HomeMy WebLinkAbout2016-05-13 GHI Mayo Report Addendum FinalGEO-HYDRO, INC
Consulting in Geology and Hydrogeology
1928 E. 14th Avenue
Denver, Colorado 80206
(303)322-3171
EXPERT REPORT OF
MARK A. HUTSON, PG
Mayo Steam Electric Plant
Roxboro, NC
Addendum No. 1
Prepared for:
Southern Environmental Law Center
601 West Rosemary Street
Suite 220
Chapel Hill, NC 27516-2356
May 2016
GEO-HYDRO, INC
1. Summary of Opinions Formed
Based upon my review of the available information I have formed the following opinions on closure
of the coal ash basin at the Mayo Steam Electric Plant (Mayo).
Coal ash stored in the Mayo ash basin is the source of contamination detected in surface
water and groundwater resources.
2. Capping the waste within the footprint of the Mayo ash basin will not be protective of
groundwater and surface water quality downgradient of the basin.
3. Monitored Natural Attenuation (MNA) is not a viable remedial option for impacted
groundwater and surface water downgradient of the Mayo ash basin.
4. Capping coal ash located within the Mayo ash basin will not be protective of surface water
quality in Crutchfield Branch.
Removal of the coal ash will reduce the concentration and extent of groundwater and surface
water contaminants.
6. The coal combustion residual impoundment risk classification proposed by North Carolina
Department of Environmental Quality (NCDEQ) improperly minimizes protection of
environmental quality.
7. The Groundwater Flow and Transport Model Does Not Reflect Real -World Conditions.
The background and rationale behind each of these opinions are described in this report.
GEO-HYDRO, INC
2. Introduction
On February 26, 2016 I completed and submitted my original expert report on the Mayo Steam
Electric plant (Mayo). Subsequent to submittal of my report the Corrective Action Plan (CAP) Part
2 was released for public review and comment. On behalf of the Southern Environmental Law
Center, the documents that I have now reviewed includes; the Groundwater Assessment Plan
(SynTerra, 2015a), Comprehensive Site Assessment (SynTerra, 2015b), the CAP Part 1 (SynTerra,
2015c), the CAP Part 2 (SynTerra, 2016), the Draft Proposed Risk Classifications (NCDEQ, 2016),
and the National Pollutant Discharge Elimination System (NPDES) permit for Mayo (North Carolina
Department of Environment and Natural Resources (NCDENR, 2009). I have also reviewed raw
groundwater model input and output files provided by Duke to facilitate understanding of the model
that informs the CAP Parts 1 and 2. This addendum to my report provides additional observations
and opinions formed that are new to this document or have recently been recognized.
2
GEO-HYDRO, INC
3. Qualifications
This section is unchanged from my expert report dated February 2016.
GEO-HYDRO, INC
4. Site Background
The Mayo CAP Part 21 indicates that ash may be capped with a "low permeability material", not
necessarily a synthetic cap. The selection of a cap design is a critical component needed to evaluate
the amount of precipitation that will infiltrate into the ash. Even new synthetic caps routinely allow
some small amount of infiltration through faults in the fabric and/or seams. Infiltration through a low
permeability soil cap would be expected to allow more infiltration than a synthetic cap, at least
initially. Evaluation of realistic infiltration through the specific materials that are to be used to
construct the cap must be performed in order to accurately predict the volume of infiltration and the
volume of leachate that can be expected to migrate out of the basin. While MODFLOW does not offer
cap performance evaluations, other programs such as Hydrologic Evaluation of Landfill Performance
(HELP) should be used to generate reasonable infiltration values for use in MODFLOW.
The same section of the document acknowledges for the first time the fundamental truth that water will
continue to enter into the ash basin, even if a cap is installed 2. The document states,
"Drains will be an integral part of the cap -in-place closure to provide a flow path for water
entering the system as precipitation, run-off, or groundwater discharging into the basin from
upgradient areas and to limit/prevent the development of impoundment of water in and
upstream of the capped ash."
This document indicates that drains will be used to control water that would otherwise enter the ash
from the sides but provides no information about how such a drain system might be designed,
constructed, and operated to control infiltration of precipitation from above, inflow of groundwater
from the sides, and discharge of contaminated groundwater from the basin. However, neither the
volume of water that is predicted to discharge into the basin from upgradient areas, nor the volume of
ash that will remain saturated and continue to generate leachate has been reported by Duke. Without
knowledge of these factors, the claim that the drains would be effective is unsupported.
Determination of Background Water Quality
Accurate and appropriate determination of background water quality is of critical importance to
evaluation of groundwater impacts from the ash basins. Provisional background concentrations for
groundwater, as well as other media, were proposed in the CAP Part 1. The CAP Part 2 reported
analytical results from rounds 1 thru 4 of background groundwater sampling, but further evaluation of
results and background concentrations was deferred to future reports. Groundwater monitoring results
are currently being compared to provisional background concentrations that are based on an
incomplete data set.
1 SynTerra (2016), p. 6-6, Section 6.3
2SynTerra (2016), p. 6-6, Section 6.3
2
GEO-HYDRO, INC
Review of the updated analytical data from background monitoring wells provided in the CAP Part 23
show that presumed background well MW- 11BR contained very high concentrations of hexavalent
chromium (13.6 ug/1) and vanadium (12.1 ug/1) at the time of the December 2015 sampling event. The
detected hexavalent chromium in well MW- 11BR was over 2 orders of magnitude greater than the
USEPA Tapwater Regional Screening Level4 of 0.035 ug/1. The detected vanadium concentration of
vanadium in this same well is an order of magnitude higher than the 0.3 ug/1 Interim Maximum
Allowable Concentration (IMAC) and 3 ug/1 greater than the previously detected high background
value that was also measured at well MW -1 113R. The repeatedly high values of ash -related
constituents in this well appears to indicate that this well may be constructed in a different geologic
formation, or at in materials that differ significantly from the other background monitoring wells. This
well specifically, and all of the background wells in general, must be evaluated to determine if each
well is appropriate for inclusion in the background monitoring program.
Further, the values identified as the provisional background are either calculated values or are the
highest value recorded at any of the various locations, both of which are dependent on the distribution
of concentrations obtained during sampling. Water quality monitoring results are easily influenced by
sampling methods and techniques and can easily provide spurious data. Even the background data set
that is being developed by sampling new background wells around the Mayo site is capable of
producing outliers that if not removed could skew the background concentration. The complete data
set needs to be evaluated for sampling or analytical problems and tested to eliminate outliers that could
skew the data prior calculating or selecting background concentrations.
3 SynTerra (2016), Appendix A-1
4 SynTerra (2015b), Table 2-5
5
GEO-HYDRO, INC
5. Opinion 1: Coal Ash Stored in the Mayo Ash Basin is the Source
of Contamination Detected in Surface Water and Groundwater
This section is unchanged from my expert report dated February 2016.
0
GEO-HYDRO, INC
6. Opinion 2: Capping Coal Ash Located Within the Mayo Ash
Basin Will Not Protect Groundwater Quality Downgradient of the
Basin
The CAMA process proposed designation of the Mayo site as low-risk creates the possibility that
Duke Energy (Duke) could pursue closure of the Mayo impoundment by capping the disposed ash in
place. Capping the waste within the footprint of the ash basin will not be protective of groundwater
quality downgradient of the basin.
Environmental contaminants contained in coal ash are leached into groundwater when precipitation
infiltrates through the waste or, when groundwater flows through waste that has been placed below
the water table. In the case of the Mayo ash basin, both of these processes are currently acting to
create the contaminated ash porewater, groundwater, and surface water that discharge into
Crutchfield Branch and contaminate surface water downstream of the impoundment. The cap -in-
place remedy would likely reduce the amount of water that enters the waste from precipitation. This
remedy would however do nothing to reduce the amount of water that flows laterally into the basin
from surrounding geologic materials, through the capped waste, and eventually into Crutchfield
Branch.
The CAP and its included modeling results show that much of the coal ash in the Mayo ash basin
will remain submerged below the water table under a cap -in-place remedy. The thickness of
saturated waste that would remain saturated can be estimated by comparing the pre -impoundment
topographic map of the basin (Figure 1) with the calculated hydraulic head map for the cap in place
option (Figure 5). The topographic map of the buried valley shows that natural land surface below
the impoundment is located at approximately 400 feet above mean sea level (msl) near the center of
the basin. The calculated hydraulic head in the basin after ash is capped in place is predicted by the
model to be approximately 460 to 470 feet msl near the center of the basin. Therefore, the model of
the cap -in-place scenario predicts 60 to 70 feet of saturated ash would remain in portions of the
basin. The data presented in these figures is consistent with the known hydrogeology of the site.
Groundwater will continue to flow into the ash basin from adjacent areas and some infiltration
through the cap would continue to occur. Groundwater that flows through the ash will continue to
leach metals from the ash and transport those metals down -gradient before discharging into
Crutchfield Branch.
Further, the results of the groundwater modeling exercise undertaken as part of this program indicate
that removal of the Mayo coal ash is by far the most effective option for improving groundwater
quality and preventing future discharges to Crutchfield Branch. The modeling results indicate that
7
GEO-HYDRO, INC
removal of the ash significantly reduces the size and concentration of the boron plume in the
saprolite and transition zones, something that capping -in-place does not achieve.
The Mayo CAP Part 25 indicates that ash may be capped with a "low permeability material", not
necessarily a synthetic cap. The selection of a cap design is a critical component needed to evaluate
the amount of precipitation that will infiltrate into the ash. Even new synthetic caps routinely allow
some small amount of infiltration through faults in the fabric and/or seams. Infiltration through a
low permeability soil cap would be expected to allow more infiltration than a synthetic cap, at least
initially. Evaluation of realistic infiltration through the specific materials that are to be used to
construct the cap must be performed in order to accurately predict the volume of infiltration and the
volume of leachate that can be expected to migrate out of the basin. While MODFLOW does not
offer cap performance evaluations, other programs such as Hydrologic Evaluation of Landfill
Performance (HELP) should be used to generate reasonable infiltration values for use in
MODFLOW. Even though there are serious flaws in the groundwater modeling that artificially
restrict groundwater impacts under the cap -in-place scenario (see Opinion 7), the modeling effort
confirms that ash excavation is the option that is most protective of the environment and is a
permanent solution.
The same section of the of the document acknowledges for the first time the fundamental truth that
water will continue to enter into the ash basin, even if a cap is installed 6. The document states,
"Drains will be an integral part of the cap -in-place closure to provide a flow path for water
entering the system as precipitation, run-off, or groundwater discharging into the basin from
upgradient areas and to limit/prevent the development of impoundment of water in and
upstream of the capped ash."
Neither the volume of water that can be expected to discharge into the basin from upgradient areas,
nor the volume of ash predicted to remain saturated and continue to generate leachate has been
reported by Duke. This document indicates that drains will be used to control water that would
otherwise enter the ash from the sides but provides no information about how such a drain system
might be designed, constructed, and operated to control infiltration of precipitation from above,
inflow of groundwater from the sides, and discharge of contaminated groundwater from the basin.
Without knowledge of these factors, the claim that the drains would be effective is unsupported.
5 SynTerra (2016), p. 6-6, Section 6.3
6 SynTerra (2016), p. 6-6, Section 6.3
GEO-HYDRO, INC
7. Opinion 3: Monitored Natural Attenuation Is Not An Acceptable
Groundwater Remediation Strategy at Mayo
The CAP7 indicates that Duke may evaluate MNA as a potential groundwater remedy for certain
area of the Mayo site. The CAP attempts to make it appear that MNA is a viable remedial option for
impacted groundwater and surface water downgradient of the Mayo ash basin. However, MNA is
not a viable closure option for this site for several reasons:
• Duke Energy has not proposed removal of the waste for disposal in a secure
location. Modeling presented in this document shows that most of the ash would
remain saturated after capping. Saturated ash will continue to leach metals into
groundwater that will flow toward and eventually discharge into Crutchfield
Branch. As a practical matter, in the absence of removal all sources of
contamination cannot be controlled
• Many of the ash -related constituents in groundwater at this site neither degrade
nor attenuate. The Geochemical Site Conceptual Model$ states that boron best
represents the extent of impact to groundwater because it "does not sorb or
precipitate within the ash or on aquifer materials." Because of this, the Mayo site
would not be eligible for MNA.
• Crutchfield Branch, which receives the contaminated groundwater discharge at
Mayo, currently exceeds surface water standards for several parameters.
Sampling has detected ash -related metals at concentrations above background and
relevant NCAC 2B and/or 2L standards, including boron, cobalt, copper, iron,
manganese, thallium, vanadium, and zinc9. These surface water exceedances are
likely to continue since the majority of flow in Crutchfield Branch is associated
with drainage from the ash basin10 and the groundwater model shows that a
substantial amount of ash will continue to be saturated with groundwater that
eventually flows to Crutchfield Branch, even if the basin is capped.
7 SynTerra, 2015b, p.5-1
S SynTerra, 2015b, Section 3.2
9 SynTerra, 2015b, Table 2-14
10 SynTerra, 2015b, Page 4-13, Section 4.4.1
0
GEO-HYDRO, INC
8. Opinion 4: Capping Coal Ash Located Within the Mayo Ash
Basin Will Not Be Protective of Surface Water Quality in
Crutchfield Branch
Prior to construction of the Mayo plant Crutchfield Branch was a free-flowing perennial stream that
occupied the valley through which it flowed. The coal ash basin was constructed by damming the
stream and used to segregate and store coal combustion wastes that were allowed to settle in the
basin. If coal ash located within the Mayo ash basin is capped in place, the surface water in
Crutchfield Branch will continue to be polluted and a significant portion of Crutchfield Branch will
remain buried by waste.
Crutchfield Branch, which receives the contaminated groundwater discharge at Mayo, currently
exceeds surface water standards for several parameters. Sampling has detected ash -related metals at
concentrations above background and relevant NCAC 2B and/or 2L standards, including boron,
cobalt, copper, iron, manganese, thallium, vanadium, and zinc". These surface water exceedances
will continue since the majority of flow in Crutchfield Branch is associated with drainage from the
ash basin 12 and the groundwater model shows that a substantial amount of ash will continue to be
saturated with groundwater that eventually flows to Crutchfield Branch, even if the basin is capped
(See Opinion #2). Coal ash must be separated from the groundwater and surface water flow
systems if further contamination of these resources is to be avoided
The surface water quality discussion in the CAP 13 indicates that concentrations of ash -related
constituents in Crutchfield Branch decrease with distance downstream from the ash basin due to
"attenuation by dilution." This is problematic because, of course, there is so such process as
attenuation by dilution. Attenuation and dilution are two different processes. Attenuation occurs
when contaminants interact with the soil or sediments, and contaminants are removed from the
water. Dilution does not remove or treat any contaminants; instead, the concentration, but not the
amount, of a pollutant is reduced by diluting the polluted plume with water containing a lower
concentration of the contaminant. Thus, dilution and attenuation are distinctly different processes.
The Cap -In -Place groundwater model scenario assumes that a 100% effective cap is installed over
the ash and that infiltration into the ash from above is zero. The Mayo CAP Part 214 indicates that
ash may be capped with a "low permeability material", not necessarily a synthetic cap. The
selection of a cap design is a critical component needed to evaluate the amount of precipitation that
will infiltrate into the ash. Even new synthetic caps routinely allow some small amount of
11 SynTerra, 2015b, Table 2-14
12 SynTerra, 2015b, Page 4-13, Section 4.4.1
13 SynTerra, 2015b, Section 4.4.2
14 SynTerra (2016), p. 6-6, Section 6.3
10
GEO-HYDRO, INC
infiltration through faults in the fabric and/or seams. Infiltration through a low permeability soil cap
would be expected to allow more infiltration than a synthetic cap, at least initially. Evaluation of
realistic infiltration through the specific materials that are to be used to construct the cap must be
performed in order to accurately predict the volume of infiltration and the volume of leachate that
can be expected to migrate out of the basin. While MODFLOW does not offer cap performance
evaluations, other programs such as Hydrologic Evaluation of Landfill Performance (HELP) should
be used to generate reasonable infiltration values for use in MODFLOW. Even though there are
serious flaws in the groundwater modeling that artificially restrict groundwater impacts under the
cap -in-place scenario (see Opinion 7), the modeling effort confirms that ash excavation is the option
that is most protective of the environment and is a permanent solution.
The same section of the of the document acknowledges for the first time the fundamental truth that
water will continue to enter into the ash basin' 5. The document states,
"Drains will be an integral part of the cap -in-place closure to provide a flow path for water
entering the system as precipitation, run-off, or groundwater discharging into the basin from
upgradient areas and to limit/prevent the development of impoundment of water in and
upstream of the capped ash."
Neither the volume of water that can be expected to discharge into the basin from upgradient areas,
nor the volume of ash that is predicted to remain saturated and continue to generate leachate has
been reported. This document indicates that drains will be used to control water that would
otherwise enter the ash from the sides but provides no information about how such a drain system
might be designed, constructed, and operated to control infiltration of precipitation from above,
inflow of groundwater from the sides, and discharge of contaminated groundwater from the basin.
Without knowledge of these factors, the claim that the drains would be effective is unsupported.
The CAP attempts to make it appear that MNA is a viable remedial option for impacted groundwater
and surface water downgradient of the Mayo ash basin. In order to be an appropriate remedy MNA
would have to demonstrate that the contaminant has the capacity to degrade or attenuate under site-
specific conditions. The Geochemical Site Conceptual Model 16 states that boron best represents the
extent of impact to groundwater because it "does not sorb or precipitate within the ash or on aquifer
materials." The properties of the contaminants at Mayo would therefore render the site ineligible for
remediation by MNA, even if attenuation by dilution were a real process. In addition, capping the
coal ash in the in place would leave a significant portion of Crutchfield Branch buried in waste.
Now that the Mayo waste handling system is no longer going to be used there is no justification for
leaving the existing stream covered in coal ash.
15 SynTerra (2016), p. 6-6, Section 6.3
16 SynTerra, 2015b, Section 3.2
11
GEO-HYDRO, INC
9. Opinion 5: Removal of the Coal Ash Will Remove the Source and
Reduce the Concentration and Extent of Groundwater and
Surface Water Contaminants
This section is unchanged from my expert report dated February 2016.
12
GEO-HYDRO, INC
10. Opinion 6: The Coal Combustion Residual Impoundment Risk
Classification Proposed by NCDEQ Improperly Minimizes
Protection of Environmental Quality
This section is unchanged from my expert report dated February 2016.
13
GEO-HYDRO, INC
11. Opinion 7: The Groundwater Flow and Transport Model Does
Not Reflect Real -World Conditions
Groundwater models can be useful tools that can be employed to evaluate alternative actions at
waste disposal site such as Mayo. This usefulness however, is predicated on the model being
constructed in a manner that faithfully recreates actual field conditions. However, the model
includes fundamental inconsistencies between observed and modeled conditions that cast doubt on
results and make it likely that the nature and extent of groundwater contamination is understated.
Each of the identified issues with the current model are identified separately below.
Groundwater Flow Model
MODFLOW Model Geometry Artificially Isolates the Ash
The definition of the inappropriately isolates the ash. The ash in layers 1-4 has no lateral continuity
with any of the adjacent natural materials. The layers extend laterally to the edge of the ash basin
and terminate at a no flow boundary. When a 100% effective cap is assumed, the only flow in the
ash of layers 1-4 is vertically into or out of the underlying saprolite through the bottom of layer 4
(and top of layer S). This prohibition of lateral flow at the edges of the ash pond into/from the
enclosing saprolite is compounded by the corresponding imposition of horizontal -to -vertical
anisotropy for coal ash hydraulic conductivity. As modeled, the vertical flow between the ash and
saprolite is constrained by hydraulic conductivity that is only 20% of that which would exist were
the connection between the materials to occur horizontally under the same gradients. The conditions
included in this model create the equivalent of a 100% effective liner on the top and sides and an
80% effective liner at the bottom of the ash. This is, of course, not reflective of real—world
conditions.
MODFLOW Model Is Overlv Constrained
The lateral boundary conditions for layer 7 (base of the saprolite/transition layers) are almost
everywhere head -defined. The imposition of fixed heads across significant portions of a model,
particularly the lateral boundaries is always risky modeling procedure. Unless there is abundant
groundwater data or a known physical condition for such constraint, the modeler runs the risk of
over -constraining the model and generating results that represent an artificial world, not the
mathematical approximation of the real world.
The use of heads to calibrate a model that is highly constrained by fixed heads is, in essence, circular
reasoning. Calibration is meaningful only if calibration is achieved using a measurable parameter
that is not significantly constrained as a boundary condition. If the model is substantially
constrained by assigned heads, a calibration based upon flows, for example, will allow
discrimination between a physically significant model and one that is simply an impressive
14
GEO-HYDRO, INC
calculation of conditions that don't exist in the real world. For this model, acceptable head
calibration is demonstrated, but it has no significance because that head distribution was dictated by
boundary conditions. And, there are no site data which can be used to calibrate how closely this
mathematical computation matches independent parameters in the real world. Streams were
simulated as drains with exceptionally high conductance. Functionally, this means they too are fixed
head boundaries, contributing to the over -constraining of the model with respect to heads.
MODFLOW Model Does Not Accurately Simulate Vertical Gradients
A critical measure of whether or not a model is properly designed and implemented is whether flows
occur in the right direction. Relying solely on impressive statistics for head calibration can mask a
model that is remarkably wrong on flow directions. The baseline model for this site is just such a
case. When one compares the directions of vertical flow17 that are simulated with the directions of
flow that are observed in the data from nested wells, the model is almost always in error. Where
vertical flow is observed to be upward, the simulation describes it as downward. Where vertical
flow is observed to be downward, the simulation describes it as upward. Inability of the
MODFLOW model to simulate appropriate vertical flow directions is another indication that the
model is overly constrained and raises serious doubt about the adequacy of this model to support
critical remedial decision making.
Fate and Transport Modeling Scenarios
Cap -In -Place Source Concentrations Are Unjustifiable
The contaminant transport simulations of the Cap -In -Place scenario is badly flawed. The change
from constant concentration in the ash (fixed source) in the current condition simulation to variable
concentration is completely unjustifiable. In MT3DMS, variable concentration cells add no
contaminants to the groundwater. By changing the cells from constant to variable concentration the
modeler assumes that any ash that remains saturated is somehow incapable of producing
contamination for any constituent. The shift in the MT3DMS source from leachate producer to inert
media has no justification.
Stopping, or approximately stopping, vertical infiltration into the ash has no impact on the capacity
of the ash to continue to produce leachate. Since some un -reported thickness of ash is expected to
remain saturated as a result of lateral or vertical flow from the surrounding strata, the ash will
continue to produce the same leachate quality that it did historically. Non -zero infiltration will
similarly generate contaminant -bearing leachate as it drains downward through unsaturated ash.
17 SynTerra , 2016, Appendix B, Table 1
15
GEO-HYDRO, INC
References
NCDEQ, 2016, Coal Combustion Residual Impoundment Risk Classifications, January 2016.
NCDENR, 2009, Permit to Discharge Wastewater Under the National Pollutant Discharge
Elimination System, Permit NC0038377, October, 2009
SynTerra, 2014, Groundwater Assessment Work Plan for Mayo Steam Electric Plant, September
2014
SynTerra, 2015a, Comprehensive Site Assessment Report, Mayo Steam Electric Plant, September
2015.
SynTerra, 2015b, Corrective Action Plan, Part 1, Mayo Steam Electric Plant, December 2015.
SynTerra, 2016, Corrective Action Plan, Part 2, Mayo Steam Electric Plant, February 2016.
United States Geological Survey, Cluster Springs VA. — N.C., 7.5 Minute Topographic Map, 1968,
photorevised 1987.
16
GEO-HYDRO, INC
Figures
This section is unchanged from my expert report dated February 2016.