20160742 Ver 2_Monitoring Plan_20180918_20190519

Cedar Cliff Hydroelectric Development Auxiliary Spillway Upgrade Project
Proposed Duke Energy Water Quality Monitoring Plan
Project Overview
In 2014, the FERC established the Inflow Design Flood (IDF) for Cedar Cliff development as
the Probable Maximum Flood (PMF). Prior to the FERC notice, the OF had been 40
percent of the PMF. The existing spillway discharge capacity is insufficient to pass the PMF
without overtopping the dam and resulting in potential failure of the structure. Engineering
design efforts are underway to expand the existing auxiliary spillway (width and depth) and
install Hydroplus-Fusegates as the control section. In addition, the existing parapet wall will
be replaced with a PMF Wall to create additional storage for the Cedar Cliff
reservoir. Studies have indicated the expanded auxiliary spillway, Fusegates, and PMF
Wall provide the necessary measures to safely pass the OF without overtopping Cedar Cliff
Dam.
The construction project is scheduled to begin July, 2019 and be completed December,
2020. The present plan specifies excavating material (approximately 283,200 cubic yards)
from the mountain hillside east of the current fuse plug (Figure 1). A gravel filter berm at the
lower end of the auxiliary spill channel will provide sediment and erosion control from the
excavations.
During construction, the lake will be lowered 30 ft to accommodate the construction activities
including staging the excavated material on the foot print of the existing fuse plug approach
channel. The excavated material will be loaded onto barges and spoiled into Cedar Cliff
Lake upstream of the dam (Figure 2) (for discussion and review of submerged disposal, see
HDR 2018a). As specified in the USACE 404/401 permit, a 3-5 ft floating turbidity barrier will
be installed at all work areas that are in, or adjacent to surface waters.
As discussed by HDR (2018a), pyrite (FeS2) was identified in rock exposures at the site and
in the rock core from boreholes drilled for the subsurface investigation (HDR 2017).
Subsequent petrographic analyses of metasedimentary rocks of the Tallulah Falls Formation
(TFFm) collected from the rock cores found that the garnet mica schist, mica schist, and
schistose biotite gneiss lithologies contained 2 percent to 7 percent pyrite by volume (HDR
2017). Based on the boreholes drilled during the geological/ geotechnical site investigation
for the Auxiliary Spillway upgrades, approximately 26 percent of the total excavated material
(73,600 cubic yards) will be made up of these three rock lithologies.
Even though there are no known instances of acid -drainage related to the metasedimentary
rocks of the TFFm in the region surrounding the site; rocks with greater than 1 percent pyrite
and/or pyrrhotite by volume are considered to be potentially acid -producing. Pyrite can react
in the presence of atmospheric oxygen and water to form ferrous sulfate and sulfuric acid
(2FeS2 + 702 + 2H2O -> 2FeSO4 + 2H2SO4). The stoichiometry of complete oxidation of one
mole of pyrite would produce 4 equivalents of hydrogen ions. Although some acid -drainage
is produced by natural weathering, construction activities can expose large volumes of rock
containing sulfide minerals to oxidizing conditions. The oxidation of pyrite and subsequent
acid production increase significantly when exposed to atmospheric oxygen and particle size
becomes smaller and smaller (Pugh et.al. 1984). Subsequent leaching of the oxidation
products by rainfall/groundwater result in the formation of acid drainage which is

Proposed Duke Energy Monitoring Plan
characterized by low pH values, high concentrations of sulfate, and mobilization of metals
such as iron, aluminum, and manganese.
Bosch and Meckenstock (2012) have suggested anaerobic nitrate -dependent microbial
pyrite oxidation may also contribute to acid production. Geochemical and stable isotope field
data from anaerobic aquifers indicate that denitrification was associated with pyrite
oxidation. Laboratory studies were more ambiguous, but suggested that this process may
occur under some conditions. The molecular mechanism of pyrite oxidation coupled with
denitrification is not well understood.
Factors affecting the amount of Acid -Production
• Percent Pyrite in Rock
• Morphology and grain size of the iron sulfide minerals
• Particle size of excavated material (surface area of disturbed or spoiled rock)
• Presence of certain iron bacteria (primarily Thiobacillus ferrooxidans)
• Presence of alkalinity producing material
• Oxygen concentration
• Nitrate
In addition to the factors affecting the amount of acid production, the characteristics of
Cedar Cliff Reservoir (Table 1) also influence the amount of oxidation products in the
reservoir and downstream reaches. For example, the wide range of pH values reflected the
poor buffering capacity of the low alkalinity water as biological activity metabolized carbon
dioxide. In addition, the very low ionic strength, as measured by conductivity, also influenced
the ability to measure accurate pH values as well as indicating the poor buffering capacity.
The relatively high oxygen concentrations in the lake would favor higher rates of pyrite
oxidation, but these rates would be very low compared to exposure to atmospheric oxygen.
The low nitrate concentrations would limit the amount of de -nitrification associated with
anaerobic nitrate -dependent microbial pyrite oxidation. Probably the most significant factor
limiting the accumulation of pyrite oxidation products in the lake is the very high flushing rate
of the reservoir (Table 1). The average retention time of the water at full pool is 15.3 days;
at a 30 foot drawdown the average retention time of the water is reduced to 8.1 days. Also,
at the 30 foot drawdown, the only way to maintain the lake level is by the hydro operation,
which draws water from the bottom of the lake further reducing the accumulation of
oxidation products in the lake.
Even though HDR (2017), HDR (2018a), and HDR (2018b) have discussed the project in
detail and have projected minimum, if any, acidification impacts from pyrite oxidation, and
the lake characteristics also suggest a lack of accumulation of acidic water, the potential
does exist for an alteration of the water quality. The proposed monitoring program is
designed to assess any direct changes in water quality due to pyrite oxidation.
Page 2 of 8

Proposed Duke Energy Monitoring Plan
Table 1. Cedar Cliff Reservoir Characteristics
Parameter Units
Full Pond (Spillway Elevation)
2330 ft-msl
Tainter Gate Elevation
2305 ft-msl
30 Foot Drawdown Elevation
2300 ft-msl
Cedar Cliff Hydro intake Elevation
2202 ft-msl
Proposed Elevation of Top of Rock Spoil
2250 ft-msl
Lake Volume below Full Pond
6320 ac -ft
Lake Volume below Tainter Gate
3742 ac -ft
Lake Volume below 30 foot drawdown
3386 ac -ft
Lake Volume below hydro intake
42 ac -ft
Mean Outflow'
209.0 cfs
Mean Retention Time (full pond)
15.3 days
Mean Retention Time (30 foot drawdown)
8.1 days
Max Depth at Dam (full pond)
148 ft
Max Depth mid -lake sampling (full pond)
106 ft
pH range2
5.7-8.0
Mean Alkalinity2
0.13 meq/1
Conductivity Rangel
14-20 uSi
Mean Oxygen below 2250 ft-ms12
7.8 mg/l
Minimum observed oxygen2
4.2 mg/l
Mean Nitrate2
0.019 mg/l
1 Calculated from Cedar Cliff operations (1953-2013)
2 from NCDENR data reported in NP&L (2000)
Figure 1. Cedar Cliff Spill Channels and Proposed Excavation and Construction Areas with
Proposed Water Quality Sampling Sites (pink circles)
Page 3 of 8

Proposed Duke Energy Monitoring Plan
Figure 2. Bathymetric Map of Cedar Cliff Reservoir Showing the Proposed Spoil Areas, the Hydro Intake, and the Downstream
Reach with Proposed Water Quality Sampling Sites (red circles)
A CEdareIrr powarhwseTunrrel Ilei Cedar Cliff Fake Bathymetry
_czaoctBM 713 -Foot Contours
l i i i i i i i i i i i i i a i i h war rye ras are n ua 6urN y f— -e-- t. w,UMM
:o3 -*-ab: 9}:�rrr' Nat-. Ca Vz 8trt Pac MAD93 U5 Bracy Fbd hlouernber ��.5
Page 4 of 8

Proposed Duke Energy Monitoring Plan
Cedar Cliff Monitoring Rationale
Most environmental monitoring programs directly assess water quality or employ various
indices for biological impacts, usually macroi nve rte b rates in streams. The proposed
water quality assessment is designed to directly address the potential impact of the rock
spoil in the reservoir, the spill channel, and the Tuckasegee River below the excavation
site.
The biological communities, particularly macro -invertebrates, are influenced by many
factors and rarely exhibit a direct cause and effect of a perturbation. He et.al. (2015),
Svitok et.al. (2014), Gray and Delaney (2008), and DeNicola and Stapleton (2016) have
reviewed the use of macro -invertebrate communities to assess acid mine drainage with
mixed results. Unless the pH was extremely low or significant iron oxide precipitation
was covering the substrates, many diversity indices, biological integrities, density and
taxonomic diversities, and various community metrics applied to macroinvertebrate
populations showed highly variable results in streams with acid -mine drainage.
The NCDENR (2011) reported good to excellent benthic communities in streams with
low pH (mean 5.98, range 5.4-6.9). These data suggest that the use of benthic
communities to assess the impact of low pH values was limited. Macro -invertebrate
communities are extremely difficult to assess in the deeper portions of reservoirs and are
probably non-existent in the dry spill channel. Cedar Cliff hydro discharge into the
riverine habitat is very different than the Lake Habitat downstream of Bear Creek Hydro.
For these reasons, macro -invertebrate assessments are not recommended to monitor
potential impacts from potential pyrite oxidation at Cedar Cliff.
Unlike biological assessments, water quality measurements specifically designed to
detect any chemical alterations have the advantage of:
• Direct correlation to pyrite oxidation
• Relatively rapid analysis of data
• Trends are readily assessed
• Various oxidation and acidification pathways are directly elucidated
• Treatment options (if necessary) may be evaluated and tested during the
excavation process rather than relying on long-term mitigation
Each water sample would measure the reactants and products of the pyrite oxidation
reactions. These analyses will be used to document the pH and pyrite oxidation
products in the lake, area leachates, and de -nitrification prior to and during the
construction project.
In addition to periodic water sampling, continual recording of conductivity would provide
a record of overall ionic change in the water. If oxidation products (ions) increase, the
very low conductivity of the Cedar Cliff water should reflect an increase in the ionic
strength and provide a record of the degree of change (if any).
Page 5 of 8

Proposed Duke Energy Monitoring Plan
Chemical Basis for Water Quality Monitoring
Major pH Buffering in Surface Water (Alkalinity): (Stumm and Morgan, 1981)
H20<=> H+ + OH -
0O2+ H20<=> H2CO3 <=> H+ + HCO3'- <=> H+ +CO3 2
Chemistry of Pyrite Oxidation (Stumm and Morgan, 1981)
Pyrrhotite reacts with oxygen and water to produce reduced iron and sulfuric acid
Fe(1-x)S + (2-0.5x)02+ xH20=> (1-x)Fe2+ +SO4 2- + 2xH+
Pyrite reacts with oxygen and water to produce reduced iron and sulfuric acid
2FeS2 (s) + 702 + 2H2O => 2Fe2+ + 4SO42- + 4H+ [2FeSO4 + 2H2SO4]
Reduced ferrous iron reacts with oxygen and acid to produce Ferric iron
(rate limiting except if biologically catalyzed)
Fee+ + 1/402 + H+ => Fe 3+ + 1/2H20
Ferric Iron reacts with water to form iron hydroxide (yellow -red precipitate) and acid
Fe3+ + 3H2O => Fe(OH)3 (s) + 3H+
Excess ferric iron reacts with pyrite and water to form ferrous iron and sulfuric acid
FeS2 (s) +14Fe3+ + 8H2O => 15Fe2+ + 2SO42- + 16H+
Table 2. Proposed Chemical Analysis of Water Samples Pending Detailed Review of
Procedures to Achieve Low Limits of Detection
Parameter
Chemical
Tentative Analytical Method
Field Analysis Lab Analysis
Symbol
Whole Water Sample
Alkalinity
HCO3'- C032-
n/a-
Titration (0.025N HCI),
Inflection end-point
Total Iron
Fe
n/a--
Digestion, ICP
Raw Water, ICP
Aluminum
Al
n/a-
Raw Water, ICP
Manganese
Mn
n/a-
Raw Water, ICP
Calcium
Cat+
n/a-
Raw Water, ICP
pH
H+
Low conductivity
- n/a-
electrodes
Turbidity
n/a
Hach 2100Q Portable
n/a-
Turbidimeter
Ferrous Iron
Feel
1, 10 Phenanthroline
n/a-
colorimetric
Iron Hydroxide
Fe(OH)3
absorbance
n/a-
Field Filtered Water Sample
Nitrate -Nitrite
NO3'—NO21-
n/a-
Colorimetric
Sulfate
SO42-
n/a-
Low level
ion chromatography
Page 6 of 8

Proposed Duke Energy Monitoring Plan
Water Quality Monitoring Program
The water quality monitoring program is divided into two phases, namely the reservoir
sampling and the spill channel sampling. The reservoir sampling is designed to evaluate
the slower pyrite oxidation rates due to lower oxygen concentrations associated with the
large particle spoil on the bottom of the lake and the suspension of small particles
washed off the larger rocks as they are put in the lake. The spill channel sampling is
designed to evaluate the expected higher oxidation rates due to the high oxygen content
of the atmosphere and the higher surface to volume ratios of the fine particles in the
excavated area. These fine particles would be suspended and transported down the
auxiliary spill channel during rain events.
Cedar Cliff Reservoir Samolin
Monthly 1 -meter profiles of Temperature, Dissolved Oxygen, Conductivity, pH, and
turbidity will be collected with a Hydrolab sonde fitted with a low ionic strength pH
reference electrode. The 1 -meter profiles would be taken in the reservoir at the deepest
point in the vicinity of the Cedar Cliff hydro intake (see Figure 2) and at the deepest
point, approximately 2000 feet up -stream of the in -lake spoil footprints. Water samples
(Table 2) would be taken one meter above the lake bottom and one meter from the
surface. An additional water sample would be taken from a depth corresponding to
either abnormally high turbidity or abnormally low pH values in the water column.
Additional monthly water samples would be taken in the immediate tailrace of Cedar Cliff
Hydro and, as recommended by the USACE, just upstream of the East Fork/West Fork
confluence. A recording Hobo© fresh water conductivity data logger will be placed in the
Cedar Cliff Tailrace for the duration of the project.
Cedar Cliff Spill Channel Sampling
Since the spill channel is normally dry, but does serve as a conduit for water runoff
during rain events, the water sampling has to be conducted while the channel has runoff
water in it. Therefore, the water samples would be taken during runoff events of 1/2 inch
of rain or more per 24 -hours (Figure 1). A continuous recording water level sensor and
a recording conductivity sensor will be placed in the auxiliary spill channel upstream of
the sediment berm location prior to excavation'. The data from a recording rain gage
located on Cedar Cliff dam will be correlated with the continuous water level data from
the auxiliary spill channel. After the sediment berm is installed, a second set of water
level and conductivity sensors will be added downstream of the berm. The data will
document runoff events with the associated ionic strength throughout the project.
Data Review and Reporting Requirements
Reservoir, tailrace, and spill channel sampling activities began in July 2018 to establish
pre -construction conditions. Sampling will continue throughout the duration of the
project, which is anticipated to be completed by March 2021. Duke Energy will consult
with applicable state and federal regulatory agencies to determine if potential
remediation measures should be implemented based on water quality monitoring results
during construction.
' The sediment berm will be constructed immediately prior to excavation and will not be in-place during
the pre -construction phase.
Page 7 of 8

Proposed Duke Energy Monitoring Plan
References
Bosch, J. and R. U. Meckenstock. 2012. Rates and potential mechanism of anaerobic
nitrate -dependent microbial pyrite oxidation. Biochemical Society Transactions
Volume 40, part 6.
DeNicola, D.M. and M.G. Stapleton. 2016. Using Macroin vertebrates to assess
ecological integrity of streams remediated for acid mine drainage. Restoration
Ecology 24:5, 656-667.
Gray, N.F.and E.Delaney. 2008. Comparison of benthic macroin vertebrate indices for
the assessment of the impact of acid mine drainage on an Irish river below an
abandoned Cu -S mine. Environ Pollut. 155:1, 31-40.
He, F., W.Jiang, T.Tang and Q.Cai. 2015. Assessing impact of acid mine drainage on
benthic macroinvertebrates: can functional diversity metrics be used as
indicators?. Journal of Freshwater Ecology, 30:4, 513-524.
HDR. 2017. Geological and Geotechnical Subsurface Investigation, East Fork Hydroelectric
Project and Cedar Cliff Development (FERC No. 2698). Tuckasegee, North
Carolina, Report for Duke Energy of the Carolinas, LLC.
HDR. 2018a. Cedar Cliff Rock Spoil Evaluation. East Fork Hydroelectric Project
and Cedar Cliff Development (FERC No. 2698) Tuckasegee, North Carolina
HDR. 2018b. Cedar Cliff Hydroelectric Development Auxiliary Spillway Upgrade Project
Status Update and Permitting Process. Power Point Presentation for Agency
Briefing January 18, 2018
Nantahala Power and Light. 2000. FERC Relicensing First Stage Consultation Package.
East Fork Hydroelectric Project, FERC Project No. 2608 -NC
North Carolina Division of Water Quality. 2011. Basin -wide Assessment Report Little
Tennessee River Basin. Water Quality Section, Division of Water Quality, North
Carolina Department of Environment and Natural Resources. Raleigh, NC.
Pugh, C.E., L.R. Hossner, and J.B. Dixon. 1984. Oxidation rate of iron sulfides as affected by
surface area, morphology, oxygen concentration, and autotrophic bacteria. Soil
Science. 137:5, pp. 309-314.
Stumm, W. and J.J. Morgan. 1981. Aquatic Chemistry. John Wiley & Sons, Inc.,New York,
NY, 780p.
Svitok, M., M. Novikmec, P. Bitusik, B.Masa, J. Obona, M.Ocadlik 5 and E.Michalkova.
2014. Benthic Communities of Low -Order Streams Affected by Acid Mine
Drainages: A Case Study from Central Europe. Water 6,1312-1338.
Page 8 of 8