater contamination because the presently -known
distribution of boron at Buck could only be replicated by retarding the migration of boron. "
(Campbell and Spruill, 2016a p. 17, 2016b, p 17).
"...Duke's claim that they are not responsible for any contamination detected in area
water -supply wells relies primarily upon the general absence of boron, which we opine is not the
sole indicatory of coal ash -related contamination. " (Campbell and Spruill, 2016a p. 4, 2016b, p4).
"The foundation of most of the conclusions in the expert reports of Drs. Bradley, Andrews,
and Hennet is the assumption that boron is a conservative tracer that does not adsorb significantly
to soil (i.e., the boron retardation factor is on the order of 1). However, they neglected to use any
of the numerous site-specific soil -water boron sorption measurements to support their claims and
ignored the results of the CAP2 model calibrated retardation factor results at each site which
were determined by Duke's Energy's own consultants. As I detail in this section, the site-specific
soil -water partition coefficient measurements, Kd, at the Allen, Cliffside, and Marshall sites
correspond to boron retardation factors, Rd, ranging from about 10 to 30.... " (Cosler, 2016, page
21).
I did not ignore the extensive data collected for the CSA and CAP reports in developing
my opinion regarding the use of boron as a tracer of migration from the ash basins. In formulating
my opinion regarding the appropriateness of boron as a tracer, I evaluated the soil sorption studies
that were conducted as part of the CAP studies (batch and column tests), and the spatial distribution
of boron and other constituents of interest in the vicinity of the ash basins.
The results section of the soil sorption study report for Allen and Marshall, which were
conducted at the University of North Carolina — Charlotte and included in the CAP 1 and CAP2
reports, stated the following:
"Batch and column Kdfor B was almost negligible. " (Allen, CAP 1, Appendix D, page 6).
"Batch and column Kdfor B was almost negligible. " (Marshall, CAP 1, Appendix D, page
6).
The results of the column tests that were conducted on soils from the Allen and Marshall
stations, in general, were that influent and effluent concentrations from the columns were nearly
identical for the duration of the tests. These results indicate that retardation of boron is negligible.
If retardation was significant, the effluent concentration would be smaller than the influent
concentration during the early part of the tests. The results of column tests for boron and vanadium
2

2.2- bVbVDObnr02 9� V220CIV1E2' JWC'
with a soil sample from monitoring well AB -8D at the Marshall station are shown below to
illustrate the negligible retardation exhibited by boron in the test and the significant retardation
exhibited by vanadium.
Boron - Column AB - 8 D
Vmmdium - C.1— AB - 1 HR
500
0 .
10400
of:14•.
.. ... •.s,
200 � �•
• V'inEmuml
300
••
. Vln Feed
•Bm Lfrl_a
5
—•— Kd 115 mLlg
200 . H F¢d
100
—Kd I?0 mLI,
—— Kd 135 mL/g
100
0 j
0o
s0 100
15C
0 511 101) 150 200
Pore�•olumzspusszd
Pore volumesp—d
Figure 19': Vanadium column Ki -AB - 1 B&
Figure 47: Boron column Kd - AB - 8 D
In the column test of the soil from AB -8D, the boron concentration in the effluent was
nearly identical to the boron concentration in the influent to the column almost immediately upon
start of the test, whereas for vanadium, the concentration in the effluent was negligible until more
than 50 pore volume had been passed through the column. The results for vanadium indicate that
during the passage of the first 50 pore volumes through the column that almost all of the vanadium
in the influent was attaching itself to the soil grains within the column; this indicates significant
retardation.
The batch tests that were performed with soils from the stations were conducted 1) with
soil samples that had been disaggregated, homogenized and air dried and sieved to remove
particles larger than 2 mm, and 2) with bedrock samples that were fragmented until an approximate
grain size of 2.0 to 0.3 mm was achieved. The particle size distribution in these tests were biased
low relative to that along the primary groundwater migration pathways.
The results of the batch tests indicate that boron is significantly less retarded than the other
constituents of interest. The median value of Kd (soil -water partition coefficient, which is
proportional to retardation) for the constituents of interest tested at the sites evaluated by Dr. Cosler
(Allen, Marshall and Cliffside) are listed on the table below based on the tabulations in Appendix
C of the CAP2 reports. The median values for the boron Kd at each of the stations is significantly
smaller than the Kd's for the other constituents of interest. This indicates that boron is significantly
less retarded than the other constituents of interest that were tested. Thus, boron is an appropriate
tracer of groundwater migration for the constituents of interest at the stations. Boron, rather than
sulfate, was used as a tracer of groundwater migration of ash related constituents because boron
concentrations are elevated much more above background levels than sulfate, and because sulfate
generally is not a constituent of concern in groundwater.
7

2.2- bVbVDObnr02 9� V220CIV1E2' IMC'
LSU
Constituent of
Interest
Kd (L/g)
Allen
Cliffside
Marshall
Arsenic
0.824
0.459
0.326
Boron
0.000
0.001
0.001
Cadmium
0.125
0.642
0.172
Chromium
0.332
0.173
0.342
Molybdenum
0.288
0.0.3 8
0.36
Selenium
0.698
0.046
0.147
Thallium
0.111
0.152
0.556
Vanadium
3.638
0.390
0.238
Dr. Cosler opines that the retardation factor for boron is in the range of 10 to 30. These
retardation factors would indicate that the apparent rate of boron migration in groundwater is 10
to 30 times slower than the rate of groundwater migration. Based on my experience, these are very
large retardation factors and are inconsistent with the spatial distribution of boron in groundwater
in the vicinity of the ash basins. If in fact the retardation factor for boron was in the range of 10
to 30, the current spatial distribution of boron in groundwater would be much more limited in the
vicinity of the ash basins than currently is observed.
Boron has been used as a tracer in a number of groundwater studies. For example, the U.S.
Geological Survey used boron as a tracer for a sewage plume on Cape Cod. In that study, it was
concluded that boron was a good tracer because: 1) the concentration of boron in treated sewage
is much higher than in groundwater and 2) the boron moves readily in the aquifer (LeBlanc, 1984).
Similarly, in the ash basins boron concentrations in the porewater are much higher than in
background groundwater, and boron moves readily in the underlying groundwater systems.

2.2- bVbVDObnrO2 9� V22OCIV1E2' JWC'
Section 4
Groundwater Discharge to Surface Water
Dr. Cosler states regarding the Allen Site (Cosler, 2016, page 37):
"Dr. Andrews asserts that there is no possibility for groundwater flow beneath the
Catawba River in either the overburden or bedrock aquifers beyond the western shoreline,
but he ignores all of the vertical hydraulic gradient and hydraulic conductivity
measurements that contradict his opinion. " and
"In fact, a downward hydraulic gradient beneath the river (from overburden to the bedrock
aquifer) is precisely what all the available site-specific data show" (Cosler, 2016, page 37)
I do not opine, and have not opined in the past, that there is no possibility for flow in either
the overburden or bedrock units beyond the western shoreline of the Catawba River. In my expert
report, I opined that "Groundwater in the vicinity of the ash basin does not flow under the Catawba
River. " In addition, I stated in my expert report that "Groundwater in the vicinity of the ash basins
flows toward and into the Catawba River.... " (Andrews, 2016a, page 12). I still hold these
opinions.
My opinion is that some groundwater, which originates as recharge at the ash basins, flows
under the western shoreline of the Catawba River and discharges into the Catawba River between
the western and eastern shorelines. In addition, my opinion is that groundwater in the overburden
and/or bedrock units does not flow from the ash basins under the entire width of the river and then
under the eastern shoreline of the Catawba River. In developing my opinions, I considered all of
the vertical hydraulic gradient data and the hydraulic conductivity data.
Dr. Cosler incorrectly states the location of site-specific monitoring locations. There are
no monitoring locations located beneath the river; all of the monitoring locations are located to
the west of the west bank of the river. In the area to the west of the river, based on fundamental
hydrogeologic principles, it is expected that hydraulic gradients from the water table (shallow)
zone would be downward toward the deep zone. Hydraulic gradients are downward, in large part,
because the water levels in the ponds within the ash basins are approximately 70 feet greater than
the water level in the adjacent river. This, though, does not imply that groundwater flow is
downward beneath the river. In my expert report, I stated: "At the river however, the water table
is equivalent to river stage, which is lower than water levels in the deep and bedrock wells at the
facility. Thus, groundwater flow from the ash basin system is upward to the river, where it
discharges. " (Andrews, 2016a, page 12).
Dr. Cosler further states "strong downward groundwater flow components from deep
overburden to bedrock were measured at each deep/bedrock monitoring well cluster (GWA-5,
GWA-3, and GWA-1)... " (Cosler, 2016, page 37).
This statement is incorrect based on water levels on May 2, 2016. The water levels
measured on this date are listed on table below.
9

2.2- bVbVDObnr02 9� V220CIV1E2' JWC'
SSU
Groundwater Levels — ay 2, 2016
feet, MSL)
Shallow
Deep
Bedrock
GWA-5 568.48
571.82
571.79
GWA-3 572.28
574.56
574.38
GWA-1 617.37
616.78
616.69
On May 2, 2016, groundwater levels in all of the wells listed on the table were well above
the river level that was reported to be about 565.5 feet, MSL. This indicates a strong hydraulic
gradient toward the river. At monitoring well clusters GWA-5 and GWA-3, there were strong
upward hydraulic gradients from the deep zone to the shallow zone. At all of the monitoring well
clusters, the water levels in the deep and bedrock wells differed by less than 0.2 feet, clearly
indicating that there is not a strong hydraulic gradient from the deep zone to the bedrock unit at
the locations of these well clusters. All of these well clusters are located to the west of the river
shoreline, and thus do not provide direct information on hydraulic gradients beneath the river.
As noted above, at the river, the water table is equivalent to river stage, and based on water levels
in the deep and bedrock wells west of the river, it can be concluded with near certainty that
hydraulic gradients beneath the river are upward.
Dr. Cosler stated in reference to water levels at GWA-5 that "...consistent downward
flow is also observed near to the lined inactive ash basin (monitoring well cluster GWA-5), where
there is no ponded water to cause groundwater -table mounding and the inactive ash basin
contains an impermeable liner that prevents any groundwater recharge beneath the basin
footprint. " (Cosler, 2016, p. 37).
This statement is incorrect. Groundwater levels at GWA-5 have been consistently upward
from the deep zone to the shallow zone (refer to table above). This indicates that there is an upward
hydraulic gradient (not a downward hydraulic gradient) at this location, consistent with the well
cluster's setting downgradient of the lined retired ash basin ash landfill.
Dr. Cosler also stated "In the AB-9S/9D well cluster Dr. Andrews incorrectly states that
the flow direction in overburden is upward... " (Cosler, 2016, page 38).
In preparing my expert report, I referenced Figures 2-2 and 2-3 in the CAP2 report that
depict groundwater levels in September 2015 in the shallow and deep zones, respectively. The
water level in AB -9S is posted as 565.68 feet MSL on Figure 2-2 and the water level in AB -91) is
posted as 568.86 feet MSL. Based on the data posted on these figures, I concluded that
groundwater levels were lower in the shallow zone than in the deep zone, and thus there was an
upward hydraulic gradient.
In summary, Dr. Cosler relies on hydraulic gradients from the shallow to deep zone west
of the river shoreline to infer that groundwater is not discharging to the river between the shoreline
and the compliance boundary in the river. He states in his report that "...the CSA does not contain
any hydraulic gradient data that demonstrates that groundwater will ultimately discharge into the
Catawba River before reaching the Compliance Boundary" (Cosler, 2016, pp. 38-39).
It is literally correct that there are no measured hydraulic gradient data from beneath the
river in the CSA report. As noted previously, no monitoring wells were constructed in the river.
10

2.2- bVbVDObnr02 9� V220CIV1E2' IMC'
There are, though, abundant hydraulic data that can be used to estimate the hydraulic gradients
under the river. Water levels in deep and bedrock wells located adjacent to the river are higher
than water levels in the river; these data are contained in the CSA and CAP reports. In addition,
from basic hydrogeologic principles it follows that the water table at the river is equivalent to the
river stage. Therefore, hydraulic gradient beneath the river from the deep zone are upward toward
the river (deep water levels are higher than river level, thus upward hydraulic gradient). As a
result, one-dimensional model calculations that ignore groundwater discharge to the river are not
reliable.
11

2.2- bVbVDObnr02 9� V220CIV1E2' IMC'
Section 5
Monitoring Results from Private Wells
In my expert reports, I opined that the results from the sampling of a large number of private
wells within one-half mile of the stations provides an excellent database of groundwater quality in
the bedrock in the vicinity of the stations (Andrews, 2016a,b,c,d). These data are appropriately
used to determine if the groundwater produced for domestic uses meets applicable groundwater
quality standards.
Dr. Cosler in his supplemental report is very critical of the use of data from these wells for
understanding if groundwater at the well has been adversely impacted as the result of migration of
ash related constituents from the ash basins (Cosler, 2016). In my opinion, the results of analyses
of constituents of concern in groundwater samples from private wells provide a direct means of
ascertaining whether or not the private wells have been impacted by constituents of concern that
have migrated from the ash basin.
Dr. Cosler states that "Sample dilution is a major problem with the drinking -water
samples... " and further opines that the "...coal -ash contaminant plumes in bedrock originating at
the Ash Ponds would be on the order of a few feet thick.... " (Cosler, 2016, page 3). Based on these
observations, he opines that since the typical private well is open to several hundred feet of the
bedrock, the concentrations of constituents of interest in a water sample collected from the well
will not represent the maximum concentration at any interval within the well. Regardless, the
concentrations of constituents of interest in the water sample represent the concentrations in the
water pumped from the well for domestic uses.
Dr. Cosler is incorrect in concluding that a typical coal -ash contaminant plume in bedrock
originating at the Ash Ponds would be on the order of a few feet thick. His opinion regarding
plume thickness is based on the use of "....a well established mathematical solution... to describe
the dispersion of contaminant concentrations emanating from a point source into groundwater
across a three-dimensional space.... " (Cosler, 2016, p. 5). This mathematical solution is based
on contamination emanating from a point source in a uniform flow field. The ash basins are not
point sources; rather they are distributed sources of significant areal extent. For example, at the
Allen station the current footprint of the ash basins is approximately 322 acres. In addition, in the
bedrock there is not a uniform flow fields. Rather, there is a complex flow field with groundwater
flow concentrated along fractures. As a result, the use of the "well established mathematical
solution " does not provide a meaningful estimate of plume width in bedrock, if one was to occur.
Dr. Cosler also opines "Using standard mathematical analyses, it is possible to calculate
the magnitude of the dilution factor..." such that the maximum concentration of a constituent of
interest in a plume intersecting the open borehole of a private well can be calculated for well
sampling data (Cosler, 2016, p. 8). He uses this approach to evaluate the results from the private
well sampling at the Allen station. He concludes that "...the initial boron sampling results (Table
1) suggest that the true maximum concentrations in the bedrock aquifer for the sampling date
ranged from values on the order of 2, 000 to 8, 000 ug/L... " (Cosler, 2016, p. 11). This calculation
of the "true" maximum concentration in the bedrock in my opinion is meaningless. There are two
major problems with this approach: 1) it assumes that the boron concentrations reported in the
12

2.2- bVbVDObnr02 9� V220CIV1E2' IMC'
�S
initial well sampling, in which boron was analyzed with high detection limits, reflects boron from
the ash basins and not background conditions and/or analytical bias, and 2) it assumes the bedrock
has homogeneous aquifer properties. In my opinion, the boron concentrations measured in the
resampling of the private wells, and not those reported for the initial sampling, reflect the actual
boron concentrations in groundwater pumped from the private wells. In addition, the boron
concentrations reported in the resampling lie within the range of background concentrations.
Therefore, it is not possible to calculate meaningfully the maximum possible boron concentration
in an interval within the open hole of a private well, as it is not possible to determine how much
of the boron in any well is potentially from the ash basin rather than reflecting background
conditions.
The dilution formula is based on equal groundwater inflow during well pumping along the
entire length of the open borehole. This is not probable in a fractured bedrock as groundwater flow
is almost entirely through fractures that are irregularly distributed with depth. In addition, studies
by the U.S. Geological Survey in North Carolina suggest that average hydraulic conductivity
decreases significantly with depth (Daniel and others, 1989). The dilution formula, as discussed
above, also assumes a very thin plume. As a result, not considering the variable distribution of
hydraulic conductivity with depth, the natures of the fractures and underestimating potential plume
thickness all result the calculation of a dilution factor that likely greatly overstates the actual
dilution factor.
In summary, my opinion is that the groundwater quality data from the sampling of a large
number of private wells within one-half mile of the stations provides an excellent database of
groundwater quality in the bedrock near the stations. These data are appropriately used to
determine if the groundwater produced for domestic uses meets applicable groundwater quality
standards. The methods used by Dr. Cosler to extrapolate maximum concentrations within the
bedrock at any areal location are fundamentally flawed.
13

2.2- bVbVDObnr02 9� V220CIV1E2' JWC'
Section 6
Other Opinions
Slug Tests
Drs. Campbell and Spruill opine that slug tests are only useful for estimating hydraulic
conductivity of clay. They state (Campbell and Spruill, 2016a p. 5, 2016b):
"We opine that the concept of slug testing was intended for use in low permeability
materials (e.g., clay) to produce order -of -magnitude estimate of hydraulic conductivity. "
This opinion is incorrect as slug tests can provide reliable estimates of hydraulic
conductivity over a wide range of material types including not only clays but also silts and sands.
In fact, slug tests are commonly used to estimate the hydraulic conductivity of silts and sands. The
ASTM standard for slug testing (D4044-96) does not include any guidance regarding the range of
hydraulic conductivities of materials in which slug tests are to be conducted. Doe and
Remer (198 1) describe the ranges of applicability of "typical" well testing techniques. For test
zones with lengths ranging from I in to 10 in, their suggested range of applicability of "standard"
slug tests extends from about 10-8 cm/sec
to 10-2 cm/sec. To put this range in
Rp1Ys 17ncmYolWlaEE • R M K ''� a
perspective, this range is superimposed on
a chart of typical hydraulic conductivitiesi0' '0 ' ,PY
as shown to the right (Freeze and '
IPY
Cherry, 1979). I I I Tw I°"3 I
I 10"r Ia, 4P"a K7'
The range of applicability of slug
tests, as shown on the figure, is relatively
large. The range is clearly and definitely
not restricted to low permeability
materials. It is important to note that the
general guidance of Hsieh (1998) and Doe
and Remer (1981) is based on practical
considerations for "standard" tests. The
analyses that has been developed to
interpret slug tests do not make any
distinctions between low and high
permeability formations (Cooper et al.,
1967, Hyder et al., 1994).
Typical hydraullc conductivity Value, (Furze and ['.'hetet', 1474)
.vllh 13or and Rrm (178U) ranrr of.ppllrahlltIV for "slaudard" slue Irrh
In summary, it is my opinion that slug tests are appropriate for evaluating the hydraulic
conductivity in the subsurface materials adjacent to the screened intervals of the monitoring wells
at the stations that I have evaluated. Slug tests were conducted at most of the monitoring wells
installed at these stations, and the data from these slug tests provide an appropriate database for
determining a reliable estimate of hydraulic conductivity. It is incorrect to opine that slug tests do
not provide appropriate data for evaluating the hydraulic conductivity of the subsurface materials
at these stations.
14
--
--
---
--
l
1P"a
IP-•
10"b
I
i
1°"
1°"E
10y
p.s
1° 1
I
1°-r
to-
se
le
1
�
ld°
10""
10-i
Iwo
I
t°"4
IP"4
0 r
10`Y
i
'7} 1
WA
10-141-
--
-----------/
v,
1C'r
Typical hydraullc conductivity Value, (Furze and ['.'hetet', 1474)
.vllh 13or and Rrm (178U) ranrr of.ppllrahlltIV for "slaudard" slue Irrh
In summary, it is my opinion that slug tests are appropriate for evaluating the hydraulic
conductivity in the subsurface materials adjacent to the screened intervals of the monitoring wells
at the stations that I have evaluated. Slug tests were conducted at most of the monitoring wells
installed at these stations, and the data from these slug tests provide an appropriate database for
determining a reliable estimate of hydraulic conductivity. It is incorrect to opine that slug tests do
not provide appropriate data for evaluating the hydraulic conductivity of the subsurface materials
at these stations.
14

SSU
Monitored Natural Attenuation
2.2- bVbVDObnr02 9� V220CIV1E2' JWC'
The primary source control measures proposed for the Allen, Buck, Belews Creek, and
Marshall stations is dewatering of the ash basins and placement of engineered caps over the ash
basins. These source control measures will significantly reduce the infiltration of ash related
constituents into groundwater, will greatly reduce the volume of saturated ash, and will alter
groundwater flow directions by lowering groundwater levels near the ash basins. The combined
result will be a reduction in groundwater contamination and rates of migration of constituents of
interest in groundwater. In addition to the primary source control measures, monitored natural
attenuation (MNA) is recommend as the main supplemental corrective action for groundwater. If
the monitoring determines that MNA is insufficient for restoration of groundwater quality, then
additional corrective actions will be considered and implemented, if warranted.
Dr. Parette has opined that MNA is not appropriate for the ash basins. In his supplemental
report he states "dilution and dispersion generally are not appropriate as primary MNA
mechanisms .... " and "the potential use of MNA at these sites is contrary to the U.S. EPA statement
that `EPA generally expects that MNA will only be appropriate for sites that have low potential
for contaminant migration" (Parette, 2016a, p. 1; 2016b, p. 1).
Dr. Parette in developing his opinions regarding the appropriateness of MNA considered
MNA to be the primary groundwater remedy and ignored the significant changes in saturated ash
and groundwater flow directions and magnitude that will occur as the result of dewatering of the
ash basins and capping of the basins. Rather, he implicitly assumed that groundwater conditions
following implementation of a cap -in-place remedy would be similar to existing conditions.
The main constituents of interest in groundwater at the ash basins, with the exception of
boron, are significantly attenuated by geochemical processes and surface interactions in addition
to dilution and dispersion (refer to table in Section 3 of this report). The spatial distribution of
these other constituents of interest demonstrate that they have low mobility under existing
conditions, and future conditions with a cap -in-place remedy would further reduce their mobility.
Therefore, under EPA's guidance these compounds are appropriate for an MNA remedy.
The primary attenuation processes affecting boron migration are dilution and dispersion.
Boron, though, leaches out of the ash at concentrations that are generally below the 21, Standard.
For example, at the Allen site, 14 ash samples were tested by the Synthetic Leaching Test
Procedure, and the maximum concentration that leached from any of the samples was only 444
ug/L; less than the 2L Standard of 700 ug/L. The leaching data suggest that elevated
concentrations of boron will rapidly flush out of the remaining saturated ash after a cap -in-place
remedy is implemented. As a result, dilution and dispersion are operative attenuation processes
needed to achieve the groundwater standard only in the short term. Overall, therefore, the
groundwater conditions near the ash basins are appropriate for an MNA remedy for the
constituents of interest.
Flux
Dr. Parette opines that my opinions with regard to flux (or mass discharge) are largely
irrelevant (Parette, 2016a, p. 2; 2016b, p. 2). The foundation for his opinion is that groundwater
standards are based upon concentration of the constituent and not flux.
15

2.2- bVbVDObnr02 9� V220CIV1E2' IMC'
�S
At each of the stations for which I prepared an expert report, groundwater is flowing
currently from the ash basins toward a surface water body where it discharges. In addition, if Cap-
in -place remedies are implemented at these stations, groundwater will continue to flow from
beneath the ash basins toward and into surface water bodies. At the Allen station, groundwater is
flowing toward and into the Catawba River; at the Buck station, groundwater is flowing toward
and into the Yadkin River; at the Belews Creek Station groundwater is flowing toward and into
tributaries of the Dan River; and at the Marshall Station, groundwater is flowing toward and into
Lake Norman.
The quantification of the effect that the groundwater discharges to surface water have on
surface water concentrations, and will have in the future, requires an understanding of mass flux.
The resulting concentrations of constituents of interest in surface water are directly proportional
to mass flux, thus a significant reduction in mass flux, even if it does not alter maximum
groundwater concentrations, will have a significant effect on concentrations of constituents of
interest in surface water. Therefore, clearly mass flux is relevant and appropriate to discuss in the
context of understanding groundwater and surface water conditions in the vicinity of the ash
basins.
Allen Station — May 2006 Water Levels
Dr. Coster relies on a May 2006 water level map of the Allen station to opine that
groundwater flow from the ash basins is westerly toward numerous private and public water supply
maps (Coster, 2016, p. 15). This contour map was prepared with water level data from very few
monitoring locations. Current knowledge of water levels in the vicinity of the ash basins, based on
a much more extensive water level data sets collected as part of the CSA studies, clearly
demonstrate that the May 2006 water level contour map did not correctly portray groundwater
levels and groundwater flow conditions. The water level maps in the CSA and CAP documents
present the current understanding of groundwater flow conditions near the ash basins, and are the
best available information for understanding groundwater flow directions. Thus, it is inappropriate
to use the May 2006 contour map to represent groundwater flow conditions.
Groundwater Flow Velocity
Dr. Coster states in one part of his supplemental expert report that the pore velocity of
groundwater is equal to the hydraulic conductivity multiplied by the hydraulic gradient divided by
the effective porosity (Coster, 2016, p. 39). In an earlier part of this report, though, he states that
"...steady-state groundwater flow rates and directions are only dependent on hydraulic
conductivity, hydraulic head (effects of groundwater pressure and elevation), and the size of the
area through which the groundwater flows.... " and not effective porosity (Coster, 2016, p. 30).
Conventionally, when a hydrogeologist refers to "groundwater flow rate" he is referencing the rate
at which groundwater migrates in the subsurface. The rate at which groundwater migrates in the
subsurface is a function of the effective porosity, and the smaller the effective porosity the faster
the flow rate. Therefore, it is incorrect to imply that the "groundwater flow rate" is not a function
of the effective porosity.
16

SSU
RCRA Statistical Guidelines
2.2- bVbVDObnr02 9� V220CIV1E2' JWC'
Dr. Cosler states that RCRA guidance regarding sampling frequency requires a separate
analysis of plume velocity for each constituent of interest, and that I failed to perform such an
evaluation in my expert report (Cosier, 2016, pp. 39-41). EPA guidance clearly states that in
determining sampling interval that it is appropriate to ensure that physically independent or distinct
volumes of groundwater are collected on each sampling trip (EPA, 2009, page 14-20). The
guidance does not state that sampling interval should be based on apparent rate of contaminant
migration as opined by Dr. Cosler. In addition, Dr. Cosler states that .... "the actual physical
process being monitored during groundwater sampling is contaminant plume migration. " (Cosler
2016, p. 39). Contrary to Dr. Cosler's opinion, groundwater sampling provides information on
chemical concentrations in the groundwater sample. The sampling, in and of itself, provides no
information on physical processes.
Marshall Station — Monitoring Well AL -1S
In reference to monitoring well AL -1S at the
Marshall station that is located adjacent to a tributary of
the Catawba River, Dr. Cosler states "The unnamed
tributary is a very small intermittent stream and exerts
minimal hydraulic influence on the underlying
aquifer. " (Cosler, 2016, p. 43). In fact, the tributary
forms an arm of Lake Norman near AL -1S and the
water level in the tributary is equal to the water level in , ;
Lake Norman. As a result, there are strong upward D
hydraulic gradients beneath this arm of Lake Norman
and it is incorrect to conclude that the tributary exerts minimal hydraulic influence on the
underlying aquifer.
In summary, since Dr. Cosler did not consider the discharge of groundwater to surface
water near AL -1 S, the calculations made using the analytical transport model of concentrations of
boron at the compliance boundary downgradient of AL -1S are incorrect. With the analytical
transport model, Dr. Cosler calculated that the boron concentrations at the compliance boundary
downgradient of AL -1S exceeds the 2L Standard. My opinion is that the boron concentration in
groundwater at the compliance boundary downgradient of AL -IS does not exceed the standard
because groundwater is discharging to the arm of Lake Norman near AL -1 S.
17

2.2- bVbVDObnr02 9� V220CIV1E2' IMC'
Section 7
References
Andrews, C.B. 2016a. Expert Report of Charles B. Andrew. Allen Steam Station, Belmont, NC.
June 30.
Andrews, C.B. 2016b. Expert Report of Charles B. Andrew. Buck Steam Station, Salisbury, NC.
June 30.
Andrews, C.B. 2016c. Expert Report of Charles B. Andrew. Belews Creek Steam Station, Belews
Creek, NC. August 1.
Andrews, C.B. 2016d. Expert Report of Charles B. Andrew. Marshall Steam Station, Terrell, NC.
August 1.
Butler, J.J. 1998. The Design, Performance, and Analysis of Slug Tests: Lewis Publishers, Boca
Raton, Florida.
Butler, J.J., E.J. Garnett, and J.M. Healey. 2003. Analysis of Slug Tests in Formations of High
Hydraulic Conductivity: Ground Water 41, no. 5: 620-630.
Campbell, S.K., and R.K. Spruill. 2016a. Expert Report Addendum #1, Buck Steam Station, 1555
Dukeville Road, Salisbury, NC 28146. (Federal Case). August 30.
Campbell, S.K., and R.K. Spruill. 2016b. Expert Report Addendum #2, Buck Steam Station, 1555
Dukeville Road, Salisbury, NC 28146. (State Case). August 30.
Catawba Riverkeeper. 2016. SCE&G and Catawba Riverkeeper Reach Settlement on Coal Ash
Storage. http://www.catawbariverkeeper.org/issues/coal-ash-1/sce-g-and-catawba-
riverkeeper-reach-settlement-on-coal-ash-storage. September 18.
Cooper, H.H., J.D. Bredehoeft, and I.S. Papadopulos. 1967. Response of a Finite -Diameter Well
to an Instantaneous Charge of Water: Water Resources Research 3, no. 1: 263-269.
Cosler, D.J. 2016. Supplemental Expert Report of Douglas J. Cosler, Ph.D., P.E. Allen, Cliffside,
and Marshall Steam Station Ash Basins North Carolina. August 30.
Daniel, C.C., D.G. Smith, and J.L. Eimers. 1989. Hydrogeology and Simulation of Ground -Water
Flow in the Thick Regolith -Fractured Crystalline Rock Aquifer System of Indian Creek
Basin, North Carolina. Water Supply Paper 2341-C. U.S Geological Survey.
Doe, T., and J. Remer. 1981. Analysis of Constant -Head Well Tests in Nonporous Fractured
Rock. Proceedings of the Third Invitational Well -Testing Symposium: "Well Testing in
Low Permeability Environments". March 26-28, 1980. Berkeley, California, Lawrence
Berkeley Laboratory. 84-89. March.
Freeze, R.A., and J.A. Cherry. 1979. Groundwater. Englewood Cliffs, NJ: Prentice -Hall, Inc.
Geosyntec Consultants. 2014a. Amended Closure Plan. Wastewater Ash Ponds. Grainger
Generating Station. Conway, South Carolina. January.
Geosyntec Consultants. 2014b. Revised - Amended Closure Plan. Wastewater Ash Ponds.
Grainger Generating Station. Conway, South Carolina. August.
IN

2.2- bVbVDObnr02 9� V220CIV1E2' IMC'
HDR Engineering Inc of the Carolinas. 2016a. Comprehensive Site Assessment Supplement 2.
Allen Steam Station Ash Basin. August 2.
HDR Engineering Inc of the Carolinas. 2016b. Comprehensive Site Assessment Supplement 2.
Buck Steam Station Ash Basin. August 2.
HDR Engineering Inc of the Carolinas. 2016c. Comprehensive Site Assessment Supplement 2.
Belews Creek Steam Station Ash Basin. August 11.
HDR Engineering Inc of the Carolinas. 2016d. Comprehensive Site Assessment Supplement 2.
Marshall Steam Station Ash Basin. August 4.
Hsieh, P.A. 1998. Scale Effects in Fluid Flow through Fractured Geologic Media. In Scale
Dependence and Scale Invariance in Hydrology. Sposito, G. ed.: Cambridge University
Press. 335-353.
Hutson, M.A. 2016a. Supplemental Expert Report of Mark A. Hutson, PG. Belews Creek Steam
Station Ash Basin Belews Creek, NC. August.
Hutson, M.A. 2016b. Videotaped Deposition of Mark A. Hutson, PG. State of North Carolina, ex
rel. North Carolina Department of Environmental Quality, and Roanoke River Basin
Association, Sierra Club, Waterkeeper Alliance, Cape Fear River Watch, Inc., Sound
Rivers, Inc. and Winyah Rivers Foundation vs. Duke Energy Progress, LLC. and State of
North Carolina, ex rel. North Carolina Department of Environmental Quality, and Catawba
Riverkeeper Foundation, Inc., Waterkeeper Alliance Mountaintrue, Appalachian Voices,
Yadkin Riverkeeper, Inc., Dan River Basin Association, and Southern Alliance for Clean
Energy v. Duke Energy Carolinas, LLC. 13 -CVS -11032 and 13 -CVS -14461. State of
North Carolina, County of Wake and State of North Carolina, County of Mecklenburg,
248.
Hyder, Z., J.J. Butler, C.D. McElwee, and W. Liu. 1994. Slug Tests in Partially Penetrating Wells:
Water Resources Research 30, no. 11: 2945-2957.
LeBlanc, D.R. 1984. Sewage Plume in a Sand and Gravel Aquifer, Cape Cod, Massachusetts.
Water -Supply Paper 2218. U.S. Geological Survey.
Mixon, A.C., and GEI Consultants Inc. 2016. 2015 Annual Landfill Inspection Report for the
Wateree Station Class III Landfill in Eastover, SC. January 15.
Parette, R. 2016a. Supplemental Opinions on the Appropriateness of Monitored Natural
Attenuation in Conjunction with Cap -in -Place at the Allen, Belews Creek, Buck, Cliffside,
Marshall, Mayo, and Roxboro Steam Stations. August 30.
Parette, R. 2016b. Supplemental Opinions on the Appropriateness of Monitored Natural
Attenuation in Conjunction with Cap -in -Place at the Allen, Belews Creek, Buck, Cliffside,
Marshall, Mayo, and Roxboro Steam Stations. (Federal Case). August 30.
Santee Cooper. 2014. Letter from Susan W. Jackson to Chris Forrest, SCDHEC Bureau of Water
and Gary Stewart, SCDHEC Bureau of Land and Waste Management, Regarding: South
Carolina Public Service Authority. Santee Cooper Grainger Generating Station, Horry
County, NPDES Permit# SC0001104; Site ID# 00367, NPDES Groundwater Semi-annual
and Compliance Report for 2014. May 28.
19

2.2- bVbVDObnr02 9� V220CIV1E2' IMC'
Santee Cooper. 2016a. Grainger Generating Station Stacks Fall. www.santeecooper.com/about-
santee-cooper/news-releases/news-items/grain _ger-generating-station-stacks-fall. aspx.
February 7.
Santee Cooper. 2016b. Letter from Susan W. Jackson to S.C. Department of Health and
Environmental Control, Regarding: Santee Cooper Grainger Station - NPDES Permit#
SC0001104, NPDES Groundwater and Surface Water Semi-annual Report for 2016. May
26.
South Carolina Electric & Gas. 2013a. Wateree Station Semi -Annual Status Report July -
December 2012. January.
South Carolina Electric & Gas. 2013b. Wateree Station Semi -Annual Status Report January - June
2013. July.
South Carolina Electric & Gas. 2014a. Wateree Station Semi -Annual Status Report July -
December 2013. January.
South Carolina Electric & Gas. 2014b. Wateree Station Semi -Annual Status Report January - June
2014. July.
South Carolina Electric & Gas. 2015. Wateree Station Semi -Annual Status Report January - June
2015. July.
South Carolina Electric & Gas. 2016a. Wateree Station Semi -Annual Status Report July -
December 2015. January.
South Carolina Electric & Gas. 2016b. Wateree Station Semi -Annual Status Report January - June
2016. July.
Southern Environmental Law Center. 2016. South Carolina Coal Ash Removal Ahead of
Schedule While North Carolina Stalls. www.southemenvironment.org/news-and-press/news-
feed/south-carolina-coal-ash-removal-ahead-of-schedule-while-north-carolina-stal. January 25.
20