20030147 Ver 2_Substrate Characterization_20070511

Yadkin-Pee Dee River Hydroelectric Project
FERC No. 2206
Substrate Characterization of the Tillery and
Blewett Falls Hydroelectric Plants Tailwater
Areas
Water Resources Work Group
Issue No. 10 -Sediment Transport
PROGRESS ENERGY
APRIL 2006
D 2006 Progress Energy

TABLE OF CONTENTS
Section Title
ACRONYM LIST .................................................................................................... AL-I
EXECUTIVE SUMMARY ..........................................................................................ES-I
SECTION I -INTRODUCTION .................................................................................... I - I
SECTION 2 -STUDY OBJECTIVES ............................................................................. 2-I
SECTION 3 -SITE DESCRIPTION ............................................................................... 3-I
No.
3.1 General Locale Description ........................................................................................... 3-1
3.1.1 Tillery Development .......................................................................................3-1
3.1.2 Blewett Falls Development ............................................................................. 3-3
3.2 Study Sites for Substrate Characterization ....................................................................3-4
3.2.1 Pee Dee River Tailwater Area below Tillery Hydroelectric Plant ................. 3-4
3.2.2 Pee Dee River Tailwater Area below Blewett Falls Hydroelectric Plant....... 3-4
SECTION 4 -METHODS ............................................................................................ 4-I
4.1 Aerial Photography Image Acquisition
4.2 Pre-Processing of Aerial Photography .
4.2.1 Image Scanning .........................................................
4.2.2 Georeferencing of Digital Photographic Mosaics.....
4.2.3 Orthorectification of Aerial Photography .................
4.3 Substrate Classification ...........................................................
4.3.1 Substrate Classification Scheme ...............................
4.3.2 Digitizing Substrate Categories ................................
4.4 Accuracy Assessment of Digitized Substrate Categories........
4.4.1 Ground-truthing Field Data Collection .....................
4.4.2 Error Matrices ...........................................................
4-1
4-1
4-2
4-2
4-2
4-3
4-3
4-4
4-7
4-7
4-8
SECTION 5 -RESULTS AND DISCUSSION .................................................................. 5-I
5.1 Accuracy Assessment of Substrate Classifications ....................................................... 5-1
5.1.1 Tillery Hydroelectric Plant Tailwater Area Study Site ................................... 5-1
5.1.2 Blewett Falls Hydroelectric Plant Tailwater Area Study Site ........................ 5-3
5.2 Tillery Hydroelectric Plant Tailwater Area Substrate Classification ............................ 5-3
5.3 Blewett Falls Hydroelectric Plant Tailwater Area Substrate Classification ................. 5-8
i

TABLE OF CONTENTS
(Continued)
Section
Title
No.
SECTION 6 - SITMMARY ........................................................................................... 6-1
SECTION 7 -REFERENCES ........................................................................................ 7-1
APPENDICES
APPENDIX A - MODIFIED WENTWORTH CLASSIFICATION FOR SUBSTRATE PARTICLE
SIZE UTILIZED DURING THE SUBSTRATE CHARACTERIZATION STUDY
OF THE TAILWATER AREAS OF THE TILLERY AND BLEWETT FALLS
HYDROELECTRIC PLANTS DURING 2005
ii

LIST OF FIGURES
Title Page No.
Figure 3-1 Yadkin-Pee Dee River Project (FERC No. 2206) location map ..........................3-2
Figure 3-2 Map of substrate characterization study sites (red outlined areas) below the
Tillery and Blewett Falls Hydroelectric Plants (1:24,000 scale USGS
topographical) ...................................................................................................... 3-5
Figure 4-1 Bedrock outcrops and areas of cobble/gravel detailed in high-resolution aerial
orthophotography and in GIS-derived polygons. The unclassified image on
the left provides an example of such areas from the Blewett Falls
Hydroelectric Plant study site. The image on the right shows the GIS-derived
delineations of the bedrock and cobble/gravel substrate categories polygons.
A discrete polygon area delineated as a substrate category was termed a
habitat area ........................................................................................................... 4-5
Figure 4-2 Excerpt from aerial photography at the Tillery Hydroelectric Plant tailwater
area study site illustrating the typical depth and water clarity conditions.
This example was representative of the majority of the study site ..................... . 4-5
Figure 4-3 Excerpt from aerial photography at the Blewett Falls Hydroelectric Plant
tailwater area study site illustrating the typical depth and water clarity
conditions. This example was representative of the majority of the study site. .4-6
Figure 4-4 An example of a turbid, deep run area in the Blewett Falls Hydroelectric
Plant tailwater study site where visual interpretation of substrate was
difficult. Training points (red dots) used for substrate classification of this
area and are shown overlayed onto the photographic image .............................. . 4-6
Figure 4-5 Area in the Blewett Falls Hydroelectric Plant tailrace without aerial mosaic
photography coverage. Training points (red dots) used for substrate
classification of this area are shown overlayed onto the image .......................... . 4-7
Figure 5-1 Tillery Hydroelectric Plant tailwater area study site showing mosaic of
orthophotographs and the study site boundary (red line) ................................... . 5-5
Figure 5-2 Substrate classification map of the Tillery Hydroelectric Planttailwater area
study site ............................................................................................................. . 5-6
Figure 5-3 Blewett Falls Hydroelectric Plant tailwater area study site showing mosaic of
orthophotographs and the study site boundary (red line) ................................... . 5-9
Figure 5-4 Substrate classification map of the Blewett Falls Hydroelectric Plant tailwater
area study site ..................................................................................................... 5-10
iii

LIST OF TABLES
Table
Title
No.
Table 4-1 Scanner settings used during photographic image scanning ............................ ... 4-2
Table 4-2 Substrate categories used in the characterization of study sites located in the
tailwater areas of the Tillery and Blewett Falls Hydroelectric Plants during
2005 ................................................................................................................... ... 4-3
Table 5-1 Error matrix and accuracy assessment of substrate categories in the Tillery
Hydroelectric Plant tailwater area study site .................................................... ... 5-2
Table 5-2 Error matrix and accuracy assessment of substrate categories in the Blewett
Falls Hydroelectric Plant tailwater area study site ........................................... ... 5-4
Table 5-3 Area estimates of substrate categories in the Tillery Hydroelectric Plant
tailwater area study site ..................................................................................... ... 5-7
Table 5-4 Area estimates of substrate/structure categories in the Blewett Falls
Hydroelectric Plant tailwater area study site .................................................... . 5-11
iv

Acronym List
Federal/State Agencies
Advisory Council on Historic Preservation (ACHP)
Federal Aviation Administration (FAA)
Federal Energy Regulatory Commission (FERC)
National Park Service (NPS)
National Marine Fisheries Service (NMFS)
National Oceanic and Atmospheric Administration (NOAA)
National Resource Conservation Service (MRCS) formerly known as Soil Conservation Service
National Weather Service (NWS)
North Carolina Department of Environment and Natural Resources (NCDENR)
North Carolina Environmental Management Commission (NCEMC)
North Carolina Department of Natural and Economic Resources, Division of Environmental
Management (NCDEM)
North Carolina Division of Parks and Recreation (NCDPR)
North Carolina Division of Water Resources (NCDWR)
North Carolina Division of Water Quality (NCDW~
North Carolina Natural Heritage Program (NCNHP)
North Carolina State Historic Preservation Officer (NCSHPO)
North Carolina Wildlife Resources Commission (NCWRC)
South Carolina Department of Natural Resources (SCDNR)
South Carolina Department of Health and Environmental Control (SCDHEC)
State Historic Preservation Office (SHPO)
U. S. Army Corps of Engineers (ACOE)
U. S. Department of Interior (DOI)
U.S. Environmental Protection Agency (USEPA)
U.S. Fish and Wildlife Service (USFWS)
U.S. Geological Survey (USGS)
U. S. Department of Agriculture (USDA)
U.S. Forest Service (LTSFS)
Other Entities
Alcoa Power Generating, Inc., Yadkin Division (APGI)
Appalachian State University (ASU)
Progress Energy (Progress)
University of North Carolina at Chapel Hill (UNCCH)
Facilities/Places
Yadkin -Pee Dee River Project (entire two-development projectinduding both powerhouses, dams
and impoundments)
Blewett Falls Development (when referring to dam, powerhouse and impoundment)
Blewett Falls Dam (when referring to the structure)
Blewett Falls Hydroelectric Plant (when referring to the powerhouse)
AL-1

List
Blewett Falls Lake (when referring to the impoundment)
Tillery Development (when referring to dam, powerhouse and impoundment)
Tillery Dam (when referring to the structure)
Tillery Hydroelectric Plant (when referring to the powerhouse)
Lake Tillery (when referring to the impoundment)
Documents
401 Water Quality Certification (401 WQC)
Draft Environmental Assessment (DEA)
Environmental Assessment (EA)
Environmental Impact Statement (EIS)
Final Environmental Assessment (FEA)
Initial Consultation Document (ICD)
Memorandum of Agreement (MOA)
National Wetland Inventory (NWI)
Notice of Intent (NOI)
Notice of Proposed Rulemaking (NOPR)
Preliminary Draft Environmental Assessment (PDEA)
Programmatic Agreement (PA)
Scoping Document (SD)
Shoreline Management Plan (SMP)
Laws/Regulations
Clean Water Act (CWA)
Code of Federal Regulations (CFR)
Electric Consumers Protection Act (ECPA)
Endangered Species Act (ESA)
Federal Power Act (FPA)
Fish and Wildlife Coordination Act (FWCA)
National Environmental Policy Act (NEPA)
National Historic Preservation Act (NHPA)
Terminology
Alternative Relicensing Process (ALP)
Cubic feet per second (cfs)
Degrees Celsius (C)
Degrees Fahrenheit (F)
Dissolved oxygen (DO)
Feet (ft)
Gallons per day (gpd)
Geographic Information Systems (GIS)
Gigawatt Hour (GWh)
Global Positioning System (GPS)
AL-2

List
Grams (g)
Horsepower (hp)
Kilogram (kg)
Kilowatts (kW)
Kilowatt-hours (kWh)
Mean Sea Level (msl)
Megawatt (MW)
Megawatt-hours (MWh)
Micrograms per liter (µg/L)
Milligrams per liter (mg/L)
Millimeter (mm)
Million gallons per day (mgd)
National Geodetic Vertical Datum (NGVD)
National Wetlands Inventory (NWI)
Non-governmental Organizations (NGOs)
Ounces (oz.)
Outstanding Remarkable Value (ORV)
Parts per billion (ppb)
Parts per million (ppm)
Pounds (lbs.)
Power Factor (p.£)
Probable Maximum Flood (PMF)
Project Inflow Design Flood (IDF)
Rare, Threatened, and Endangered Species (RTE)
Ready for Environmental Assessment (REA)
Resource Work Groups(RWG)
Revolutions per Minute (rpm)
Rights-of--way (ROW)
Stakeholders (federal and state resource agencies, NGOs, and other interested parties)
Volts (V)
AL-3

Executive Summary
Progress Energy is currently relicensing the Tillery and Blewett Falls developments (i.e., Yadkin-
Pee Dee River Hydroelectric Project No. 2206) with the Federal Regulatory Commission (FERC).
As part of the relicensing process, Progress Energy established Resource Work Groups (RWG)
during May 2003 to identify environmental issues associated with Project operations and develop
study plans, if necessary, specific to Project lands and associated lakes and tailwater area. The
Water Resources Work Group identified the need to document existing gravel and cobble bars in the
immediate tailwater areas below the Tillery and Blewett Falls Hydroelectric Plants and determine
the stability or persistence of these substrate types through time (i.e., Progress Energy [2004] and
Water RWG Issue No. 10, "Sediment Transport").
The study objective was to assess the distribution and extent of the existing substrate types,
including gravel and cobble bars, in the immediate tailwater areas below the Tillery and Blewett
Falls Hydroelectric Plants. Each substrate type was classified and delineated from high-resolution
aerial photography using Geographical Information System (GIS) software. The study also
evaluated the spatial patterns of classified substrate types within each power plant tailwater study
area.
Low altitude, high-resolution aerial photography was acquired with a fixed wing aircraft at each
study area during February 2005. The aerial photographs were scanned, orthorectified with
established ground elevation control data, and then compositedto produce one georeferenced mosaic
digital image for each study area. Substrates were delineated and classified in a digital environment
using GIS software. Aground-truthing accuracy assessment was conducted by comparing known
field substrate classifications from GPS established transects to the classified GIS electronic image.
A total of 617 substrate type areas encompassing 112 acres were mapped within the Tillery
Hydroelectric Planttailwater study area Bedrock outcrops were the mostfrequently mapped habitat
category numerically (554 areas) but only comprised 7 percent of total mapped acreage. Smaller
particle substrates or a mixture of smaller and larger particle substrates were the most prevalent on
an areabasis within the Tillery Hydroelectric Planttailwater study area. Cobble/graveUboulder and
cobble/graveUbedrockwerethe most prevalent substrate categories in the Tillery Hydroelectric Plant
tailwater area and comprised 28 and 27 acres, respectively. These two substrate categories
represented about half ofthe total mapped area. Riprap, cobble/graveUsand, and cobble/gravel with
emergent andterrestrial vegetation were minor categories combiningfor approximately 1 percent of
the total area.
A total of 644 substrate or structure type areas encompassing 184 acres were mapped within the
Blewett Falls Hydroelectric Plant tailwater study area. Bedrock outcrops were the mostfrequently
mapped habitat category numerically, consisting of 552 mapped areas. Bedrock/boulder with
emergent and terrestrial vegetation was the second numerically dominant substrate category (22
areas) but only comprised 4 percent of the total mapped area Boulder/cobble, cobble/gravel,
cobble/graveUboulder, and cobble/graveUsilt were the neat numerically dominant substrate
categories with 12 to 14 mapped areas, respectively. Boulder/cobble and bedrockoutcrops were the
mostprevalent substrate categories on an areabasis; comprising 38 and 41 acres, respectively, in the
Blewett Falls Hydroelectric Plant study location. On a percentage basis, bedrock outcrops and
boulder/cobble represented 43 percent of the total mapped area. Smaller particle substrates or
ES-1

Executive
mixtures of small and large particle substrates (e.g., cobble/gravel, bedrock/boulder/cobble, and
cobble/graveUboulder)rnllectively comprised 63 percent ofthe total study area The island category
(i.e., Big Island) accounted for 6 percent ofthe total mapped area in the Blewett Falls Hydroelectric
Plant study site.
Based on these results, cobble and gravel substrates or mixtures of these substrates with larger
substrates such as boulder and bedrock were prevalent in the immediate tailwater areas below the
Tillery and Blewett Falls Hydroelectric Plants. There was no evidence of substantial scouring or
armoring of the river channel given that areas with smaller substrate particles or mixtures of smaller
and coarser particle substrates were frequently mapped within each power plant tailwater area
ES-2

Section 1 -Introduction
Progress Energy is currently relicensing the Tillery and Blewett Falls developments (i.e., Yadkin-
Pee Dee River Hydroelectric Project No. 2206) with the FERC. As part ofthe relicensing process,
Progress Energy established RWG during May 2003 to identify environmental issues associated
with Project operations and develop study plans, if necessary, specific to Project lands and
associated lakes and tailwater areas. The Water RWG identified the need for a sedimenttransport
study to be conducted in the Projecttailwater areas (i.e., Progress Energy [2004] and Water RWG
Issue No. 10 "Sediment Transport"). Gravel recruitment downstream of each hydroelectric dam was
mentioned by RWG members as a potential issue. Based on visual observations, Progress Energy
contended that the amount of gravel/cobble (i.e., smaller particle substrates) presentbelow each dam
did not appear to be a limiting factor for inhabitation or spawning by fish or other aquatic organisms,
such as mussels or other macroinvertebrate species. However, after further discussions at Water
RWG meetings, Progress Energy agreed to conduct an empirical-based study to determine the
distribution and extent of all substrate types, including gravel and cobble, in the immediate tailwater
areas below the Tillery and Blewett Falls Hydroelectric Plants.
Characterizing substrates is importantfor three reasons. First, many fish and other aquatic species
require specific substrate types for spawning and inhabitation. Gravel and cobble, in particular,
provide a substantial amount of surface areafor deposition of eggs into redds or adheringto exposed
surfaces. These substrate types also provide the micro-habitat conditions and interstitial spaces
needed by many fish and aquatic invertebrate species. Second, the substrate composition determines
the roughness of stream channel, and roughness has a large influence on the channel hydraulics
(water depth, width, and current velocity) of stream habitat. Third, substrate composition, including
the degree of embeddedness, can indicate localized and broader watershed anthropogenic influences
on stream habitat quality. For example, small particle composition may reflect land surface
disturbances such as forestry and agriculture practices (Bain and Stevenson 1999).
1-1

Section 2 -Study Objectives
The purpose of this study was to document existing substrate types in the immediate tailwater areas
below the Tillery and Blewett Falls Hydroelectric Plants. The immediate tailwater area
encompassed the entire river channel from the hydroelectric power plant and dam extending at least
one linear mile downstream.
The specific objectives of this study were to: (1) provide a detailed characterization of the aquatic
habitat substrate types (i.e., substrate types and frequency of occurrence) below the Tillery and
Blewett Falls Hydroelectric Plants, with emphasis on smaller substrate types such as sand, gravel,
and cobble; (2) quantify each substrate type on an area basis; and (3) provide a documented,
repeatable GIS-based high-resolution aerial orthophotography methodology for substrate mapping
and classification.
2-1

Section 3 -Site Description
3.1 General Locale Description
The Yadkin-Pee Dee River Project is located in south-central North Carolina (Figure 3-1). The
Yadkin-Pee River basin is the second largest in North Carolina covering 7,213 mil as measured at
the North Carolina-South Carolina state line (North Carolina Division of Water Quality [NCDWQ]
2002). The Yadkin-Pee Dee River originates near the town of Blowing Rock and flows
northeasterly for approximately 100 miles from the Blue Ridge Mountains into the Piedmont
physiographical region. As the riverturns southeast, it enters an area in Central North Carolinathat
has experienced considerable urban growth. This growing urban area is known as the Piedmont
Crescent (Appalachian State University [ASU] 1999). Just to the south of the Piedmont Crescent,
the region enters an area known as the Uwharrie Lakes Region. This region is named for the chain
of six reservoirs located along this reach of the Yadkin-Pee Dee River, two of which are Lake
Tillery and Blewett Falls Lake. It is in this region that the Uwharrie River joins the Yadkin River at
the upper end of Lake Tillery to form the Pee Dee River.
The flow of the Yadkin-Pee Dee River is regulated by a federal flood control development and six
hydroelectric developments on the main stem of the river (Figure 3-1). The first development,
traveling downstream fromthe headwaters, is the W. Scott Kerr Dam, afederal flood control project.
The next four developments make up the Yadkin Project. These four hydroelectric developments,
High Rock, Tuckertown, Narrows, and Falls (FERC No. 219'x, are owned and operated by Alcoa
Power Generating, Inc., Yadkin Division (APGI), and are located along a38-mile stretch ofthe river
(River Mile [RM] 272 to 234). High Rock Reservoir is operated as a storage reservoir and serves as
the principal storage and water regulation facility for the lower Yadkin-Pee Dee River (APGI 2002).
The next two hydroelectric developments on the river, located at RMs 218 and 188 are the Tillery
and Blewett Falls developments, which constitute Progress Energy's Yadkin-Pee Dee River Project
(FERC No. 2206). The primary purpose of this Project is to provide peaking and load-following
generation. Its ability to provide such benefits and meet other flow-related needs is largely
dependent on the schedule of flows being released from upstream reservoirs. Currently, an
agreement between APGI and Progress Energy governs the release of waters from APGI
developments to Progress Energy developments. Additional Project-related information is discussed
in the Initial Consultation Document for the Project (Progress Energy 2003).
3.1.1 Tillery Development
Construction of the Tillery Development began in 1926 and was completed in 1928. Lake Tillery
extends approximately 15 miles upstream to the tailrace of the Falls Project powerhouse. At the
normal maximum operating elevation of 277.3 ft', Lake Tillery has an average depth of 23.6 ft and a
maximum depth of approximately 71 ft atthe dam. The depth ofthe intake extends from 39 to 62 ft
(12 to 19 m) below the lake surface at normal maximum pool elevation. The lake has a reservoir
' NAVD 88 datum. Unless otherwise noted, all data are NAVD 88 datum. The NAVD 88 datum is 0.9 ft lower than
the 1929 NGVD datum (NAD 291.
3-1

Section 3 Site
,_T
';;, ,
-r---. r _
a'-----~.r,~.,,_;E--..____ - _- Virginia ~_ _ ~ -.----
Yadkin River
:. W. Kerr Scott ~
Dam
~'
_ __ Salem Greenstwro 8uni^~ ,n
.- ,
:'\ Hi~h~~ > Durham ..
~'' 'y n
i
~ Uwharrie River T
rT 9 _- ,
_ - ~' High Rock Dam ~ ';,.., ,, - ~ '~~ ~..
Q Tuckertown Dam ,~ l
c7
Narrows Dam , ; ;a. `Falls Dam
Ctiarlotte t?n
- ~:_~_~--- ~~~ ,~., ~,, Tillery Dam T
L .~~ '.1.,1''~
-___...~ ~~ ~ ~ F,;~'- ` Blewett Falls Dam
~~ .. ~
Fayettewllc
~:
-. -- ! f'.~-. ~~
Rocky River
North Ca_rolina_
- - ------~- --~,r:.~
South Carolina
~ ~`
n,.~,3:. ~'~ ~ ~
~~-, ~ Pee Dee River
~~
-- //
,_ ~IColumbia
~~~~ ,~~~ Sumter
_ ~ ~' ~~'
~~ ~
~ ~ \ ~~ Myrtb,/
~:•Augusta 4 ~ Beach
1L,non r
i
Winyah Bay ;
~. Atlantic
! ~ Ocean
,\
- ~ \~\C~\ 't, ,
~ .,. 1
---•--"---•.L ------- ----- ~~~"~`~ Rivers -Froevray System ~.
~~~,ee.ip„~ _ -Rivers abng tho Yadkin-Poe Dee UrbanAreas
=~w,r-'"s`oWnc« ..: `,^_ I ~ Lakes I~
_ ~ '--q ~ Lakes along the Yadkin-Pee Dee
y
~` '"~ smes~i ow,rr,mai,aa 1 u~ ~
\l -' Regional Location Map v'~~: n~,oreYaei.irm )MStw me r w~c«neki• ~.-
Figure 3-1 1 adkin-Pee Dee Ri~rei• Project (FERC No. 2206} location maia.
3-2

Section 3 Site Descripti
surface area of approximately 5,697 acres and a total volume of 5.85 x 109 ft3. The average
retention time for the lake is approximately 8.3 days at normal maximum operating pool elevation
based on 1983 to 2000 inflow data Lake Tillery is dendritic in shape with several large tributary
arms (e.g., Mountain Creek, Jacobs Creek, and Cedar Creek complex). The lake has a shoreline
length of approximately 118 miles with 55 percent of the shoreline in residential or commercial
development (Progress Energy 2003). The remaining shoreline areas are forested, pasture, or
agricultural.
The inflows into the Lake Tillery consist primarily of the outflow from the APGPs Falls
Development and the inflow from the Uwharrie River. The Uwharrie River contributes about 8
percent of total inflow volume, based on watershed area size. The current Progress Energy license
allows for drawdowns at Lake Tillery of up to 22 ft below full pond. However, over the past several
years, Progress Energy has voluntarily made its best efforts to operate Lake Tillery within a 4-ft
range under normal circumstances and much ofthe time operating within a 2-ft range except during
FERC-required inspection and maintenance periods (12-ft drawdown) (Progress Energy 2003).
Outflows from the Lake Tillery flow into Blewett Falls Lake after passing through a 17-mile reach
of the Pee Dee River. Under normal circumstances, it takes approximately eight hours for releases
from the Tillery Hydroelectric Plant to be observed at the Blewett Falls Hydroelectric Plant
powerhouse (Progress Energy 2003).
3.1.2 Blewett Falls Development
The Blewett Falls Hydroelectric Plant dam and powerhouse are located 17 miles upstream of the
North Carolina-South Carolina state border. The construction of this facility began in 1905 and
brought into service in 1912. The normal maximum pool elevation is 177.2 ft' and the reservoir
extends approximately 11 miles upstream. The average lake depth is 10.8 ft and the maximum depth
is approximately 35 ft. The depth of the intake extends from 20 to 33 ft (6 to 10 m) below the lake
surface atnormal maximum pool elevation. The surface area ofthe lake at normal operating level is
approximately 2,866 acres with atotal volume of 1.35 x 109 ft3. The mean hydraulic retention time
is 1.9 days at normal maximum operating pool elevation based on 1983 to 2000 inflow data.
Blewett Falls Lake is relatively narrow with few large tributary arms. The lake has a shoreline
length of approximately 46 miles including mainland and island shoreline areas. The Blewett Falls
Lake shoreline is relatively undeveloped with surrounding lands mainly forested (Progress Energy
2003). Some limited residential development exists alongthe lake shoreline, primarily in the lower
lake.
The Blewett Falls Hydroelectric Plant is operated in coordination with the upstream Tillery
Hydroelectric Plant. The hydraulic capacity of Blewett Falls Lake is significantly less than Lake
Tillery; therefore, Blewett Falls Lake must anticipate flows from Lake Tillery generation and begin
generating in advance of flows reaching the lake. The normal operation of the Blewett Falls
Hydroelectric Plant results in a daily drawdown of approximately 2 to 3 ft below the normal
maximum operating level. This drawdown provides storage capacity needed to regulate flows from
Lake Tillery. The Blewett Falls Hydroelectric Plant generating units normally begin operation atthe
same time that the Tillery Hydroelectric Plant begins generation. Generation at Blewett Falls
Hydroelectric Plant is usually stopped by midnightto allow the reservoir to refill. This operation is
consistent year round and varies only with seasonal availability of water (Progress Energy 2003).
3-3

Section 3 Site Descripti
Periodic maintenance can require the lowering ofthe reservoir levels atboth power plants. At Lake
Tillery, drawdowns are typically associated with the maintenance ofthe steel spillway gates, repairs
to the trashrack system, or repairs to the upstream slope of the earthen embankment. Drawdowns
required at the Blewett Falls Lake are similar to the Lake Tillery except that the most frequent
maintenance requirement is to service the 4-ft-high, wooden flashboards atop the spillway. During
periods of high flow, such as those encountered with the passing of tropical storm systems during
September of 2004, damage or loss ofthese flashboards may occur and repairs require the lake to be
drawn down about 4 to 5 ft over a period of time (Progress Energy 2003).
3.2 Study Sites for Substrate Characterization
3.2.1 Pee Dee River Tailwater Area below Tillery Hydroelectric Plant
The study areafor the Pee Dee River below the Tillery Hydroelectric Plant (Reach 1) was a 1-mile
(1.61 km) segment of the tailwater area immediately below the power plant (Figure 3-2). This
tailwater area segment can be qualitatively characterized as consisting of shallow (1 to 2 m depth)
runs and shoals with substrate of varying sizes, including bedrock outcroppings. Channel widths
ranged from 527 to 1,271 ft with an average width of 903 ft within the study area.
3.2.2 Pee Dee River Tailwater Area below Blewett Falls Hydroelectric Plant
The study area for the Pee Dee River below the Blewett Falls Hydroelectric Plant (Reach 2) was a
1.5-mile segment (2.41 km) of the tailwater area immediately below the power plant with the lower
study boundary located atthe Cartledge Creek confluence (Figure 3-2). This tailwater area segment
can be qualitatively characterized as consisting of shallow to moderately deep runs and shoal habitat
(1 to 3 m depth) with an island (i.e., Big Island) and side channel complex. There is awide range of
substrate sizes present in the study area and bedrock outcroppings are quite common. Channel
widths ranged from 661 to 1,538 ft with an average width of 976 ft within the main river channel in
the study area. The powerhouse tailrace channel ranged in width from 248 to 600 ft with an average
width of 442 ft.
3-4

Section
Site
3-~
Figure 3-2 Map of substrate characterization study sites (red outlinecl areas) below the
Tillery and Blewett Falls Hydroelectric Plants (1:24,000 scale USGS
topographical)

Section 4 -Methods
The field data collection phase of the study was conducted during January through March 2005.
Field data collected included aerial photography of each study site followed with ground-truthing of
substrate types using a transect-grid method. The study was divided into four stages: (1) image
acquisition which consisted of low altitude (300-500 ft) aerial photography of the study sites under
low flow (i.e., no power plant operations and minimum dam spillage) to acquire GPS-based images
for digitizing; (2) preprocessing of the photography imagery which included scanning the aerial
photographs into aphoto-mosaic composite of the study site, orthorectifyingthe imagery to known
ground elevation control points, and georeferencing the images to a common latitude-longitude
coordinate system; (3) substrate classification which consisted of GIS digitization of the river
bottom substrates from the aerial photography imagery and ground-truthing field work using a
scientifically accepted substrate classification system; and (4) accuracy assessmentwhich consisted
of developing a quantifiable estimate of the accuracy of the GIS-derived substrate classification
maps. The accuracy assessment of the substrate mapping classification is presented in Section 5.1
prior to discussion of the substrate characterization results.
4.1 Aerial Photography Image Acquisition
KUCERA International was contracted by Progress Energy to acquire the low altitude, high-
resolution aerial photography. The aerial photography was conducted by KUCERA using afixed-
wingaircraft onFebruary 19, 2005, at the Blewett Falls Hydroelectric Planttailwater study site and
on February 26, 2005, at the Tillery Hydroelectric Planttailwater study site. The target scale for the
hard copy natural color photographs (9 x 9 inches) was 1 inch on the map = 200 ft on the ground.
Scanned resolution of the aerial photography imagery resulted in 0.1 ft pixel resolution.
The aerial photography of each study site was performed under low flow conditions when the power
plants were not operating and the only flows in the study site were dam spillage, power plant wicket
gate leakage, and tributary inflows. Flows were estimatedto be approximately 100 cfs at the Tillery
Hydroelectric Plant and approximately 300 cfs at the Blewett Falls Hydroelectric Plant on the dates
of aerial photography image acquisition. Both hydroelectric plants were off-line (i.e., no power
generation) for 12-24 hours prior to image acquisition to allow water levels to decrease to low flow
conditions and to help increase water clarity. Weather conditions, including cloud cover, wind
speeds, and the time of day (i. e., sun angle and light reflection) were also taken into consideration
during the image acquisition.
4.2 Pre-Processing of Aerial Photography
KUCERA performed all preprocessing steps for the aerial photography. The preprocessing steps
included scanning the natural color aerial photographs, orthorectifying the images to known
elevation ground controls, georeferencingthe images to an established latitude-longitude coordinate
grid system, and creating the final digital photograph mosaics. Preprocessing of the data included
established quality control guidelines to ensure the final photographs were accurate in ground
elevation and geographical location.
4-1

Section 4 Methods
4.2.1 Image Scanning
The hard copy negatives were true color scanned into TIFF digital format using a Z/I PhotoScan
2002 first order (two micron precision + accuracy) photogrammetric scanner. The images were
scanned at 24 bits per pixel with a compression ratio of -1.12. The scanner settings are shown in
Table 4-1.
Table 4-1 Scanner settings used during photographic image scanning.
Scanner: Z/I PhotoScan 2002
Type: TIFF
Mode: True color
Width: 16862
Height: 16861
Bits Per Pixel: 24
Colors: 16777216
DPIX: 1814 (14u)
DPIY: 1814 (14u)
Width (Inches): 9.2955
Height (Inches): 9.2949
Compression Ratio: -1.12
4.2.2 Georeferencing of Digital Photographic Mosaics
The photograph images were georeferenced after the hard copy aerial photography imagery was
scanned into a digital format to create a photographic mosaic of the entire study site.
Georeferencing is the process oftransforming the images into a common reference datum and makes
the images useable in a GIS format (Jensen 199. The reference datum used in this study was
NAD83 with feet as the distance unit. Atthe time of aerial photograph exposures, the six orientation
parameters (i.e., Latitude, Longitude, Altitude, Omega, Phi, and Kappa rotations) were recorded
from the on-board GPS/Inertial Measurement Unit (IMU) system. The GPS/IMU data was further
refined through a differential correction with input from a GPS ground base station with the final
orientation parameters output translated to the North Carolina State Plane coordinate system. To
further refine the orientation parameters, the scanned aerial imagery was combined with the post-
processed GPS/IMU data and an automated Aerial Triangulation adjustment was performed to
generate tie-in points for the entire mosaic block of photographs. The addition of the tie-in points
between adjaoent photography exposures strengthened the geometry of the entire photo-mosaic
block to maximize the aocuraoy of the triangulated orientation parameters.
4.2.3 Orthorectification of Aerial Photography
The topographical variations in the earth's surface (i.e., elevation and slope aspect) and the tilt ofthe
on-board camera can affect the distance and how the features are displayed on the aerial photograph
images. Distortion of photographs becomes more of an issue with more topographically diverse
landscape features. As a result, actual linear distances may not be uniformly represented on the
photograph. Orthorectification isthe systematic process used to remove these sources of distortion
and to equilibrate photographic units to actual linear distances. Once an aerial photo has been
orthorectified, it is commonly referred to as an orthophotograph.
4-2

Section 4 Methods
Digital Elevation Models (DEMs) were used forthe orthorectification of photographs in this study.
Resampling of the DEM image involved warping the photograph image so that distance and area
were uniform in relationship to the actual, "real world" linear measurements. In short, with the
resampled DEM photo, an inch on the photographic image was corrected to measure the same
distance, particularly on steep terrain, as it really measures with on-the-ground survey methods.
After the photograph images were orthorectified, the resulting digital data were used with vector and
raster data ofthe same coordinate plane system for accurate assessments of substrate composition.
4.3 Substrate Classification
4.3.1 Substrate Classification Scheme
The aquatic habitat substrate classification was based on the modified Wentworth scale for
classifying substrate particle (originally Cummins 1962 as cited in McMahon et al. 1996)
(Appendix A). However, modifications to the substrate classification were necessary for this study
based on field observations of actual substrate composition (i.e., mixtures of substrate classes) and
the resolution of the aerial photographs (Table 4-2). First, and most importantly, the heterogeneity
of substrate types required multiple combinations of these classes, particularly with regard to
mixtures of gravel and cobble with other substrate types. There were numerous instances where
cobble, gravel, or a combination ofthese two substrate types were overlaid orformed pockets on top
of bedrock. Another example was substrate consisting of a mixture of gravel and cobble
interspersed among small to large boulders and bedrock. Second, there was difficulty in
distinguishing differences between some of the smaller substrate classes such as pebble (16 to
63 mm) and gravel (2 to 16 mm) from the aerial orthophotography. In this case, both classes were
lumped together and classified as gravel. Finally, the presence of exposed bedrock and structure
features specific to a study site (i.e., Big Island below Blewett Falls Hydroelectric Plant and dam
apron) warranted stand-alone categories.
Table 4-2 Substrate categories used in the characterization of study sites located in the
tailwater areas of the Tillery and Blewett Falls Hydroelectric Plants during
2005.
Number Substrate Category Applicable Study Site
1 Cobble/gravel Both study sites
2 Cobble/graveUsand Both study sites
3 Cobble/graveUsilt Blewett Falls Hydroelectric Plant study site only
4 Cobble/graveUboulder Both study sites
5 Cobble/graveUbedrock/boulder Tillery Hydroelectric Plant study site only
6 Cobble/graveUbedrock Tillery Hydroelectric Plant study site only
7 Cobble/gravel with emergent and terrestrial vegetation Both study sites
8 Bedrock/gravel Tillery Hydroelectric Plant study site only
9 Bedrock/boulder/cobble Blewett Falls Hydroelectric Plant study site only
10 Boulder/cobble Blewett Falls Hydroelectric Plant study site only
11 Bedrock/boulder Both study sites
12 Bedrock/boulder with emergent and terrestrial Blewett Falls Hydroelectric Plant study site only
vegetation
13 Bedrock Both study sites
14 Riprap (introduced shoreline stabilization material) Both study sites
15 Island Blewett Falls Hydroelectric Plant study site only
16 Blewett Falls Hydroelectric Plant dam apron area Blewett Falls Hydroelectric Plant study site only
4-3

Section 4 Methods
Sixteen substrate or structure categories were defined for both the Tillery and Blewett Falls
Hydroelectric Plant study sites (Table 4-2). Seven of the 16 substrate categories were common to
both study sites while the remaining nine categories were specific to either study site. For example,
there was a combination of bedrock gravel present at the Tillery Hydroelectric Plant study site, but
this substrate category was not observed at the Blewett Falls Hydroelectric Plant study site. The
mixed categories were visually estimated to be equal mixtures of each substrate class arthe leading
substrate class inthe category was considered predominantwith the other substrate class comprising
smaller percentages of a particular category. All substrate categories were based on field visual
observations during the ground-truthing and accuracy assessment phases of the study. No
quantitative assessments were made with standardized sieves to determine actual fractions of each
substrate class within a particular mixed category.
4.3.2 Digitizing Substrate Categories
The most common method to delineate substrate categories from digital photography is to visually
categorize the substrate on a computer screen and then manually digitize areas of similar
composition using GIS software. This method is often called "heads-up"digitizing, and it is widely
used for selective capture of digital data (Longley et al. 2001). "Heads-up" digitizing was used for
this study and areas of similar substrate composition were determined by the color, shape, texture,
and patterns from the digital photograph. These areas were outlined digitally using GIS software to
create a continuous coverage of polygons across the study area. Each polygon was assigned a
substrate category after the digitizing was completed, and the assigned category was verified with
the ground-truthing data. Each polygon area was termed a habitat area for the identified substrate
category and these habitat areas varied in size and shape.
Many polygon substrate classifications were obvious due to their nature, such as exposed bedrock,
and cobble/gravel bars (Figure 4-1). The shallow water conditions (usually less than 1 m depth) and
high-resolution photography usually provided suitable conditions for visual identification of
substrate types (Figures 4-2 and 4-3). In some instances (i.e., deep run areas with depths greater
than one meter, very turbid areas, and a few areas outside of the mosaic photography coverage),
substrate classification was achieved by overlaying substrate classification datafrom GPS transect
training points collected during the ground-truthing phase of the study (Figures 4-4 and 4-5). To
eliminate any potential bias, these training points were excluded from the accuracy assessment.
4-4

Section 4
I~lethods
f
t
:t
~1
1~
ti
.'
Figure 4-1 Bedrock outcrops and areas of cobblelgravel detailed in high-resolution
aerial orthophotography and in GIS-derived polygons. The unclassified
image on the left provides an example of such areas from the Ble~vett Falls
Hydroelectric Plant study site. The image on the right shows the GIS-
derived delineations of the bedrock and cobble/gravel substrate categories
polygons. A discrete polygon area delineated as a substrate category was
termed a habit<~t area.
Figure 4-2 Excerpt from aerial photography at the Tillery Hydroelectric Plant tailwater
area study site illustrating the typical depth and water clarity conditions.
This example was representative of the majority of the study site.
4-5

Section 4 Methods
-- - ~ r.
- ~ ~
•
f.
~`
r i / N,
r`4 p ~ 1
rr
+-.
<-,J - - ~i i- ~
.- •
h' - . -.
>,
r
~:~JI
~.i1I
-1
Y~
7
Yr"a'd~
Figure 4-3 Excerpt from aerial photography at the Blewett Falls Hydroelech•ic Plant
tailwater area study site illush•ating the hTpical depth and water clarity
conditions. This example was representative of the majority of the study site.
~~`! 1~~ -
# k A ~~ F ~-7r~~y~.y..
a `'
~"~ ~ ~ ~ ~
4
~~ It. u i 1 _ i•'
o ~ 'v . :~~ u~ ~y`j4p ' `
.4. , ~g w
" 4'r tp .vet ~' ~ r
• ~
b
~ ~ ~/ a i
• _ t°. r - ! ` tn. ~ `~y`~}~'~ raj ."~ ~:.rtit~
.Its ~~ ~'~ ..~-- ~ ~ psi .F" 4
L. ~I~ ~;
Figure 4-4 An example of a turbid, deep run area in the Blewett Falls Hydroelectric
Plant taitr~~ater study site ~•here visual interpretation of substrate was
difficult. Training points (red dots) used for substrate classification of this
area and are shown overlayed onto the photographic image.
4-6

Section 4 Methods
~~- R 1 ~.
x ~ -~ L_ 'a
•_t•1'.:
• rte: ~{'
}

t

~ ~

•
*

V~. •
~
Figure 4-5 Area in the Blewett Falls Hydroelectric Plant tailrace without aerial mosaic
photography Coverage. Training points (red dots) used for substrate
classification of this area are shown overlayed onto the image.
4.4 Accuracy Assessment of Digitized Substrate Categories
4.1.1 Ground-truthing Field Data Collection
Grotuld reference paints obtained during the ground-tnitlung field data collection were used to
provide an estimate ofthe accuracy of the "heads-up" digitizing method of substrate classification.
Transects were established across the river chatmel from bank to bar~° at each study site. These
transects were located approximately every 500 ft begimung below each hydroelectric datn and
extending downstream forthe entire length of the study site. Ground reference paints Gvere collected
at approximately 50 ft inter4~als along each established tr<~nsect (Figures 4-4 and 4-5).
Transect and ground reference point data were acquired using a Trimble GeoXT GPS unit tenth sub-
meter accuracy (i.e., 1 m or approximately 3.3 ft radius arotuicl the mapped GPS point). A minimum
of 30 GPS latitude-longihule readings were recorded at one second inten~als for each ground
reference point along a transect. The primary substrate type (i.e., > 50 percent of substrate
composition) was deternuned visually gild physically in an area approximately one meter (~. ~ ft)
around the GPS sun~ey rod wlule collecting the latitude aild longitude positions. The second most
abundant substrate t5g3e in the sun~eyed groluid reference point was identified as the secondary
substrate type. Any other minor substrates present were also noted.
4-7

Section 4 Methods
4.4.2 Error Matrices
An error matrix for the digitized thematic maps for each study site was generated using 494 ground
reference points for both study sites and 8 substrate assessment classes per study site. An error
matrix is a square array of numbers set out in rows and columns that express the number of sample
units (pixels, clusters, or polygons) assigned to aparticular category in one classification relative to
the number of sample units assigned to a particular category in another classification (Congalton
1991; Fitzgerald and Lees 1994; Khorram et al. 1999).
An error matrix is an effective way to represent map accuracy in that the individual accuracies for
each category (class) are described along with both omission and commission errors. In addition to
these accuracy measures, the overall accuracy (does not account for data source errors), the
producer's accuracy and the user's accuracy were produced for the accuracy assessment. The
producer's accuracy is a referenced-based accuracy that is computed by looking at the predictions
produced for a class and determiningthe percentage of correct predictions. The user's accuracy is a
map-based accuracy that is computed by looking atthe reference datafor a class and determining the
percentage of correction predictions for the samples. The accuracy assessment produced a level of
confidence in the "heads-up" digitizing approach ofthe aerial photographs under the environmental
conditions present at the time that the aerial photographs were taken at each study site.
4-8

Section 5 -Results and Discussion
5.1 Accuracy Assessment of Substrate Classifications
Water clarity conditions duringthe aerial photography were somewhatturbid due to inflow resulting
from precipitation events in the river basin prior to the study (Figures 4-2 and 4-3). Turbidity ranged
from 7.2 to 38 NTU at the Tillery Hydroelectric Plant study site and from 21 to 38 NTU at the
Blewett Falls Hydroelectric Plant study site on the dates thatthe aerial photographs were taken. For
reference, the North Carolina water quality standard for turbidity is 50 NTU for streams (NCDWQ
2004). A large portion of the substrate was either exposed or submerged in very shallow water (1 to
2 ft) on the dates that the aerial photographs were taken so visual classification was achievable even
with the reduced water clarity (Figure 4-1). As mentioned under Section 4.3.2, the areas of
uncertainty regarding substrate classification were areas that were either too deep and/or too turbid
to correctly categorize the substrate. The ground-truthing approach using transects and ground
reference points was used to validate the "heads-up" digitizing approach and also overcome any
difficulties in areas where this approach was problematic due to depth or reduced water clarity.
5.1.1 Tillery Hydroelectric Plant Tailwater Area Study Site
The classification accuracy assessmentresults and associated error matrix for substrate categories at
the Tillery Hydroelectric Plant tailwater area study site are shown in Table 5-1. Two substrate
categories cobble/gravel with emergent and terrestrial vegetation and riprap were not included
in the accuracy assessment. These substrate categories were infrequently mapped and consisted of
only a few small areas within the study site (approximately 0.5 acre). These substrate categories
were not present at any of the ground truthing transects.
The overall substrate classification accuracy, which includedthe user's and producer's accuracy, was
95 percent for the Tillery Hydroelectric Plant tailwater area study site (Table 5.1). The user's
accuracy, defined as the proportion of each substrate category which was correctly identified by the
person performing the "heads-up" digitizing, ranged from 86 percent (bedrock category) to
100 percent (cobble/gravel, cobble/graveUbedrock/boulder,rnbble/graveUsand, and bedrock/boulder
categories). In other words, approximately 86 percent of the area classified as bedrock using the
"heads-up"digitizing method was actually bedrock based on the ground-truthing data. It should be
noted that only seven ground reference points were observed for this class and six of these points
were correctly classified. The bedrock in the study area was more difficult to delineate from the
aerial photographs than bedrock at the Blewett Falls Hydroelectric Plant tailwater area study site
because ofthe: (1) lack ofboulder outcrop exposure in the Tillery Hydroelectric Planttailwater area
study site; (2) striated, fractured pattern of bedrock in the Tillery Hydroelectric Plant study site
versus the clumped pattern of bedrock in the Blewett Falls Hydroelectric Plant study site; and
(3) fractured nature of many bedrock outcrops in both study sites which often transitioned into large
boulders. The majority of substrate categories (i.e., seven of the eight categories) in the Tillery
Hydroelectric Plant tailwater area study site exceeded 90 percent for user's accuracy.
5-1

Section 5 Results and Discussions
Table 5-1 Error matrix and accurac y assessment of substrate categories in the Tille ry Hydroelectric Plant tailwater area study site.
Ground-frothing Substrate Category

Digitized Map Substrate
Cobble / Cobble /
Cobble /
Cobble /
Number of
User's

Category Cobble /
Gravel/ Gravel /
Gravel/
Gravel/
Bedrock Bedrock / Bedrock /
Pixels
Accuracy
Gravel Bedrock/ Boulder Gravel
Bedrock Boulder Sand
Boulder
Cobble/Gravel 31 0 0 0 0 0 0 0 31 100.0%
Cobble/GraveUBedrock 0 51 0 3 0 0 0 0 54 94.4%
Cobble /Gravel /Bedrock /

Boulder 0 0 24 0 0 0 0 0 24 100.0%
Cobble/GraveUBoulder 0 1 0 37 0 0 3 0 41 90.2%
Cobble/GraveUSand 0 0 0 0 7 0 0 0 7 100.0%
Bedrock 0 0 0 1 0 6 0 0 7 85.7%
Bedrock/Boulder 0 0 0 0 0 0 11 0 11 100.0%
Bedrock/Gravel 1 0 0 0 0 0 1 19 21 90.5%
Number of ground truth

reference pixels 32 52 24 41 7 6 15 19 196
Producers Accuracy 96.9% 98.1% 100.0% 90.2% 100.0% 100.0% 73.3% 100.0% 94.9%
5-2

Section 5 Results and Discussions
The producer's accuracy, defined as the proportion of the entire study site which was correctly
classified by the person performing the "heads-up" digitizing, ranged from 73 percent
(bedrock/boulder category) to 100 percent (cobble/graveUbedrock/boulder, cobble/graveUsand,
bedrock gravel, and bedrock categories). Similar to the user's accuracy results, the majority of
substrate categories in the Tillery Hydroelectric Plant study site exceeded 90 percentfor producer's
accuracy (Table 5-1).
5.1.2 Blewett Falls Hydroelectric Plant Tailwater Area Study Site
The classification accuracy results and associated error matrixfor substrate categories atthe Blewett
Falls Hydroelectric Planttailwater area study site are shown in Table 5-2. Five substrate or structure
categories were not included in the accuracy assessment because they were not encountered along
established transects during the ground-truthing field work. The excluded substrate or structure
categories were: (1) bedrock/boulder with emergent and terrestrial vegetation; (2) cobble/gravel
with emergent and terrestrial vegetation; (3) riprap; (4) island; and (5) Blewett Falls Hydroelectric
Plant dam apron. These substrate or structure categories comprised approximately 20 acres within
the study site. The island was considered to be an obvious delineation, coupled with the factthat it
was heavily wooded which affected GPS satellite reception, so no reference points were collected
duringthe ground-truthing. The island comprised 58 percent ofthe 20 acres of substrate or structure
categories that were not included in the accuracy assessment.
The overall classification accuracy ofthe Blewett Falls Hydroelectric Planttailwater area study site
was 94 percent. The user's accuracy ranged from 83 percent (cobble/gravel/sand) to 100 percent
(cobble/graveUsilt). For the cobble/graveUsand substrate category, it should be noted that only six
reference points were observed forthis category with five points correctly classified. The remaining
substrate categories had a user's accuracy greater than 90 percent (Table 5-2). The producer's
accuracy ranged from 83 percent (cobble/graveUsand) to 100 percent (cobble/gravel/boulder).
5.2 Tillery Hydroelectric Plant Tailwater Area Substrate Classification
The unclassified map of the Tillery Hydroelectric Plant tailwater study site and the resulting
classified, thematic map showing digitized substrate categories are shown in Figures 5-1 and 5-2,
respectively. Ten substrate categories comprising 617 substrate areas and 112 acres were mapped in
the study site (Table 5-3). The substrate composition was diverse in the study site, particularly
mixtures of the smaller substrates cobble and gravel (Figure 5-2). Substrates comprised of cobble,
gravel or mixtures of cobble and gravel with other substrates comprised approximately 80 percent of
the study site area (Table 5-3).
Bedrock outcrops were the mostfrequently mapped habitattype on anumerical basis (554 substrate
areas) but only comprised 7 percent oftotal area acreage (Table 5-3). Cobble/graveUboulder was the
second numerically dominant substrate type (24 substrate areas) followed by cobble/gravel
(13 substrate areas), and cobble/graveUbedrock (12 substrate areas). The other substrate categories
comprised the remaining 14 substrate areas that were mapped.
5-3

Section 5 Results and Discussions
Table 5-2 Error matrix and accuracy assessment of substrate categories in the Blewett Falls Hydroelectric Plant tailwater area study
site.

Digitized Map Substrate

Category

Cobble /

Gravel
Cobble /

Gravel /
Boulder Ground-frothing
Cobble / Cobble /

Gravel / Gravel /
Sand Silt Substrate Category

Bedrock /
Bedrock
Boulder
Bedrock /

Boulder /
Cobble

Boulder /

Cobble
Number

of Pixels
User's

Accuracy
Cobble /Gravel 40 0 1 0 0 0 0 2 43 93.0%
Cobble /Gravel /Boulder 2 29 0 0 0 0 0 0 31 93.6%
Cobble /Gravel /Sand 0 0 5 0 0 0 0 1 6 83.3%
Cobble /Gravel /Silt 0 0 0 15 0 0 0 0 15 100.0%
Bedrock 1 0 0 1 67 0 0 2 71 94.4%
Bedrock /Boulder 0 0 0 0 0 20 0 1 21 95.2%
Bedrock /Boulder /Cobble 2 0 0 0 0 0 29 0 31 93.6%
Boulder /Cobble 1 0 0 0 1 1 1 76 80 95.0%
Number of ground truth reference
pixels 46 29 6 16 68 21 30 82 298
Producers Accuracy 87.0% 100.0% 83.3% 93.8% 98.5% 95.2% 96.7% 92.7% 94.3%
5-4

Fig~ire 5-1 Tillery Hydroelectric Plant tailwater area study site showing mosaic of
or8iopl-oto~•aplis and the study site boundary (red line}.
5-5

Section 5
Results and Discussions

Flow
Legend
^.. Studv Boundary
Cobble_Gravel
Cobble_Gravel_Bedrock_Boultler
Cobble_Gravel_Bedrock
K Cobble_Gravel_Boulder
Cobble_Gravel_Sand
Bedrock
K Betlrock_Boulder
Betlrock_Gravel
K Cobble_Gravel_w emerg veg
Riprap
0 250 WO 1,000 Feet
Figure 5-2 Substrate classification map of the Tillery Hydroelectric Plant tailwater area
study site.
5-6

Section 5 Results and Discussions
Table 5-3 Area estimates of substrate categories in the Tillery Hydroelectric Plant
tailwater area study site.
Number of Minimum Maximum Total Percent
Total
Substrate Category Mapped Square Square Square of Total
Acres
Areas Feet Feet Feet Acreage
Cobble /gravel 13 1,524 177,553 813,652 18.7 16.8%
Cobble/gravel/bedrock 12 1,268 401,627 1,158,488 26.6 23.8%
Cobble /gravel /bedrock / 2 122,077 326,543 448,620 103 9.2%
boulder
Cobble /gravel /boulder 24 32 778,218 1,240,047 28.5 25.5%
Cobble /gravel /sand 3 4,389 8,908 21,603 0.5 < 1.0%
Bedrock 554 3 53,531 318,294 73 6.5%
Bedrock/boulder 4 13,189 466,395 624,943 143 12.8%
Bedrock/gravel 3 39,443 116,442 211,629 4.9 4.4%
Cobble /gravel with emergent
and terrestrial vegetation 1 450 450 450 < 0.1 < 1.0%
Riprap 1 23,445 23,445 23,445 0.5 <1.0%
Totalt 617 4,861,172 111.6 100.0%
1 Total area estimates may vary from summation of columns due to rounding
Cobble/graveUbedrock and cobble/graveUboulder were the mostprevalent substrate categories on an
area basis, comprising 27 and 28 acres, respectively. These two substrate categories comprised
almost half of the total mapped area in the Tillery Hydroelectric Plant tailwater area study site
(Table 5-3). Cobble/gravel was the third most prevalent substrate type on an areabasis consisting of
19 acres followed closely bybedrock/boulder (14 acres). These two substrate categories accounted
for 30 percent of the total mapped area within the study site. Riprap, cobble/graveUsand, and
cobble/gravel with emergent and terrestrial vegetation were minor substrate categories comprising
approximately 1 percent of the total acreage (Table 5-3).
The spatial distribution of substrates indicated a diverse mixture of substrate types throughout the
Tillery Hydroelectric Plant tailwater area study site (Figure 5-2). There was a large expanse of
cobble/gravel alongthe west shoreline ofthe study site. This cobble/gravel bar extended alongtwo-
thirds ofthe study site. Visual observations indicated this cobble/gravel bar persisted alongthe west
shoreline for another one to two miles downstream of the study site. Another large area of
cobble/gravel was present on the east shoreline near the midpoint of the study site. Mixtures of
cobble/gravel with boulder and/or bedrock were dispersed throughout the study site along channel
margins and at mid channel areas. Bedrock outcroppings were found throughoutthe study site with
the greatest concentration located in the upper third of the study site just downstream of the Tillery
Hydroelectric Plant. Bedrock/boulder substrate was located at the base of Tillery Hydroelectric
Plant Dam and aY the powerhouse and extended downstream on the east shoreline (Figure 5-2). This
coarse substrate area was located in a high energy zone where dam spillage and power plant
discharges would tend to scour out smaller substrate types. However, the scour zone was confined
to the immediate area in front of these structures. Large areas of cobble/graveUboulder and
cobble/graveUbedrockwere located in close proximity ofthese bedrock/boulder areas, immediately
downstream of the hydroelectric plant. Very small substrate types such as sand and silt were
not prevalent in the study site. There were a few small patches of sand intermixed with
cobble/gravel (cobble/graveUsand category) located along the west shoreline.
5-7

Section 5 Results and Discussions
5.3 Blewett Falls Hydroelectric Plant Tailwater Area Substrate Classification
The unclassified map of the Blewett Falls Hydroelectric Planttailwaterstudy site and the resulting
classified, thematic map showing digitized substrate categories are shown in Figures 5-3 and 5-4,
respectively. Eleven substrate categories and two special areas (i.e., Blewett Falls Hydroelectric
Plant dam apron and island areas) were mapped in the study site. These 13 categories comprised
644 substrate or structure areas and 184 acres in the study site (Table 5-4). The substrate
composition was also diverse in the Blewett Falls Hydroelectric Plant tailwater area study site,
although bedrock was more predominant due to the study site's location in the Fall Line zone.
Mixtures of the smaller substrates (cobble, gravel, sand, and silt) were prevalent in the study site and
accounted for over 21 percent ofthe total acreage (Table 5-4 and Figure 5-4). Substrates comprised
of cobble, gravel or mixtures of cobble and gravel with other substrates comprised approximately
62 percent of the study site area.
Bedrock outcrops were the most frequently mapped substrate category, on a numerical basis,
consisting of 552 areas (Table 5-4). Bedrock/bouldertyith emergent and terrestrial vegetation was
the second numerically dominant substrate category (22 areas) but only comprised 7 percent ofthe
total acreage in the study site. Boulder/cobble, cobble/gravel, cobble/graveUboulder, and
cobble/graveUsilt were the next most frequently observed substrate categories with the number of
mapped areas between 12 to 14 areas.
On an area basis, boulder/cobble and bedrock were the most prevalent substrate categories
comprising 38 and 41 acres, respectively, and represented approximately 43 percent of the total
acreage in the study site (Table 5-4). Bedrock/boulder/cobble and cobble/gravel were the next most
prevalent substrate categories accounting for almost equal amounts of acreage (22 and 23 acres,
respectively). These two latter substrate categories comprised 25 percent ofthe total acreage in the
study site. Cobble/gravel with emergent and terrestrial vegetation, riprap, and the Blewett Falls
Hydroelectric Plant dam apron structure were minor categories accountingfor less than 1 percent of
the total mapped area. The island comprised 6 percent of the total mapped area.
The spatial distribution of substrates also indicated a diverse mixture of substrate types throughout
the Blewett Falls Hydroelectric Plant tailwater area study site (Figure 5-4). There was a large
expanse of cobble/gravel along the west shoreline in association with the Big Island and side
channel complex (Figure 5-4). The cobble/gravel bar encompassed the entire island. As mentioned
previously, bedrock was prevalent in the study reach due to the study site's location in the Fall Line
zone and the large shoal complex located in the study site. Bedrock outcrops were concentrated in
the mid channel and east shoreline areas, particularly in the lower half of the study site. Smaller
substrates (sand and silt) were more frequently observed in the Blewett Falls Hydroelectric Plant
tailwater areawhen compared to the Tillery Hydroelectric Planttailwater area (Figures 5-2 and 5-4).
There were mixed categories ofthese two smaller substrates with cobble/gravel on the east and west
shorelines inclose proximity to the powerhouse and dam. The presence ofthese smaller substrates
suggested downstream transport ofthese substrates from Blewett Falls Lake and the Pee Dee River
5-8

Section 5 Results and Discussions
5-9
Figure 5-3 Ble~;rett Falls Hydroelectric Plant tailwater area study site showutg mosaic of
ortho~hotogi•aphs and the study site boundary (red line).

Section 5 Results and Discussions
5-10
Figure 5-4 Substrate classification map of the Blewett Falls Hydroelectric Plant
tailwater area study site.

Section 5 Results and Discussions
Table 5-4 Area estimates of substrate/structure categories in the Blewett Falls
Hydroelectric Plant tailwater area study site.

Substrate/Structure Category Number of
Mapped
Area Minimum
Square
Feet Maximum
Square
Feet Total
Square
Feet Total

Acres Percent of
Total
Acreage
Cobble/gravel 13 404 716,717 1,010,032 23.2 12.6%
Cobble/graveUboulder 12 61 437,005 768,561 17.6 9.6%
Cobble/graveUsand 1 225,764 225,764 225,764 5.2 2.8%
Cobble/graveUsilt 12 256 98,467 375,705 8.6 4.7%
Bedrock 552 4 582,851 1,785,921 41.0 22.3%
Bedrock/boulder 2 79,058 291,518 370,576 8.5 4.6%
Bedrock/boulder/cobble 5 157 672,244 970,509 22.3 12.1%
Boulder/cobble 14 219 829,552 1,644,891 37.8 20.6%
Bedrock/bouldertyith emergent
and terrestrial vegetation 22 25 229,689 306,826 7.0 3.8%
Cobble/gravel with emergent and
terrestrial vegetation 8 42 6,227 15,977 0.4 < 1.0%
Riprap 1 12,002 12,002 12,002 0.3 <1.0%
Blewett Falls Hydroelectric Plant
dam apron 1 13,981 13,981 13,981 0.3 < 1.0%
Island 1 499,734 499,734 499,734 11.5 6.3%
Totalt 644 8,000,479 183.7 100.0%
1 Total area estimates may vary from summation of columns due to rounding
and associated tributaries, most notably the Rocky River. An unnamed tributary on the east
shoreline was also likely responsible for some of the silt deposition in that mapped area.
Bedbrock/boulder/cobble and boulder/cobble were dispersed throughout the study site, especially
mid channel areas (Figure 5-4). Bedrock/boulderand bedrock/boulderwith emergent and terrestrial
vegetation were prevalent in frontthe Blewett Hydroelectric Plant dam. This area was a high energy
zone during dam spillage events and most small sediments were scoured out ofthis immediate area.
However, there was a large area ofbedrock/cobble and smaller patches ofcobble/gravel just below
this area indicated the scour zone was not extensive.
5-11

Section 6 -Summary
This study characterized the substrate types and distribution in the immediate tailwater areas below
the Tillery and Blewett Falls Hydroelectric Plants. This study was performed to address concerns
raised by Water RWG members during relicensing scoping meetings held in 2003. The concern was
that the recruitment of smaller substrates, mainly gravel and cobble, maybe limited in the tailwater
areas of each hydroelectric plant. Substrate was characterized and delineatedfrom a combination of
high-resolution aerial photography using GIS computer software and ground-trnthing field
measurements. The study also evaluated spatial patterns in substrate composition in each tailwater
area study site.
The two tailwater study areas were defined as: (1) a 1-mile reach ofthe Pee Dee River immediately
below the Tillery Hydroelectric Plant; and (2) a 1.5-mile reach of the Pee Dee River below the
Blewett Falls Hydroelectric Plant to the Cartledge Creek confluence. The Blewett Falls
Hydroelectric Plant study site was longer to include the entire Big Island shoal-island-side channel
complex which has been a focal point in other relicensing studies (e.g., instream flow, fisheries,
aquatic invertebrates, freshwater mussels, and water quality studies). Low altitude, high-resolution
aerial photography (1 inch onphotograph = 200 ft on ground) was acquired at each ofthe two study
sites in February 2005. Scanned resolution ofthe aerial photography imagery resulted in 0.1 ftpixel
resolution.
The photographs were scanned, orthorectified to known ground control points, and composited to
produce one georeferenced mosaic digital image for each tailwater area study site. Substrates were
characterized and delineated using "heads-ups" manual digitizing methods with GIS software.
Ground-trnthing field work, using atransect ground reference point method, was used to verify the
"heads-up" digitizing of the aerial photography. An accuracy assessment was conducted by
comparingthe ground-trnthingfield observations to the classified image. The accuracy assessment
was a quality control method to ensure the substrate classification and digitizing was accurate. The
overall classification accuracy exceeded 94 percent at both tailwater areas study sites. User's
accuracy, defined as the proportion of each substrate category which was correctly identified by the
person performingthe "heads-up" digitizing, ranged from 83 to 100 percentfor substrate categories
at both tailwater areas study sites. Producer's accuracy, defined as the proportion ofthe entire study
site which was correctly classified by the person performingthe "heads-up" digitizing, ranged from
73 to 100 percentfor substrate categories atboth study sites. The "heads-up" digitizing approach of
the high-resolution aerial photography, coupled with theground-trnthing field data, attributed to the
good results of the accuracy assessment.
There was a diverse mixture of substrate types present in each tailwater area study site. The mixed,
heterogeneous nature of the substrate present in both study sites resulted in various substrate
combinations within the defined substrate categories for this study. Ten substrate categories
comprising 617 substrate areas and 112 acres were mapped in the Tillery Hydroelectric Plant
tailwater area study site. There were eleven substrate categories and two special features (Blewett
Falls Hydroelectric Plant dam apron and Big Island) mapped in the Blewett Falls Hydroelectric Plant
study site. These 13 substrate or structure categories comprised 664 mapped areas and 184 acres in
the Blewett Falls Hydroelectric Plant study site. Seven of the 16 categories (both study sites
combined) were common to both study sites while the remaining nine categories were specific to
either study site.
6-1

Section 6
Bedrock outcrops were the mostfrequently mapped habitattype on anumerical basis (554 substrate
areas) in the Tillery Hydroelectric Plant study site but only comprised 7 percent of total area
acreage. Cobble/graveUboulder was the second numerically dominant substrate type (24 substrate
areas) followed by cobble/gravel (13 substrate areas), and cobble/gravel/bedrock (12 substrate
areas). The other substrate categories comprised the remaining 14 substrate areas thatwere mapped
in the Tillery Hydroelectric Plant tailwater area study site.
Cobble/graveUbedrock and cobble/graveUboulder were the mostprevalent substrate categories on an
area basis, comprising 27 and 28 acres, respectively in the Tillery Hydroelectric Plant study site.
These two substrate categories comprised almost half of the total mapped area (112 acres).
Cobble/gravel was the third most prevalent substrate type on an area basis consisting of 19 acres
followed closely bybedrock/boulder (14 acres). These two substrate sites accounted for 30 percent
of the total mapped area within the Tillery Hydroelectric Plant study site. Cobble/graveUsand,
cobble/gravel with emergent and terrestrial vegetation, and riprap were minor substrate categories
comprising approximately 1 percent of the total acreage.
In the Blewett Falls Hydroelectric Plant tailwater study site, bedrock outcrops were the most
frequently mapped substrate category, on a numerical basis, consisting of 552 areas.
Bedrock/boulder with emergent and terrestrial vegetation was the second numerically dominant
substrate category (22 areas) but only comprised 7 percent of the total acreage in the Blewett Falls
Hydroelectric Plant study site. Boulder/cobble, cobble/gravel, cobble/graveUboulder, and
cobble/graveUsiltwerethe nextmostfrequentlyobwrved substrate categories with 12 to 14 mapped
areas.
On an area basis, boulder/cobble and bedrock were the most prevalent substrate categories
comprising 38 and 41 acres, respectively within the Blewett Falls Hydroelectric Plant study site.
These two substrate categories represented approximately 43 percent ofthe total acreage (184 acres)
in the study site. Bedrock/boulder/cobble and cobble/gravel were the neatmostprevalent substrate
categories accounting for almost equal amounts of acreage (22 and 23 acres, respectively). These
two latter substrate categories comprised 25 percent of the total acreage in the Blewett Falls
Hydroelectric Plant study site. Cobble/gravel with emergent vegetation, riprap, and the Blewett
Falls Hydroelectric Plant dam apron structure were minor categories accounting for less than
1 percent of the total mapped area
The presence and operation of each hydroelectric development has not resulted in river channel
"armoring" of the immediate tailwater area below each power plant. River channel armoring is
typically characterized as a complete absence of smaller substrates such as sand and gravel with only
larger coarse substrate types present such as bedrock and boulders. Smaller substrate categories,
which included combinations of silt, sand, gravel, and cobble, comprised 18 and 22 percent,
respectively, in the Tillery and Blewett Falls Hydroelectric Plants tailwater areas. Additionally,
combinations of cobble and gravel with boulder and bedrock substrates, accounted for 42 and
63 percent ofthe total mapped areas in the Tillery and Blewett Falls Hydroelectric Plants study sites,
respectively.
Recruitment rates of gravel and other small particle substrate types into the immediate tailwater
areas were not determined by this study. Moreover, no comparable historical data exists to
6-2

Section 6
determine the stability or persistence of small particle substrate types over the life span of the
Project. However, the prevalence of smaller substrate types throughout each power planttailwater
study site suggested that the flow regimes of each hydroelectric development have not been of a
magnitude to scour away smaller particle substrates.
6-3

Section 7 -References
Alcoa Power Generating, Inc. 2002. Yadkin River Hydroelectric Project FERC No. 2197 NC.
Project Relicensing Initial Consultation Document. September 2002. Alcoa Power
Generating, Inc., Yadkin Division, Badin, North Carolina.
Appalachian State University. 1999. North Carolina's Central Park: Assessing Tourism and
Outdoor Recreation in the Uwharrie Lakes Region. Appalachian State University,
September 1999.
Bain, M. B. and N. J. Stevenson. 1999. Aquatic habitat assessment, common methods. American
Fisheries Society. 216p.
Collier, M., R.H. Webb, and J.C. Schmidt. 1996. Dams and rivers: a primer on the downstream
effects of dams. U.S. Geological Society. 94p.
Congalton, R. 1991. A review of assessingthe accuracy of classifications ofremotely sensed data.
Remote sensing of environment. Vol. 37. ISSN: 0034-4257.
Fitzgerald, R.W. and B. G. Lees. 1994. Assessing the classification accuracy ofmultisource remote
sensing data. Remote sensing of the environment. Vol. 47, pp. 362-368.
Jensen, J.R. 1996. Introductory digital image processing: aremote sensing perspective. Prentice-
Hall, Inc. ISBN 0-13-205840-5.
Khorram, S., G.S. Biging, N.R. Chrisman, D.R. Colby, R.G. Congalton, J.E. Dobson, R.L.
Ferguson, M.F. Goodchild, J.R. Jensen, and T.H. Mace. 1999. Accuracy assessment of
remote sensing-derived change detection. Monograph. America Society ofPhotogrammetry
and Remote Sensing. 68p.
Longley, P.A., M.F. Goodchild, D.J. Maguire, and D.W. Rhind. 2001. Geographic information
systems and science. John Wiley & Sons.
McMahon, T.E., A.V. Zale, and D.J. Orth. 1996. Aquatic habitat measurements. Pages 83-120 in
B. R. Murphy and D. W. Willis (eds.). Fisheries techniques. Second edition. American
Fisheries Society, Bethesda, Maryland.
North Carolina Division of Water Quality. 2002. Basinwide assessment report. Yadkin River
Basin. June 2002. North Carolina Department of Environment, Health, and Natural
Resources, Division of Water Quality, Water Quality Section, Environmental Services
Branch, Raleigh, North Carolina.
2004. NC DENR-Division of Water Quality "Redbook". Surface waters and wetlands
standards. NC Administrative Code 15A NCAC 02B.0100, .0200 & .0300. Amended
effective: August 1, 2004. North Carolina Department of Environment and Natural
Resources, Division of Water Quality, Raleigh, North Carolina.
7-1

Section 7
References
Progress Energy. 2003. Initial Consultation Document. Yadkin-Pee Dee River Project FERC No.
2006. Submitted by Progress Energy, Raleigh, North Carolina, February 2003.
. 2004. RWG meeting summary notes, templates, and study plans. Yadkin-Pee Dee River
Project FERC No. 2206. January 2004. Progress Energy.
7-2

APPENDICES

APPENDIX A
MODIFIED WENTWORTH CLASSIFICATION FOR SUBSTRATE
PARTICLE SIZE UTILIZED DURING THE SUBSTRATE
CHARACTERIZATION STUDY OF THE TAILWATER AREAS OF THE
TILLERY AND BLEWETT FALLS HYDROELECTRIC PLANTS DURING
2005

Appendix A modified Wentworth classification for substrate particle size utilized during the
substrate characterization study of the tailwater areas of the Tillery and Blewett
Falls Hydroelectric Plants during 2005.
Substrate Classification Particle Size Range
(mm)
Boulder > 256
Cobble 64-256
Pebble 32-64
16-32
Gravel 8-16
4-8
2-4
Very coarse sand 1-2
Coarse sand 0.5-1
Medium sand 0.25-0.5
Fine sand 0.125-0.25
Very fine sand 0.0625-0.125
Silt 0.0039-0.0625
Clay < 0.0039
Appendix A - 1