EJBD
EPA
600-
7-
81-
043
Repository Material
Permanent Collection
November 1980
USEPA
U& tKA
S???/s and Cnemical Libraries
EPA West Bldg Room 3340
Mailcode 3404T
1301 Constitution Ave NW
Washington DC 20004
202-566-0556
THE OHIO RIVER BASIN ENERGY STUDY
ENERGY FACILITY SITING MODEL
Volume I
Methodology
By
Gary L. Fowler
University of Illinois at Chicago Circle
Chicago, Illinois 60680
Robert E. Bailey
Steven I. Gordon
The Ohio State University
Columbus, Ohio 43210
Steven I). Jansen
University of Illinois at Chicago Circle
Chicago, Illinois 60680
J. C. Randolph
Indiana University
Bloomington, Indiana 47405
W. W. Jones
Indiana University
Bloomington, Indiana 47405
EJBD
ARCHIVE
EPA
600-
7-
81-
043
Prepared for:
The Ohio River Basin Energy Study (ORBES) Phase II
Grant Nos. EPA R805588, R805589, and R805609
and Subcontract under Prime Contract EPA R805588
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
-------
CONTENTS
Figures v
Tables ix
Acknowledgement xl
1. Section I 1
Introduction 1
2. Section II 4
Siting Electrical Generating Capacity in the ORBES Region . . 4
Siting Procedures 4
General Methodology 4
Siting Coal Fired and Nuclear-Fueled Generating Units. 6
Site Selection in the ORBES Region 7
Trends in Siting in the ORBES Region 9
Changes in the Schedule of Planned Additions 19
3. Section III 22
The Ohio River Basin Energy Facility Siting Model 22
Regional-Scale Energy Fac-lity Siting Models 22
Regional Energy Supply 24
Definition of Site Suitability 25
Allocation of Additional Facilities 26
The ORBES Regional Siting Model 27
Scenario and Siting Policies 30
4. Section IV 33
Scenario Unit Additions and Spatial Allocation Procedures . . 33
Calculating Scenario Unit Additions 33
Projected Electrical Generating Unit Additions for ORBES
Scenarios 38
Schedule of Capacity Additions 38
Spatial Allocation Procedures for Scenario Unit Additions . . 43
Site Specific Locations for Scenario Unit Additions 46
5. Section V 48
Siting Issues and Site Suitability 48
Site Suitability Model 48
Siting Issues 51
Ambient Air Quality 51
Site Suitability: Ambient Air Quality 56
Water Availability 61
Site Suitability for Water Availability 68
Land Use and Ecological Systems 71
Seismic Suitability 77
Population Distribution 81
Definition of Siting Weights 84
iii
-------
Definition of Site Suitability for Basic Scenarios . . 89
Coal-based, Base Case Environmental Control
Scenarios 93
Coal-based, Strict Environmental Control Scenarios . . 97
Nuclear Emphasis, Base Case Environmental Controls . . 97
6. Section 6 106
Siting Patterns for ORBES Scenarios 106
Coal Emphasis, Conventional Technology 106
Scenario 2: Base Case Environmental Controls 106
Scenario 1: Strict Environmental Controls 108
Scenario 7a and 7b: Very High Energy Growth 108
Scenario 2a: Coal-fired Export
Fuel Substitution and Conservation
Scenario 3: Alternate Technology
Scenario 4: Natural Gas Emphasis 114
Scenario 6: Conservation (Very Low Energy Growth) . . 117
Scenario 2c: Nuclear Emphasis 117
Siting Patterns for Special Policy Analysis 120
Scenario la: Very Strict Air Quality Controls .... 120
Procedure 120
Exclusionary Screening and Site Suitability 123
Siting Pattern i23
Scenario Ic: Agricultural Lands Protection 127
197
Procedure *-*•'
Exclusionary Screening and Site Suitability 128
Scenario Ic: Agricultural Lands Protection Policy . . 128
Bibliography "6
Appendices 14-*
A. Sited Capacity Additions, 1976 through 2000 144
B. Capacity Removals, 1976 through 2000 151
C. Air Quality Data for ORBES Counties, 1977 165
D. Counties Excluded as Sites for Coal-fired Scenario
Unit Additions, Base Case Environmental Controls 171
E. Counties Excluded as Sites for Coal-fired Scenario
Unit Additions, Strict Environmental Controls 172
F. Counties Excluded as Sites for Nuclear-fueled Scenario
Unit Additions 176
G. ECAR Region Site Inventory 179
iv
-------
FIGURES
Number PaEe
1 The Ohio River Basin Energy Study region 2
2 Generalized site selection and evaluation process 5
3 Total installed coal-fired electrical generating
capacity, 1975, six-state region 11
4 Total installed nuclear-fueled electrical generating
capacity, 1975, six-state region 12
5 Total proposed coal-fired electrical generating capacity,
1976-1985, six-state region 13
6 Total proposed nuclear-fueled electrical generating
capacity, 1976-1985, six-state region 14
7 Total coal-fired electrical generating capacity, 1985,
six-state region ....................... 18
8 Total nuclear-fueled electrical generating capacity, 1985,
six-state region ....................... 20
9 Ohio River Basin Energy Study facility siting model ....... 28
10 Schedule of electrical generating capacity additions
for the ORBES region, 1976-2000 (coal and nuclear
plants only) ......................... 41
11 Growth in installed electrical generating capacity for
the ORBES region, 1976-2000 (coal and nuclear plants
only ............................. 44
12 A scheme for scoring a site in terms of air quality ....... 57
13 Prevention of significant deterioration, sulfur dioxide
(802), base case environmental controls ............ 59
14 Prevention of significant deterioration, total suspended
particulates (TSP) , base case environmental controls ..... 60
15 Prevention of significant deterioration, sulfur dioxide
, strict environmental controls ............. 62
-------
Number Page
16 Prevention of significant deterioration, total suspended
particulates (TSP), strict environmental controls 63
17 Ambient air quality component, base case environmental
controls 64
18 Ambient air quality component, strict environmental
controls 65
19 Water availability component 70
20 Natural, scenic and recreational areas 74
21 Sensitive and protected environments 75
22 Agricultural and ecological productivity 76
23 Ownership and management of forest lands 78
24 Ecological systems and land use component 79
25 Seismic suitability component 83
26 Population distribution component 86
27 Definition of weights for siting components 88
28 Site suitability index, coal-fired scenario unit additions,
base case environmental controls 94
29 Counties excluded as sites for coal-fired scenario unit
additions, base case environmental controls 96
30 Site suitability index, coal-fired scenario unit additions,
strict environmental controls 98
31 Counties excluded as sites for coal-fired scenario unit
additions, strict environmental controls 100
32 Site suitability index, nuclear-fueled scenario unit
additions 101
33 Counties excluded as sites for nuclear-fueled scenario unit
additions 103
34 Scenario 2: Conventional technology, base case controls,
Total proposed coal-fired generating capacity additions,
1976-85, plus scenario unit additions, 1986-2000 107
vi
-------
Number Page
35 Scenario 2: Conventional technology, base case controls.
Total proposed nuclear generating capacity additions, 1976-85,
no scenario unit additions, 1986-2000 109
36 Scenario 1: Conventional technology, strict controls.
Total proposed coal-fired generating capacity additions,
1976-85, plus scenario unit additions, 1986-2000 110
37 Scenario 7a: 35 year life, conventional technology, base
case, high electrical energy growth. Total proposed coal-
fired generating capacity additions, 1976-85, plus scenario
unit additions, 1986-2000 112
38 Scenario 7b: 45 year life, conventional technology, base
case, high electrical energy growth. Total proposed coal-
fired generating capacity additions, 1976-85, plus scenario
unit additions, 1986-2000 113
39 Scenario 2: Conventional technology, base case controls,
coal-fired export. Total proposed coal-fired generating
capacity additions, 1976-85, plus scenario unit additions,
1986-2000 115
40 Scenario 3: Alternate technology, base case controls. Total
proposed coal-fired generating capacity additions, 1976-85,
plus scenario unit additions, 1986-2000 116
41 Scenario A: Conventional technology, natural gas emphasis,
base case controls. Total proposed coal-fired generating
capacity additions, 1976-85, plus scenario unit additions,
1986-2000 118
42 Scenario 6: Conventional technology, base case controls,
very low energy growth. Total proposed coal-fired generating
capacity additions, 1976-85, plus scenario unit additions . . . 119
43 Scenario 2c: Conventional technology, base case controls,
nuclear emphasis. Total proposed nuclear-fueled generating
capacity additions, 1976-85, plus scenario unit additions,
1986-2000 121
44 Counties excluded as candidate sites: Scenarios la and Ib,
very strict air quality control policies 125
45 Scenario la: Very strict air quality controls, dispersed
siting 126
46 Counties excluded as sites: Scenario Ic and Id, agricultural
lands protection policy 129
vii
-------
Number Page
47 Ecological systems and land use component, agricultural
lands protection policy 130
48 Site suitability index, agricultural lands protection
policy 131
49 Scenario Ic: Agricultural lands protection, dispersed
siting 133
viii
-------
TABLES
Number Pflge
1 Summary of electrical generating capacity, 1975 and changes,
1976-1985, six-state region .................. '10
2 Size, in MWe, of coal-fired and nuclear-fueled electrical
generating units and sites, six-state region (steam units
only) ............................. 16
3 Description of scenarios and siting policies ........... 31
4 Methodology for calculating unsited electrical generating
unit additions ......................... 34
5 Unsited electrical generating unit additions ........... 35
6 1974 baseline data, 1985 and 2000 scenario 1 solution to the
ORBES energy demand model ................... 37
7 Projected number of coal-fired and nuclear-fueled electrical
generating facility scenario unit additions to be sited in
the ORBES region, 1986-2000 .................. 39
8 Schedules for projected installed capacity (MWe) in the ORBES
region for twelve scenarios (coal and nuclear units only) . . . 42
9 Definition of primary variables used in determining site
suitability .......................... *»9
10 Summary of clean air act amendments of 1977, prevention of
significant deterioration (PSD) ................ 53
11 Definition of the air quality component ............. 58
12 Cooling water requirements for scenario unit additions
(in CFS/unit) ......................... 67
13 Water availability component scores ............... 69
14 Definition of the land use and ecological systems
component ........................... ' *•
15 Definition of seismic suitability scores from seismic
suitability zones ....................... 82
ix
-------
Number Page
16 Suitability scores for population distribution
component 85
17 Weights for siting components and variables, coal-fired
electricity generating facilities, in the Ohio River
Basin Energy Study Region 90
18 Weights for siting components and variables for nuclear-
fueled electricity generating facilities in the Ohio
river Basin Energy Study Region 91
19 Summary of counties excluded as sites for coal-fired
scenario unit additions, base case environmental controls ... 95
20 Summary of counties excluded as sites for coal-fired
scenario unit additions, strict environmental controls 99
21 Summary of counties excluded as sites for nuclear-fueled
scenario unit additions 102
22 Summary of counties excluded as sites for coal-fired
scenario unit additions in scenarios la and Ib:
strict air quality controls, and in scenarios Ic and
Id: agricultural lands protection policy 124
-------
ACKNOWLEDGEMENT
During Phase II of the Ohio River Basin Energy Study (ORBES), a Siting
Group, which I chaired, and which was composed of members of the ORBES Core
Team and Advisory Committee, met regularly to define the siting issues, de-
velop the model and translate the energy development scenarios into geograph-
ical patterns of future electrical generating capacity. The Core Team
members were: Robert E. Bailey and Steven I. Gordon, the Ohio State Univer-
sity; and J. C. Randolph, Indiana University. Bailey was responsible for
the air quality component of the siting model; Randolph was responsible
for the land use and ecological systems component. Advisory Committee
members who participated were: Sy Ali, Public Service Indiana; John Barcalow,
Illinois Power Company; Dana E. Limes, Columbus and Southern Ohio Electric
Company; Owen A. Lentz, Executive Director, ECAR; and Charles Tillotson, Swit-
zerland County, Indiana.
Steven D. Jansen, University of Illinois at Chicago Circle, and W. W.
Jones, Indiana Univeristy, made significant contributions to the work of the
Siting Group. Jansen developed procedures for defining the number of gener-
ating unit additions for each scenario, and scheduling their on-line dates.
Jones was responsible for expanding the model's capacity for land use and eco-
logical systems analysis. Larry Wong, Indiana University, programmed the site
suitability component of the model and did the runs for each scenario.
The maps and other graphic materials were produced by the Chicago Area
Geographical Information Study (CAGIS) and the Cartographic Laboratory, De-
partment of Geography, University of Illinois at Chicago Circle. Raymond
Brod, Eric D. Heckman, Ruth Laski and David Merrill were involved. Steven
D. Jansen coordinated the cartographic work.
Ms. Claudette Eldridge, University of Illinois at Chicago Circle, orga-
nized and produced this report.
I deeply appreciate the continued dedication, interest and skills of each
person involved in the work of the Siting Group.
Gary L. Fowler
University of Illinois at Chicago Circle
November, 1980
xi
-------
SECTION 1
INTRODUCTION
The purpose of power plant siting models in regional technology assess-
ments of energy development is to translate energy-related policies into a
geographical pattern of impacts that can be assessed and evaluated. Given
an aggregate level of future energy demand, or production, from which a spe-
cific number and type of generating unit additions can be determined, the
additions must be distributed within a region in a consistent manner that
is explicitly related to other scenario elements. Candidate sites, usually
counties, are defined by exclusionary criteria and ranked according to their
suitability as future sites for electrical generating units. Siting patterns
may vary by scenario and, at county scale, are highly dependent on assumptions
about energy technologies, resource requirements and environmental policy.
Regional technology assessments are most useful if they evaluate the
impact that are associated with different scenarios and related sets of poli-
cies. Estimates of the changes in impacts that result from different policy
options provide important, if not essential, information to policymakers (cf.
Fowler, 1977; White and Hall, 1978). These options may include policies that
affect the geographical distribution of energy facilities such as electrical
generating units. Changes in policy that directly or indirectly affect the
relative location of capacity additions may significantly change the nature
of the resultant impacts. Whereas other regional assessments focus on a
single future siting pattern, the ORBES project analyzes several.
The Ohio River Basin Energy Study (ORBES) siting model is specifically
designed for regional policy analysis. The region includes 423 counties
in a six-state area that focuses on the Ohio River main stem (Figure 1).
Policies that indirectly affect siting generating unit additions include pro-
jections of the future production of electricity; fuel type and technologies
that will meet the demand; and the resource requirements of capacity additions.
Policies that directly affect siting include items such as the exclusionary
requirements of technology to regulations, and preferences for one type of
distribution (e.g., dispersed siting or power parks) over another. The ORBES
scenarios incorporate both types of policies so that changes in impacts can
be systematically evaluated. The direct effect of environmental control poli-
cies with respect to air quality, water availability, and ecological systems
and land use is of particular concern.
The siting model has several important characteristics.
• Different sets of policies that directly or indirectly affect
siting patterns are analyzed systematically. The pattern and
type of changes in impacts that is the result of different
siting patterns can be isolated and evaluated.
1
-------
Figure i. THE OHIO RIVER BASIN ENERGY STUDY REGION
ILLINOIS
LOCATOR MAP FOB
STUDY REGION
Seal* ton!to*
-------
• Announced utility plans for generating capacity additions in
the near-term (i.e., 1976-1985) are combined with alternate
scenarios of long-term (1986-2000) development. The base
case scenario assumes that current and near-term behavior
and policies will dominate the region's energy future. Other
scenarios project different long-term futures which, when com-
pared with the base case, are used to evaluate the impact of
different policy options.
• Sites are defined and evaluated with respect to regional
issues, resources and values. Indices of land use and eco-
logical systems, for example, are included as siting issues.
The relative importance of these other variables that affect
the geographical distribution of generating unit additions,
are defined by knowledgeable people.
The siting model is limited to base-loaded steam generating units that use
conventional coal and nuclear fuels. Some scenarios assume alternate tech-
nologies and fuels. The alternatives affect the geographical distribution
of coal-fired and nuclear-fueled generating units only to the extent that
fewer conventional units need to be sited. The geographical distribution
of alternate technologies is, in effect, unknown. Other models may be de-
veloped for these technologies in future assessments.
This report consists of two volumes. The methodology is presented in
this volume. An analysis of siting patterns and procedures for coal-fired
and nuclear-fueled generating units in the ORBES region is followed by defi-
nition of siting issues, and the methodology used in transforming the issues
and related policies into future geographies of electricity supply and dis-
tribution. The siting patterns that are developed for the basic set of
scenarios are compared. The second volume (Fowler et al, 1980) contains
detailed lists of on-line dates and county-level sites for each scenario
that was developed for the ORBES assessment.
-------
SECTION 2
SITING ELECTRICAL GENERATING CAPACITY IN THE ORBES REGION
Siting electrical generating units is an integral part of capacity plan-
ning (Womeldorff, 1978). Given a load forecast, an array of available capac-
ity and alternative technologies, a utility makes decisions about scheduling
capacity additions at the best available sites within its service area, or
adjacent service areas. Site selection takes into consideration existing
environmental and regulatory constraints, as well as plant design and system
economics. Planning always involves some uncertainty about the effect of
changes that may occur during the 8 to 15 years before a unit is in service.
The distribution of electrical generating capacity in the ORBES region
in 1975 is the baseline for the technology assessment. Announced utility
plans for the next decade (1976-1985) represent a projection of near-tarm
changes in the geography of electrical generating facilities in the region.
As such, it is a guide to the near-term impacts of aggregate patterns of
energy development and to policy issues. In the long-term, capacity addi-
tions are sited according to alternate development scenarios that systemat-
ically change policies that affect the distribution of electrical generating
units.
SITING PROCEDURES
General Methodology
There is no single best method or set of criteria for selecting and evalu-
ating sites for capacity additions. Siting procedures depend upon a utility's,
experience and situation, including the available technological choices; the
regulatory environment within which decisions are made; and the resources
within the region of interest. A synthesis of the procedures that utilities
use in locating nuclear-fueled units, however, suggests a generalized site
selection process that incorporates, in sequential form, the basic steps in-
volved (Figure 2).1 The sequence leads from a systematic screening at macro-
geographic, or regional, scale to an evaluation of a few proposed sites at
micro-geographic scale. At each step, specific criteria are used to evaluate
places as sites for capacity additions. System planning; safety concerns;
engineering characteristics of the plants; environmental, institutional and
regulatory constraints; and economic factors are issues in the siting pro-
cess. The emphasis generally shifts from initial concerns about system plan-
ning to engineering, environmental and regulatory concerns in Stage 2, and to
engineering and environmental issues only in Stage 3. Also, the degree of
detail in the data required to evaluate sites increases as the number of pos-
sible sites decrease.
The determination of need for additional generating capacity, and defi-
nition of the technological alternatives available to meet that need, is the
first step in the siting process. The utility's objective is then to select
-------
Figure 2. GENERALIZED SITE SELECTION AND EVALUATION PROCESS
STAGE 1
O*iwmin«f ion of
Oflttrtnmjtion of
C*nd>dat» &tn
PropowcJ S*t«
SOURCE: Calvert, Heilman and Smith (1974).
-------
the best site from among those that are available to it within its region of
interest. The region of "interest (ROI) is the geographical area within which
the utility could conceivably locate capacity additions. It may be the util-
ity service area; the service areas of adjacent utilities; the combined ser-
vice areas of pooled utility groups; or a state. The definition of the ROI
is essentially a political issue (Keeney et al, 1978). The size of the re-
gions has increased because the political problems of siting generating units
have tended to favor more remote locations, and technological improvements
have made long-distance transmission of power more feasible (cf. Morrill,
1977).
The objective of Stage 1 is to reduce the ROI to a relatively small group
of candidate areas that are likely to have a number of suitable sites. Exclu-
sionary screening to determine general environmental suitability is the method
most commonly used. Areas are excluded if they fail to meet some minimum
performance standard or resource requirement. Exclusionary thresholds for
selected technological and regulatory criteria can be defined in objective
quantitative terms. However, not all siting issues have exclusionary charac-
teristics.
The objective of Stage 2 is to select a relatively small number of sites
within the candidate areas that can be licensed and developed. The roster of
candidate sites can include the inventory of sites that have been evaliuted
previously, or existing sites that can accomodate additional generating ca-
pacity. The comparative evaluation of sites according to multiple criteria
is the method most commonly used at this stage. This usually involves using
numerical scoring procedures, and weighting each criterion according to Its
relative importance in siting, in order to evaluate and rank candidate sites.
Whereas secondary data are generally adequate for Stage 1, more detailed data,
some of which may need to be collected from primary sources, is frequently
required in Stage 2 and 3.
The objective of Stage 3 is to select the proposed site from among a few
high-ranking candidates. Usually, the methodologies used to evaluate the
candidate sites are more complex than those used in previous stages, and more
detailed site-specific data are required. The siting methodologies incorpo-
rate methods of weighing economic costs, engineering and environmental impacts,
and other relevant siting criteria in order to justify the selection of one
site from among a relatively few alternatives. Frequently, the selection is
based upon the comparative evaluation and ranking of the candidate sites
identified in Stage 2.
Siting Coal-Fired and Nuclear-Fueled Generating Units
Coal-fired generating units account for the majority of the electricity
produced in the United States, as well as the ORBES region. Recent national
energy policies have emphasized the increased use of coal to fuel electric
plants. Coincidentally, the environmental and institutional constraints on
licensing, siting and operating coal-fired units have increased. The para-
dox of this situation is that relatively little attention has been given to
establishing standard siting and licensing procedures for coal-fired units
that are comparable to those for nuclear facilities (Feldman, 1978; Williams,
1978).
-------
Resource constraints and environmental regulations are the most important
considerations in siting coal-fired capacity additions. These include ambi-
ent air quality standards; water quality standards; and ecological issues such
as rare or endangered species, other unique habitats, and public or protected
natural lands (Envirosphere Company, 1977). The availability of an adequate
supply of cooling water, accessibility by barge or rail transport for coal,
and compatible land uses are also important criteria. Coal-fired generating
units must compete for scarce natural resources and, in portions of the ORBES
region, with highly productive land uses such as agriculture. Siting issues
also include problems from residuals such as air pollution and solid waste
management (Calzonetti, 1979).
The siting procedure for nuclear-fueled units is more clearly defined
because of the NRC's licensing process. According to 10 CFR 100, nuclear
reactors are expected to (10 CFR, Part 100, p. 544):
... reflect through their design, construction and operation
an extremely low probability for accidents that could result
in release of significant quantities of radioactive fission
products. In addition, the site location and engineered fea-
tures included as safeguards against the hazardous consequences
of an accident, should one occur, should ensure a low risk of
public exposure.
Subsequent regulatory guidelines specify the factors to be considered and
their definition for siting. These factors include reactor design; popula-
tion distribution and density near the site, and distance from population
centers; and physical characteristics of the site, such as seismology, me-
teorology, geology and hydrology.
In choosing candidate sites, the availability of cooling waters, site
geology, accessibility and land use are the most important criteria that
utilities consider (U.S. NRC, 1976). Meteorology, population density and
distribution, seismology, transmission requirements and aesthetics are less
important. Whereas system planning is important in Stage 1 regional screen-
ing, emphasis shifts to a larger number of engineering, environmental and
institutional criteria in determining candidate sites. A long list of en-
gineering and environmental criteria are used in Stage 3 to evaluate and
select proposed sites. Here the focus is on design and site characteristics,
as most other issues have been satisfactorily resolved in earlier stages of
the site selection process.
SITE SELECTION IN THE ORBES REGION
The site selection processes of utilities in the ORBES region follow
general patterns. The service area is their primary region of interest. In
some cases, such as Indianapolis Power and Light Company (IPALCO), the ser-
vice area is small with little, if any, possibility that additional coal-
fired generating units can be located in it (Saper and Hartnett, 1978, pp.
10-20). At the other extreme is American Electric Power (AEP), with member
companies whose service areas are in several states and a wide range of
siting opportunities. Most utilities in the ORBES region, however, have
-------
service areas that are comprised of all or portions of relative large numbers
of contiguous counties within state boundaries. Their ROI's include adjacent
service areas or the entire state in which they are located (cf. Elkins and
DiNunno, 1975; Soyland Power Cooperative, 1980).
The siting criteria that are most important in selecting a site for a
particular type of generating unit will vary according to the environment and
resources that are available in the ROI. For example, Louisville Gas and
Electric's (LG&E) decision to site four coal-fired units (a total of 2,304
MWe) in Trimble County, Kentucky was based upon five factors (USEPA, 1978).
1. Ample acreage for plant facilities and solid waste disposal.
2. Easy access to the Ohio River main stem for a cooling water
supply and barge transport for coal.
3. Near existing transmission line tie-in.
4. Low concentrations of sulfur dioxide (802) in the area.
5. Located near major population concentrations in the northern
part of LGE's service area.
At larger scale, ambient air quality is the most important environmental con-
sideration, followed by ecological criteria (primarily endangered species),
water availability, geotechnical factors (mined areas and geological hazards),
land use and accessibility (Elkins and DiNunno, 1975; and Soyland Power Coop-
erative, 1980, p. C-19). Accessibility to transmission lines is the most
aspect of the latter issue because all but a few of the counties in the ORBES
region have railway lines or waterways that can be used for coal barge traffic.
Nuclear reactor siting also follows general procedures (Laney and Gustaf-
son, 1979). Although reactor safety issues dominate the initial considerations,
environmental issues have exerted an increased influence on site selection
and evaluation since 1970 as regulatory requirements have become more complex.
In Illinois, for example, sites that are well-connected to the utility grid;
are near large supplies of cooling water; and are relatively remote from
densely-populated areas are preferred. Physical characteristics of the site,
ecological impacts and land use compatibility are also of concern. Locating
nuclear reactors in the prime agricultural lands of northern and central Il-
linois is definitely an issue. The utility's problem is to minimize the costs
of acquiring and developing land for new sites and transmitting the electric-
ity while maximizing safety and system reliability.
Siting in the ORBES region is primarily the responsibility of utilities
operating within the framework of state policies and procedures. A majority
of the states in the nation have introduced diverse legislation designed to
increase the state's role in siting process (cf. Southern States Nuclear
Board, 1978; Williams, 1978). In the ORBES region, Kentucky and Ohio h.ive
specific comprehensive procedures that dea] with siting electrical f.ici 1 i l u-s.4
Kentucky has a system of multiple approvals for electrical power generators
-------
(except those that are municipally owned) and transmission line >_ 400 kv.
The Ohio Power Plant Siting Commission is the lead agency in a one-stop pro-
cess that includes all electrical generating facilities (oil, coal and nuclear
fuel) of _> 50 MWe, electrical transmission lines of >_ 125 kv and gas trans-
mission lines _> 125 psi. Other states do not have overall policies or pro-
cedures for siting electrical energy facilities, although task forces have
studied the issue, and legislation has been introduced in Indiana and Illinois
(Nelson and Mitchell, 1979). The Ohio River Valley Water Sanitary Commission
(ORSANCO) has proposed a regional siting authority for all energy facilities
in its member states, which includes all of those in the ORBES region (ORSANCO,
1979). This initiative, as well as others that propose state siting authori-
ties, have yet to be adopted.
TRENDS IN SITING IN THE ORBES REGION
The six states in the ORBES region had an estimated 116,524 MWe of in-
stalled electrical generating capacity in 1975 (Table 1). Coal provided 76%
of the region's total capacity with oil (11.7%) and nuclear-fueled units
(7.4%) the next most important sources. Ohio, Illinois and Pennsylvania had
the majority of the total 88,602 MWe installed coal-fired capacity. The seven
nuclear reactors located in northern Illinois were the only such units in the
ORBES region. No other state had installed nuclear capacity in 1975. The
total installed capacity in a state is a function of total state population
(r2 = 0.99, with 1970 population data). Kentucky and West Virginia were the
exceptions, as each exported almost twice the amount of electricity consumed
in the state. Electricity produced in the ORBES region is also exported
across the region's boundaries to the non-ORBES portion of each state subre-
gion (Page, 1979, Appendix A).
The majority of the total 1975 generating capacity was located along the
Ohio River main stem (28.7%) and its major tributaries (31.8%).5 All of this
capacity came from coal-fired units, most of which were concentrated along the
Ohio upstream from Louisville, Kentucky (Figure 3). Approximately 60% of the
state of Ohio's capacity, and most of that in Kentucky, Pennsylvania and West
Virginia were along the Ohio and its tributaries. Other concentrations of
electric generating capacity were either in or near major load centers, such
as metropolitan areas, or along the Great Lakes outside of the ORBES region.
Nuclear unit additions were concentrated in areas that already had nuclear
capacity (Figure 4). These were relatively close to large metropolitan areas,
especially Chicago.
Electrical generating capacity in the six-state region is expected to
nearly double from 1976 to 1985. Fifty-three percent of the net 56,361 MWe
projected increase is from coal-fired generating units, with nuclear-fueled
units accounting for 44% of the total (Figures 5 and 6). Nearly one-half of
the scheduled additions are expected to be in Illinois and Pennsylvania, pri-
marily because of the large increases in nuclear-fueled capacity in these
states. Coal-fired units account for all of the scheduled additions in Ken-
tucky and West Virginia, and three-quarters of those in Indiana. The majority
(58%) of the total coal-fired capacity additions, as well as the additions in
Kentucky, Ohio and West Virginia, are sited along the Ohio River main stem.
Another 30% of the total will be located on tributaries of the Ohio; nearly
-------
Table i. SUMMARY OF ELECTRICITY GENERATING CAPACITY, 1975. AND CHANGES, 1976-1985
SIX-STATE REGION
Fuel Type. MWe
State
Illinois
Indiana
Kentucky
Ohio
Pennsylvania
West Virginia
TOTAL
Z Total
1985 Capacity MWe
1975-1985 MWe Change, Net
1975-1985, Z Change
Coal Nuclear
15,801 5,717
13,104
10,948
21,266
15,517 2,904
11,966
88,602 8,621
75.9 7.4
118,572 33,240
29.970 24.619
+ 33.8 + 285.6
Petroleum
4,058
1.166
121
2,700
5,818
12
13,875
11.9
14,087
212
+ 1.5
Natural Gas
204
110
128
71
19
_
532
0.4
456
76
- 14.3
Hydro8
34
114
679
1
1,717
205
2,750
2.4
3,037
387
+ 10.4
Otherb
197
324
-
626
815
387
2,349
2.0
3,493
1.144
-i- 32.8
MWe
Total
26.011
14,818
11,876
24,664
26,790
12.570
116,729
100
172,885
56,156
+ 48.1
Z Total
22.3
12.7
10.2
21.1
22.9
10.8
100.0
SOURCE: S. D. Jansen (1978).
Includes hydro and pumped storage.
Includes refuse, waste heat, multi-fueled and unknown fuel types.
-------
Figure 3. TOTAL INSTALLED COAL-FIRED ELECTRICAL GENERATING CAPACITY. 1975
SIX-STATE REGION
YTT
- 5100.
- 3000.
- 2000.
1000.
250. - 500.
100. - 250.
1. - 100.
0. - 0.
MEGRWflTTS
-------
Figure 4. TOTAL INSTALLED NUCLEAR-FUELED ELECTRICAL GENERATING CAPACITY. 1975
SIX-STATE REGION
3000. - 5>IOO.
S^ooo. - 3000.
1000. - 3000.
g]500. - 1000.
EJ250. - 500.
£3 100. - ?50.
Hi. - 100.
GO. - o.
MEGAWATTS
-------
Figure s. TOTAL PROPOSED COAL-FIRED ELECTRICAL GENERATING
CAPACITY ADDITIONS, 1976 1985
SIX-STATE REGION
3000. - 5100.
2000. - 3000.
1000. - ?000.
500. - 1000.
- 500.
- 250.
100.
0.
HEGHMRTTS
-------
Figure e. TOTAL PROPOSED NUCLEAR -FUELED GENERATING
CAPACITY ADDITIONS. 1978 1985
SIX-STATE REGION
3000. - 5400.
2000. - 3000.
1000. - 2000.
500. - 1000.
250. - 500.
100. - 250.
I . - HJO.
0. - 0.
MEGWMflTIS
-------
one-half of Indiana's scheduled capacity additions will be at sites along
the Wabash and White Rivers.
Nuclear-fueled units comprise a large portion of the total scheduled
capacity additions in Illinois, Ohio and Pennsylvania. Eighty percent of the
total 24,619 MWe of new nuclear capacity are for sites located outside of the
ORBES region. Although Indiana and Ohio have scheduled nuclear capacity ad-
ditions for the first time, the majority of the nuclear expansion continues
to be in parts of the region that had nuclear capacity in 1975. Nuclear
plants are located outside the major coal-producing areas, many of which have
seismic risks. Neither Kentucky nor West Virginia scheduled nuclear-fueled
units through 1985, thus continuing their preference for coal-fired electric
generating capacity.
Several trends in siting coal-fired and nuclear-fueled generating capac-
ity additions through 1985 in the ORBES region are apparent.
1. The size of electric generating units and plant sites is
expected to increase (Table 2).
This trend has accelerated rapidly since about 1960,
both in the ORBES region (Saper and Hartnett, 1979, pp.
10-21) and in the United States (Cirillo et al., 1977).
The generating unit additions and plant sites in the ORBES
region, however, are more than twice as large as the national
average (in 1974). As a consequence of these trends, fewer
sites are required for capacity additions.
2. The majority of the coal-fired capacity additions are sched-
uled for new and larger sites. Nuclear-fueled units are also
on_ new sites, although they are not significantly larger than
In"l975.
Sixty percent of the coal-fired units in the region and
96% of the nuclear-fueled units are scheduled for new sites.
By comparison, 75% of the coal-fired units in the nation are
scheduled for new sites (Cirillo et al, 1977). The East Cen-
tral Area Reliability (ECAR) Council has identified 12 of the
36 current sites in its area as "expandable," i.e. they can
physically accomodate some generating capacity additions.
According to Burwell, Ohanian and Weinberg (1979), three of
the four nuclear reactor sites in ORBES states can also ac-
comodate additional capacity.
Many of the coal-fired sites cannot easily accomodate
additional capacity because of air quality constraints in
heavily industrialized urban areas. Many of these are rela-
tively old, smaller units that can burn oil or other fuels.
In other cases, especially in the ORBES region, large units
can be added to existing sites if they burn low-sulfur western
coal, or if they add pollution control technologies. Few sites
are actually closed when existing units are retired, as utili-
ties retain them for use by types of capacity that use other
fuels.
15
-------
Table 2. SIZE, IN MWe, OF COAL-FIRED AND NUCLEAR-FUELED ELECTRICAL
GENERATING UNITS AND SITES, SIX-STATE REG TON
(Steam Units Only)
Fuel Size, in MWe
Coal Nuclear Mean Maximum
1975
Operating
Units •
Sites •
Units •
Sites •
190 1,300
546 2,932
862 1,098
1,724 2,19f>
1976-1985
Planned
Units •
Sites •
Units •
Sites •
549 1,300
1,349 2,751
~~l,02Ja 1,205
l,688a 2,410
Minimum
1
2
209
818
20
480
60b
810
SOURCE: Jansen (1978).
aMean site calculated excluding 60 MWe Shippingport experimental
Light Water Breeder Reactor.
Shippingport experimental Light Water Breeder Reactor.
16
-------
3. Capacity additions, especially coal-fired units, are sited
away from metropolitan areas and other concentrations of
population.
The majority of the coal-fired capacity additions in the
ORBES region are scheduled for areas with relatively low pop-
ulation density along the Ohio River main stem and its major
tributaries, as well as near the region's coal resources. At
national scale, the trend is toward mine-mouth siting with
most additions located at sites within 50 miles of adequate
coal supplies (Cirillo et al, 1977). Mine-mouth siting in
the ORBES region is an attractive option in the region pro-
vided that other resources are available.
Most nuclear-fueled capacity additions are located in
parts of the region that already have nuclear capacity.
Others are along the Ohio River main stem. As in the nation,
nuclear reactors in the ORBES region are actually sited closer
to population concentrations than are coal-fired units. In
this sense, issues of environmental constraints, especially
with regard to air quality, and public health and safety are
relative.
4. Joint ownership of generating unit capacity additions is ex-
pected to increase.
Three companies, plus Ohio Valley Electric Corporation,
in the ECAR region have generation that is located outside
of their service area. This represents approximately six
percent of the total generation. Elsewhere, capacity addi-
tions are scheduled for sites within the service area of the
utility that owns them, or that has majority control. How-
ever, joint ownership is common in the ORBES region. Approx-
imately 38% of the scheduled capacity additions from 1976 to
1985 will be in joint ownership, which will result in an in-
crease from 23.4% of the total MWe in 1975 to 30.5% of the
total in 1985. Although jointly-owned units are concentrated
in the eastern part of the region, the practice is expected
to spread to each of the ORBES states. Except for Pennsyl-
vania, all jointly-owned units in the region have at least
one out-of-state partner.
By 1985, 61% of the total coal-fired and nuclear-fueled generating capac-
ity is expected to be located in Illinois, Pennsylvania and Ohio. The most
significant growth is projected to be along the Ohio River main stem, where
the concentration of generating capacity would increase to 34.5% of the six-
state total; and along the Ohio's tributaries, where the capacity would in-
crease to 26.5% of the total. This is primarily because of the location of
new coal-fired capacity adidtions in Ohio, Indiana and Pennsylvania. These
states are expected to contain 54% of the total 11,572 MWe of coal-fired
generating capacity in 1985 (Figure 7). Almost three-quarters will be in the
Ohio River basin, with 40% located on the main stem, (as compared to 32%
17
-------
Figure 7. TOTAL COAL-FIRED ELECTRICAL GENERATING CAPACITY. 1985
SIX-STATE REGION
00
3000. - 5100.
3000. - 3000.
1000. - 3000.
500. - 1000.
250. - 500.
100. - 350.
I. - 100.
0. - 0.
HEGflMOTTS
-------
located there In 1975) and 34% on the tributaries. This projected growth
would significantly increase the concentration of electric generating capac-
ity along the main stem of the Ohio River between Portsmouth, Ohio and Louis-
ville, Kentucky.
Most of the nuclear electric generating capacity projected for 1985
will be located in Illinois and Pennsylvania (Figure 8). These two states
are expected to have 76% of the total 33,240 MWe nuclear capacity in ]985.
Only three sites (Zimmer in Ohio; Marble Hill in Indiana; and Beaver Valley
in Pennsylvania) will be along the Ohio River main stem. These three sites
would constitute 15% of the total nuclear capacity. The remainder is ex-
pected to be located outside of the Ohio River drainage basin in northern
Illinois and eastern Pennsylvania.
CHANGES IN THE SCHEDULE OF PLANNED ADDITIONS
The utilities constantly revise their announced plans for capacity addi-
tions. Deferrals and reduced commitments for new electrical generating capac-
ity already had begun to result in delaying construction schedules and on-line
dates in 1975 (Old, 1976; Rittenhouse, 1976). The net effect of these changes
over a one year period was to reduce the expected 1985 installed capacity by
1,926 MWe. While the MWe of postponed coal units was approximately equal to
the raegawattage of newly announced plants, the expected nuclear capacity had
a net reduction of 2,158 MWe because postponements were not compensated for
by newly announced units (Saper and Hartnett, 1979). Subsequently, the extent
of slippage increased, especially for nuclear-fueled units. The National Coal
Association has reported that in 1979, 57% of new coal-fired capacity, and 71%
of new nuclear-fueled capacity, experienced delays. The most common reasons
for the delays are:°
... revisions in forecast demand for electricity, delays in
siting or licensing, problems with preparation of environ-
mental data or financial uncertainties.
The changes in scheduled capacity additions in the six-state ORBES region,
especially in Illinois, Indiana, eastern Ohio and Pennsylvania, have followed
the general national slowdown in new power plant construction.
19
-------
Figure 8. TOTAL NUCLEAR-FUELED ELECTRICAL GENERATING CAPACITY.19B5
SIX-STATE REGION
ro
O
3000. - 5MOO.
^QOQ. - 3OOO.
1000. - 3000.
500. - 1000.
250. - 500.
100. - 250.
I. - 100.
0. - 0.
HEGRHRTIS
-------
FOOTNOTES
The generalized siting process represents a synthesis of the procedures
used by 26 electric utilities for siting nuclear reactors in 1973-1974. The
NRC stated that there was no reason to assume that the basic procedures had
changed substantially since then (U.S. NRC, 1976).
2Detailed evaluations of siting methodologies are in: Keeney et al (1978);
and Hobbs and Voelker (1978).
3The NRC currently is considering adopting new rules that will affect
siting. These rules are intended to reflect the experience gained since the
original siting regulations were published in 1962. They are meant to apply
to facilities for which an application for construction permit is filed after
October 1, 1979 (Energy Users Report. August 7, 1980, p. 10).
These are reviewed in detail by McLaughlin (1980).
5Seventy-two of the 524 counties in the six-state region border the main
stem of the Ohio River, and 276 are along major tributaries (i.e., the Wabash,
Great Miami, Scioto, Muskingum, Allegheny, Monongahela and Kanawha). The re-
mainder of the counties in the six states are in basins that drain to the
Mississippi River (Illinois River), the Great Lakes and the Atlantic Ocean
(Susquehanna and Delaware Rivers).
See Appendix G.
According to Burwell, Ohanian and Weigberg (1979), the Clinton site in
Illinois cannot be expanded beyond current utility plans. Other nuclear sites
in the ORBES region can accomodate additional capacity.
8Energy Users Report, December 20, 1979, p. 12. According to a recent
DOE analysis of construction delays of coal-fired generating units in the
first nine months of 1979, utilities in the ORBES region had a net loss of
26,245 MW-M, all of which was accounted for by delays in Pennsylvania and
Ohio. On-line dates were advanced for units in Kentucky and Indiana (U.S.
Department of Energy, Economic Regulatory Administration, 1980).
21
-------
SECTION 3
THE OHIO RIVER BASIN ENERGY FACILITY SITING MODEL
Forecasting the future geographical arrangement of energy facilities Ls
an essential step in converting energy supply scenarios into a geographical
pattern of impacts that can be assessed. Economic and technological factors
dominate decisions about the location of large facilities at national scale
(Willbanks and Calzonetti, 1977). At regional scale, the consideration of
environmental impacts and resource use are important influences in the future
geography of electrical generating units and similar energy facilities.
The ORBES siting model is designed for use in regional technology assess-
ments of energy development scenarios that emphasize electrical generation.
It incorportates selected features of other regional siting models into a pro-
cess that simulates current and projected siting practices of utilities in the
region, and that is sensitive to regional resources and values. Scenario pol-
icies that change these conditions may also result in significantly different
siting patterns. The assessment of changes in impacts will help to anticipate
the political geography of future energy supply policy.
REGIONAL-SCALE ENERGY FACILITY SITING MODELS
The role of energy facility siting models in regional technology assess-
ments is to translate energy supply scenarios into a geographical pattern of
impacts. These scenarios, and their associated policies, generally specify
levels of energy production for some future date, or dates, as well as the
distribution of that production among several sources and technologies. The
scenarios generally describe conditions at national, and sometimes regional,
scale. Environmental assessment models, however, generally require more pre-
cise geographical locations of projected facilities in order to provide useful
information about their cumulative impact under different policies. The crit-
ical issue is geographical scale (Meir, 1977c; Palmadeo, 1976).
Within a region, the problem is how to distribute relatively large numbers
of hypothetical energy facilities in a manner that is consistent with scenario
policies, energy technology mix, and the available resources and regional values.
Several comprehensive models have been developed to solve this problem. One
group, which Church and Hillsman (1979) refer to as "regional siting policy
models," define optimal siting patterns to meet the constraints imposed by pub-
lic policy and regulations (Eagles, Cohen and ReVelle, 1980; Meir, 1977a and
1977b- and Provenzano, 1978). Baseline siting simulation models constitute a
second group (Davis et a], 1978; Dobson, 1979; and Van Horn, Liroff and Hirata).
These models project plausible, rather than optimal, facility locations in con-
sideration of public policy and regulatory constraints. Although neither ap-
proach will predict actual sites for individual facilities, they will provide
the basis for assessments of the trade-offs that may be involved under different
22
-------
policy options. This type of information is necessary for informed decision-
making about the political geography of energy supply policy.
Each model has certain parameters that are specified by the user. These
parameters, and their definition in the models reviewed here, are
1. The regions-of-interest (ROI) are generally composed of
contiguous states located in the eastern part of the United
States.
The multi-state regions also coincide with the extent of
power pools or other energy planning regions. Individual states
and subregions of a state may also be significant for energy
planning.
2. Most scenarios project current (i.e., mid-1970s) conditions
to the year 2000 or 2020.
The "future" is divided into two periods. In the near-
term—for example, 1975 to 1985—the geographical distribution
of electrical generating facilities depends upon existing and
announced utility plans. Beyond the mid-point, the siting
model is responsible for locating capacity additions that
are required to meet final year production.
Most of the models are static. That is, they describe
siting patterns for particular years—1975, 1985 and 2000 or
2020. Consequently, impact assessment depends upon the defi-
nition of incremental change, for each scenario, from one end
year to another. The Utility Simulation Model (Van Horn,
Liroff and Hirata) is the only siting model to schedule elec-
trical generating capacity additions on a continuous, year-by-
year basis.
3. The energy supply technologies are large, central station base-
load steam electrical generating units that are fueled by either
coal or uranium.
The units range in size from 850 MWe to 1100 MWe, and may
vary in terms of cooling options. Otherwise, they are conven-
tional technologies. The type and mix of technologies that a
scenario projects are important because they affect the policy
issues and options, as well as the evaluation of resource re-
quirements and the range of expected environmental impacts.
Alternate technologies are usually considered to the extent
that they may reduce the amount of energy that conventional
facilities must supply.
4. "Sites" are either counties, minor civil division, or small
rectangular cells based upon conventional map grid coordinates.
23
-------
The "site" is the smallest common geographical area in
the model. It is the unit for which data on siting variables
are gathered, and in which facilities are located. Compared
with the counties, the cells in rectangular grids are smaller
and uniform in size and shape. They provide a consistent base
for collecting and analyzing environmental information. How-
ever, a wider range of socioeconomic data are available for
counties.
5. The suitability of sites for new energy facilities are evaluated
according to issues related to the technological characterisLics of
a facility^ its resource requirements, and regulatory cpnstraints
that affect its location and operation.
Air quality, water availability, land use and ecological
impacts and fuel resources are common issues in siting coal-fired
generating units. Public health and safety is important also
for nuclear-fueled units. These are general concerns throughout
each region and can be defined for large areas. Wherens soi-io-
economic issues are represented to some degree, public accepta-
bility is not considered in the siting models.
Regional Energy Supply
Energy scenarios specify total regional energy supply, or production,
for some future time and the mix of technologies that will provide it. In
order to be useful for siting models, this information must be disaggregated
within a framework of smaller geographical areas in the region. States,
economic regions (e.g., Bureau of Economic Analysis regions) and utility ser-
vice areas have been used, with the total population in each subregion as the
most common denominator for disaggregating regional energy supply projections.
Within each subregion, supply may be assigned to one or several "load centers"
(usually cities and metropolitan areas), also on the basis of population.
The exchange of energy across subregional boundaries is not considered.
Existing generating capacity and planned additions represent a porLJon
of the future geographies of energy supply facilities. In most cases, these
are considered to be sufficient to meet supply in the near-term, e.g. from
1975 to 1985. In the long-term, most if not all of the supply presumdbly
will be provided according to scenarios that specify the mix and characteris-
tics of technologies, as well as policies that may significantly alter siting
patterns. The ANL (1978) model assumes that the sites of current (1975) and
planned (1985) capacity will have the same size and type of plants in 2000 and
2020. They argue that utilities will add new units of the same or larger
size if older plants are retired from existing sites rather than compete for
scarce new locations. The schedule of unit retirements can be a significant
factor in the long-term, especially if "new" units are assigned a different
fuel. This is the case in the Northeast region, where a large portion of the
existing facilities use oil and may be replaced by coal-fired units (Meier,
1977a).
-------
Definition of Site Suitability
Definition of the suitability of sites as locations for generating unit
additions usually involves three steps:
1. Specific criteria are selected that can be used to define
the compatability of site characteristics with technological
and regulatory siting issues.
2. The region is screened to exclude those sites that are not
likely to have a suitable location for a given facility.
3. The remaining candidate sites are compared with one another
in order to define their relative suitability as locations
for future facilities.
The usual procedure is to project current conditions into the future, and then
change the definition of suitability to accomodate different technologies; reg-
ulatory policies, especially with respect to environmental controls; or even
resource availability, expecially coal production. This is accomplished by
the selection of exclusionary criteria and the relative importance that is
given to each factor in defining the suitability of candidate sites.
The definition of site suitability is sensitive to the choice of criteria
that represent the siting issues, and their translation into measures of com-
patability with a particular technology or policy. Most models depend upon
a small a priori list of criteria that are closely related to the technological
characterisitcs of the facility and regulatory issues, and that can be measured
with readily-available data. The issue of water availability, for example, is
measured by the consumptive use or withdrawal requirements of a particular
cooling system relative to the low flow stream volume under drought conditions,
In turn, the value of a criteria at each site is assessed on a common scale
in terms of its "compatability" with the technology and regulatory environment.
[f safety is an issue for nuclear-fueled units, densely-populated areas are
less-desireable (i.e., less 'compatible') as sites than are sparsely-populated
areas. The expected relationship is inverse. In the case of water availabil-
ity, however, locations that are close to streams that have large consistent
flows are highly valued (i.e., more 'compatible') as sites for all types of
technologies. The choice of campatability scales can vary considerably unless
technological constraints of the facility or regulatory rule-making sets some
standard of performance, or resource requirement that define threshold values
for siting criteria.
Exclusionary screening uses threshold values to filter out those sites
that are not likely to contain suitable facility locations.^ The translation
of these values into siting constraints may vary widely, especially where
proximity to an area that is unlikely to contain a suitable site is involved.
The issue of separation distances and buffer zones in relationship to air qual-
ity issues is a case in point. The definition of exclusionary criteria and
thresholds may change to reflect different policies and regulations. Conse-
quently, they may significantly affect the geographical distribution and cha-
racteristics of candidate regions, as well as the nature and concentration of
25
-------
the resultant impacts.
Finally, the candidate sites are compared with one another according to
the selected criteria in order to determine their suitability as locations for
new facilities. Weights assigned to each criteria reflect their relative
importance in the siting decision. Models that concentrate on the influence
of a single criteria, such as water availability, usually assign weights on
an a priori basis. Models that are concerned with decisions that involve
numerous factors are more likely to rely on the consensus of expert panels
to define the relative importance of each. Optimization models, on the other
hand, define suitability in terms of objective functions such as minimizing
costs of transportation (of electricity or coal), augmenting water supplies,
or environmental impacts. Because of importance weights, definition of site
suitability can incorporate the effects of specific technological and regula-
tory changes that affect new facilities, as well as general shifts in the
policy environment.
The way in which the models define site suitability is subject to three
criticisms. First, the procedure is basically judgemental. Technical jus-
tifications are necessary to support each decision, although some models
are ambiguous. The importance weights are supposed to reflect group values,
although siting "experts" dominate the panels. Second, none of the models
actually determine whether or not different sets of policies and regulatory
decisions create significantly different suitability patterns. Keeney et al
(1979) report that whereas lists of the most suitable, and the least suitable
sites are not likely to change, the rankings of sites in the medium suitability
range can shift significantly. This is important in the long-term, as new
facilities are more likely to be located in such places as the few best sites
are developed. Third, the siting criteria generally do not change through
time. Except for population projections, current conditions are assumed for
all future time periods. They do not change, even to reflect the synergestic
affects of incremental siting decisions.
Allocation of Additional Facilities
In the third phase of the siting models, the number and type of new facil-
ities that are necessary to satisfy total supply are allocated to locations
within each subregion, or siting region. For each technology, the facilities
are allocated according to:
1. Site suitability scores7
2. Proximity of a site to a load center or fuel resource
(e.g., mine-mouth siting for coal-fired units), con-
strained by one siting criteria (usually water availa-
bility) or site suitability scores.
3. Objective functions, such as minimization of transmission
or transportation costs or environmental impacts.
Each model also sets a maximum total capacity that can be located at a single
site. The result is to project a type of "dispersed" siting policy into the
26
-------
future.8 The maximum can be increased to simulate the concentration of facil-
ities in energy parks (Argonne National Laboratory, 1977a). Because all new
facilities must be located in the region to which they are assigned, none of
the models account for practices of joint ownership of facilities in other
states.
The result is a geographical pattern of energy supply facilities in the
region that includes two sets of facilities. One is sited according to exist-
ing or near-term conditions as evaluated by utility planners. The other is
developed by the siting model according to technologies and policies that are
integral parts of long-term energy supply scenarios. Teknekron's Utility Sim-
ulation Model provides a year-by-year schedule of on-line dates for new facil-
ities in addition to the announced utility schedules for planned units. All
other models provide only aggregate patterns for a particular year.
THE ORBES REGIONAL SITING MODEL
The ORBES regional energy facility siting model is a hierarchical, linear-
weighted model that allocates, at county level and according to scenario poli-
cies, base-loaded coal-fired and nuclear-fueled electrical generating units in
addition to those that are already in service or are planned in order to reach
some future total regional energy supply (Figure 9). The ORBES model incorpo-
rates selected features of other regional assessment siting models within a
framework that is designed to facilitate policy analysis. The model has three
interrelated modules:
1. Disaggregation of total regional energy supply, by fuel
type and technology mix, to siting regions.
2. Definition of candidate site suitability, by fuel type
and scenario energy policies.
3. Allocation of generating unit capacity additions, by fuel
type, to county-level sites according to scenario siting
policies.
Policy changes can be simulated provided that they are functionally related by
policy issues to some aspect of the model. Because the process is determinist-
ic, the incremental changes in impacts that result from policy changes that
affect siting can be estimated.
The siting model depends upon regional energy supply scenarios for three
pieces of information. They are:
1. Total regional energy supply, by fuel type and technology,
for some future year(s).
2. Technological characteristics of the generating units that
are to be sited in the long-term (i.e., beyond utility plans).
3. Policies that may affect site suitability or siting procedures.
27
-------
Figure 9. OHIO RIVER BASIN ENERGY STUDY ENERGY FACILITY SITING MODEL
28
-------
The total regional energy supply data, along with existing and planned capacity
additions, provides the basis for calculating the amount, type and subreglonal
distribution of projected, unmet demand within the region of interest. The
technological characteristics of the scenario unit additions are used to cal-
culate the number that are to be sited, as well as to help define the siting
issues and data requirements. Siting issues include those considerations that
are relevant to the location of scenario unit additions of concern to the as-
sessment and the policies it addresses.
The final production of energy from electrical utilities in the ORBES re-
gion in the year 2000 is allocated to state subregions on the basis of the dis-
tribution of projected supply, by fuel type, in 1985. The existing generating
capacity in 1975, and scheduled capacity additions from 1976 to 1985 for which
county-level sites have been announced, is then subtracted from the total re-
quired capacity in the year 2000 to determine the total unsited capacity addi-
tions. Announced and expected retirements add to the total. The total unsited
capacity addition for each state subregion is translated into the number of
standard base-loaded coal-fired and nuclear-fueled scenario units, as specified
by the scenarios, that is to be located according to the site suitability of
ORBES counties and the allocation procedures for each scenario. Electricity
generation with alternative fuels and technologies, and the impact conservation
measures, are considered to the extent that they reduce the capacity additions
in conventional technologies.
Siting issues represent scenario policies and technologies that affect the
resource requirements of the electrical generating units or the regulatory en-
vironment within which they operate. The primary issues are: air quality;
water availability; land use and ecological systems; seismic suitability;and
public health and safety. Each of these components is represented by one or
more specific criteria for which quantitative data are collected at county
scale. Threshold values that define some minimum expected resource requirement
or performance level for a given criterion are used to exclude from considera-
tion those counties in which the likelihood of finding a suitable site is low.
The remaining candidate counties are evaluated according to their relative suit-
ability as sites for either coal-fired or nuclear-fueled scenario unit addi-
tions. Site suitability is determined by a two-step, hierarchical linearly
weighted model. Each site is given a standardized score for each criterion,
and weights derived from an expert panel are used to indicate the relative im-
portance of each major component in siting decisions in the region. The result
is a set of descriptions of site suitability for candidate counties across the
region that varies by fuel type and environmental policies.
The unsited generating unit additions are then allocated on a state-by-
state basis to candidate counties according to their suitability indices sub-
ject to locational constraints and scheduling patterns. In most scenarios,
existing and announced utility plans provide sufficient capacity additions to
meet supply projections in the near-term, i.e. from 1975 to 1985. Scenario
unit additions provide most of the additional supply required from 1986 to
2000. In the near-term, the impact assessment focuses upon utility projections,
and is consistent among most scenarios. In the long-term, impacts are directly
related to scenario policies that change the level of supply or its distribution
29
-------
within the region; or any component involved in defining site suitability.
The comparison between such policies. The results may suggest either a ch.-mge
in scenario policies or the redefinition of siting issues.
SCENARIOS AND SITING POLICIES
Two basic groups of scenarios are considered (Table 3).^ The first as-
sumes that all scenario unit additions will use conventional technology, with
coal as the primary fuel. No nuclear-fueled units are sited after 1985 except
those that the utilities had announced in 1975. Scenario 2, which assumes
high rates of economic growth, base case environmental control policies and
other current conditions, including siting policies, is the point of reference
for the coal-based scenarios. Whereas growth rate assumptions affect the num-
ber of scenario units that need to be added to projected additions, environ-
mental controls and siting policies directly affect the geographical distribu-
tion of those units within the ORBES region. Scenarios la, Ib, Ic and Id are
specifically designed to assess the impacts of selected changes in environmen-
tal and siting policies within the context of strict environmental controls.
In the case of Scenario 2a, additional scenario units that are dedicated to
export electricity to the Northeast are located in the eastern part of the
region.
The second group of scenarios emphasizes fuel substitution and conserva-
tion. Scenario 2c emphasizes nuclear-fueled capacity additions after 1985.
Others assume that other fuels (Scenarios 3 and 4) or conservation (Scenario
6) will dominate energy supply in the long-term. These scenarios have the
same environmental controls and siting policies as Scenario 2. The number of
coal-fired scenario unit additions that are sited, however, differs signifi-
cantly. In some cases, this has the effect of changing the schedule of on-line
dates for electrical generating capacity additions.
30
-------
Table 3. BBSdimoH or SCENARIOS AID SITING POLICIES*
rrlBary Fuel
Scenario Technology 1986-2000
growth Ratea
Iniriy
Bconoalc
Environmental
Control Policy
Siting
Policy
Conventional
CM!
Bl|h
Strict
Strict
1*
Ib
Ic
Id
[Very atrlngent] Dlapareed
•It quality J Concentrated
[ Agricultural 1 Dlepereed
(land* protection Concentrated
2
Conventional
Coal
High
Bate Caie
Bau Caa*
2d
21
2a
2a2
Coal-fired
exporta
Lax air quality
atandarda
Once-through
cooling for
planta on Ohio
River aaln aten
Once-through
cooling for
planta on Ohio
River aain ate«
Coal-fired
exports
2c
Conventional Nuclear
High
Bale Caie
Baie Caae
2b
2bl
Nuclear-fueled
exporta
Once-through
cooling for
planta on Ohio
River aaln aten
Nuclear-fueled
exports
Alternative Alternative
High
Baae Caie
Baie Case
4 Conventional Natural Caa
High Baae Caie Biae Caie
3 Conventional Coal
Low Baae Caie Baae Caae
Very High
6 Conventional Coal Very Low High Baae Caie Baae Caie
7 Conventional Coal
High High Baae Caae Baae Caae
Laaal ealaalona
diapatch
•The baalc acenarloe are ancloaed in boxee. followed by other acenarloa that are de-
rived fro* the* in order to aaieia changea in iBpacti that Bight occur aa the remit of
apeclfle policy optima. The policy optima are apeclfied In the deacriptlona of derlve-
tlve ecenirloa.
31
-------
FOOTNOTES
siting models developed by Argonne National Laboratory (1978) Brook-
haven National Laboratory (Meier, 1977) and Oak Ridge National Laboratory
(Davis et al, 1978) were used in the National Coal Utilization Assessment. The
Argonne model incorporates previous work on SITE (Frigerio et al, 1975) where-
as Oak Ridge used previous work for the Maryland Power Plant siting Program
(Dobson, 1979).
2The Argonne model distinguishes coal-fired units by source of coal (e.g.,
in-state coal and imported high sulfur and low sulfur coal). A special siting
pattern for Illinois is developed for a high energy growth scenario using Illi-
nois coal. Coal gasification plants are sited in Illinois and Indiana, and
coal liquifaction plants are sited in the region.
^Economic costs generally concern either coal transportation or electric-
ity transmission, whereas population density is an indicator for 'social1 im-
pacts. Socioeconomic issues are not defined as a matter of resource use,
technological constraints or regulatory decisions.
AThis procedure is identical to the exclusionary screening and comparative
evaluation procedures used by utilities except that the regions of interest and
'sites' are larger, and several technologies may be involved. See the discus-
sion in Section 2.
5Contrary to the practice of other models, Eagles, Cohon and ReVelle
(1979) include exclusionary criteria that are not used in comparative evalua-
tion. Most are land use and ecological criteria.
6This is especially true of the ORNL model, which defines site suitability
by a linear-weighted model. A large number of siting criteria are used to de-
fine the issues, and then are assigned weights that are intended to reflect
different siting objectives (Dobson, 1969). Although these may be expected to
reflect the values of different groups, only siting experts participated.
7Most models are deterministic in the sense that they assign new facili-
ties to sites according to rank order on the site suitability scale. In the
Utility Simulation Model (Van Horn, Liroff and Hirata, 1980), site (county)
weights are converted into cumulative probabilities and generating unit addi-
tions are allocated by a Monte Carlo method.
8The ORNL model (Davis et al, 1978) also uses the maximum to redistribute
'excess' planned capacity when disaggregating energy supply scenarios. This
creates an even more dispersed geography of energy supply in the long-term.
9The scenarios are discussed in detail in Page and Stukel (forthcoming).
32
-------
SECTION 4
SCENARIO UNIT ADDITIONS AND SPATIAL ALLOCATION PROCEDURES
The energy and fuel demand model projects the total electricity produc-
tion for each scenario, by fuel type, for the ORBES region in the year 2000.
Generating units that are in service in 1975 supply a portion of the total
production in each scenario. The number of additional coal-fired and nuclear-
fueled generation units that is required to satisfy the incremental produc-
tion from 1976-2000 is calculated on the assumption that it will be supplied
by a combination of generating units for which utilities have announced on-
line dates after 1975, and a sufficient number of standard generating units
for the 1986-2000 period to account for the necessary capacity additions as
well as the retirement of older units. The schedule of capacity additions
also combines the announced utility plans with a linear schedule of the sce-
nario units added after 1986.
Procedures for the geographical distribution of generating capacity addi-
tions at county scale also combines announced utility plans and scenario
models. The locations of the announced capacity additions are generally
known. The scenario unit additions, however, are allocated to counties
within state subregions according to a procedure that takes into considera-
tion their relative suitability as candidate sites for coal-fired and nuclear-
fueled units.
CALCULATING SCENARIO UNIT ADDITIONS
A standardized procedure for calculating capacity additions for each sce-
nario was used to determine the number of coal-fired and nuclear-fueled sce-
nario unit additions that will need to be sited in each state subregion
(Tables 4 and 5). Information on sited electrical generating capacity in
1975, and near-term (1976-1985) changes in capacity, are from the Electrical
Generating Unit Inventory (EGUI) (Jansen, 1978). Sited capacity in 1985 is
calculated by adding the 1976-1985 additions and removals (negative signs
are for removals) to the 1975 capacity, according to electric utility plans
announced at the end of 1976 and reported in the EGUI.
Sited capacity in the year 2000 is calculated by adding the 1986-2000
additions and removals to the 1985 figures, assuming an average useful life
of 35 years for units that had no announced retirement dates. Because com-
prehensive data on planned capacity additions and removals were available
only through 1986, the useful life of existing units was estimated from data
in the EGUI. An analysis of actual and projected retirement dates in rela-
tion to on-line dates for electrical generating units in the study region
indicated that units were retired after an average 35 years of on-line ser-
vice. ^ Consequently, generating units that had no announced retirement date
were removed after 35 years of service. Units that had an announced retire-
ment date were allowed to remain in service until that date, and units that
had neither an on-line date nor a retirement date were retired in 1985.
33
-------
Table 4. MFTHODOLOCY FOR CALCULATING UNSITED ELECTRICAL GENERATING UNIT ADDITIONS
Worksheet
Column Number
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
Title
SUBREC10N
FUEL
1975 SITED CAPACITY, MWe
1976-1985 SITED
ADDITIONS. MWe
1976-1985 REMOVALS. MWe
1985 SITED CAPACITY, MWe
STATE SHARE. Z
1986-2000 SITED
ADDITIONS. MWe
1986-2000 REMOVALS. MWe
2000 SITED CAPACITY. MWe
2000 SCENARIO x 10 15 Btu
Source
or
Method of Calculation
ORbES portion of each state
Coal (650 MWe sitlnfi increments)
Nuclear (1000 MWe siting Increments)
Electrical Generating Unit Inventory (ECU!)*
EGUI
ECU I, assuming 35 uear unit life
(3) + (4) + (5) - (6)
(State MWe1985/Total MWe1985>c N * 10° ' <7>
ECU1
ECUI. assuming 35 year unit life
(6) + (8) + (9) - (10)
(Total). - 0.95 x fossil electric utilities production from
(12)
(13)
(14)
(15)
energy and fuel dccer.d r-odel""
(Totol)N • nuclear electric utilities product lor fron cmrgv
and fuel model**
iN - (Total)CiB x
- (11)
2000 SCENARIO, MWe
2000 UNSITED CAPACITY, MWe
2000 UNSITED UNITS, I
2000 RESIDUAL, MWe
13
For coal ((ID/I 49 x 10Btu/MWe) at 50Z capacity f.icior
For nuclear KlD/1.94 x 1010Btu/MWe] at 651 capacity factor
(12) - (10) • (13)
For coal (13)/650 MWe rounded to nearest integer
For nuclear (131/1000 MWe rounded to neon-si Integer
For coal [(14) x 650 MUe) - (13) - (15)
For nuclear 1(14) x 1000 MUe] - (13) - (IS)
*Jansen (1978).
**Page, Gllnore and Hewlnga (1980).
-------
Tabl« S. UNSITED ELECTRICAL CLNERAT1NC UNIT ADDITIONS
(Jl
SUBRECION
(1)
ILLINOIS
INDIANA
KENTUCKY
OHIO
PENNSYLVANIA
WEST VIRGINIA
TOTAL
FUEL
(2)
Coal
Nuclear
Coal
Nuclear
Coal
Nuclear
Coal
Nuclear
Coal
Nuclear
Coal
Nuclear
Coal
Nuclear
1975
SITED
CAPACITY
HUe
(3)
10,512
1.865
10.114
™~
10.948
—
17.034
9.691
—
11.966
70.265
1.865
1976 -
SITED
ADDITIONS
NWe
(4)
4.399
4.056
8.951
2.260
8.880
—
3,927
810
6.134
1.830
2.552
^™
34,843
8.956
1985
REMOVALS
HUe
(5)
-511
-534
—
-837
—
-1.438
—
-336
—
-582
—
-4.238
1985
SITED
CAPACITY
MVe
(6)
14,400
5.921
1B.S11
2.260
18.991
—
19.523
810
15.489
1.830
13,936
~~~
100.870
10.821
STATE
SHARE
Z
(7)
14.28
54.7
18.37
20.9
18.83
0.0
19.35
7.5
15.16
16 9
13.82
0.0
100.01
100 0
1986 -
SITED
ADDITIONS
HUe
(8)
1.273
—
1.000
—
2.150
—
810
—
—
—
4.423
810
2000
REMOVALS
MUe
19)
-1.631
-209
-5,272
—
-4,962
—
-5,160
—
-2,525
-1,670
—
-25.220
-209
2000
SITED SCENARIO
HUe X 1015Btu MUe
(10) (11) (12)
12.042
5.712
14.259
2.260
16.179
14.363
1.620
12,964
1,830
10.266
—
80.073
11.422
UNSITEO UNSITED
HUe f HUe
(13) (14) (15)
Column does not add to total due to rounding.
-------
The ORBES energy and fuel demand model (Page, Gilmove and Hewings, 1980)
provides the total regional electricity production for each scenario (Table
6, for example). The total production figures for the fossil electric utili-
ties sector and the nuclear electric utilities sector are converted to quad-
rillion Btu's (quads).2 Fuel-specific electric production totals in the year
2000 then were apportioned to state subregions according to the states' pro-
jected shares of electricity production in 1985. This procedure preserves
the state subregions1 relative rank in the region with respect to their use
of coal and uranium as electricity generating fuels. It also assumes Lhat
Kentucky and West Virignia will not acquire nuclear-fueled generating capac-
ity by the year 2000, as neither state had nuclear capacity planned for opera-
tion by 1985. Similarly, Illinois will maintain its lead in nuclear-fueled
capacity because it had more than one-half of the region's nuclear capacity
planned for 1985. The distribution of future coal-fired generating capacity
was similarly apportioned. For example, Ohio is allocated the largest share
of coal-fired capacity from 1985-2000 because it had more coal-fired capacity
planned for 1985 than any other state subregion.
The scenario electric utilities production figures for the year 2000 are
converted to megawatts (MWe) of generating capacity. The conversion takes
into account the Btu's in a megawatt-hour of electricity (3.412 x 10° Btu/MWH);
the number of hours in a year (8760 hr/yr); and the capacity factor (CF) of
the generating equipment. Although the first two numbers are invariant, CF,
which is defined as the ratio of the average load on a piece of generating
equipment to the equipment's capacity rating, varies widely. Estimated CFs
for coal-fired generating units in the region averaged 53.3% in 1976 and 50.9%
in 1977. Nuclear-fueled units had CF's that averaged 54.7% in 1976 and 58.3%
in 1977. Sample data for units in the region with capacities of 400 MWe or
greater for 1967-1975 show a steady decline in CF's for coal-fired units from
a range of 60-80% for state-wide values in 1967 to 40-60% in 1975 (USDOE,
1978). In systems that have a mixture of coal-fired and nuclear-fueled gen-
eration, the nuclear units commonly are loaded first and, consequently, have
higher CF's than coal-fired units in the same system. Uranium is less expen-
sive than coal on the basis of the energy content of the fuel, and nuclear-
fueled generating equipment is more expensive than coal-fired equipment so
it is to a utility's economic advantage to generate all the electricity it
can from its nuclear units. Utility plans in the region indicate a trend
toward a mixture of coal-fired and nuclear-fueled generation. Consequently,
coal-fired scenario unit additions are assigned a CF of 50% and nuclear-fueled
scenario unit additions are assigned a CF of 65%.
The conversion from quads of electricity production to MWe of generating
capacity used conversion factors based on the Btu/MWH, hr/yr and the CF's. Thus,
coal-fired capacity was the result of dividing the coal entries in Table 5,
Column (11) by 1.94 x 1010 Btu/MWe. The total MWe of capacity needed to sup-
ply the electricity production that the scenarios project, less the sited
capacity either in place or planned, is the unsited capacity that must be
allocated to sites, according to some scheduling pattern, in order to assess
the impacts in the long-term, i.e. from 1986-2000. This unsited capacity in
each fuel type is then divided into a number of standard scenario unit addi-
tions. Coal-fired units are assigned a nominal electrical generating capacity
of 650 MWe and nuclear-fueled units are assigned a nominal capacity of 1000
36
-------
Table 6. 1974 BASELINE DATA, 1985 AND 2000 SCENARIO 1 SOLUTIONS TO THE ORBES ENERGY DEMAND MODEL
(Sectors 1-24 are in Trillion Btu's while Industries 25-67 are in Millions of 1967 Dollars)
1 1 nj t UrnunU
1
2
3
C
6
7
c
V
111
11
12
1 3
15
16
1?
1 h
19
20
61
62
63
64
65
66
67
cOtfl • i n i I-)
crude j.etroleu»» Tas
shjle oil
qas l t led C»a I
so 1 v»ri t • re f i ne j cf»al
rcl'rt |.etrul»ur. products
n^itur|nel» elec. util's
or»- reuuc t ion tee-Ik tucks
ch«o.icdl feedstocks
Mater transiort
air transport
truck/ bus transport
full t r j n s t o r t
ou t c t r ans i>or t
• isc. ,: h e r ii a J uses
re.il estate
hotels, loonna* jnd •ausearnts
nisc liustncss ond personal serv.
advert i s in i
autu repair
•edical and educational serv.
nonprofit organisations
1V/4
L
li
U
0
U
I)
ll
0
•J
U
0
0
U
12
0
0
U
'1
240
U
V55V
1751
22u1
22
10/2
4630
1298
1Vh5
0
U
0
0
0
0
0
0
(j
0
0
0
0
17
0
0
0
0
331
U
12543
22VS
2MC
29
1407
6U76
1/U3
c. 000
U
U
U
U
0
U
n
0
U
II
U
0
0
26
U
U
u
U
5 1.6
U
20145
3&VO
463V
47
225V
9 /5/
2735
lu l o I pr ouuC t ion
1 W4
10964
6b3
C
C
0
3302
C
10/4
?3
C
13
215
325
15
26
45
1 7
254
11VC
V4^/
2KV5
3712
1 167
1006
4409
1684
19b'j
15254
722
0
0
0
2175
3273
0
1587
210
0
35
2H2
434
19
34
c'l
2 1
3*9
1542
1244/
38U2
4M7n
1562
2112
5/86
2210
2100
21.682
7C3
C
0
0
2V75
0
2233
210
C
35
39C
7U4
21
4V
HC
33
532
215C
1V1V3
5/54
72hV
2177
3189
•»IMO
3499
tutftl consult pt ion
1 V74
4842
542V
C
0
C
31fl6
2176
C
814
in
C
1C
215
325
15
26
45
1 7
254
119C
14106
3577
5i.15
2215
1t>/9
4754
1492
19c5
6736
573K
0
0
0
32e7
2157
0
1083
143
It
24
2r2
434
1V
54
60
It
349
1542
1*507
4f ">9
65V2
2915
2471
6258
1V5V
2000
V133
5507
0
0
0
35U4
1961
0
16V2
15V
C
if
390
704
'el
4V
to
53
532
2150
2^37
/11 1
VJi4V
4061
3/31
10005
3102
SOURCE: Page, Gilmore and Hewings (1980, Appendix A, Table A.I).
-------
MWe. These values are consistent with recent trends in utility unit size
selection in the ORBES region.
PROJECTED ELECTRICAL GENERATING UNIT ADDITIONS FOR ORBES SCENARIOS
The number of unsited electrical generating units in each state subregion
was obtained by dividing the unsited coal-fired (or nuclear-fueled, when appro-
priate) electric generating capacity by the nominal unit size, and rounding
to the nearest integer. The number of unsited units calculated on the basis
of state shares was adjusted to match the number of unsited units calculated
for the regional total, as rounding errors sometimes resulted in discrepancies
between the sum of the state subregions and the regional total. This adjust-
ment minimized the residuals for each subregion subject to the constraint that
the total residual for the region also be minimized.
The projected number of coal-fired and nuclear-fueled scenario unit addi-
tions were calculated for each scenario (Table 7).4 The final electric demand
in the year 2000, and the choice of technology and fuel mix, are the principal
determinants of the variations in the number of scenario unit additions. In
Scenarios 2a and 2b, the assumption that additional generating capacity will
be distributed according to state shares was changed according to policies re-
lated to the export of electricity to the Northeast. In Scenario 7a and in
variations of those scenarios that have very strict environmental control pol-
icies, some of the coal-fired units assigned to Ohio are sited in other states.
Otherwise, environmental controls do not affect the number of scenario unit
additions or their distribution among the states.
SCHEDULE OF CAPACITY ADDITIONS
The schedule of on-line dates for capacity additions through 1985, and
for the few units that are planned beyond 1985, is based upon announced utility
plans. In most instances, the locations and tentative on-line dates for all
planned additions are known. Although scheduling in the long-term is more
difficult, the timing and order of plant construction for the additional units
required to meet projected scenario demand in the year 2000 are necessary for
impact assessment. The schedule chosen may be desireable for policy analysis,
the base case should approximate utility behavior. Changes in the announced
plans for near-term (i.e., 1976-1985) capacity additions should be minimized
whereas the schedules for scenario unit additions post-1985 should follow the
aggregate pattern for planned units in the region and the state subregions.
Scheduling new plant construction is an integral part of utility system
planning (cf. Poldasek, 1977). Load forecast, reserve margins and the average
size of electric generating units are the primary variables. If the reserve
margin falls below the level considered necessary to meet projected peak loads,
the utility may either purchase power from neighboring utilities or consider
installing new capacity. If the utility decides to install new capacity, the
schedule depends upon a projection of when the deficit in the reserve margin
will occur, and the construction schedule of neighboring utilities. Employment
scheduling is incorporated into the planning for capacity additions. Because
reserve margins are a function of the size of units that a utility operates,
larger utilities may install new capacity more frequently than smaller utilities,
38
-------
Table 7. PROJECTED NUMBER OF COAL-FTRED AND NUCLEAR-FUELED ELECTRICAL GENERATING SCENARIO
UNIT ADDITIONS TO BE SITED IN THE OHIO RIVER BASIN ENERGY STUDY REGION
1986-2000
Projected Number of Scenario
State
Subregion
Illinois
U)
vo
Indiana
Kentucky
Ohio
Pennsylvania
West Virginia
Total Number
of Units
la
Coal
13
18
16
20
14
lit
95
2b
Coal
13
18
16
20
1A
1A
95
2ac
Coal
13
18
18
33
19
25
126
Coal
13
18
16
20
14
1A
95
2bd
Nuclear
1
1
0
10
8
0
20
Coal
A
6
A
8
A
6
32
Unit Additions, by Scenario and Fuel Type
2c
Nuclear
19
7
0
2
5
0
33
3
Coal
9
13
11
14
9
10
66
A
Coal
A
7
5
8
A
6
34
5
Coal
10
15
13
17
11
12
78
5a
Coal
17
2A
22
26
18
19
126
6
Coal
2
A
2
6
2
A
20
7a
Coal
20
29
32
32
18
28
159
7b
Coal
18
25
28
32
16
25
1AA
aScenarios la, Ib, Ic and Id have the same number of scenario unit additions.
Scenarios 2d and 21 have the same number of scenario unit additions.
Scenario 2a2 has the same number of scenario unit additions.
d
Scenario 2b1 has the same number of scenario unit additions.
-------
They will spread their construction commitments through time, although several
units at the same site may be scheduled for consecutive years. They will also
prefer to install new capacity within their service areas, although joint-
ownership and a shortage of suitable sites are factors that a utility may
consider in siting capacity additions in other service areas. Load forecasts
are restricted to the demand in a utility's service area, whereas the location
of the supply is becoming more flexible through arrangements such as joint
ownership.
At regional scale, planned capacity additions, 1976 through 1985, are
scheduled linearly with respect to time (Figure 10). The correlation is r = .99,
with an estimated annual increment of A,543 MWe. The schedules of planned cum-
ulative additions for state subregions follow a similar pattern, although the
strength of the relationship is slightly less in states that have smaller in-
crements; e.g., Kentucky (r = .97), Ohio (r = .97) and Pennsylvania (r = .96).
West Virginia has planned additions of only 2,552 MWe, with on-line dates of
1979 and 1980. The aggregate pattern, however, is unambiguously linear at re-
gional scale and for five of the six states.
The schedule of on-line dates for capacity additions in the high electric
energy growth scenarios combines announced utility plans, 1976-1985, with sce-
nario unit additions distributed linearly from 1986-2000. Between 1975 and
1985, utility plans call for an additional 43,799 MWe of installed capacity
in the ORBES region, for a 1985 total of 111,691 MWe (Table 8). The calculated
annual increment of additions over the period 1977-1985 is 4,543 MWe for the
high growth scenarios.5 After 1985, the increment varies for each scenario,
depending upon final electric demand in 2000. The low electric energy growth
scenarios pose a special problem. In Scenario 6, the projected demand for
electricity in the year 2000 is less than planned capacity for 1985. Scenario
4 final demand is only slightly more than that. Rather than prematurely re-
tiring existing or planned generating units from service, the on-line dates
of selected planned additions are delayed in order to conform to a linear pat-
tern in which a single annual increment of additions is applied to the 1976-
2000 period.6 This assumes a pattern of slippage in the on-line dates of
planned capacity additions similar to that observed in the region as utility
plans are adjusted to lower projections of demand for electricity.
The growth curves of total installed electrical generating capacity for
the ORBES region combine the schedule of planned additions through 1985, and
linear additions for each scenario to the year 2000 (Figure 11). The ORBES
region had 72,130 MWe of coal-fired and nuclear-fueled electrical generating
capacity in 1975. Subsequently, the growth curves follow separate paths to
1985 capacities of 111,691 MWe for the high growth scenarios, and 96,387 MWe
for the low growth scenarios. Beyond 1985, the curves diverge further in re-
sponse to the differences in installed capacity additions that are necessary
in order to supply the electrical energy demands that each scenario specifies
for the year 2000. Except for scenarios 2a and 2b, 5a, and 7a and 7b, the
slope of the installed capacity curves falls off smoothly toward the year
2000, primarily because of capacity retirements.
40
-------
Figure 10
I— "1
SCHEDULE OF ELECTRICAL GENERATING CAPACITY ADDITIONS
FOR THE OIKS REGION. U7S-2000
(COAL AND NUCLEAR PLANTS ONLY)
Scenario additions
(plus planned additions]
40 -
1>76
Year
-------
Table 8. SCHEDULES FOR PROJECTED INSTALLED CAPACITY (MWe)
IN THE ORBES REGION FOR TWELVE SCENARIOS
(COAL AND NUCLEAR UNITS ONLY)
.Projected Capacity8
Annual Increments Cumulative Additions0
Scenario 1985C 2000 1977-1985d 1986-2000 1976-1985 1976-2000
1
2
2a
2b
2c
3
111.691
153.
153,
173.
173,
145,
134,
4 96,387 113,
Ell, 691
_
142,
173,
6 96,387 104,
7a (35 yr. life) f 1 f
111.691 178,
7b (45 yr. life) 1 J I
245
245
395
245
295
395
4,543
4
4
5
5
3
3
595 2,843 2
195 |~~ ~~| 3
4.543
395 |_ J
495 2,843 2
H
4,543 ]
J 6
.466
,466
,809
.799
,936
,209
43.
.843 28,
,729 |~~
43.
,809 j^
,236 28,
.239 f
43,
,589 I
799
110.
110.
130,
130.
103,
91.
495 71,
~~| 99,
799
_J 130.
495 62,
799
152.
142,
782
782
932
782
832
932
132
734
932
032
382
632
"installed capacity In 1975 was 72,130 MWe.
Rounding In annual Increments creates small differences from figures shown when calculating cumulative
additions.
CPlanned capacity in 1985 was 111,691 MWe.
^Increment of 2,908 MWe used for 1976.
-------
SPATIAL ALLOCATION PROCEDURES FOR SCENARIO UNIT ADDITIONS
States and portions of states that are in the ORBES region are the geo-
graphical units within which scenario unit additions are distributed. This
choice of siting regions is consistent with the objectives of approximating
current practices, and assessing policies at relevant geopolitical scales.
Electric utilities clearly prefer to locate generating capacity in their
own service areas. This has the advantage of reducing transmission costs to
load centers. However, utilities may also evaluate sites in adjacent service
areas or elsehwere in their state, especially as the environmental and politi-
cal constraints on siting new capacity additions increase. Locating new gen-
erating capacity in adjacent service areas, and entering into joint ownership
agreements for capacity additions that may be located elsewhere, is limited
primarily to planned additions.
Furthermore, the state is the lowest level of government that has any
significant regulatory control over the siting process. Public service com-
missions, siting authorities and other permitting agencies generally affect
the location of new generating units within state boundaries and, except for
a few cases, utility service area aboundaries coincide with state lines (Saper
and Hartnett, 1980, p. 7). Federal agencies, such as the U.S. Environmental
Protection Agency, have jurisdiction in selected areas that transcend state
boundaries. Interstate commissions also may have some authority. However,
their influence is at a larger geographical scale. The geographical alloca-
tion of scenario unit additions within state subregions can incorporate poli-
cies that reflect major subregional variations in resources that affect the
suitability of counties as sites for new generating units.
Within a state or state subregion, the geographical distribution of
electric generating capacity is specified at county scale. This is consistent
with other regional assessment models, and offers a sufficient level of geo-
graphical detail for most types of impact assessment. The location of all
existing units, and the county location of most planned additions, are known.
For each scenario, the specified number of additional coal-fired and nuclear-
fueled units are allocated to counties on state-by-state basis according to
five general rules:
1. Scenario unit additions are allocated, by fuel type, two (2)
units at a time, within each state or state subregion accord-
ing to the rank order site suitability indices of the candidate
counties.
Coal-fired and nuclear-fueled units are treated separately,
as counties may have different suitability indices for each fuel
type because of their different resource requirements. Therefore,
coal-fired and nuclear-fueled units may be located in the same
or adjacent counties. Allocating two (2) units of each fuel
type in a single siting decision is common utility practice,
and consistent with ORBES scenario policies. Multiple units
are usually scheduled for alternate years.
-------
Figure 11
190
170
160
130
c
1
I
1 110
70
GROWTH IN INSTALLED aECTMCAL GENERATING CAPACITY
FOR THE ORBES REGION, 1976-2000
(COAL AND NUCLEAR PLANTS ONLY)
Scenario additions
(plus planned additions)
1975
1980
1985
1990
Scenario
2a, 2b, Sa
1.2
2c
5
4
S
1995
2000
Year
-------
2. If two or more candidate counties in a state have the same site
suitability index, the order of siting within the group is ran-
dom. Siting proceeds within that group before scenario units
are allocated to the county, or group of counties, with the
next highest suitability index.
Adjacent counties may share the same general resources.
They may be approximately equally suitable as sites for new
generating unit additions. This is common at the regional
screening stage of site evaluation. However, relatively few
counties in the ORBES region have the same site suitability
indexes as defined by the ORBES siting model.
3. Scenario unit additions continue to be sited in the state sub-
region until the total number of units allocated to that state
are located. A county may be selected more than once provided
that its total sited electrical generating capacity (i.e., ex-
isting units, planned capacity additions and scenario unit addi-
tions) does not exceed 2600 MWe for coal-fired, and 4,000 MWe
for nuclear-fueled units.
The maxima, which are equivalent to four scenario unit
additions of each fuel type, are consistent with utility guide-
lines. They allow for concentrations of electrical generation
capacity in the most suitable counties, with both coal-fired
and nuclear-fueled capacity in those counties that are highly
suitable for either fuel type. The maxima can be increased
to simulate a policy of "power parks" or "energy centers," or
decreased to simulate a dispersed siting policy.
4. If there are scenario unit additions that cannot be sited in
the state subregion to which they are allocated, this "excess"
is sited in an adjacent state, or states, after the scenario
unit additions assigned to that state, or states, have been
sited.
Adjacent states are defined as those having common boun-
daries.' The excess capacity is sited in the adjacent states
according to their shares in the estimated electricity exports
from the ORBES region in 1974 (Page, 1979, Appendix B). In
effect, this simulates the common pattern of out-of-state part-
ners in the joint ownership of recently constructed and planned
large electrical generating unit additions in the ORBES region.
5. Scenario unit additions that cannot be sited in any state sub-
region will be located "outside" of the ORBES region.
These units are not included in the impact assessment.
The allocation of scenario unit additions in several scenarios vary from
the standard procedure in order to incorporate specific siting policies, or to
reflect changes in policies that are inherent to the scenarios. These scenarios
are:
45
-------
Scenario la: Very Strict Air Quality Controls, Dispersed Siting
Scenario Ib: Very Strict Air Quality Controls, Concentrated Siting
Scenario Ic: Agricultural Lands Protection, Dispersed Siting
Scenario Id: Agricultural Lands Protection, Concentrated Siting
Scenario 2c: Conventional Technology, Base Case Controls, Nuclear
Emphasis
The procedures used in each are defined in subsequent sections. Also, Sce-
nario 7: Conventional Technology, Base Case Controls, High Electricity Energy
Growth, 35 Year Plant Life, as well as scenarios la and Ib have "excess" sce-
nario units that are sited in adjacent states.
SITE SPECIFIC LOCATIONS FOR SCENARIO UNIT ADDITIONS
Certain impact assessment models require more specific locations for sce-
nario unit additions. These include site-specific models of ecological and
social impacts, as well as the models for calculating water quantity and qual-
ity impacts, especially for alternative power plant cooling systems. The exact
locations of all existing and most planned capacity additions are known (Jansen,
1978). For scenario units that are added after 1985, the places at which the
units might be located within a quarter county, if that county should be se-
lected as a future site, are identified. Each quadrant of each candidate county
was evaluated subjectively to determine "preferred" locations. Each quadrant
was evaluated according to selected criteria. Those quadrants that border or
are relatively close to a river; have small areas in public lands and few nat-
ural or unique areas; do not include large urban areas; are accessible to rail,
road or water transportation; and have topographically suitable land sufficient
for a scenario unit are preferred. Nonattainment areas for TSP and S02 are
excluded, as are existing power plant sites that cannot accomodate additional
units. In counties that have meager water resources, the preferred quadrants
are those that have relatively large drainage areas exclusive of lakes and
reservoirs most of which are associated with state parks and wildlife areas.
In most instances, the preferred quadrants as well as the most plausible sites
within them were readily identifiable.
-------
FOOTNOTES
Retirement age, or average useful life assumption, has important impli-
cations in calculating the number of generating unit additions that are nec-
essary to meet projected final electric demand. As the useful life increases,
the number of units that are required to meet a given level of demand de-
creases. The significance of these assumptions are analyzed in scenarios
7a and 7b, where air quality and fiscal impacts of meeting electric demand
by increasing useful plant life from 35 to 45 years are compared.
2The fossil electric utilities production figure was corrected to re-
move the contribution from peaking units that generate electricity from
natural gas and petroleum products. Peaking units contribute about five
percent of the total fossil electric production in the region. The remainder
is from coal-fired electricity generation.
The residuals represent the difference between the number of scenario
generating units to be sited and the amount of capacity specified in the sce-
nario. The differences, which are the result of the fixed MWe size of scenario
unit additions, are calculated by multiplying the scenario unit size by the
number of unsited units and then subtracting the unsited capacity in Table 5,
Column (13). A minus sign (-) in column (15) indicates that insufficient MWe
has been sited; a plus sign (+) menas that more MWe capacity has been sited
than the projected scenario demand.
Tables that detail the calculations used in determining the number of
scenario unit additions for each scenario are in: Fowler et al (1980).
The cumulative additions are calculated by multiplying the annual in-
crement by the number of years in the period, and adding the fixed increment
of 2908 MWe for 1976.
The annual increment was calculated for scenario 4 and then modified
slightly for the period 1986-2000 to accomodate the lower energy growth rate
of scenario 6. The on-line dates for planned additions through 1985 were re-
scheduled over the entire 1976-2000 period in scenarios 4 and 6.
The adjacent states are:
Illinois
Indiana
Kentucky
Ohio
Pennsylvania
West Virginia
Indiana, Kentucky
Illinois, Ohio, Kentucky
Illinois, Indiana, Ohio, West Virginia
Indiana, Kentucky, West Virginia, Pennsylvania
Ohio, West Virginia
Ohio, Pennsylvania, Kentucky
47
-------
SECTION 5
SITING ISSUES AND SITE SUITABILITY
The ORBES siting methodology includes a hierarchial linear weighted model
for determining the suitability of counties as sites for standard coal-fired
and nuclear-fueled capacity additions under different scenario policies. The
siting criteria are defined in terms of the resource requirements of standard
plants and the regulatory constraints included in the description of current,
base case and strict environmental controls. These criteria are then assigned
weights according to their relative importance to power plant siting in the
ORBES region. The weights, which are based upon a Nominal Group Process tech-
nique exercise, are used to define site suitability indices for siting the
scenario unit additions that, in addition to planned units, are needed to
meet the total electricity production in the ORBES scenarios.
SITE SUITABILITY MODEL
The ORBES siting model includes five components, two of which are composed
of several variables (Table 9). Each represents an issue of resource availa-
bility or regulatory constraint that is significant to siting coal-fired or
nuclear-fueled electricity generating facilities in the ORBES region. The
choice of issues, as well as the definition of primary variables used to de-
termine site suitability, was the result of a review of the general process
for siting coal-fired and nuclear-fueled generating units, in consideration
of the resource base in the ORBES region and the policies of concern to the
assessment.
Water availability and air quality are examples of components. The air
quality component includes two variables, the maximum 24 hour ambient sulfur
dioxide (S02) concentration (pg/M3) and the maximum 24 hour ambient total
suspended particulate (TSP) concentration (yg/M2). Each of the variables
used in the suitability formula are transformed or mapped onto a 0-10 scale.
The minimum resource requirements of the standard plants and regulatory re-
quirements are presented by these scores. Each variable within a component
is then weighted on a scale of 0 to 1, and each component can have a weight
of 0 to 10 according to its relative importance in the siting process.
Scenario policies that affect site suitability can be incorporated into
the siting methodology by the choice of relevant components and variables as
well as by the scores and weights assigned to them. For example, a policy
that says that particulate concentrations are much more important than S02,
and that water availability is of equal to air quality, could have a set of
weights:
48
-------
Table 9. DEFINI1IO:: UF PKIMARY VARIABLES USED IK DbTEKMINItiC Sill. SUITABILITY
Criteria for Exclusion3
Issue
Consideration
Base Case Environmental Controls Strict Environmental Controls
VO
AMBIENT AIR QUALITY0
WATER AVAILABILITY
ECOLOGICAL SYSTEMS
AND LAND USE
HEALTH AND SAFETY
Reduction in pollutant l-oeatlon relative to
concentrations nonattainment areasc
Cuunty designated nonattalnment
area, primary standard"*
Prevention of significant Allowable Increments as Majority of county In mandatory
deterioration In PSD regulations Class 1 area
("Thermal pollution and
• • I acqulslon of cooling Low flow availability
|_water
• • Natural, scenic and re- Extent of public lands All of county In public lands6
creatlonal areas
• • Sensitive and protected Unique natural areas
environments
• • Agricultural and ecolog- Extent nf Class I and
leal productivity Class II soils
• • Ownership and management Extent of non-federal
of forest lands forest
County contains nonattainment areas,
primary and secondary standards
Majority of county In mandatory
Class I area
Capacity additions exceed allow-
able Increments (24 hr. max.)
Majority of county In public lands
( Radiation exposure
Population distribution
Seismic suitability Distance from capable
faults
Population density Z. SOD people
per square mile'
Majorlty.of county In Seismic
Zone III
*The definition of exclusionary criteria varies according to scenario policies.
According to the 1977 Clean Air Act Amendments.
°For SOj and TSP
dFederal Register. 43. No 43 (March 3. 1978): 896J-B853, ind No 194 «0clob. r i. 1978): 45993-46019.
Actual ownership
Total area, including designated purchase .ire.n.
-------
Wso2 • wn • °-3
WTSP = W21 ' °'7
Xl =I2 =5
The challenge is to relate policy statemnets to weights for those variables
that are included in the definition of each scenario.
The mathematical description of the model follows:
C., = the absolute component index for the ith county and
^ kth component.
W = weighting factor for the ith criteria of the kth
^ component.
X.., = numerical ranking for the ith criteria within the
1-' kth component for the jth county.
I, = the importance of the kth component.
S. = the absolute suitability index for the jth county.
S = the maximum suitability.
CJ"aX = the maximum component index.
The equations that use these definitions to define site suitability are:
M
k=l
50
-------
i° vi
M k~1
The domain for each of the parameters is:
- Jk - k
0 < W., < 1 suqh that I W. = 1 for each k
o 1 x..k i 10
0 < I < 10
~ k "~
0 , s. <
This model combines exclusionary screening with a comparative evaluation of
candidate sites for coal-fired and nuclear-fueled generating unit additions
under different environmental control scenarios. It is similar to other base
line assessment siting models, such as developed by ORNL (Davis et al, 1979).
It differs from them in terms of its inclusion of a larger number of environ-
mental variables, and its use in translating scenario policies into unique
siting patterns for use in impact assessments.
SITING ISSUES
Ambient Air Quality
Ambient air quality is an issue of fundamental importance in siting coal-
fired electrical generating units. Large coal-fired units are major stationary
sources of air pollutant emissions, especially sulfur dioxide and particulate
matter that is discharged as fly ash. They may contribute significantly to the
deterioration of ambient air quality in regions where they are concentrated, or
in distant areas affected by long-range pollutant transport. With respect to
issues or ambient air quality, siting coal-fired generating units is subject to
provisions of the Clean Air Act (42 U.S.C. 7400 et. seq.) applicable to station-
ary sources, and the attainment of National Air Quality Standards (NAAQS).
NAAQS are expressions of the allowable levels of concentration of specific
pollutants in the ambient air. Currently, NAAQS have been established for six
"criteria" pollutants: sulfur dioxide; particulate matter; carbon monoxide;
photochemical oxidants; hydrocarbons; and nitrogen dioxide. The primary stan-
dard is that level necessary to protect the public health. The secondary
51
-------
standard is that level necessary to protect the public welfare from adverse
effects of any pollutant. In cases where the standards for a pollutant dif-
fer, the secondary standard is always the most rigorous. Some provisions of
the Act are designed to improve ambient air quality in places that do not
meet the NAAQS, whereas others are designed to prevent deterioration of air
quality in places that exceed the standards. Both are significant factors
in siting new energy facilities.
A non-attainment area includes, for any pollutant, areas designated by
the State, or any area that is shown by monitored data or air quality modeling,
to exceed any ambient air quality standard for such pollutants (McHugh, 1978
and Grant, 1979). The object of the regulations for nonattainment areas is
to improve ambient air quality by reducing emissions from existing sources and
by severely restricting new source construction in or near the areas. An "emis-
sions offset" policy applies to most major new construction or modifications
of sources in nonattainment areas, including replacement of existing sources.
In order to obtain a permit, the applicant must show that:
1. The new source achieves the "lowest achievable emission
rate" (LAER) for that type source.
2. All of the company's existing sources in the region^ are
in compliance with their respective emission requirements.
3. Sufficient reduction of pollutants to be emitted are ob-
tained from other sources in the region to more than offset
the emissions from the new source.
4. The emission offset and the proposed new source emission
levels will provide a net air quality benefit to the af-
fected area, not just the region as a whole.
Regulations for the prevention of significant deterioration (PSD) govern
new source construction in areas with ambient air quality that is equal to or
better than that required by NAAQS (Table 10). Under the EPA regulatory
scheme, these "clean air" areas are placed in one of three classes, each of
which has maximum allowable increments of net air pollution increases for
particulate matter (or, total suspended particulates—TSP) and sulfur dioxide
(S02) permitted for each class up to a level considered significant for that
area. The Increments are based roughly on a percentage of the NAAQS for u.icli
pollutant. Thus, in Class I areas, ambient levels of TSP and SC^ may hi- in-
creased above the baseline concentration by an amount equivalent to about 27.
of the NAAQS. Class II allows a 25% increase, and Class III a 50% incrunso.
However, ambient air quality cannot exceed NAAQS in any case.
Certain Federal lands, national parks, wilderness areas, international
parks and memorial parks are classified as mandatory Class I areas. They
cannot be reclassified. Other clean air regions are in Class II. States
have the authority to redesignate any area as Class I. Certain areas can be
redesignated as Class III after public hearings and extensive review by state
and federal agencies. In order to locate a major new source in a PSD area,
52
-------
Table 10. SUMMARY OF CLEAN AIR ACT AMENDMENTS OF 1977,
PREVENTION OF SIGNIFICANT DETERIORATION (PSD)
MAXIMUM ALLOWABLE
MM DETERIORATION INCREMENTS
HAttATOAY CLASS I
MM DETERIORATION AREAS
MAXIMUM ALLOWABLE
CONCENTRATION INCREASE
fOLLUTANT
10, (I HR)
10, (2* M«)
SO, (ANNUAL)
PAATICULATE (It Ml)
PARTICULATE (ANNUAL)
EXCEPT FOR THC ANNUAL VALUES, THE MAXIMUM ALLOWABLE INCREMENT
CAN II CKIEDEO OWE PER YEAR.
IT JKOULO IE NOTED THAT PSD CONCENTMTION INCREASES FOR CERTAIN
AREAS ARE NANDATEO AS CLASS I AS SHOWN IN THE NEXT BOI
OF SPECIAL IMPORTANCE TO UTILITIES IS THE IBPACT OF ANY FUTURE
LEGISLATION NAMING NEW NATIONAL PARKS 01 UILOERNESS AREAS IF
THESE PARKS OR WILKERHESS AREAS ARC LESS THAN 10.000 ACRCS IN
SIIC. THEY COU10 COMCIVAULT BE CLASSIFIED CLASS III THOSE
AREAS LARGER THAU IO.ODD ACHES ON ONLT BE CLASSIFIED AS CLASS
I OR CLASS II IT WOULD APPEAR ADVANTAGEOUS FOR UTILITIES TO
PRESS FOR AREAS LESS THAN 10.000 ACRES IN ODDER TO GAIN OPTIONS
OF CLASSIFYING PORTIONS OF REGIONS AS CLASS III
, UPON OETIRNINAIION If THt FEDERAL I AMI MANAGER
THAT THC AIR QUALITY IMPACT ON A FEDERAL CLASS I
ARIA IS ACCEPTABLE. A PEMIT COULD BE ISSUED
ALLOWING CLASS I INCREMENTS TO IE EUCCIOEO THE
MAXIMUM AILOWAILE CONCENTRATION INCREASES ABOVE
IASELINE FOR THIS CASI WOULD IE LIMITED AS FOLLOWS
POLLUTANT
SO, (i HR)
SO, |2« NR)
I, (ANNUAL)
ALLOWABLE INCREMENT
SO,
PARTICULATE (It HR)
PARTICULATE (ANNUAL)
in
ji
10
J7
I)
A VARIANCE TO ONLY THE ) HR ARO 2* NR SO, STAN-
DARDS APPLICABLE TO CLASS I ARIAS CAN SE OBTAINED
AFTER AN AfFIRHATIVt DEMONSTRATION TO THt
GOVERNOR AND ACHIEVEMENT OF CONCURRENCE FROM THE
FEDERAL LAND MANAGER IN THIS (VINT. THE VARIMCl
WOULD ALLOW THE FXCEEDAHCE OF THE SO] STANDARDS
FOR A PERIOD OF NOT NIMC THAN It DAY* DURING AHY
ANNUAL PERIOD IN ADDITION. THE KAHIMUN ALLOWABLE
INCRCMEHT WOULD IE LlnlTED AS FOLLOWS
INCREMENT (ug/«»
HIGH TERRAIN
AREAS
SO, (] HR)
SO, (It NR)
121
Cl
THE FOLLOWING ARCAI ARE DESIGNATED AS CLASS I AND ARE NOT SUBJECT TO REDES ICHATIOH-
. INTERNATIONAL PARKS
. RATIONAL WILDERNESS AREAS > 5000 ACRES
. NATIONAL MEMORIAL PARRS > SOOO ACRES
. NATIONAL PARKS > 6000 ACRES WHICH EXISTED AS OF AUGUST 7. 1977
. ALL AREAS WHICH HAD BEEN REOESIGNATEO CLASS I UNDER REGULATIONS EXISTING PRIOR TO THE CLEAN AIR ACT AMENDMENTS OF H77
VISIBILITY PROTECTION
PROVISION
CPA MUST PROMULGATE REGULATIONS WITHIN 21 MONTHS TO ASSURE REASONABLE PROGRESS TOWARDS PREVENTING IMPAIRMENT OF VISIBILITY
IN MANDATORY CLASS I FEDERAL AREAS
ALL ELECTRIC GENERATING UNITS WITH THE POTENTIAL TO EMIT 150 TONS OR MORE OF ANY POLLUTANT AND WHICH HAVE A HIAT INPUT MORE
THAN TJO MILLION BTU'S/HR ARE AFFECTED BY THE VISIBILITY PROVISION
(LKTAIC GENERATING STATIONS WHICH HAVE A CAPACITY LESS THAN 7» MW'S AND WHICH HAVE BEEN IN OPERATION FOR MORE THAN 15 YEARS
AS OF AUGUST 7. 1977. ARE EXEMPTED FRON THE REQUIREMENTS OF THE VISIBILITY PROVISION
EPA WILL PROMULGATE REGULATIONS DIRECTLY APPLICABLE TO ELECTRIC GENERATING UNITS HAVING A CAPACITY IN EXCESS OF 750 NW
IN DETERMINING BOTH "REASONABLE PROGRESS" TOWARDS EllMINATINC VISIBILITY IMPAIRMENT AND "BEST AVAILABLE RETROFIT TECHNOLOGY"
THE ADMINISTRATOR MUST CONSIDER FACTORS SUCH AS COST. NON-AIR QUALITY ENVIRONMENTAL AND ENERGY IMPACTS. THE REMAINING USEFUL
LIFE OF THE SOURCE AND. AS APPROPRIATE. THE DECREE OF IMPROVEMENT IN VISIBILITY WHICH MIGHT IE EIPECTEO
PSD - PERMIT PROGRAM
REQUIREMENTS
THE LAW STATES THAT NO MAJOR EMITTING FACILITY CAN BE CONSTRUCTED AFTER AUGUST 7. 1977 UNLESS. A PERMIT ASSURING COMPLIANCE
WITM PSD REQUIREMENTS. IS OBTAINED FROM THE EPA OR STATE. WHICHEVER IS APPROPRIATE
THE PERMIT APPLICATION MUST IHCLUDE lYEAA'SUORTH OF AMBIENT AIR QUALITY MONITORING DATA UNLESS A WAIYER ALLOWING A SHORT
PERIOD OF FIELD DATA IS OBTAINCO FRON THE STATE
THE APPLICANT MUST DEMONSTRATE THAT EMISSIONS FROn CONSTRUCTION OR OPERATION OF THE PROPOSED FACILITY WILL NOT RESULT III
CONTRAVENTION OF NATIONAL AMBIENT AIR QUALITY STANDARDS. MAXIMUM ALLOWABLE NON-DETERIORATION INCREMENTS AND ANY OTHER
APPLICABLE [MISSION STANDARD OR STANDARD OF PERFORMANCE IN ADDITION. THE APPLICANT MUST OCHONSTMTE THAT (I) THE REST
AVAILABLE CONTROL TECHNOLOGY IS BEING EMPLOYED. (2) THE FACILITY EMISSIONS ARE COMPLYING WITH ALL REQUIREMENTS RELATED TO
MANDATORY CLASS I AREAS. AND THAT (1) AN ANALYSIS OF THESE POTENTIAL AIR QUALITY IMPACTS ASSOCIATED WITH GROWTH RELATED TO
THE FACILITY HAS BEEN PERFORMED
EPA WILL IE PROMULGATING REGULATIONS REGARDING ACCEPTABLE AIR QUALITY MODELS FOR USE IN THE REQUIRED DISPERSION ANALYSES.
A HEARING MUST BE HELD REGARDING THE PERMIT APPLICATION. AT WHICH TINE THE POTENTIAL AIR QUALITY IMPACTS. CONTROL TECHNOLOGY.
ALTERNATIVES AND OTHER APPROPRIATE CONSIDERATIONS CAN BE RAISED
A PERMIT MUST BE GRANTED OR DENIED WITHIN ONE YEAR AFTER THE DATE OF FILING A COMPLETED APPLICATION.
OTWR POLLUTANTS
IN ADDITION TO THE NON-DETERIORATION INCREMENTS FOR SO, AND PAMTICULATES. THE CPA IS MANDATED TO COHMJCT A STUDY MO Wl rMIN
2 YEARS TO PROMULGATE REGULATIONS TO PREVENT THE SIGNIFICANT DETERIORATION OF AIR QUALITY ASSOCIATED WITH EMISSIONS OF
HYDROCARBONS. CAR10H HOWHIDE, PHOTOCHEMICAL OIIOANIS. AMD NITROGEN DIOIIDE
EPA DUST ALSO PROMULGATE NON-DETERIORATION REGULATIONS FOR ANY OTHER POLLUTANT FOR WHICH IT ESTABLISHES NATIONAL AHIIENT AIR
QUALITY STANDARDS.
SOURCE: Envirosphere Company (1977).
53
-------
the applicant must undergo a review for S02 and TSP to demonstrate that emis-
sions from the new source will not exceed the allowable increments in that
particular area as well as adjacent areas.
Although nonattainment and PSD provisions apply to specific geographical
areas, the boundaries are not absolute with respect to the review processes
(cf. McHugh, 1978). A new source that wishes to locate in a nonattainment
area, for example, might, in addition to obtaining an emissions offset, also
need a PSD permit if the air quality in adjacent clean air regions might be
adversely affected in any way. Conversely, a new source that wishes to lo-
cate in a clean air region might, in addition to a PSD permit, be required to
obtain an emissions offset if air quality modeling indicates that it will im-
pact at all on the nonattainment area. An additional constraint in the PSD
regions is the visibility impairment provisions, which are directed at elimi-
nating and preventing any impairment of visibility in the mandatory Class I
areas. The effect is to add a buffer zone to the Class I areas. The geo-
graphical range that may be used in the review procedures may be extensive.
It may also involve extraterritorial sources, in which case provisions govern-
ing the prevention of interstate air pollution apply. These require a State
to prohibit any new or existing source from emitting pollutants that would in-
terfere with the attainment or maintenance of any NAAQS in a neighboring state.
Thus, whether a source locates in a nonattainment area or PSD region, it must
meet standards to prevent impact upon ambient air quality of neighboring areas,
as well as interstate impacts.
The Clean Air Act has introduced considerable "uncertainty" into the
siting of coal-fired electrical generating facilities (Grant, 1979). Non-
attainment provisions make it diffucult to locate new units near major load
centers, as most such centers are in industrialized urban and metropolitan
areas that do not meet NAAQS. They also frequently have relatively small
coal-fired units that were built prior to 1970, and that are not likely to be
replaced by new units when they are retired. The nonattainment policy is a
fairly stringent land control measure for all types of new developments, in-
cluding energy facilities.
States in the ORBES region have tried to limit nonattainment areas to the
smallest geographical units possible. Nonattainment areas for SC>2 and TSP in
Illinois and Indiana, for example, are drawn along township lines or other
minor civil divisions (Grant, 1978; Illinois-Power Company, 1979). Although
the geography of the nonattainment areas across the ORBES region varies con-
siderably, two trends are apparent:
1. The number and size of nonattainment areas for TSP are
greater than for SC^-
2. The number and size of nonattainment areas increases from
west to east, and north of the Ohio River.
These reflect a combination of actual pollutant concentrations as well as
methods used to define nonattainment areas.2
-------
Most new and proposed coal-fired generating units are sited in PSD areas.
Although it is easier to obtain a PSD permit than to locate in nonattainment
areas, the constraints are considerable nonetheless. Class I areas and sur-
rounding areas are virtually excluded as sites. Coal-fired units must locate
at some distance from the boundaries of Class I areas, although the exact dis-
tance will depend on meteorology and terrain. The provisions for visibility
protection promise to extend the buffer zone. The ORBES region only has four
mandatory Class 1 areas. Two are in Kentucky (Mammoth Cave National Park,
Edmonson County; and Beaver Creek National Wilderness Area, Menifee County)
and two are in West Virginia (Otter Creek National Wilderness Area, Randolph
County; and Dolly Sods National Wilderness Area, Tucker County).
However, a large number of Class II areas are potential candidates for re-
designation to Class I. The Shawnee National Forest and Crab Orchard Wildlife
Sanctuary in southern Illinois are examples (Grant, 1979). In addition to in-
creasing the geographical extent of Class I areas, they could restrict sites
for coal-fired plants in coal-producing areas where mine-mouth locations might
be desirable.
PSD policies also constrain the number and size of coal-fired units in
Class II areas. An ANL study (Garvey et al, 1978) concludes that a max-
imum of 2700 MWe can be located at one site in flat or moderate terrain in
Class II areas. Unit size may be reduced in rugged terrain, such as Appala-
chian Kentucky and West Virginia, where emissions will be trapped. The max-
imum allowable increments also raise the question of separation distances,
as well as the optimum geographical distribution of coal-fired units within
Class II areas (Equitable Environmental Health, 1976; Envirosphere, 1978).
Issues of proximity may be especially critical where state boundaries are
along the Ohio River and its major tributaries. The concentration of new
sites in Boone County, Kentucky and Switzerland County, Indiana has resulted
in a dispute over the PSD permit for the Indianapolis Power and Light Company's
(IPALCO) Patriot p]ant. Where PSD provisions apply, applicants must ensure
that increments will be available during the construction period. Other
disputes concerned with the interstate pollution provisions of the Clean
Air Act involve Ohio, Pennsylvania and West Virginia.3
Air quality issues are represented in the siting model with respect
to the nonattainment and PSD provisions of the Clean Air Act for S02 and TSP.
Nonattainment provisions are considered as exclusionary criteria. Depending
on environmental control policies, counties are excluded from consideration
as candidate sites if they contain, or are designated, nonattainment areas
for any standard. This is consistent with current siting practice in the
ORBES region, and with the difficulty of constructing new sources in nonattain-
ment areas. PSD Class I areas are also excluded as sites because of the small
allowable increment. Siting coal-fired generating units in the Class I areas
is unlikely, and locating relatively close to them may even be difficult.
The extent of the areas excluded for nonattainment areas and PSD Class I areas
indirectly accounts for buffer zones, separation distances and other aspects
of location relative to attainment and nonattainment areas.
The suitability of other counties as sites for coal-fired units is based
on estimates of the increment of clean air that remains after a standard coal-
55
-------
fired scenario unit addition is sited (Figure 12). Given the NAAQS or whatever
limit is allowed by law, the clean air increment is the amount that remains
after the pollutants associated with a scenario unit addition are added to
the background ambient concentrations in a county. Each county is then a
score on a scale of 0 to 10, with all pollutant levels normalized to a base of
10 depending on the concentration limits allowed by law. As PSD Class I areas
are excluded, all counties that are assigned suitability scores for the air
quality component are in PSD Class II regions.
Average annual ambient concentrations for each pollutant represented the
state of air quality in an area (county) before a scenario unit addition was
sited. These baseline concentrations were estimated from monitoring data for
each ORBES state subregion for 1977.4 For counties that had no monitor, the
ambient concentration was estimated using either a geographically weighted,
linear interpolation between monitoring stations or simply the concentration
in an adjacent county. The ambient concentrations after siting a scenario
unit addition was then estimated, taking into consideration persistent wind
conditions for five meteorological subregions and a calculational model con-
sisting of the Guossian model (Kark and Warner, 1976) and the plume rise
models of Holland, TVA-Concurve and Briggs.5 The maximum concentration was
calculated for each subregion for stability classes B, C, D, E and F. The
worst case was selected for each subregion.
The increment of air quality that remain is used to score a county for
each pollutant (Table 11). If the ambient concentrations that exist after a
unit is sited exceeds NAAQS (or whatever the law allows), the A is
negative. Counties with negative scores are excluded from consideration.
If standards are not violated, the A is positive. In general, the larger
the positive A the more suitable the county with respect to air quality.
State standards are used as the ambient concentrations allowed by law. Pri-
mary standards are used for base case environmental control scenarios and
secondary standards for strict environmental controls.
Site Suitability; Ambient Air Quality
In base case environmental control scenarios, the majority of the ORBES
region has relatively high suitability scores with respect to S02 (Figure 13).
Counties in western Pennsylvania and the West Virginia panhandle have the low-
est score; i.e., the smallest allowable increment of clean air. Elsewhere,
the only other areas that have some problems with S02 are in northeastern
Ohio, southern Illinois and selected metropolitan regions (e.g., Indianapolis
and Terre Haute, Indiana; Louisville, Kentucky; Cincinnati, Ohio; and Charleston,
West Virginia). Compared with S02, however, TSP contributes more to lov;
suitability with respect to air quality (Figure U). The majority of the
ORBES region has low suitability scores. The lowest are in western Pennsyl-
vania, southeastern Kentucky and in the East St. Louis metropolitan area.
The distribution of the scores, however, suggests that TSP is a much more
ubiquitous pollutant than S02-
Because secondary standards are used for strict environmental control sce-
narios, the suitability scores decrease and cover a larger number of counties.
Western Pennsylvania counties continue to have the lowest suitability with rc-
56
-------
Figure 12. A SCHEME FOR SCORING A SITE IN TERMS OF AIR QUALITY
Ambient
Pollutant
Concentration
Scoring
Scale
National Ambient
Air Quality Standard
(NAAQS) or whatever
limit is allowed by law
Ambient concentration
(estimated) after
facility is sited
Increment of air quality
that is left. This A
is used for scoring.
Ambient concentration
(estimated) before
facility is sited
. . 6
- - 7
• - 8
. . 9
ID-
Note: A score of 10 implies no pollutant concentration.
A score of <0 implies other sources of pollutant
must be curtailed before facility is sited.
57
-------
Table 11. DEFINITION OF THE AIR QUALITY COMPONENT
- V + (X2jl - ATSP> + X3jl
= I 1/3 (As + ATSp) + 1/3 (Xljl + X2jl + X3jl)
-KA + 1/3 (Xyl + X2jl
Illinois : Ag = 1.84, ATgp = 0.2, -KA = -0.68.*
Pennsylvania : A - 1.84, ATSP = 0.2, -KA = -0.68.*
Ohio : As = 2.58, A = 0.2, -K = -0.93.*
West Virginia: Ag = 1.84, ATgp = 0.2, -KA = -0.68.*
Kentucky : A = 1.84, ATgp = 0.2, -Kft = -0.68.*
*After a scenario unit is sited, the A for air quality is decreased by
a fixed amount for each unit added. The A's are constant values for a sce-
nario unit addition located in the meteorological subregions of the ORBES
region.
58
-------
VO
Figure 13. PREVENTION OF SIGNIFICANT DETERIORATION
SULFUR DIOXIDE (S02)
BASE CASE ENVIRONMENTAL CONTROLS
ALLOWABLE INCREMENT: RaATIVE SCORES
9. - 10.
8. - 9.
7. - 8.
6. - 7.
3. - 6.
MCPMCO FOR OHIO RIVER BASM ENERGY STUDY
BTCACS/UCC. FTWWJARY. 1980
-------
Figure 14. PREVENTION OF SIGNIFICANT DETERIORATION
TOTAL SUSPENDED PARTICULATES CTSP)
BASE CASE ENVIRONMENTAL CONTROLS
ALLOWABLE INCREMENT: RELATIVE SCORES
9. - 10.
8. - 9.
7. - 8.
6. - 7.
3. - 6.
PKPABED ran OHIO RKR BASH oeter STUDY
BY OCB/UCC. rCBMMMT. 1MO
-------
spect to S02 (Figure 15). The range of counties with less than the maximum
scores, however, expands to include large parts of eastern Ohio and West
Virginia; central and southern Indiana; and southern and west central Illinois.
However, TSP continues to be the dominant pollutant (Figure 16). The majority
of the counties in the ORBES region are in the lowest suitability category.
These include areas of intensive agriculture, coal mining and the major metro-
politan industrial areas.
S02 and TSP are of equal importance in determining the suitability of
counties as sites for coal-fired generating unit additions with respect to
the ambient air quality component. In base case environmental control sce-
narios, all of the counties in the ORBES region have medium to high suitabil-
ity (-Figure 17). Western Pennsylvania and a few scattered counties, or groups
of counties, have medium-high suitability. Most are counties that have some
problem with both S02 and TSP. Elsewhere, concentrations of TSP are primarily
responsible for reducing suitability scores. In the strict environmental
control scenarios, the general effect of secondary standards is to reduce
the suitability scores by at least one class (Figure 18). Selected areas
are even less suitable. These include the Pittsburgh metropolitan area;
most other metropolitan areas with relatively high S02 background concentra-
tions; and a block of counties in west central Illinois with relatively high
TSP concentrations.
Water Availability
Conventional methods of generating electricity from either coal or nuclear
fuel require large quantities of water. These methods are based on the steam-
electric cycle in which heat from the combustion of coal or from the fission
of uranium is used to heat water to steam. The steam is expanded through a
turbine which drives a generator to produce electricity. Closure of the steam-
electric cycle is accomplished by condensing the steam to liquid water for re-
circulation back through the system. A relatively small amount of water is
required as the working fluid but large quantities are needed as cooling water
to condense the steam. In fact, the cooling water requirements are so large
in comparison to the working fluid requirements that the latter may be neglect-
ed for siting purposes without significantly affecting the result.
The amount of water required for a site depends on the fuel type of the
generating units, the number and size of the units, and the cooling technology.
Each of these variables is prescribed within narrow limits for the ORBES sce-
narios. Except for certain variations on the base case scenario that allow
once-through cooling on the Ohio River main stem, cooling technology is lim-
ited to wet (evaporative) cooling towers. This is in accordance with USEPA
regulations and guidelines (CFR 40, Part 423) issued in 1974 to implement the
1972 Amendments to the Federal Water Pollution Control Act. These regulations
designate closed-cycle cooling (cooling towers) as "best available technology"
for control of thermal effluents. Reynolds (1980) gives further details on the
history and implications of these regulations.
Nuclear-fueled generating units require considerably greater amounts of
cooling water than coal-fired units. The difference is attributable to the
61
-------
Figure is. PREVENTION OF SIGNIFICANT DETERIORATION
SULFUR DIOXIDE (S02>
STRICT ENVIRONMENTAL CONTROLS
ALLOWABLE INCREMENT: RELATIVE SCORES
9. - 10.
8. - 9.
7. - 8.
6. - 7.
3. - 6.
PtEPAKD FOR OH» »NO» BAS* DCRGY SWOT
0T CAGB/UCC. rtHKMPY. 1MO
-------
Figure 16. PREVENTION OF SIGNIFICANT DETERIORATION
TOTAL SUSPENDED PARTICULATES CTSP)
, STRICT ENVIRONMENTAL CONTROLS
CO
ALLOWABLE INCREMENT: RELATIVE SCORES
MEPAKD FOR OMO MVO) BASM ENERGY STUDY
er cAOK/incc. rtwttMRY. i»eo
-------
Figure 17. AMBIENT AIR QUALITY COMPONENT
BASE CASE ENVIRONMENTAL CONTROLS
INDEX VALUE: RELATIVE SCORES
09. - in-
08. - 9.
• 5*. - B'.
MKPAIKD rOR OHO RIVOt BASH CHOttff STUDY
er oos/uec. nsmiARY. i»»o
-------
Figure is. AMBIENT AIR QUALITY COMPONENT
STRICT ENVIRONMENTAL CONTROLS
Ul
INDEX VALUE: RELATIVE SCORES
9. - 10.
8. - 9.
7. - 8.
6. - 7.
5. - 6.
WEPAHCO FOR OHIO MVOI 8ASM CNOWY STUDY
tT CACK/UCC. rtBWAFY. 1980
-------
typically higher thermal conversion efficiency of coal-fired units (38%
efficiency) compared with that of nuclear-fueled units (33% efficiency) and
the fact that coal-fired units generally release 15 to 20% of their waste
heat directly to the atmosphere with the flue gases, while nuclear-fueled
units release all but 0 to 50% of their waste that heat through cooling con-
densers (Pigford et al., 1974). Together, these differences cause nuclear-
fueled units to consume 39 to 50% more cooling waters than coal-fired
units on a per MWe basis (Harte and El-Gassein, 1978, p. 628). Comparable
figures from Brill et al. (1980) indicate consumptive loss to be nearly 70%
greater for nuclear units.
Table 12 compares water withdrawals with consumptive loss for standard
coal-fired and nuclear-fueled scenario units under different cooling technolo-
gies. Withdrawal is the amount of water that must be taken from the source
water body for cooling purposes. Consumption refers to the amount of water
lost from the local hydrologic system as a result of cooling system operation.
Consumption is always less than withdrawal but the difference between the two
varies as a function of the type of cooling, as well as other factors. Con-
sumption can be calculated as the difference between withdrawals from and
returns to local water bodies. In the case of wet cooling towers, consumption
is due to evaporative cooling loss to the atmosphere because this water can-
not be expected to be returned to the local hydrologic system. Once-through
cooling consumed significantly less water but withdraws considerably more
water than wet towers or ponds to accomplish the same cooling.
Water resources are also in demand for purposes other than power gen-
eration. The available surface water supplies are required to serve other
industries and municipal needs while furnishing sufficient flow to maintain
navigation and a healthy aquatic environment. Consequently, power generation
can only be allocated a certain portion of the available water. Competition
for use of the local water resources is generally most acute during periods
of low flow. Thus, a valid measure of water availability must reflect low
flow conditions. The usual measure is the low flow that persists for seven
days and can be expected to occur once every ten years — i.e., 7-day/10-year
(7Q10) low flow. Selection of the 7-day duration low flow is usually based
on evidence that aquatic organisms often can tolerate several days of stress
but not weeks or months (Hynes, 1970).
Consistently reliable quantitative data on water availability for the
ORBKS region are available for streamflow only. Comparable data are not avail-
able for groundwater or potential reservoir yields. Consequently, potential
methods of augmenting streamflow, (such as constructing reservoirs and pumping
groundwater or stream water) and using dry cooling towers to reducing cooling
water requirements are accounted for in the ORBES siting model by giving coun-
ties with low 7Q10 low flow values relatively low suitability scores for water
availability rather than excluding them from consideration as candidate sites.
This simulates electric utility company behavior. Utilities prefer to locate
capacity additions immediately adjacent to a large supply of water with suf-
ficient yQio l°w flow. If that is inconvenient, a site may be selected on
a smaller body of water where some type of augmentation may be necessary in
order to generate at peak capacity. If utilities are forced to move further
66
-------
Table 12. COOLING WATER REQUIREMENTS FOR SCENARIO UNIT ADDITIONS
(in CFS/Unit)
Cooling Technology
Withdrawal
Consumption
Coal-fired Units (650 MWe/unit)
Wet Towers
Once-through Cooling
Ponds
16.6
910.0
16.6
10.4
6.5
10.4
Nuclear-fueled Units (1000 MWe/unit)
Wet Towers
Once-through Cooling
Ponds
43.0
2000.0
43.0
27.0
15.0
27.0
SOURCE: Brill et al., 1980.
67
-------
inland, it may be necessary to utilize groundwater, pump stream water over long
distances (25-30 miles) or install dry cooling towers. Each successive option
would cost the utility more in terms of capital and operating outlays, and
risk problems with impacts on land use and ecological systems. The siting
model simulates these "costs" by assigning lower suitability scores to areas
that must depend upon streamflow augmentation.
Site Suitability for Water Availability
Suitability scores for the water availability component of the siting
model are assigned on the basis of 7Q10 low flow values for streamflow (Table
13). Each county in the ORBES region is assigned a single -jQ^Q low flow value
by summing values based on Brill et al (1980) for each gauged stream within
or adjacent to the county. For many counties, yQ^Q values are estimated be-
cause no gauges are located along stream reaches within or adjacent to the
counties. In these cases, a linear extrapolation between appropriate gauging
stations is used except when topographic maps indicate contributions from tri-
butaries justify modification of the linear extrapolation estimate.
Site suitability scores vary directly with the amount of cooling water
available. Counties with less than or equal 10 cfs for yQ^Q low Elow are not
excluded as a generating site although this is not enough water to supply the
consumptive requirements of a 650 MWe coal-fired unit operating at maximum
output. Augmentation is taken into consideration at the low end of the suit-
ability range. The cooling water consumption requirements for an electrical
generating unit cannot claim all available stream flow. As a guide, no more
than 10% of the yQxo low flow can be used. In the ORBES siting model, two or
more scenario unit additions can be sited in a county only if the water avail-
ability score for the county is greater than four for coal-fired units, or
greater than five for nuclear-fueled units. Streamflow augmentation or altern-
ative cooling technologies could support multiple units in counties with lower
scores but these alternatives are judged to be prohibitively expensive.
The pattern of relative county suitability scores for the water availa-
bility component in the ORBES region varies by location on the stream network
(Figure 19). Counties in the highest suitability range (scores = 8-10) are
found along the Ohio River below Huntington, West Virginia and along the
Mississippi River on the western borders of Illinois and Kentucky. Counties
in the next suitability range (scores = 6-8) are located along the Allegheny
and Monongahela Rivers in Pennsylvania; along sides of the Ohio River from
Pittsburgh, Pennsylvania to Huntington, West Virginia; along the Kanawha and
New Rivers throughout their lengths in West Virginia; along the lower Wabash
River in Indiana and Illinois; and on the Illinois and Rock rivers in Illinois.
Counties in the low end of the range suitable for more than one generating
unit (scores = 4-6) are found further upstream on the Allegheny River in
Pennsylvania and the Monongahela River in Pennsylvania and West Virginia; on
the Mahoning, the Muskingum, the Scioto and the Miami rivers in Ohio; along
the Green River and the Kentucky River in Kentucky; on the Wabash, both forks
of the White, and the Kankakee Rivers in Indiana; and along the Wabash, Kan-
kakee, and Sangamon Rivers in Illinois. Counties with lower suitability scores
occupy the remaining upland areas of the ORBES region.
68
-------
Table 13. WATER AVAILABILITY COMPONENT SCORES
Flow
for County (cfs) Score
10,000.1
5,000.1
1,000.1
200.1
100.1
50.1
20.1
10.1
< 10
> 20,000.1
- 20,000
- 10,000
- 5,000
- 1 ,000
200
100
50
20
10
9
8
7
6
5
4
3
2
1
69
-------
Figure 19. WATER AVAILABILITY COMPONENT
INDEX VALUE: RELATIVE SCORES
mPAKD FOR OHIO RIVER BASK CNERGY STUDY
IT CAGS/UCC. FEBRUARY. I WO
-------
Land Use And Ecological Systems
The land use requirements for electrical generating facilities can be
substantial.6 At a coal-fired facility, land is required for the main boiler
unit, cooling towers or ponds, coal storage, ash disposal and roads associated
with the facility. For six (6) coal-fired facilities under construction in
the ORBES region, utility land ownership averaged 1,050 acres per 650 MWe
capacity. Using this figure, present (1976) land use at energy conversion
facility sites in the ORBES region is estimated at 140,673 acres. If land
requirements for high voltage transmission line rights-of-way (estimated at
600,000 acres in 1976) are also considered, land use conversion is even high-
er. Nuclear-fueled generating units and associated facilities, including space
for fuel storage and the exclusionary area surrounding the reactor site, also
use a significant land area. Land is converted to energy-related use for at
least the life of the plant. Consequently, change in land use and ecological
systems are important issues in power plant siting.
A variety of regulatory legislation and agency policies contain provisions
that are relevant to the impacts of generating unit additions on land use and
ecological systems. These include the National Historic Preservation Act
(PL 89-665), The Endangered Species Act (PL 93-205), The Federal Land Policy
and Management Act (PL 94-579) and the National Forest Management Act (PL
94-588). Their effect has been to increase the range of impacts that are con-
sidered in environmental reviews, and the relative importance of each to actions
such as power plant siting. The definition of land use and ecological systems
that is sensitive to environmental decisions depends, to a large extent, on
the scarcity of resources. As a resource becomes more scarce, its value as
an element of land quality can also increase.
Four indices are selected to represent siting issues relevant to land use
and ecological systems in the ORBES region (Table 14). These are: natural,
scenic and recreational areas; sensitive and protected environments; agri-
cultural and ecological productivity; and the ownership and management of
forest lands. Each represents a resource that is important to subregions in
the ORBES area. Agricultural and ecological productivity, for example, is of
central importance in prime agricultural lands, which are concentrated north
of the Ohio River whereas conflict involving forest lands are more likely to
occur in the southern and eastern portions of the region. Each variable also
has reliable county-level data for all six ORBES states, as uniform data are
essential in making comparisons among the indices. Absolute values for each
variable are normalized on a scale of 0 to 10, with an index of "uniqueness"
used for sensitive and protected environments (Randolph and Jones, 1980).
The weights assigned to each parameter can reflect evaluations of the rela-
tive importance of the resources, as well as scenario policy issues. They
are combined to form a suitability index for the land use and ecological sys-
tems component.
Xl Natural, Scenic and Recreational Areas
The extent of public lands is used as the measure of
natural, scenic and recreational areas. Public lands are all
state and federally-owned lands that are managed for special
71
-------
Table 14. DEFINITION OF THE LAND USE AND ECOLOGICAL SYSTEMS COMPONENT
Ll
Natural,Scenic and
Sensitive and
Protected Environments
Agricultural and
Ecological Productivity
Forest Lands
Ownership and Management
IS}
% County in
X.
1
Public
0
5.1
10.1
15.1
20.1
25.1
30.1
35.1
40.1
45.1
Component
Lands
- 5 =
- 10 =
- 15 =
- 20 =
- 25 =
- 30 =
- 35 =
- 40 =
- 45 =
- 50 =
Index
1
2
3
4
5
6
7
8
9
10
(CJ:
X2 = 3
£ ^iNij
i=l
where U. is a unique-
ness coefficient
1 = normal
2 = medium
3 = high
NJ; is number of areas
in each category.
)— UY 4- U Y 4-UY 4-UY
3 1*1 j 2A2j w3A3j 4 '
% County in
Class I &
X, = 0 -
11 -
21 -
31 -
41 -
51 -
61 -
71 -
81 -
91 -
4
« - * ,V
II Soils
10 =
20 =
30 =
40 =
50 =
60 =
70 =
80 =
90 =
100 =
ij
1
2
3
4
5
6
7
8
9
10
% County in
Non-Federal Forest
x4 = o -
11 -
21 -
31 -
41 -
51 -
61 -
. 71 -
81 -
91 -
10 =
20 =
30 =
40 =
50 =
60 -
70 =
80 =
90 =
100 =
1
2
3
4
5
6
7
8
9
10
where W = weighting factor for the ith criterion
X = numerical ranking for the ith criterion in the jth county
-------
uses. They include all state and federal parks, forests,
wildlife areas, public hunting and fishing areas, historical
landmarks, and government installations such as Fort Knox,
Kentucky and The Crane Naval Munitions Depot in Indiana.
The majority of the public lands are concentrated in the
southern part of the ORBES region (Figure 20). Parks, forests
and wildlife areas account for the majority of counties that
have relatively high scores. Counties that have >_ 50% of their
area in public lands are assigned a score of 0.0, which excludes
them from consideration.
j
2 Sensitive and Protected Environments
An index of "uniqueness" for natural areas is used as the
measure of sensitive and protected environments. These include
state nature preserves and other natural areas that are gen-
erally recognized by state and academic authorities as having
important ecological significance. Counties that have the high-
est uniqueness scores are in Illinois, Ohio and Pennsylvania
(Figure 21).7 They may be less suitable as sites for new elec-
trical generating units because of the need to consider the
impacts that might result from facility location or design.
However, because sensitive and protected environments generally
do not occupy large areas, they are not considered to be exclu-
sionary criteria.
^
3 Agricultural and Ecological Productivity
The extent of Class I and Class II soils is the measure of
agricultural and ecological productivity. The distribution of
Class I and Class II soils defines the potentially most produc-
tive parts of the agricultural lands in the ORBES region. These
prime agricultural lands account for 39% of the total area in the
region and 72% of the agricultural lands. Corn is the most im-
portant crop, with much smaller acreages in soybeans and winter
wheat. These and other conventional grains are important sources
of food and feed. The U.S. Department of Agriculture (USDA),
USEPA and other federal and state agencies have recently adopted
policies designed for the preservation of agricultural land. The
conversion of farm land for energy related activities, such as
coal mining and power plant siting, is a special concern in the
prime agricultural lands of the ORBES region.^
The largest extent of prime agricultural land is located in
a wedge from central Illinois through central Indiana and west
central Ohio (Figure 22). This highly productive, relatively
level farmland is devoted primarily to corn, soybeans and other
cereal grains that are used for feed and food. Counties around
the periphery of this wedge, and in the southwestern part of
Indiana and western Kentucky, have smaller portions of their land
73
-------
Figure 20. NATURAL. SCENIC, AND RECREATIONAL AREAS
EXTENT OF PUBLIC LANDS: RELATIVE SCORES
88. - 10.
6. - 8.
HO. - 2.
PREPARED rO? OHIO DTVtR BASH ENERGY STUDY
BY c»cis/ucc, rrawjAifY. IMC
-------
Figure 21.
SENSITIVE AND PROTECTED ENVIRONMENTS
UNIQUE NATURAL AREAS: RELATIVE SCORES
B. - 10.
6. - 8.
H. - 6.
2. - 1.
0. - 2.
PKPAKO TOT OHM RNCX BASH ENCR6Y STUOr
IT CAOS/UCC. rnnwApy. two
-------
Figure 22. AGRICULTURAL AND ECOLOGICAL PRODUCTIVITY
EXTENT OF CLASS I * CLASS H SOILS: RELATIVE SCORES
8. - 10.
6. - 8.
4. - 6.
2. - 4.
0. - 2.
MCPARED f OR OHIO PVER QASM ENERGY STUDY
C. FEBRUARY. I MO
-------
area in Class I and II soils. Except for specialized agricul-
tural areas, such as the Bluegrass Basin in north central Ken-
tucky, the remainder of the southern and eastern part of the
region has relatively little high quality land. Counties that
have relatively high suitability scores are considered to be
less suitable as sites for new generating units than those with
low suitability scores. However, agricultural and ecological
productivity is not a sufficient condition to exclude a county
from consideration.
Y
4 Ownership and Management of Forest Lands
The extent of non-federal forests is the measure of the
ownership and management of forest lands. Federal forests are
not included, as they are represented in the public lands cate-
gory. The distribution of non-federal forests is the mirror
image of the map of Class I and Class II soils (Figure 23). The
majority of the forests are in the Appalachian areas of eastern
Kentucky and West Virginia, southeastern Ohio and western Penn-
sylvania. Kentucky has the largest number of acres whereas West
Virginia has the largest proportion of its land area in forests.
Most of the forests are in small, privately-owned tracts. Forest
products, primarily hardwoods, are used for manufactured wood
products.
In combination, the impact of ecological systems and land use criteria
on reducing the suitability of counties as sites for coal-fired and nuclear-
fueled scenario unit additions is greatest in Illinois, Indiana and Ohio, and
in the extreme eastern part of the region (Figure 24). Illinois has the
largest number of counties with relatively low (<_ 6) suitability scores. Com-
pared with other states, the geography of ecological and land use resources
in Illinois is complex. Prime agricultural lands is the dominant factor in
reducing the suitability of counties in Indiana and western Ohio, whereas
natural areas and forest lands are most important in Pennsylvania and West
Virginia. Ecological systems and land use factors add constraints to site
selection in these counties, but they alone are not sufficient under cur-
rent practice to exclude a county from consideration.
Seismic Suitability
Seismic suitability is an important consideration in siting nuclear-
fueled electricity generating units. The safe operation and shutdown of nucle-
ar reactors under the stress of earthquake vibrations is at issue. Seismic
criteria are included in evaluating the physical characteristics of proposed
sites (Title 10, CFR, Part 100). The applicant for a construction permit is
required to perform certain specified engineering and geologic investigations
to determine: 1) the maximum vibratory ground motion produced by the strong-
est earthquake that could potentially affect the site; 2) whether and to what
extent the proposed nuclear power plant should be designed for surface fault-
ing; and 3) the potential for the site to be exposed to seismically induced
water waves or floods. Even if seismic site characteristics are unfavorable,
77
-------
Figure 23. OWNERSHIP AND MANAGEMENT OF FEDERAL FORESTS
oo
EXTENT OF NON-FEDERAL FOREST: RELATIVE SCORES
OB. - 10.
He. - a.
HO. - 2.
PfiCPAREO "OR OHIO RIVER BASH ENERGY STUDY
BTCAGK/UCC rCBRUARY. 1*80
-------
vO
Figure 24. ECOLOGICAL SYSTEMS AND LAND USE COMPONENT
STRICT ENVIRONMENTAL CONTROLS
INDEX VALUE: RELATIVE SCORES
HS. - 10.
3]6. - 8.
U. - 6.
_2. - 4.
|0. - 2.
PREPARED rOR OHIO SIVER 8ASN CNCKGV STUDY
BY CACK/UCC. TEBRUAFT. 19BO
-------
the proposed site may be approved if facility design includes appropriate and
adequate compensating engineering safeguards. The guiding principle in the
NRC's determination as to seismic suitability of a proposed nuclear generating
station is whether or not the proposed design is adequate to provide for the
safe shutdown of the reactor under the worst credible earthquake or fault
conditions that could affect the site.
Electric utility companies consider seismic criteria as economic factors.
The consulting geogolists that are required for site investigation in active
or potentially active seismic zones, and capital outlays and engineering for
improved seismic design, can be very expensive. Last minute licensing delays
resulting from inadequate preliminary geologic — geotechnical siting studies
have, in several cases, contributed to abandoning certain proposed sites
(McClure, Jr. and Hatheway, 1979, p. 6). Consequently, utilities tend to avoid
siting nuclear reactor zones that are seismically very active and to review
carefully the costs and benefits of sites in areas that have even occasional
significant seismic activity.
Earthquakes in the central United States have several characteristics
that distinguish them from their western counterparts. They occur infre-
quently; none has produced surface breakage in historic times; their seismic
wave energy shows much smaller anelastic attenuation; and, as a result of the
first two reasons, less is known about them (Nuttli, 1979, p. 92). Three
broad areas of the ORBES region are accompanied by some degree of seismic
risk. These are:
1. The southwestern part of the region, including southern
Illinois, western Kentucky and southwestern Indiana.
2. A small area in the northcentral portion of the region,
including ten counties in west-central Ohio.
3. The eastern part of the region, including portions of west-
ern Pennsylvania, West Virginia and a few counties in south-
eastern Kentucky.
The most significant of these is the area in the southwestern part of the
region. It includes parts of the New Madrid and the St. Francois seismic
zones, and the entire Wabash Valley seismic zone.
The New Madrid seismic zone has been by far the most active seismic re-
gion in the central United States during the last 200 years (Nuttli, 1979, p.
68). The three principal shocks that occurred in this portion of southeast
Missouri in 1811 and 1812 had body-wave magnitudes greater than 7.0. Minor
damage was experienced as far north as Lake Michigan and as far east as West
Virginia.^ Two other areas have histories of earthquake activity. One is
centered on Shelby County, Ohio and the other is the Appalachian Plateau in
the eastern part of the region. Although neither of these has historically
experienced earthquakes of sufficient magnitude to preclude nuclear genera-
ting stations, the need for conservative, and thus more expensive, seismic
design could influence the decision to site in these areas.
80
-------
Seismic criteria are included in the ORBES siting model to simulate the
decision-making process of electric utility companies in evaluating sites
for nuclear reactors. Certain areas are excluded from consideration as can-
didate sites for nuclear units and other areas were given relatively unfavor-
able suitability scores depending on the degree of difficulty one might en-
counter in searching for a site that would be acceptable to the NRC and
could be built on at reasonable cost.
Suitability is based on a county's location with respect to the three
seismic zones that the NRC used in coarse screening to identify potential
nuclear energy center siting regions (U.S. Nuclear Regulatory Commission,
1976, pp. 2-6, 2-7). These zones are:
• Zone I: includes areas of low seismicity with no known capable
faults. It is expected that seismically suitable sites can be found
with little difficulty.
• Zone II: includes areas with moderate seismicity and complex geo-
logical structures, having numerous, old, incapable faults; and
areas close to zones of high seismic risk, that may lead to con-
troversial risk assessment. Detailed site-specific studies would
be necessary to determine geologic and seismic site suitability.
• Zone III: is characterized by high seismicity, accompanied in
most cases by intense, recent faulting. In general, the cost and
time required for investigation of site suitability makes it im-
practical to consider these areas for nuclear power plants.
The relative seismic suitability zones are depicted by the NRC on a
series of regional maps that are used to assign seismic suitability scores
to counties for the ORBES siting model (Table 15). For example, a county
that is located entirely in Zone III is assigned a score of 0.0, which ex-
cludes it from consideration for a nuclear reactor unit. Other counties
are assigned standard scores depending on their location relative to the
three seismic zones.
Counties in southern Illinois, southwestern Indiana and extreme western
Kentucky that are entirely in seismic Zone II or are in Zone II and III, are
either excluded from consideration as candidate sites or have very low suit-
ability scores (Figure 25). They are bordered by counties that are more suit-
able with respect to seismicity, although potential sites for nuclear reactors
in these areas usually are subject to careful evaluation. Another large area
of moderate seismic suitability is located in the Appalachian Plateau of West
Virginia and western Pennsylvania. The areas of low seismic suitability are
geographically coincident with the location of extensive coal reserves. Al-
though no nuclear reactors are located in these areas, a large number of coal-
fired generating units are in service and others are planned.
Population Distribution
Federal regulations encourage siting nuclear reactors away from large
concentrations of population. Title 10 CFR, Part 100 specifically includes pop-
81
-------
Table 35. DEFINITION OF SEISMIC SUITABILITY SCORES FROM
RELATIVE SEISMIC SUITABILITY ZONES
County
Location
Entirely in Zone
In both Zones
Entirely in Zone
In both Zones
Entirely in Zone
Relative Seismic
Suitability Zone3
I
I-II
II
II-III
III
Relative
Score
10.0
7.5
5.0
2.5
0.0
aSource: U.S. Nuclear Regulatory Commission, 1976, pp. 2-1 and 2-6.
82
-------
Figure 25. SEISMIC SUITABILITY COMPONENT INDEX
oo
w
INDEX VALUE: RELATIVE SCORES
8. - 10.
6. - 8.
H. - 6.
2. - M.
0. - 2.
MCPAfttD FOR OMO RIVE* BASM ENOCT STUOT
ITCAGS/UCC. rcMttARY. i«w
-------
ulation distribution characteristics among the factors to be considered Ln
determining site acceptability. Public health and safety is the issue. Al-
though the NRC does not define levels of acceptability for population charac-
teristics, locating reactors away from densely-populated areas is considered
to be the most important siting constraint in states such as Illinois (Laney
and Gustafson, 1979). The problem is to minimize transportation and land
acquisition costs while maximizing system safety and reliability. During
the 1970's, the trend in nuclear reactor siting has been to locations away
from densely-populated areas with access to the utility grid and adequate
water supplies.
The NRC has developed a technique for describing population characteris-
tics that can be used in evaluating alternative sites for nuclear reactors.
The site population factor (SPF) is an index that weights the cumulative pop-
ulation with a function that decreases with increasing distance from the pro-
posed reactor site (Kohler, Kenneke and Grimes, 1975). This is consistent
with the idea that risk to an individual decreases as distance from the
source of radioactivity increases. The distance factor is derived from an
analysis of meteorological dispersion data; and the population distribution
is normalized to an area with a uniform density of 1000 persons per square
mile. With a bounding radius of 30 miles, a SPF = 0.3 is equivalent to 300
persons per square mile distributed uniformly out to a distance of 30 roUes
from the proposed reactor site.
SPF contours have been drawn for the contiguous United States using 1970
residential population data (Kohler, Kenneke and Grimes, 1975). The maps
assumed locations at the intersections of each 0.1 degree latitude and longi-
tude lines for densely populated areas, and at the intersection of each 0.25
degree lines for low density areas. At this scale, the SPF maps clearly out-
line the major cities and their urbanized areas (Louisville, Kentucky; Indi-
anapolis, Indiana; Cincinnati-Dayton and Columbus, Ohio; and Pittsburgh,
Pennsylvania). Each occupies an area with SPF contours of 0.5 and higher.
Most smaller metropolitan areas are also shown, although they have SPF con-
tours of between 0.4 and 0.2.
The population distribution component of the siting model is based upon
1970 county population densities. Counties are rated on a scale of 1 to 10,
according to densities that are analagous to SPF's at a distance of 30 miles
(Table 16 and Figure 26). Counties with low scores have high population densities
with a score of 5 equivalent to a SPF = 0.5, or 500 persons per square mile.
Counties with the lowest suitability scores are those which have the central
cities of large metropolitan areas. Most other counties that have scores £ 8
contain the urbanized areas of large cities or are smaller metropolitan areas.
Counties with a relative score _> 5 (i.e., a SPF = _> 0.5) are excluded From con-
sideration as candidates for nuclear-fueled scenario unit additions.
DEFINITION OF SITING WEIGHTS
The definition of siting weights for the ORBES region is based upon in-
formation collected using a modification of the Nominal Group Porcess Tech-
nique (NGT). Given the specification of siting components, members of the
ORBES Core Team and Advisory Committee were asked to evaluate the relative
84
-------
Table 16. SUITABILITY SCORES FOR POPULATION
DISTRIBUTION COMPONENT
Population Density, 1970 County Score
0 - 99.999 10
100.0 - 199.999 9
200.0 - 299.999 8
300.0 - 399.999 7
400.0 - 499.999 6
500.0 - 599.999 5
600.0 - 699.999 4
700.0 - 799.999 3
800.0 - 899.999 2
> 900.0 1
85
-------
Figure 26. POPULATION DISTRIBUTION COMPONENT
00
INDEX VALUE: RELATIVE SCORES
DB. - 10.
36. - 8.
Eh- - 6.
132. - IJ.
DO. - 2.
PREPARED FOP OHIO RIVE* BASM ENERGY STUDY
BTCACIS/IMCC. rCBRtMRY. 198O
-------
importance of each criterion to siting a standard 650 MWe coal-fired or 1000
MWe nuclear-fueled electrical generating unit within the region. The func-
tional relationships of each component and variable Lo regional siting issues
had been discussed previously in presentations to the Core Team and Advisory
Committee.
Each person indicated his or her evaluation of the relative importance
of the criteria on a specially designed instrument (Figure 27). The relative
weight for each component was shown by a line drawn to the appropriate point
on a continuous graphic scale from 0 (unimportant) to 10 (most important).
According to Voelker (1977, pp. 2-3), such a scale is appropriate for a group
which is "technically qualified to make refined distinctions" among siting
criteria. The scale is also appropriate for the level of detail and accuracy
desired from the siting model.
The first round voting involved relatively little prior discussion about
the substantive issues of power plant siting. The objective was to obtain an
initial set of data. After the results were tabulated, the group reconvened
for a second round of,voting. Each person received a tabulation of the mean
and standard deviation for each criterion as well as her or his original vote.
The group discussed the importance of the siting criteria to the ORBES region
and the distribution of individual evaluations for approximately one hour
prior to the second vote. The objective of the second iteration was to im-
prove the accuracy of the group output, and to reduce the dispersion among
individual votes.
The methodology was an optimal use of the NGT technique that ORNL had
previously applied to develop the siting model for the National Coal Utiliza-
tion Assessment (Davis et al, 1979).
1. The siting components and variables had been selected to
represent issues that were especially relevant to power
plant siting in the ORBES region, and central to the
assessment's scenario policies. The issues had been thor-
oughly discussed and presented to the Core Team on several
occasions. Thus, the first three steps of the NGT process
were unnecessary (Voelker, 1977).
2. The siting criteria were evaluated with respect to locating
scenario unit additions in the ORBES region only. An eval-
uation of criteria "in general" (i.e., at national scale) is
inappropriate for a regional technology assessment.
3. The relative importance of the criteria were evaluated si-
multaneously for coal-fired and nuclear-fueled units.
No attempt was made to evaluate "site specific" criteria, as the objective
of the ORBES siting methodology is to distribute scenario unit additions at
regional scale.
87
-------
Figure 27. DEFINITION OF WEIGHTS FOR SITING COMPONENTS
SITING COMPONENTS
NUCLEAR
• AIR QUALITY •
in —
9 :
8
7 -
6 :
4
3
2
1
0
S00 •
TSP "
i.o i.o j
•
• ECOLOGICAL SYSTEMS • :
AND LAND USE :
Public Lands -
Unique Natural
Areas ;
Forest Land
Agricultural Land ;
1.0 1.0 1
• WATER AVAILABILITY «
• POPULATION DENSITY •
• SEISMIC SUITABILITY •
10
9
9
8
7
6
4
3
2
1
0
88
-------
The rank order of the siting components, and most variables, for each
fuel type were the same in Round 2 ns in Round 1 (Table 17 and 18). The
means of some criteria changed significantly. However, the standard devia-
tion decreased in all cases except one. The objective of increasing group
agreement in the second round of voting was achieved.
Ambient air quality was judged to be the most important consideration
for siting coal-fired plants. Population density and seismic suitability
were most important for nuclear-fueled units. These results were expected.
However, the fact that the ecological systems and land use component ranked
higher than water availability for both fuel types was not consistent with
the results of similar siting studies. Ecological systems and land use
variables have a more prominent role in the ORBES siting model than in those
that are national in scale. The importance assigned to this component under-
scores the sensitivity of the model to the regional characteristics of the
study area.
Although the ecological systems and land use component was considered
to be slightly more important for siting coal-fired units, the weights were
essentially the same for each round of voting and across fuel types. Even
the change in importance of the unique natural areas and agricultural lands
variables in the Round 2 voting was the same for coal-fired and nuclear-
fueled units. Water availability was considered to be almost as important
as ecological systems and land use in siting nulcear units. But for coal-
fired units, water availability was evaluated significantly lower on the
scale.
Water availability and engineering considerations were the most important
siting variables in ORNL's evaluation. Ecological considerations were next in
importance, with land use compatability identified as an issue that electric
utilities and others perceived as a central issue in site selection. How-
ever, water resources are considered to be relatively plentiful in the ORBES
region. Supplying the water requirements of standard plants is considered to
be an economic matter, and thus less restrictive than ecological, land use
and air quality criteria for siting new electricity generating facilities.
The application of NGT to the ORBES siting issues resulted in a consis-
tent set of weights that can be used with confidence in calculating site
suitability indices for siting standard coal-fired and nuclear-fueled elec-
tricity generating facilities in the study region. The weights are a type
of baseline data that are sensitive to expert evaluations of the relative
abundance of regional resources, such as water, and their importance in siting
generating capacity additions under current economic, regulatory and techno-
logical conditions. Policies that may affect power plant siting by altering
any of these assumed relationships can be simulated by systematic changes in
the weights, as well as the set of candidate counties to which they apply.
Definition Of ..Site Suitability for Basic Scenarios
Three formulas define the suitability of counties in the ORBES region as
sites for coal-fired or nuclear-fueled generating units additions in the basic
scenarios. These are:
89
-------
Table 17. WEIGHTS FOR SITING COMPONENTS AND VARIABLES FOR
COAL-FIRED ELECTRICITY GENERATING FACILITIES IN THE
OHIO RIVER BASIN ENERGY STUDY REGION
Weights
Siting Components
and Variables
Air Quality
Sulfur Dioxide (S02)
Total Suspended Farticulates
Ecological Systems and
Land Use
Public Lands
Unique Natural Areas
Forest Land
Agricultural Land
Water Availability
Population Density
Seismic Suitability
1st
Mean
(X)
9.01
0.52
(TSP) 0.48
7.55
0.20
0.27
0.19
0.34
3.79
2.46
0.97
N=23
Round
Standard
Deviation
€
1.67
0.15
0.15
1.79
0.10
0.12
0.08
0.10
1.91
2.69
2.93
2nd
Mean
(X)
9.15
0.50
0.50
7.62
0.20
0.37
0.17
0.29
3.34
2.26
0.86
N=19
Round
Standard
Deviation
^
1.04
0.14
0.14
1.55
o.io
0.17
0.08
0.11
1.94
2.31
1.93
90
-------
Table 18. WEIGHTS FOR SITING COMPONENTS AND VARIABLES
FOR NUCLEAR-FUELED ELECTRICITY GENERATING FACILITIES
IN THE OHIO RIVER BASIN ENERGY STUDY REGION
Weights
Siting Components
and Variables *
Ecological Systems and
Land Use
Public Lands
Unique Natural Areas
Forest Land
Agricultural Land
Water Availability
Population Density
Seismic Suitability
1st
Mean
(X)
7.04
0.21
0.26
0.18
0.35
6.6
8.5
8.4
N=23
Round
Standard
Deviation
4
1.82
o.io
0.11
0.78
0.16
2.1
2.06
2.00
Mean
(X)
7.05
0.21
0.36
0.18
0.28
5.57
8.86
9.06
N=19
2nd Round
Standard
Deviation
^
1.43
0.10
0.17
0.08
0.11
2.15
1.28
1.11
*Air quality was excluded from consideration in siting nuclear-fueled
facilities.
91
-------
1. Coal-based Scenarios, Base Case Environmental Controls
Sj = 9.15 [(0.5) X1;jl + (0.5) X2jl] + 7.62 [(0.20) Xlj2 (5)
-I- (0.37) X2j2 + (0.17) X3j2 + (0.29) X4j2] + 3.34(Xlj3>
9.15 + 7.62 + 3.34 + 2.26 + 0.86
2. Coal-based Scenarios. Strict Environmental Controls
S.. = 9.15 [(0.5) Xljl + (0.5) X2jl] + 7.62 [(0.20) X (6)
+ (0.37) X2j2 + (0.17) X3j2 + (0.29) X4j2] + 3.34
+ 2.26 (X) + 0.86 (
9.15 + 7.62 + 3.34 + 2.26 + 0.86
3. Nuclear Emphasis, Base Case Environmental Controls
S.. = 7.05 [(0.21) Xj^ + (0.36) X2-2 + (0.18) X3-2 (7)
(0.28) X] + 5.57 (X]J3) + 8.86 (X) + 9.06
7.05 + 5.57 + 8.86 + 9.06
In each case
Sj = the absolute suitability index for the jth county
Xlil = numerical ranking for S02 in the Air Quality Component
J for the jth county
X2jl = numerical ranking for TSP in the Air Quality Component
for the jth county
Xlj2 = numerical ranking for Natural, Scenic, and Recreatioiuil
Areas in the Land Use and Ecological Systems Component
for the jth county
X2j2 = numerical ranking for Sensitive and Protected Environments
in the Land Use and Ecological Systems Component for tho
jth county
X3j2 = numerical ranking for Agricultural and Ecological Produc-
tivity in the Land Use and Ecological Systems Component
for the jth county
92
-------
X,-2 = numerical ranking for Forest Lands Ownership and Management
in the Land Use and Ecological Systems Component for the
jth county
X,-3 = numerical ranking for the Water Availability Component for
the jth county
X,., = numerical ranking for the Population Density Component for
the jth county
X1'5 = numerical ranking for the Seismic Suitability Component for
the jth county
All components and criteria are used to define site suitability in the
coal-based scenarios with strict environmental control policies, whereas the
seismic suitability and population density components are not included in de-
fining base case environmental controls. The nuclear-based scenarios, which
exclude the air quality component, assume current environmental control poli-
cies that apply to nuclear-fueled generating units. The county-level patterns of
site suitability for each of the basic scenarios are significantly different.
Coal-based, Base Case Environmental Control Scenarios
The coal emphasis, base case environmental control scenarios, have the
largest number of counties with relatively high suitability indices (Figure
28), The most suitable counties border the Ohio River main stem and the lower
reaches of its major tributaries upstream and downstream of Louisville, Ken-
tucky. This reflects the importance of water availability relative to other
siting components, such as air quality, which is defined by primary standards.
Elsewhere, the majority of counties have better than average suitability (with
scores ranging from 6 to 8). In general, counties in Illinois, Indiana, Ohio
and Pennsylvania are more likely to be less suitable as sites than those in
Kentucky and West Virginia. Land use and ecological system criteria are more
important north of the Ohio River.
The majority of the counties in the ORBES region are candidate sites for
coal-fired scenario unit additions under base case environmental control poli-
cies (Table 19 and Figure 29). Relatively few are excluded from consideration,
as the exclusionary criteria are defined in a liberal manner. For example, an
entire county must be designated a nonattainment area, or be in public lands,
to be excluded from the list of candidate counties. This assumes that sites
can be located in counties that contain nonattainment areas or acreages of
public land. Siting is also restricted in counties (17) with > 1950 MWe
scheduled for 1985, as the addition of a 650 MWe scenario unit would exceed
the 2400 MWe maximum. The majority of the counties that are excluded are
located along the Ohio River main stem north of Louisville, and in Ohio, Penn-
sylvania and West Virginia. However, the majority of the counties with the
highest suitability scores are candidate counties.
93
-------
Figure 28. SITE SUITABILITY INDEX, COAL-FIRED SCENARIO UNIT ADDITIONS
, BASE CASE ENVIRONMENTAL CONTROLS
INDEX VALUE: RELATIVE SCORES
9. - 10.
8. - 9.
7. - 8.
6. - 7.
3. - 6.
mtPAKCD TO* OMO RMX MS* EMCKGT STUW
VTCMB/lNCC. FTMNMirr. 1MO
-------
vO
Table 19. SUMMARY OF COUNTIES EXCLUDED AS SITES FOR COAL-FIRED SCENARIO UNIT ADDITIONS
IN BASE CASE ENVIRONMENTAL CONTROL SCENARIOS
State
Subregion
ILLINOIS
INDIANA
KENTUCKY
OHIO
PENNSYLVANIA
WEST VIRGINIA
Total
Counties
Air Quality
Violation
of NAAQS3 PSD
Class Ib
S02 TSP Areas
2
5 6 2
5
1 1
1 2
14 7 4
Public
Landsc
Total
o
4-
9 2
3
11 2
Combined
Total
Excluded
2
10
5
1
3
21
Total
r\DHi7C
UK.DE.O
Counties
85
83
120
68
19
48
423
aCounty designated nonattainment area, primary standards.
^County contains mandatory Class I area.
CA11 of county in public lands; actual ownership.
-------
Figure 29. COUNTIES EXCLUDED AS SITES FOR COAL-FIRED SCENARIO UNIT ADDITIONS,
BASE CASE ENVIRONMENTAL CONTROLS
ILLINOIS
WEST
VIRGINIA
J KENTUCKY
EXCLUSIONARY CRITERION
AIR QUALITY
Non •tUiMMnt SO* •• l*r
PSO Class I
Public Lands
2 1960 MWe proposed in 1986
-------
Coal-based. Strict Environmental Control Scenarios
The site suitability indices for coal-based, strict environmental control
scenarios are lower, and the geographical pattern is more complex (Figure 30).
The sequence of highly suitable sites along the Ohio River main stem and trib-
utaries is significantly reduced, and the extent of counties with lower suit-
ability indices north of the Ohio River is expanded significantly, especially
in Illinois. These changes are primarily the result of using secondary stan-
dards to define ambient air quality, with the seismic suitability and popula-
tion distribution components adding constraints in selected counties.
The exclusionary criteria are defined more conservatively for strict en-
vironmental control policies (Table 20 and Figure 31). For example, a county
is excluded if it contains a nonattainment area for S02 or TSP, or the majority
of its area is in public lands, including the designated purchase area. Other
exclusionary criteria are the same as in the base case. The net effect is to
increase significantly the number of counties that are excluded from considera-
tion as candidate sites, and to change the distribution of the counties that
are available for scenario unit additions. Large portions of Ohio and Penn-
sylvania are excluded, as are counties along the Ohio River main stem and most
metropolitan areas. TSP is the most important exclusionary criteria. In com-
bination with the generally lower site suitability indices for strict control
policies, the exclusionary criteria significantly reduce the choice of highly
suitable sites to a few clusters of counties upstream of Cincinnati, Ohio;
from Cincinnati to Louisville, Kentucky; and downstream of LouisviJle.
Nuclear Emphasis, Base Case Environmental Controls
The pattern of site suitability for the nuclear emphasis scenarios differ
significantly from both of those that emphasize coal-fired generating units
(Figure 32). The dominance of the seismic suitability and population distri-
bution components is apparent, as well as the influence of ecological systems
and land use criteria in central Illinois, Indiana and Ohio. Three areas have
relatively high suitability indices. The most suitable counties are along the
Ohio River main stem upstream from Louisville, Kentucky. Counties on major
tributaries of the upper Ohio River, and in east central Kentucky, also have
high suitability indices. Counties along the upper Illinois River in north-
western Illinois, and in northern Indiana, are also suitable sites.
Relatively few counties are excluded as sites for nuclear-fueled scenario
unit additions (Table 22 and Figure 33). The majority of these are in seismic
Zone III in the southwestern part of the region. Counties with the majority
of their area in public lands and densely-populated counties that include the
region's largest cities account for the remaining excluded counties. Large
portions of the OR8ES region, especially along the middle and upper Ohio River
and its tributaries, have high suitability scores and are available as candi-
date sites for nuclear-fueled units.
97
-------
Figure 30. SITE SUITABILITY INDEX, COAL-FIRED SCENARIO UNIT ADDITIONS
STRICT ENVIRONMENTAL CONTROLS
00
INDEX VALUE: RELATIVE SCORES
9. - 10.
8. - 9.
7. - 8.
6. - 7.
3. - 6.
MtPMEO fOA OHIO MVOt BASK CNOCY SWOT
•Y CAGB/MCC, rowAnr. i*ao
-------
Table 20. SUMMARY OF COUNTIES EXCLUDED AS SITES FOR COAL-FTRED SCENARIO UNIT ADDITIONS
IN STRICT ENVIRONMENTAL CONTROL SCENARIOS
\O
\0
State
Subregion
ILLINOIS
INDIANA
KENTUCKY
OHIO
PENNSYLVANIA
WEST VIRGINIA
Total
Counties
Air Quality
Violation
of NAAQSa PSD
S02
3
3
9
21
5
2
..
Class Ib
TSP Areas
19
8
15 2
38
8
6 2
94 4
Combined
Total
19
8
19
43
9
8
106
Public
Lands'"
4
4
9
3
1
3
24
Combined
Total
Excluded
22
12
27
44
10
9
124
Total
ORBES
Counties
85
83
120
68
19
48
423
aCounty designated nonattainment area, primary standards.
County contains mandatory Class I area.
CA11 of county in public lands; actual ownership.
-------
Figure 31.
o
o
COUNTIES EXCLUDED AS SITES FOR COAL-FIRED SCENARIO UNIT ADDITIONS.
STRICT ENVIRONMENTAL CONTROLS
ILLINOIS
EXCLUSIONARY CRITERION
AIR QUALITY
Non attainment
PSD Clsss I
Public Lands
>19SO MW« in IMS
-------
Figure 32. SITE SUITABILITY INDEX. NUCLEAR-FUELED SCENARIO UNIT ADDITIONS
INDEX VALUE: RELATIVE SCORES
9. - 10.
8. - 9.
7. - 8.
6. - 7.
3. - 6.
PfCPMCD FOR OHIO RIVER BASM ENERGY STUDY
•TCMB/UCC. rtWNMRY. 1980
-------
Table 21. SUMMARY OF COUNTIES EXCLUDED AS SITES FOR NUCLEAR-FUELED SCENARIO UNIT ADDITIONS
State
Subregion
ILLINOIS
INDIANA
KENTUCKY
OHIO
PENNSYLVANIA
WEST VIRGINIA
Total
Counties
Seismic
Suitability8
28
4
14
46
Population
Density6
2
4
7
I
1
15
Public
Landsc
4
4
9
3
1
3
24
Combined
Total
Excluded
28
9
27
10
2
4
80
Total
ORBES
Counties
85
85
120
68
19
48
423
Majority of county within seismic Zone III.
b
County population density _> 500 persons per square mile.
Majority of county in public lands; total area, Including designated purchase area.
-------
o
OJ
Figure 33. COUNTIES EXCLUDED AS SITES FOR
NUCLEAR-FUELED SCENARIO UNIT ADDITIONS
ILLINOIS
WEST
VIRGINIA
EXCLUSIONARY CRITERION
KENTUCKY
Seismic ZWM III
PSD Class I
Public Lands
> 3000 MM proposed in 1986
-------
FOOTNOTES
The 'region1 is defined relative to the site, and to attainment and
nonattainment areas (McHugh, 1978).
2
Cf. McHugh (1978). Air quality modeling, which was used to define non-
attainment areas in Indiana, is a more conservative method that results in
fewer, and smaller, areas than using measured data, such as in Illinois.
•i
Regional air quality issues are discussed by McLaughlin (forthcoming).
The data are from the states' annual air quality reports for 1977,
except Indiana, which was for 1977-1978.
5In calculating plume rise, the Holland and TVA-Concurve equation re-
sults were averaged for speeds <_ 4 m/s; for > 4 m/s, the TVA-Concurve and
Briggs equation are averaged. Such decision was made on the basis of a TVA
plume rise report (TVA, 1968 and 1974).
V d
Holland : Ah = -§- [1.5 + .0096Q,/V d]
u n s
TVA-Concurve: Ah = 4.71 "69^
u"
C = 1.58 - 41.4 *1
Briggs : Ah = ±±^ At
u
F = gV d (T - T )/AT
s s a a
in each of these, A is in [m/s], Vg [m/s], d[m], u[m/s], Qh[kj/s], G(°K), Z(m)
and 0^ = 6.7 x 104 KJ/S
QTSp = 8.33 x 107 ug/S
H = stack height + Ah
1 17 x
Steady-state X = ±-=+ - = — for S00
max uo o 2
Y Z
104
-------
= A-87 x 10 for particulate
UO 0
Y z
where X has the units
For a detailed discussion of the impacts of energy development in the
ORBES region on land use and ecological systems, see: Randolph and Jones
(forthcoming) .
To a certain degree, the definition of sensitive and protected environ-
ments depends on the way in which states identify natural areas; see: Ran-
dolph and Jones (forthcoming).
Q
Fletcher (1980) and USEPA (1977). See also the discussion in this
report, Section 6, pp. 127-133.
9
In addition to concern over the recurrence of such an event in the
New Madrid seismic zone, the NRC has hypothesized that an equally large earth-
quake could occur in the Wabash Valley seismic zone. This conservative view
toward seisraicity in the Wabash Valley is documented in several letters from
the NRC to Illinois Power Company with respect to the Clinton Power Plant
which, at the time, was proposed for a site in De Witt County, Illinois. The
NRC contends that the Wabash Valley fault zone is structurally connected with
the New Madrid fault zone and, therefore, could experience an earthquake as
large as those at New Madrid. The Illinois Power Company contends that the
Rough Creek fault zone, which cuts across the trend of the Wabash Valley and
New Madrid fault zones, separates the two into unrelated seismic zone. The
outcome of this exchange was that the Clinton Safety Evaluation Report was
modified by Supplement No. 1 and the plant is now being built with enhanced
earthquake resistance.
Similar techniques were used to collect information on the relative
importance of siting criteria in ORNL's work with the Maryland Power Plant
Siting Program (Dobson, 1979). ORNL used the weights for coal-fired plants
from the Maryland study for siting capacity additions in the South for the
National Coal Utilization Assessment (Davis et al, 1978). Delbecq, Van de Ven
and Gustafson (1975) discuss the NGT and other Delphi techniques.
The Wilcoxon Matched-Pairs Signed-Ranks test was used to determine
whether or not the rank order of counties by site suitability was significantly
different for the basic siting scenarios (i.e., coal emphasis base case and
strict environmental controls; nuclear emphasis; and coal emphasis, very strict
air quality and agricultural lands protection policy, both with dispersed siting)
When compared with each other, the scenarios had significantly different site
suitabilities at well below the one percent level of confidence.
105
-------
SECTION 6
SITING PATTERNS FOR ORBES SCENARIOS
The siting patterns and on-line dates of capacity additions for the ORBES
scenarios are designed to facilitate impact assessments of the interrelation-
ships among different levels of energy demand, technology mix and environment-
al control policies. In the near-term, the schedule of sites and on-line dates
for capacity additions in most scenarios follow announced utility plans. The
number and type of scenario unit capacity additions that are necessary to meet
final demand in the year 2000 are added after 1985 according to the ORBES siting
model.
Two basic groups of scenarios are considered.1 The first assumes that a]J
scenario unit additions will be coal-fired. No nuclear-fueled units are sited
except those that the utilities had announced in 1975. Scenario 2, which as-
sumes base case environmental control policies and other current conditions,
is the point of reference for those scenarios that emphasize coal-fired elec-
trical generation. The second group emphasizes fuel substitution and conser-
vation. One scenario assumes an emphasis on nuclear-fueled generation, where-
as others assume that other fuels, or conservation, will dominate energy
supply and demand after 1985. A third group of scenarios derived from Scenario
1 simulate very strict air quality policies, and an agricultural lands pro-
tection policy. These are developed for purposes of special impact assessments.
COAL EMPHASIS, COUVE'CTIOrtAL TECHNOLOGY
Scenario 2; Base Case Environmental Controls
The majority of the scenario unit additions that are required to meet
electricity production in the year 2000 are sited in or adjacent to counties
that have existing and announced generating capacity (Figure 34).2 This re-
sults in the expansion of coal-fired generating units along the Ohio River
main stem upstream from Louisville, Kentucky and Cincinnati, Ohio; along the
upper Ohio River main stem in West Virginia and the coal fields of southeast-
ern Ohio; and in counties bordering the Allegheny and Monongahela Rivers in
western Pennsylvania. Scenario units are also added to existing or planned
concentrations in the lower Illinois River basin and at the confluence of the
Wabash River and the Ohio River. This scenario assumes continuation of current
trends in policies that affect siting coal-fired units, especially with regard
to environmental controls. Consequently, the scenario units are expected to
be located in proximity to existing and planned additions to coal-fired gen-
erating capacity, in areas that have sufficient resources to support new units.
106
-------
Figure 3A.
SCENARIO 2: CONVENTIONAL TECHNOLOGY, BASE CASE CONTROLS
TgTcL PROPOSED COPL-rIRED GENERATING
CPPRCITY RDDITIQNS. 1976-85
PLUS SCENARIO UNIT ADDITIONS. 1986^ 2000
OWES REGION
BOUNDARY
3000. - 5400.
2000. - 3003.
;OCO. - 2000.
500. - 1000.
£50. - 500.
100. - 250.
t. - 100.
o. - o.
(Planned additions)
Z NURwflF Of •CfeABTIO IMn
additions
-------
The geographical distribution of the capacity additions also suggests that
current environmental impacts may continue, if not intensify. The issue of air
quality is a case in point. At local scale, capacity additions are located in
or adjacent to counties that may already have significant problems of air qual-
ity degradation. Questions about rights to resources, especially such as those
raised by plants that are located close to one another across state boundaries,
may also increase. At regional scale, the majority of the scenario unit ad-
ditions are along the Ohio River main stem, with significant additions to the
concentration of plants in the eastern part of the region. Because the new
additions are in line with the prevailing winds, long-range pollutant trans-
port, acid precipitation and related issues should be examined carefully.
No nuclear-fueled scenario unit additions are sited in the scenarios
that emphasize long-term dependence on coal-fired generating capacity. The
nuclear-fueled units that are in service, under construction or announced in
utility plans are included in the assessment siting patterns (Figure 35).
Scenario 1; Strict Environmental Controls
In Scenario 1, changes in environmental policy with respect to the qual-
ity issue, have the most significant effect on siting scenario unit additions
(Figure 36). The geographical pattern is more dispersed, with a larger num-
ber of units located away from concentrations of existing and planned genera-
ting capacity. The change is especially pronounced in the eastern part of
the region. In Ohio, scenario units are located along tributaries to the Ohio
River, rather than the main stem; and in Pennsylvania, they are along the
Allegheny River. By comparison, the distribution of scenario unit additions
changes relatively little in the western part of the region. The majority of
scenario unit additions in Indiana and Kentucky are sited in or adjacent to
counties that have existing and announced electricity generating capacity.
A few units are displaced from the Ohio River main stem to its tributaries.
In Illinois, the number of units that are located in counties bordering the
lower Wabash River increases.
Compared with the base case scenario, the change to strict environmental
controls results in a shift in the geographical distribution of the coal-fired
scenario unit additions away from areas that have problems meeting air quality
standards to areas that have relatively meager water resources. In many in-
stances, especially in Ohio, some type of augmentation may be necessary to
provide the cooling water necessary for generating units. Water availability
is clearly a major economic and environmental issue. The cost of constructing
cooling ponds and reservoirs raises economic questions whereas water quality
impacts and land use change, especially in counties that have agricultural land
resources, are related issues. In addition, the question of the effect of a
more dispersed siting pattern in the eastern part of the region on ambient
air quality remains a central concern.
Scenario 7a and 7b: Very High Energy Growth
This scenario assumes an average annual growth rate in electricity that
is 3.1 times higher than in the base case. This requires siting an additional
69 scenario unit additions with a 35 year useful life, or an additional 49 units
108
-------
Figure 35. SCENARIO 2: CONVENTIONAL TECHNOLOGY. BASE CASE CONTROLS
TOTPL PROPOSED NUCLEPR GENERATING
CflPRCITT RDDITION5. 1976-85
NO SCENARIO UNIT ADDITIONS. 1388 2000
- 5400.
- 3000.
- 2000.
- 1000.
- 500.
- 250.
IOC.
MCGOWPT7S (Planned additions)
-------
Figure 36.
ORBES REGION
BOUNDARY
SCENARIO I: CONVENTIONAL TECHNOLOGY, STRICT CONTROLS
• Tnrq: PROPOSED CQPL -F 1 RED G E N E R P T I N G
CPPPCITY PDDITIONS.. 1976-85
PLUS SCBIAMO UNIT ADDITIONS. 19W-2000
- 5UOO.
- 3000.
- 20CC.
500. - 1000.
- 500.
Jj iOO. - 250.
Bl. - 100.
0. - 0.
ME CPUS' TS (Planned addition*)
2 Nwiter erf •cwiarlo wit
additions
-------
if the useful life assumption is increased to 45 years. Environmental control
policies are the same as in the base case scenario. In order to simplify the
assessment of impacts associated with a very high electrical energy growth
rate, the incremental number of scenario units is added to the siting pattern
for Scenario 2.
The siting pattern for this scenario, with 35 year useful plant life, has
three distinctive characteristics (Figure 37). First, the majority of the
scenario units are added to counties that are already identified in Scenario 2
as sites for capacity additions. Additional units are sited, sometimes to the
maximum of 2600 MWe per county. Second, scenario units are sited in areas of
meagre water resources, again primarily in the eastern part of the region.
Third, some units that are assigned to Ohio are sited in West Virginia and
Kentucky. Under strict environmental controls, the large number of scenario
units that might be allocated to Ohio do not have an adequate number of coun-
ties in the state with sufficient resources to support them even if units are
located in areas of meagre water supplies.
The siting pattern for the high electricity growth scenario with 35 year
useful plant life combines the characteristics of the base case and strict en-
vironmental controls, with the additional feature of siting excess capacity
out-of-state in the eastern part of the region. If the useful plant life is
changed to 45 years, no scenario unit additions are sited out-of-state and rel-
atively few are in areas of meager water resources (Figure 38). The geograph-
ical distribution of the planned plants and scenario unit additions is the dif-
ference between what might result from adding the scenario units necessary to
meet a very high electrical growth rate, under current conditions and strict
environmental controls, and from adding a decade to the useful life of each
generating unit in order to reduce the number of new units required.
Scenario 2a; Coal-fired Export
The coal export scenario specifies that coal-fired units will supply the
additional 20,000 MWe of installed generating capacity in the ORBES region
that is dedicated to export to the northeastern states. This requires siting
31 coal-fired scenario units in addition to those needed for Scenario 2.
Otherwise, the scenario policies are the same as in the base case.
In siting the scenario unit additions that are dedicated to export, two
assumptions were made:
1. The costs of transmitting electricity from the ORBES region
to the northeast will be minimized.
Consequently, candidate counties in the eastern part of the
ORBES region (eastern Kentucky, Ohio, Pennsylvania and West
Virginia) are favored sites. They are also located close to
major coal reserves.
2• Utilities will prefer to add generating capacity dedicated
J.V J-xi>Qr.t to existing Rites (either announced or designated
in the scenario) rather than develop new sites.
Ill
-------
t-o
Figure 37. SCENARIO 7a: 35 YEAR LIFE
CONVENTIONAL COAL EMPHASIS. BASE CASE, HIGH ELECTRICAL_ENERGY GROWTH
TOTPL PROPOSED COfiL-FIRED GENERATING
CRPR:ITY PDOITIONS. ig^e-es
PLUS SCENARIO UNIT ADDITIONS
MM-2000
3000. - 5400.
2000. - 3000.
iOOO. - 2000.
soo. - :oos.
250. - 500.
J 100. - 250.
i. - :oo.
i 10. - 0.
M£CPWfir*S (Planned additions)
2 Number of scenario unit additions
-------
Figure 38. SCENARIO 7b: 45 YEAR LIFE
CONVENTIONAL COAL EMPHASIS. BASE CASE, HIGH ELECTRICAL ENERGY GROWTH
TQTRL PROPOSED CORL-FIRED GENERATING
CfiPRCITY RDOITIONS. 1976-85
PtUS SCENARIO UNIT ADDITIONS
1 1986-2000
3000. - 5400.
2000. - 3000.
1000. - 2000.
soo. - ;ooo.
250. - 500.
iOO. - 250.
i. - ;oo.
M£OPwfiTT5 (PlMw«d additions)
2 Number of scenario unit additions
-------
Electricity dedicated to export is not intended to serve
local demand.
The 31 additional units are allocated to eastern Kentucky, Ohio, Pennsyl-
vania and West Virginia roughly in proportion to each state subregion's share
of the region's projected net exports in 1986 (Page, 1979, Appendix B). With-
in each state subregion, the capacity additions are allocated first to existing
stations that can accomodate additional capacity (assuming a maximum of 2600
MWe per county) and then to new counties consistent with the order of site
selection followed in Scenario 2. This procedure allows the impacts associa-
ted with export to be calculated as incremental changes without changing the
basic geography of ORBES electricity production in the year 2000.
The siting patterns in the western part of the ORBES region are the same
as in Scenario 2 (Figure 39). In the eastern part of the region, scenario
units are added in or adjacent to counties that have existing or planned addi-
tions in Scenario 2. These are located along the middle and upper Ohio RLver;
in the coalfields of southeastern Ohio; and along the Allegheny River in Penn-
sylvania. The effect is to increase significantly the geographical concentra-
tion of coal-fired units that will come on-line after 1985 in the eastern part
of the region. Although the siting pattern is similar to those of high energy
growth scenarios, the additional scenario units are located only in the east-
ern part of the region, and are concentrated in fewer counties.
FUEL SUBSTITUTION AND CONSERVATION
Scenario 3: Alternate Technology
In Scenario 3, alternate technologies are assumed to supply a portion
of the region's electricity production in the year 2000 that is projected by
the base case scenario. The result is that 66 rather than 95 coal-fired
scenario unit additions will be sited after 1985. Base case environmental
controls also apply to this scenario, as well as others in this set. Conse-
quently, the siting pattern will be similar to Scenario 2 except that fewer
counties will be involved.^
Coal-fired scenario unit additions are sited in 39 of the 54 counties
identified in Scenario 2 (Figure 40). The less suitable counties are excluded,
and the concentration of coal-fired electric generation is reduced somewhat
along the Ohio River main stem. Any changes in impacts that result from siting
fewer scenario unit additions can be determined by comparison with Scenario 2.
These changes can be attributed indirectly to the substitution of alternate tech-
nologies to produce electricity, although any impacts directly associated with
these technologies cannot be assessed as they are not assigned county locations.
Scenario 4; Natural Gas Emphasis
In Scenario 4, the large-scale substitution of natural gas as a fuel for
electricity generation is the reason for the significant reduction in the num-
ber of coal-fired scenario unit additions that are to be sited. The 34 units
are located in 21 counties, including those that are most suitable as sites
114
-------
Figure 39
SCENARIO 2a:
CONVENTIONAL TECHNOLOGY. BASE CASE CONTROLS, COAL-FIRED EXPORT
TGTRL PROPOSED CCR —CIR£0 i3^NE;RAT:N.:
CRPfiCITY PDDITIGNS. 1976-85
PLUS SCBIARIO UNIT ADDITIONS. 1986-2000
ORBES REGION
BOUNDARY
3000. - 5400.
2000. - 3000.
g ;cc:. - 2000.
§500. - 1033.
§250. - 500.
3 100. - 250.
] 1. - 100.
1o. - o.
(Planned additions)
2 N»i unit
additions
~> Circled numbers include
unit* for export
-------
Figure 40
SCENARIO 3:
ALTERNATE TECHNOLOGY, BASE CASE CONTROLS
TPL. PROPOSED CQPL-CIRED GENERATIN
CRPncITT RDDITIONS. 1976-85
PLUS SCBIAIBO UNIT ADDITIONS. 1989-2000
X.
OR8ES REGION
BOUNDARY
3000. - 5HOO.
2000. - 3000.
iocs. - 2000.
SOO. - 1000.
250. - 500.
100. - 350.
1. - 100.
0. - 0.
(Plwmd •ddhions)
Nuntarof scenario unH
•ooftrara
-------
for coal-fired plants. Given the environmental control policies, this repre-
sents a "better" siting pattern than in the other scenarios in the sense that
the counties selected for scenario unit additions have higher suitability
indices.
The geographical concentration of coal-fired scenario unit additions is
significantly reduced (Figure 41). In fact, the persistence of the cluster of
coal-fired units along the Ohio River main stem upstream from Louisville,
Kentucky and Cincinnati, Ohio is the most prominent feature of the pattern.
No scenario unit additions are located along the main stem downstream of
Louisville; and the concentration of new units along the upper Ohio is sig-
nificantly reduced. None of the units that is added in Illinois is along
the Illinois River. The impact of the substitution of natural gas for coal
as a fuel for electricity generation can be assessed to the extent that it
results in fewer coal-fired scenario unit additions being sited in the ORBES
region after 1985.
Scenario 6; Conservation (Very Low Energy Growth)
Energy conservation results in the most significant change in the siting
patterns for coal-fired capacity additions when compared with the base case
scenario. Because of very low energy growth rates, only 20 additional units
are required. These are located in 13 counties, each of which has the highest
site suitability rank of candidate counties in each respective state subregion
(Figure 42). The middle Ohio River main stem continues to be the core area
for capacity additions, but fewer units and counties are involved.
The geographical distribution of the proposed coal-fired capacity addi-
tions clearly dominates the siting pattern for this scenario. Because other
fuels (except nuclear) and technologies are not used to produce electricity,
the changes in impacts that result from having a relatively small number of
coal-fired scenario unit additions should be more readily identifiable.
Scenario 2c; Nuclear Emphasis
The siting pattern for Scenario 2c is dominated by nuclear-fueled sce-
nario unit additions after 1985. The distribution of the few coal-fired
scenario unit is similar to that of Scenario 4, Alternate Technologies. The
distribution of the nuclear-fueled scenario unit additions is based upon the
site suitability model for that fuel type. No nuclear-fueled units are sited
in Kentucky and West Virginia. This assumes that the current policy of lo-
cating only coal-fired generating capacity in these two states will continue.
In other state subregions, nuclear-fueled scenario unit additions are allo-
cated with preference to counties having existing or announced sites for
nuclear-fueled units that can be expanded. This assumes that utilities will
prefer to locate additional units at sites that can physically accomodate
additional capacity rather than risk the political and economic costs that
might be associated with developing new sites. The additions are allocated
to existing and announced sites so that the total site capacity does not ex-
ceed that specified by Burwell, Ohanian and Weinberg (1979) or 4000 MWe,
whichever is less.
117
-------
Figure 41
SCENARIO 4:
CONVENTIONAL TECHNOLOGY, NATURAL GAS EMPHASIS. BASE CASE CONTROLS
TOTP^ rfijfJSED'C'ORL-MRED C-ENERRTIN.:
CRPRCITY PDDITIONS, 1976-85
PLUS SCWAWO UNIT ADDITIONS. 1986-2008
OASES REGION
BOUNDARY
CD
3000. - 5400.
^ooo. - 3000.
;oco. - 2000.
500. - 1000.
250. - 500.
100. - 350.
!. - 100.
0. - 0.
MEGfiWRTTS (Planned addition*)
2 Nwntar of scenario unit
addition*
-------
Figure 42
SCENARIO 6:
CONVENTIONAL TECHNOLOGY, BASE CASE CONTROLS. VERY LOW ENERGY GROWTH
TOTflL PROPOSED CORL-FIRED GENERATING
CRPRCITY RDDITIONS, 1976-85
PLUS SCENARIO UNIT ADDITIONS. 19W 2000
OWES REGION
BOUNDARY
3000. - 5400.
3000. - 3000.
iooo. - aooo.
500. - 1000.
250. - 500.
100. - 250.
1. - 100.
0. - 0.
MECfiWRT T5 (PIMined additions)
2 Number of scenario unit addition*
-------
The nuclear-fueled scenario unit additions are concentrated in the west-
ern part of the ORBES region near existing and planned units (Figure 43). The
majority are in northwestern Illinois, where additions are sited in counties
along the middle and upper Illinois River, the Rock River, and the Mississippi
River. The counties are either in or adjacent to Commonwealth Edison s ser-
vice area; they include two of the alternate sites that Commonwealth Edison
evaluated in the Environmental Impact Statement for its Savannah plant in
Carroll County.5 In Indiana, the nuclear-fueled scenario unit additions are
along the Ohio River main stem in the southeast corner of the state with a
single unit in the northeast, where the environment is similar to, but less
suitable than, areas in northwestern Illinois.
The geographical distribution of nuclear-fueled scenario unit additions
in Illinois and Indiana outline the basic environmental issues of nuclear
siting throughout the ORBES region. Excluded from areas of high seismic risk
and population density, the plant locations shift to predominantly rural coun-
ties that have significant acreages of prime agricultural land, ecologically
sensitive areas, and problems of water availability. Illinois Power Company s
Clinton Plant (DeWitt County, Illinois) is an example of the tradeoffs between
seismic risk and water availability. On the other hand, reactions from Putnam
County residents to Commonwealth Edison's designation of the county as an al-
ternate" site for the Savannah plant shows concern over the issue of prime
agricultural land. In Indiana, the location of nuclear-fueled scenario unit
additions along the Ohio River main stem adds significantly to the concentra-
tion of electrical generating capacity in that area.
SITING PATTERNS FOR SPECIAL POLICY ANALYSIS
Scenario la: Very Strict Air Quality Controls
The basic siting pattern for Scenario 1 is based upon a moderate inter-
pretation of strict environmental controls, especially those concerning air
quality. Whereas the general effect is a more dispersed siting pattern for
scenario unit additions in the upper Ohio River Basin, new units in the middle
and lower basin are clustered in counties bordering the main stem in Indiana,
Kentucky and southwestern Ohio. A review of the siting pattern suggests Lhat
the configuration of scenario unit additions may still contribute to air
quality problems at subregional and regional scale. Some of the additions are
located in counties which, according to 1977 NADB monitor data, had less than
the full PSD increment available (Appendix C). Other units are located rela-
tively close to one another, especially along the Ohio River main stem which
raises the issue of separation distance policy within the interstate pollution
abatement provisions of the 1977 Clean Air Act Amendments. Consequently,
additional scenarios incorporate very strict air quality control P°11C1"
designed to create a more dispersed spatial distribution of new electricity
generating units.
Procedure
The siting pattern for this scenario is produced by making selected changes
in the siting model used in the strict environmental controls of Scenario 1.
120
-------
Figure 43
SCENARIO 2c:
CONVENTIONAL TECHNOLOGY. BASE CASE CONTROLS, NUCLEAR EMPHASIS
TGTRL PROPOSED NUCLERR GENERRTING
CflPRCITY RDDITIONS, 1976-85
PLUS SCENARIO UNIT ADDITIONS. 1988-2000
ORBES REGION
BOUNDARY
3000. - 5MCO.
2000. - 3000.
1000. - 2000.
500. - 1000.
250. - 500.
100. - 250.
1. - 100.
0. - 0.
MET.PHRTTS (Planned additions)
2 Number of scenario unit additions
-------
Changes in exclusionary criteria for air quality assume a more stringent
policy on air pollution performance standards; and changes in the relative
importance of several components of the siting model alter the suitability
of candidate counties as sites for coal-fired generating unit additions.
The allocation procedure for scenario units additions is also modified con-
sistent with the increased environmental constraints for siting decisions.
The changes are:
1. Exclude counties with violations of NAAQS for SCfe and TSP
and/or less than full PSD increment available, for 24 hour
and annual secondary standards (Appendix C).
This assumes a more stringent USEPA policy on air pollution
performance standards, including the addition of the PSD
increment as an exclusionary criterion for new stationary
sources.^
2. Increase the importance value of the Air Quality component
from the Delphi value of 9.15 to the maximum, 10.
This is consistent with the increased role of air quality
in evaluating the suitability of a site for generating unit
additions.
3. Increase the importance value of the Land Use and Ecological
Systems component from the Delphi value of 7.62 to the maxi-
mum, 10.
The importance of this component in evaluating site suitability
will increase because of the indirect impacts of air quality
upon productivity, and the land use conflicts that result from
the water requirements of scenario unit additions.
4. Decrease the importance value of the Water Availability com-
ponent by 50 percent, from the Delphi value of 3.34 to 1.67.
The importance of water availability in evaluating the site
suitability will decrease as more generating units are located
away from large rivers and streams in areas that will require
constructing large reservoirs for cooling water supplies. The
50 percent figure is arbitrary.
5. Allocate scenario unit additions with preference to counties
having announced utility sites which can be expanded.
Utilities will prefer to locate unit additions on sites which
can accomodate additional capacity, especially under very strict
environmental controls.°
122
-------
Exclusionary Screening and Site Suitability
Changing the exclusionary criteria significantly decreases the number
of candidate counties and changes their geographical distribution (Table 22
and Figure 44). The total number of excluded counties increased from 124 to
199. The majority of the 64 additional counties had less than the full PSD
increment available for TSP. The geographical distribution of non-attainment
counties and those counties excluded because of PSD criteria are significantly
different. However, the majority of the excluded counties failed to meet both
the 24 and the annual air quality standards. The use of the annual standard
resulted in a net addition of only 14 counties to the list.
Compared with Scenario 1, changes in the air quality exclusionary cri-
teria had their greatest impact upon the geography of candidate counties along
the middle and lower Ohio River main stem and its major tributaries in West
Virginia and Kentucky. The valley of the Monongahela (except for Greene
County) and the Kanawah, and large parts of the Licking and Green Rivers are
excluded from consideration. Along the Ohio River main stem, only 22 counties
are candidate sites for new electricity generating units. The changes are
less dramatic elsewhere. The majority of the additional counties excluded in
Indiana and Ohio were in the central eastern part of each state. Eastern Ohio
has only one county available for siting. However, the geography of candidate
counties did not change significantly in either Illinois or Pennsylvania.
Changes in the weights for the Air Quality, Land Use and Ecological Sys-
tems, and Water Availability components also produced significantly different
site suitability patterns for the scenarios (Figure 45). The magnitude of
change is greatest • among the middle and lower ranking counties in states
that have large numbers of counties in the ORBES region. This has significant
implications for siting, as many of the top-ranking counties are excluded as
sites for scenario unit additions under the very strict interpretation of air
quality criteria.
Siting Pattern
The siting pattern developed under the environmental constraints of very
strict air quality controls has the same number of scenario unit additions as
specified for Scenario 1. The geographical distribution of the scenario unit
additions, however, is significantly different (Figure 46). The majority of
the counties in which capacity additions are sited either were not selected
in Scenario 1 (22 of 64) or are in a different position in the schedule (27
of 64). The most significant changes are in Indiana and Kentucky, where the
clusters of "new" units along the middle and lower Ohio River main stem are
dispersed along the major tributaries.
In Indiana, the dispersed siting pattern is in response to significant
changes in county site suitability indices and the fact that a large number
of counties are included in ECAR's utility site inventory. The majority of
the utility sites are in counties that also have high suitability indices.
Only three counties (four units) were displaced because of the changes in ex-
clusionary criteria. However, changes in exclusionary criteria displaced 11
units in Kentucky away from the main stem to counties with lower water avail-
ability scores. Ten of the 11 Kentucky counties in which scenario units are
located are not on the list for Scenario 1.
123
-------
Table 22. SUMMARY OF COUNTIES EXCLUDED AS SITES FOR COAL-FIRED SCENARIO UNIT ADDITIONS
IN SCENARIOS la AMD Ib: VERY STRICT AIR QUALITY CONTROLS, AND IN
SCENARIOS Ic AND Id: AGRICULTURAL LANDS PROTECTION POLICY
Air Quality
Combined Total Excluded
Non-attainment PSD
State
Subregion
ILLINOIS
INDIANA
KENTUCKY
OHIO
PENNSYLVANIA
WEST VIRGINIA
Total
Counties
Violation
of NAAQS3
so2
2
3
1
3
1
1
11
TSP
17
10
39
43
8
12
129
Less than
Full PSD Class I
Increment Areas
Available3
S02 TSP
1 11
3 15
2 45 2
9 41
1 3
4 8 12
20 123 6
Combined
Totalb
20
18
60
50
8
10
166
Public
Landsc
4
4
10
3
1
3
25
Class I
and II
Soils" Scenarios
la and Ib
54 24
50 22
6 68
29 52
9
13
139 IBS
Scenarios
Ic and Id
66
62
70
64
9
13
284
Total
ORBES
Counties
85
83
120
68
19
48
423
SOURCL: Appendix C.
324 hour and annual secondary standards.
Counties that meet more than one exclusionary air quality criterion are counted only once.
Majority of county In public Lines; total area, including designated purchase area.
Majority of t-ntal area of countv in Class I and II soil?.
eCounties that meet more than one exclusionary criterion are counted only once.
-------
Figure 44. COUNTIES EXCLUDED AS CANDIDATE SITES:
SCENARIOS IA AND IB
VERY STRICT AIR QUALITY CONTROL POLICIES
ILL.NO.S
KENTUCKY
WEST
VIRGINIA
EXCLUSIONARY CRITERION
AIR QUALITY
Violations of NAAOS
for SO2 and TSP. 24 hour
and annual standards or IMS than
tfie full PSD incramsnt available
at NAM monitors in 1977
PSD Class I
III JIIIJIU
•UUliiUlUI
> 1960 MW« proposed in 1986
-------
Figure A5
SCENARIO IA:
VERY STRICT AIR QUALITY CONTROLS, DISPERSED SITING
IQTRL PROPD5FO CORL-FIRED GENERRTING
CflPRClTY RDDITIGNS, 1976-85
PLUS SCENARIO UNIT ADDITIONS, 1988-2000
ORSES REGION
BOUND
ro
3000. - 5400.
2000. - 3000,
1000. - aooo.
jSOO. - 1300.
jZBO. - 500.
] 100. - 250.
l, - 100.
o. - o.
S (tlanned
3 Number of scenario unit additio
-------
Changes in siting patterns in other QRBES states are relatively minor. In
Illinois, scenario units are added to counties along the lower Illinois River
rather than the Ohio River. This is the result of changes in county suita-
bility indices and the addition of Greene County as a future utility site.
No units are displaced because of changes in exclusionary criteria. In West
Virginia, the scenario unit additions continue to be located along the Ohio
River main stem as there are sufficient counties with high suitability indices
to accomodate capacity additions. Stability is also characteristic of the
siting patterns and schedules in Ohio and Pennsylvania, as each scenario has
the same limited number of candidate counties with similar site suitability
indices.
The siting pattern for very strict air quality controls suggests increased
environmental conflicts between air quality, water availability, and land use
and ecological systems impacts. The relocation of scenario unit additions in
Indiana and Kentucky to counties with relatively meager water resources means
that, with conventional technologies, reservoirs and ponds may be necessary
to provide the cooling water for coal-fired plants in four of the six ORBES
states. The land use requirements for conversion will increase, which in-
creases the probability of conflicts with other types of land use, including
agriculture and ecological systems.
Scenario let Agricultural Lands Protection
Agricultural land is a significant environmental resource in the ORBES
region. Its conversion to other uses because of energy development activities,
whether surface mining of coal or the location of new conversion activities,
is a source of conflict.10 If very strict air quality standards are enforced
(as in Scenarios la and Ib), conflicts associated with the location of con-
version facilities are likely to increase because more scenario unit additions
will be located in counties where reservoirs may be necessary to provide
adequate cooling water. Agricultural lands protection policies are concerned
with such conflicts, especially where prime farmlands are involved. Scenario
Ic assumes that such policies are enforced with respect to siting electrical
generating capacity additions.
Procedure
The siting pattern for this scenario is produced by making selected
changes in the siting model for Scenario 1. Some of these are the same as
changes made for the very strict air quality scenario. Others specifically
relate to an agricultural lands protection policy.
The changes are:
1. Exclude counties with violations of NAAQS for SOg and TSP
and/or less than the full PSD increments available, for 24
hour and annual secondary standards (Appendix A).
2. Exclude counties that have the majority (50 percent or more)
of their land area in Class I and II soils.
127
-------
This is consistent with the assumption that prime
farm land is the most important agricultural lands resource.
3. Increase the importance value of the Class I and II soils
from the Delphi value of 0.29 to the maximum. 1.0.
This is also consistent with the assumption that prime
farm land is the most important agricultural lands resource.
4. Increase the importance value of the Land Use and Ecological
Systems component from the Delphi value of 7.62 to the maximum,
10.0.
This is consistent with the agricultural lands protection policy.
5. Allocate scenario unit additions with preference to counties
having announced utility sites that can be expanded.
Utilities will prefer to locate unit additions on sites that
can physically accomodate additional capacity, especially under
very strict air quality and land use controls. Future land
purchases could be a difficult issue, especially where agri-
cultural lands are involved.
Exclusionary Screening and Site Suitability
The addition of prime farmland as an exclusionary criterion significantly
decreases the number of candidate counties and changes their geographical
distribution, especially in the western part of the region (Table 22, Figure
46). The total number of excluded counties increases from 188 to 284. Most
of the 96 additional counties are in Illinois (42), Indiana (40) and Ohio (12).
The effect is to exclude scenario unit additions from a broad wedge of coun-
ties across northern Illinois and Indiana into western Ohio. A smaller cluster
of excluded counties is in southwestern Indiana and along the lower Wabash
River in Illinois.
Changes in the weights for Class I and II soils, as well as for the Land
Use and Ecological Systems Component, also produced significantly different
site suitability patterns for the scenarios (Figure 47, 48). Counties along
the Ohio River main stem and other rivers are less suitable although, in most
cases, the amount of farmland is too limited to place them in the lower half
of the list of candidate counties. Conversely, the relative suitability posi-
tion of other counties is increased. These are located in the coal producing
areas of southern Illinois and Indiana, and southeastern Ohio. Changes in
site suitability indices of Kentucky, Pennsylvania and West Virginia counties
are minor.
Scenario Ic; Agricultural Lands Protection Policy
The scenario unit additions in Illinois, Indiana and Ohio are located
in counties that do not have the majority of their area in prime agricultural
lands. The majority are located in the southern part of each state, but not.
128
-------
Figure 46. COUNTIES EXCLUDED AS CANDIDATE SITES:
SCENARIOS 1C AND ID, AGRICULTURAL LANDS PROTECTION POLICY
"-LINO.S
KENTUCKY
21960 MWe in 1986
EXCLUSIONARY CRITERION
Vio(aiion» of NAAOS to SO2
Mid TSP. 24 hour and annual
•tandante. I«M AM Ml PSD
increment available al NAM
monitor* in 1ST?
PSD Claw I
Public
Class I and Clan II Soil*
-------
Figure 47. ECOLOGICAL SYSTEMS AND LAND USE COMPONENT
AGRICULTURAL LANDS PROTECTION POLICY
INDEX VALUE: RELATIVE SCORES
8. - 10.
36. - 8.
4. - 6.
2. - 1.
0. - 2.
PDCPAKD TOO OHIO RWB1 MSM OOCY STUDY
C. ITBRIMPY. 1MO
-------
Figure AS. SITE SUITABILITY INDEX
AGRICULTURAL LANDS PROTECTION POLICY
INDEX VALUE: RELATIVE SCORES
9. - 10.
38. - 9.
7. - 8.
6. - 7.
3. - 6.
HKPAKD ro« OMO RtVtF BASM DrtJHJY VUW
BY CAos/uec. rtMUAcr. i*w
-------
in counties along the Ohio River that have air quality problems. The siting
pattern in Kentucky is also changed, as interior agricultural regions, such as
the Bluegrass basin around Lexington, are less suitable as sites for capacity
additions. The most significant change, however, is the need to locate the
majority of the scenario units that are necessary to meet Ohio's demand for
electricity in West Virginia. Relatively few Ohio counties are candidates for
scenario unit additions because they do not meet threshold requirements for
air quality and agricultural lands criteria. Consequently, Ohio's "excess"
units are added to counties that are already designated as sites for capacity
additions dedicated to serve West Virginia.
The regional siting pattern suggests that implementation of an agricultur-
al lands protection policy relative to siting electricity generating units
will involve tradeoffs among air quality, water availability, and land use and
ecological systems impacts (Figure 49). Assuming that agricultural lands pro-
tection policies and strict air quality controls are at least compatible,
candidate counties will be restricted to a band in the western part of the
region between the areas of prime agricultural lands and the Ohio River main
stem and in the eastern part of the region. The candidate counties are lo-
cated in Kentucky and West Virginia. In the case of West Virginia (and perhaps
Kentucky also), that state's role as an exporter of electricity generated by
coal-fired plants might increase.
132
-------
Figure 49
SCENARIO 1C:
AGRICULTURAL LANDS PROTECTION, DISPERSED SITING
TGTflL PROPOSED COOL-FIRED GENERATING
CRPflCITT RDDITIONS, 1976-85
PLUS SCENARIO UNIT ADDITIONS, 19W-2000
ORBES REGION
BOUNDAI
U>
to
3000. - 5400.
2000. - 3000.
1000. - 3000.
500. - 1000.
]250. - 500.
3 ir-'0- - 250.
MEGP^PTTS(PI«nn«4 •ddlttma)
3 Number of •eanarie unit
(3) Scenario onit MMItionc
to supply Ohio demand
-------
FOOTNOTES
The siting patterns for the scenarios that are analyzed in detail in
the impact assessment are discussed in this section. The siting patterns
and schedules of on-line dates for all scenarios are in: Fowler et al
(1980).
o
This is to be expected, as the ORBES siting model is designed to simu-
late, at large scale, utility siting under current conditions.
3
Some rescheduling is necessary in order to accomodate the number of
scenario unit additions that are required to meet the incremental demand for
Scenario 7. Also, fewer existing and planned units are retired ( a total of
15,473 MUe) because of the 45 year useful plant life assumption.
Siting patterns for each of the scenarios that emphasizes coal-fired
electricity generating units and base case environmental control policies
have a number of sites (counties) in common. Scenarios that have more
scenario unit additions than in Scenario 2 add counties that always have
lower site suitability indices. Scenarios that have fewer additions are,
in effect, smaller subsets of counties with higher suitability indices.
Illinois Times. March 9-15, 1979, p. 3.
Two siting patterns are developed for these and the agricultural Idiuls
protection scenarios. In Scenario la and Ic, 2600 MWe is the maximum coal-
fired electrical generating capacity that can be sited in a county. This is
consistent with the 'dispersed* siting policy of Scenario 1, and permits im-
pact assessment under the changes in environmental controls only. In Sce-
nario Ib and Id, the maximum is increased to 5200 MWe. This allows generating
unit additions to be 'concentrated1 in candidate counties that are more suit-
able sites, and in which the utilities have sites that can accomodate capacity
additions. The result in each case is to locate a larger number of scenario
unit additions in the most suitable candidate counties in each state subregion.
Fewer counties are involved and, in general, the distance between them is in-
creased. Thus, the use of a policy of 'concentrated* siting to mitigate im-
pacts can be evaluated. The siting patterns and schedules of on-line dates
for each of these scenarios are in: Fowler et al (1980).
This also implies that the county is the most relevant geographical
area for air quality control decisions. The issue of separation distances
is also relevant to the geographical definition of exclusionary criteria as
well as site evaluation. Litigation involving IPALCO's Patriot plant in
Switzerland County, Indiana, is a case in point. However, the definition of
separation distances is not sufficiently precise for inclusion into the siting
model. See: Garvey et al (1977).
134
-------
well as site evaluation. Litigation involving IPALCO's Patriot plant in
Switzerland County, Indiana, is a case in point. However, the definition of
separation distances is not sufficiently precise for inclusion into the siting
model. See: Garvey et al (1977).
o
See: Appendix D.
The 1800 acre Greene County site is being purchased by Illinois Power
Company; Illinois Times. April 21-27, 1978. The Mid-America Interpool Network
(MAIN) Regional Reliability Council, which includes the ORBES portion of
Illinois, does not have a utility site inventory similar to that available
from ECAR.
The protection of agricultural land recently has become a major policy
goal at national and state level. Agencies such as the U.S. Department of
Agriculture and the USEPA (1977) have developed agricultural lands protection
policies as part of their resource development and environmental protection
activities. In the ORBES region, conflicts between farmland and energy devel-
opment, which are described in general by Fletcher (1980), focus upon the
Illinois, Indiana and Ohio state subregions (Randolph and Jones, forthcoming)
135
-------
BIBLIOGRAPHY
Argonne National Laboratory. 1977a. National Coal Utilization Assessment:
A Preliminary Assessment of the Health and Environmental Effects of Coal
Utilization in the Midwest. ANL/AA-6 Vol. 1. Argonne, Illinois: Energy
and Environmental Systems Division, Argonne National Laboratory.
. 1977b. National Coal Utilization Assessment. An^
Integrated Assessment of Increased Coal Use in the Midwest. Impacts and
Constraints. Argonne, Illinois: Energy and Environmental Division,
Argonne National Laboratory.
Brill, E. D., Jr.; Chang, S.; Fuessle, R. W.; and Lyon, R.M. Potential Water
Quantity and Quality Impacts of Power Development Scenarios on Major
Rivers in the Ohio Basin. Ohio River Basin Energy Study Phase IT, Sub-
contract under Prime Contract EPA R805588. Washington, D.C.: U.S.
Environmental Protection Agency, forthcoming.
Burwell, C.C.; Ohanian, M. J.; and Weinberg, A.M. 1979. "A Siting Policy
for an Acceptable Nuclear Future." Science, 204: 1043-1951.
Calvert, J. D., and Heilman, W. L. 1972. "New Approach to Power Plant Siting.1
Journal of the Power Division, Proceedings of the American Society of
Civil Engineers, 98: 93-102.
Calvert, J. D.; Heilman, W. L.; and Smith, H. L. 1974. Nuclear Power Plant
Siting; A Generalized Process. New York: Prepared by Commonwealth
Associates, for Atomic Industrial Forum.
Calzonetti, F. J. 1979. Impacts of Resource Conservation and Recovery Act
on the Siting of Coal Conversion Facilities in the United States. Pre-
pared for the Office of Environmental Policy Analysis, Oak Ridge National
Laboratory. Oak Ridge, Tennessee: Oak Ridge National Laboratory.
Cardi, V. P. Legal and Institutional Issues in the Ohio River Basin Energy
Study. Ohio River Basin Energy Study Phase II, Subcontract under Prime
Contract EPA R805588. Washington, D.C.: U.S. Environmental Protecting
Agency, forthcoming.
Carpenter, S. B.; Thomas, F. W.; and Gartrell, F. E. 1968. Full-scale Study
of Plume Rise at Large Electric Generating Stations. Muscles Shoals,
Alabama: Tennessee Valley Authority.
Church, R. L.; and Cohon, J. L. 1976. Multiobjective Location Analysis of
Regional Energy Facility Siting Problems. Upton, N.Y.: Brookhaven
National Laboratory.
136
-------
Church, R. L.; and Hillsman, E. L. 1979. "A Simulation Approach for Base
Line Energy Siting Analysis." In Proceedings of the Tenth Annual Pitts-
burgh Conference on Modeling and Simulation, Part III, pp. 895-902.
Pittsburgh: University of Pittsburgh.
Davis, R. M. et al. 1978. National Coal Utilization Assessment; A Prelimi-
nary Assessment of Coal Utilization in the South. ORNL/TM-6122. Oak
Ridge, Tennessee: Regional Studies Program, Oak Ridge National Laboratory.
Delbecq, A. L.; Van de Ven, A. H.; Gustafson, D. H. 1975. Group Techniques
for Program Planning. A Guide to Nominal Group and Delphi Processes.
Glenview, Illinois: Scott, Foresman and Company.
DiNunno, J. J. 1975. "Regional Approaches to Power Plant Siting in the United
States of America." In Siting of Nuclear Facilities, pp. 3-39. Vienna:
International Atomic Energy Agency.
Dobson, J. E. 1979. "A Regional Screening Procedure for Land Use Suitability
Analysis," The Geographical Review, 69: 224-235.
Eagles, T. W.; Cohon, J. L.; and ReVelle, C. 1979. Modeling Future Power
Plant Location Patterns. Applications. Final Report. Prepared by Johns
Hopkins University, Applied Physics Laboratory, for Electric Power Re-
search Institute, EPRI/EA-375. Palo Alto, California: EPRI.
Elkins, M. and DiNunno, J.V. 1975. "A Regional Site Survey for Thermal Power
Plants in the State of Ohio." Nuclear Technology. 25: 724-740.
Envirosphere Company. 1977. The Federal and State Regulatory Requirements
Affecting the Licensing of Electric Coal-Fired Generating Units. New
York: Envirosphere Company.
Equitable Environmental Health, Inc. 1976. Assessment of the Impact of En-
vironmental Protection Agency Significant Deterioration of Air Quality
Regulations on the Capacity for Siting Coal-Fired Power Plants in North-
ern and Central Illinois. Prepared for Commonwealth Edison Company.
Woodbury, New York: Equitable Environmental Health, Inc.
Kvans, V. A. 1976. "State Environmental and Siting Laws for Power Facilities."
Power Engineering, 80: 51-53.
Federal Power Commission. 1976. Factors Affecting the Electric Power Supply,
1980-85. Washington: Bureau of Power, Federal Power Commission.
Feldman, R. D. "The Siting of Coal-Fired Plants." Energy Reference Bureau,
1978.
137
-------
Fowler, G. L. 1977. "An Evaluation of Technology Assessment as a Method for
Policy Formulation in Energy Development." In Proceedings, Fourth Inter-
University Energy Conference, pp. 198-209. Chicago: Energy Resources
Center, University of Illinois at Chicago Circle.
Frigerio, N. H. et al. 1975. SITE: A Methodology for Assessment of Energy
Facility Siting Patterns. ANL/AA-Z Environmental Control Technology and
Earth Sciences (UC-11). Argonne, Illinois: Regional Studies Program,
Argonne National Laboratory.
Garvey, D. et al. 1977. The Prevention of Significant Deterioration; Impli-
cations for Energy Research and Development. Argonne, IL: Office of En-
vironmental Policy Analysis, Argonne National Laboratory.
Grant, R. J. 1978. "Impact of Clean Air Act Amendments on Power Plant Siting
- A Utility View." In Electrical Utilities in Illinois. Proceedings of
the Sixth Annual Illinois Energy Conference, pp. 66-74. Chicago: Energy
Resources Center, University of Illinois at Chicago Circle.
Gros, J. G. et al. 1975. "A Systems Approach to Nuclear Facility Siting."
In Siting of Nuclear Facilities, pp. 387-402. Vienna: International
Atomic Energy Agency.
Harte, J.; and El-Gassier, M. 1978. "Energy and Water." Science 199:623-634.
Hobbs, B. F. and Voelker, A. H. 1978. Analytical Multiobjective Decision-
making Techniques and Power Plant Siting; A Survey and Critique. ORNL
- 5288, Special, Oak Ridge, Tennessee: Oak Ridge National Laboratory.
Hynes, H. B. N. 1970. The Ecology of Running Waters. Liverpool: Liverpool
University Press.
Jansen, S. D. 1978. Electrical Generating Unit Inventory, 1976-1986: Il-
linois, Indiana, Kentucky, Ohio, Pennsylvania, and West Virginia. Ohio
River Basin Energy Study Phase II, Grant No. EPA R805590. Washington,
D.C.: U.S. Environmental Protection Agency.
Jopling, D. G. 1976. "Large Capacity Plant Sites: Problems and Opportuni-
ties." Power Engineering, 80: 36-43.
Keeney, R. L.; and Nair, K. 1977. "Nuclear Siting Using Decision Analysis."
Energy Policy. 5: 223-231.
et al. 1979. An Evaluation and Comparison of Nuclear
Powerplant Siting Methodologies. NUREG/CR-0407. SAND 78-1284. Albuquer-
que, MM: Sanda Laboratories.
Kohler, J. E.; Kenneke, A. P.; and Grimes, G. K. 1975. "Population Distri-
bution Considerations in Nuclear Power Plant Siting." Nuclear Technology,
25: 617-625.
138
-------
Laney, R. V., and Gustafson, P. F. 1978. "Nuclear Power Plant Siting in Illi-
nois." In Electric Utilities in Illinois. Proceedings of the Sixth Annu-
al Illinois Energy Conference, pp. 81-85. Chicago: Energy Resources
Center, University of Illinois at Chicago Circle.
McClure, Jr., C. R. and Hatheway, A. W. 1979. "An Overview of Nuclear Power
Plant Siting and Licensing." pp. 3-12. In Geology in the Siting of Nu-
clear Power Plants, A. W. Hatheway and C. R. McClure, Jr., eds. Reviews
of Engineering Geology, Vol. IV; Boulder, Colorado: The Geological Society
of America.
McHugh, L. A. 1978. "An Overview of the Effect of the Clean Air Act Amendment
of 1977 on Stationary Sources." In Environmental Law Seminar; Significant
New Developments in Environmental Law, pp. 1-29. Chicago: The Chicago
Bar Association.
McLaughlin, J. A. Legal and Institutional Aspects of Interstate Power Plant
Development in the Ohio River Basin Energy Study Region. Ohio River
Basin Energy Study Phase II, Subcontract under Prime Contract EPA R805588.
Washington, D.C.: U.S. Environmental Protection Agency, forthcoming.
Maryland Power Plant Siting Program. 1977. Maryland Major Facilities Study.
3 Vols; Baltimore: Energy and Coastal Zone Administration, Maryland
Department of Natural Resources.
Meier, P. M. 1977a. A Linear Programming Model for County Level Electric
Facility Siting. BNL 23388 Informal Report; Upton, New York: Brookhaven
National Laboratory.
. 1977b. "Regional Science and Energy Policy: A Methodology for
the Assessment of Coal Utilization in the Northeast." Paper presented
at the Northeast Regional Science Association Meeting, Halifax, Nova
Scotia. BNL 22742.
.; Stern, R; and McCoy, M. 1977. Issues and Siting Scenarios.
Vol. I of Coal Utilization in the Northeast. BNL 23596. Upton, New
York: Regional Energy Studies Program. Brookhaven National Laboratory.
Morell, D. 1977. "The Complex Political Problem of Siting Electrical Energy
Facilities" Paper presented at the 1977 Annual Meeting of the Midwest
Political Science Association Meeting, Chicago, Illinois.
Nair, K. et al. 1975. "An Approach to Siting Nuclear Power Plants: The
Relevance of Earthquakes, Faults and Decision Analysis," pp. 435-467.
In Siting of Nuclear Facilities. Vienna: International Atomic Energy
Agency.
. 1976. "Decision Analysis - A Methodology for Solving Complex
Problems in the Utility Industry: Benefit-Cost Analysis, Siting, and
Long-Range Planning." Proceedings of the American Power Conference, 38:
590-606.
139
-------
National Coal Association. 1977. Steam Electric Plant Factors, 1977. Wash-
ington, D.C.: National Coal Association.
1977. Steam Electric Plant Factors. 1978. Wash-
ington, D.C.: National Coal Association.
Nelson, A. J.; and Mitchell, K. G. 1979. "Conflict in Policy Initiation:
Power Plant Siting in Illinois." Paper presented at the 1979 Annual
Meeting of the Midwest Political Science Association, Chicago, Illinois.
NUS Corporation. 1973. A Regional Siting Survey for Thermal Power Plant
Sites in the State of Ohio. Prepared for Columbus and Southern Ohio
Electric Company. Rockville, Maryland: NUS Corporation.
Nuttli, 0. W. 1979. "Seismicity in the Central United States." In Geology
in the Siting of Nuclear Power Plants, A. W. Hatheway and C. R. McClure,
Jr., eds. pp. 67-93. Reviews of Engineering Geology, Vol. IV; Boulder,
Colorado: The Geological Society of America.
Ohio River Valley Water Sanitation Commission. 1979. Regional Facility
Siting in the ORSANCO States. Task Force on Major Facility Siting.
Cincinnati: ORSANCO.
. Task Force on Major Facility Siting. State and
Federal Regulations Affecting the Siting of Major Energy Related Facili-
ties. Background Paper. Cincinnati: ORSANCO, n.d.
Old, F. C. 1976. "What Happened to the Nuclear Program in 1975?" Power
Engineering. 80: 83-85.
Palmadeo, P. F. 1976. "Approaches to Regional Energy Analysis." Growth
and Change, 7: 25-32.
Page, W. P. 1979. Energy Consumption in the Ohio River Basin Energy Study
Region. 1974, by End User and Fuel Type. Ohio River Basin Energy Study
Phase II, Grant No. EPA R805585. Washington, D.C.: U.S. Environmental
Protection Agency.
Page, W. P.; Gilmore, D.; and Hewings, G. An Energy and Fuel Demand Model for
the Ohio River Basin Energy Study Region. Ohio River Basin Energy Study
Phase II, Grant No. EPA R805585 and Subcontract under Prime Contract EPA
R805588. Washington, D.C.: U.S. Environmental Protection Agency, forth-
coming.
Page, W. P., and Stukel, J. J. Integrated Technology Assessment Methodology
for the Ohio River Basin Energy Study. Ohio River Basin Energy Study
Phase II, Grant No. EPA R805588 and Subcontract under Prime Contract EPA
R805588. Washington, D.C.: U.S. Environmental Protection Agency, forth-
coming.
140
-------
Pigford, T. H. et al. 1974. Fuel Cycles for Electrical Power Generation,
Parts I and II. Towards Comprehensive Standards. Berkeley, California:
Teknekron, Inc.
Podlasek, R. J. 1977. Electric Utilities; An Assessment of the Industry
in Illinois. Springfield: Illinios Energy Resources Commission.
Provenzano, G. 1978. "A Linear Programming Model for Coordinating Power
Generation, Coal Gasification and Water Resources Development." In
Symposium Papers; Energy Modeling and Net Energy Analysis, pp. 559-579.
Chicago: Institute of Gas Technology.
Randolph, J. C. and Jones, W. W. Ohio River Basin Energy Study; Land Use
and Terrestrial Ecology. Ohio River Basin Energy Study Phase II, Grant
No. EPA R805609. Washington, D.C.: U.S. Environmental Protection Agency,
forthcoming.
Reynolds, J. Z. 1980. "Power Plant Cooling Systems: Policy Alternatives."
Science. 207 (4429): 367-372.
Richetto, J. S. 1978. "Strategy for the Siting of Electric Energy Facilities."
Public Utilities Fortnightly. 29-33.
Rossin, D. A.; and Nichols, B. L. 1975. "Site Selection and Evaluation of
Alternatives by United States Utilities." Nuclear Technology, 25: 670-4.
Saper, J. L. and Hartnett J. P. eds. 1980. The Current Status of the FAectric
Utility Industry in the Ohio River Basin Energy Study States. Ohio River
Basin Energy Study Phase II, Grant Nos. EPA R805585 and R805588. Washing-
ton, D.C.: U.S. Environmental Protection Agency.
Silvey, J. F.; Moy, K.; and Fertel, M. S. 1977. "Clean Air Act Amendments of
1977. Envirosphere Regulatory Report, 7.
Southern States Energy Board. 1978. Energy Facility Siting in the United
States. 2 Vols. Atlanta: Southern States Energy Board.
Soyland Power Cooperative. 1980. Site Selection Study for a Coal-Fired
Electric Power Generating Facility. Final Report. Prepared by Environ-
mental Sicence and Engineering, Inc., Plantec Corporation and Reynolds,
Smith and Hill. Decatur, Illinois: Soyland Power Cooperative.
Tennessee Valley Authority. 1974. Summary of TVA Atmospheric Dispersion
Modeling. Internal Report. Muscle Shoals, Alabama: Tennessee Valley
Authority.
Turner, D. B. 1970. Workbook of Atmospheric Dispersion Estimates. PB-191
482. Cincinnati, Ohio: Public Health Service, Division of Air Pollution.
U.S. Department of Energy, Economic Regulatory Administration, Office of
Utility Systems. The Effects of Delays in Construction of Coal-fired
Generating Units on Electric Utility Use of Coal, Oil and Natural Gas.
Washington, D.C.: U.S. Department of Energy.
141
-------
U.S. Department of Energy, Economic Regulatory Administration. 1979. Interim
Report on the Performance of 400 Megawatt and Larger Nuclear and Coal-Fired
Generating Units. Washington, D.C.: U.S. Department of Energy.
United States Nuclear Regulatory Commission. 1976. Resource Availability
and Site Screening. Part V of V in Nuclear Energy Center Site Survey -
1975. Washington: U.S. Nuclear Regulatory Commission.
Van Horn, A. J.; Liroff, S. D.; and Hirata, A. Selected Impacts of Electric
Utility Operations in the Ohio River Basin (1976-2000): An Application
of the Utility Simulation Model. Ohio River Basin Energy Study Phase II,
Subcontract under Prime Contract EPA R805588. Washington, D.C.: U.S.
Environmental Protection Agency, forthcoming.
Voelker, A. H. 1977. Power Plant Siting: An Application of the Nominal
Group Process Technique. ORNL/NUREG/TM-81. Oak Ridge, Tennessee: Oak
Ridge National Laboratory.
Wagner, M. 1978. "Impact of the Clean Air Act on Power Plant Siting: An
Environmental View." In Electrical Utilities in Illinois. Proceedings
of the Sixth Annual Illinois Energy Conference, pp. 75-80. Chicago:
Energy Resources Center, University of Illinois at Chicago Circle.
Warls, K. and Warner, C. F. 1976. Air Pollution—Its Origin and Control.
New York: Harper and Row.
White, I. L.; Ballard, S. C.; and Hall, T. A. 1978. "Technology Assessment
as an Energy Policy Tool." Policy Studies Journal, 76-83.
1979. Energy from the West; A Technology Assessment
of Western Energy Resource Development. Washington, D.C.: U.S. Environ-
mental Protection Agency.
Wilbanks, T. J.; and Calzonetti, F. 1977. "Forecasting the Location of New
Energy Facilities." Paper prepared for the Annual Meeting of the Asso-
ciation of American Geographers, Salt Lake City.
Williams, J. H. 1978. "Power Plant Siting Reform - Panacea or Purge?"
Public Utilities Fortnightly. 21-27.
Womeldorff, P. 1978. "Siting: A Key to Capacity Planning." Electrical
Utilities in Illinois. Proceedings of the Sixth Annual Illinois Energy
Conference, pp. 61-65. Chicago: Energy Resources Center, University
of Illinois at Chicago Circle.
142
-------
APPENDICES
143
-------
APPENDIX A
SITED CAPACITY ADDITIONS, 1976 THROUGH 2000
Column 1
Column 2
Column 3
Column 4
Column 5
Column 6
Column 7
Column 8
Column 9
UNIT ID
CO INDEX
NAME
COUNTY
MWE
STATUS
DATE
RETIRE
FUEL
The following tables (A.I through A.6) list sited capacity additions by
state subdivisions of the ORBES region. The entries represent published elec-
tric utility company plans for capacity additions as of December 31, 1976.
Nine separate pieces of information are given for each generating unit addi-
tion :
Unit Identification
Company Index
Unit Name
FIPS County Code
Capacity In MWe
Unit Status
On-line Date
Retirement Date
Primary Fuel
A period (.) in a column indicates that no information was available for that
entry. Further details concerning the interpretation of the coded data are
available in:
Steven D. Jansen, University of Illinois at Chicago Circle,
"Electrical Generating Unit Inventory, 1976-1986: Illinois,
Indiana, Kentucky, Ohio, Pennsylvania, and West Virgina,"
Ohio River Basin Energy Study Phase II, Grant No. EPA R805590
(Washington, D.C., November 1978).
In calculating the number of required scenario unit additions, it
is assumed that these planned and sited capacity additions will be built as
listed in the following tables. Only those generating unit additions for which
the county site is known are used in the calculation of scenario unit additions.
Units which have a period (.) in the column for county are unsited and are given
in the tables only to convey as much information as possible about utility plans.
-------
Table A.I. SITED CAPACITY ADDITIONS, 1976 THROUGH 2000: ILLINOIS
UNiT_.in co_.TNni-:x NAME
COUNTY MWE STATUS DATE RET IRK FUEL
COEC
COEC
COEC
COEC
COEC
COEC
COEC
ILPC
1 1. PC
ILPC
SOIP
CEIL
CEIP
CEIP
CEIP
COEC
COFC
TL.PC
TLPC
SOIP
SPFI
SPFI
SPFT
SPFT
SPFI
UNEC
WEIL.
WEIL
WEIL
CEIL
CEIL
COLLTNS
COLLINS
COLLINS
COLLTNS
COLLINS
LASALIE COUNTY
LASALLE COUNTY
CLINTON
CLINTON
HAVANA
MARION
DUCK CREEK-
NEWTON
NEWTON
NEWTON
UNSITED
UNSITED
UNSITED
UNSITED
MARION
DALLMAN
FACTORY
PLANT 4-1
REYNOLDS
UNSITED
VENICE
PEARL STATION
UNSITED
UNSIFED
DUCK CREEK
DUCK CREEK
63
63
63
63
63
99
99
39
39
125
199
57
79
79
79
99
99
«
,
199
167
167
,
167
,
119
149
,
,
57
57
515
510
500
505
505
1078
1078
950
950
450
173
400
617
600
600
600
550
600
400
173
192
50
175
50
192
220
400
20
20
500
600
U
U
u
u
LI
U
U
U
U
U
u
p
p
p
p
p
p
p
p
p
p
p
p
p
p
p
p
p
p
p
p
7804
7710
7704
7810
7904
7909
8009
8112
8406
7806
780A
8201
7712
8104
8404
8501
8504
8606
8406
8600
7806
8401
8601
8101
8606
7905
8400
8106
8406
8900
9000
F06
T06
F06
F06
F06
UR
UR
UR
UR
COl
COI
COL
COI
COL
COL
(JNK
COI.
COI.
Oil
COL
COI.
OIL
COI.
OIL
COL
HD2
COI.
COI.
COI
COI
COI.
145
-------
Table A.2. SITED CAPACITY ADDITIONS, 1976 THROUGH 2000: INDIANA
UNIT.ID C0_INDEX NAME
COUNTY MWE STATUS DATE RETIRE FUEL
1
2
3
A
15
3
4
1
1
1
2
1
2
13
13
3
2 '
3
4 ,
INME
INME
INPI.
INPL
NOIP
PSIN
PSIN
PSIN
SOIG
HEDI
HEDI
INPL
PSIN
RCMP
RCMP
RICI
SOIG
SOIG
SOIG
ROCKPORT
ROCKPORT
PETERSBURG
PETERSBURG
SCHAHFERr R. M, .
GIBSON
GIBSON
MARBLE HILL
BROUNp A. P.
MEROM
MEROM
PATRIOT
MARBLE HILL
RENSSELAER
UNKNOWN
WHITEWATER VALLEY
BROWN. A. B.
BROWN* A. B.
BROWNt A. B.
147
147
125
125
73
51
51
77
t?9
153
153
155
77
73
73
177
129
129
129
1300
1300
532
532
556
650
650
1130
265
490
490
650
1130
6
6
100
265
500
500
U
U
U
U
U
U
U
U
(J
P
P
P
P
P
P
P
P
P
P
8112
8212
7711
8204
7905
7804
7904
8201
7904
8009
8109
8504
8404
8200
8206
8507
8304
8701
9301
COI.
COI.
cm.
cm.
COI.
BtT
BIT
UR
HOL
COL
COL
COL
UR
Oil
FO?
COI.
COI
COI.
COI
146
-------
Table A.3. SITED CAPACITY ADDITIONS, 1976 THROUGH 2000: KENTUCKY
UNIT.ID CO-INDEX NAME
2
0
1
2
.5
4
I
]
2
4
1
1
o
*
]
2
3
4
1
•>
o
3
3
4
KEUC
CETU
EAKR
EAKR
LOGE
LOGE
LOGE
BIRI
BIRI
BIRI
BIRI
CIGE
CIGE
EAKR
KEPC
KEPC
KEUC
KEUC
KEUC
KEUC
LOGE
LOGE
VEHP
CIGE
LOGE
LOGE
GHENT
LAUREL
SPURLOCKr H
SPURLOCKr H
MILL CREEK
MILL CREEK-
TRIMBLE COUNTY
GREEN
GREEN
COLEMAN
STATION 4
EAST BEND
EAST BEND
UNSITED
LEWIS COUNTY
LEWIS COUNTY
GHENT
GHENT
UNSITED - SITE
UNSITED-SITF
TRIMBLE COUNTY
UNSITED
CANNFLTON
EAST BEND
TRIMBLE COUNTY
TRIMBLE COUNTY
COUNTY MWE STATUS DATE RETIRE FUEL
COL
WAI
COL
COI
COL
COL
COL
cm
COI
COI
COL
COL
COL
COL
COI
COL
COL
COL
COI
COL
COL
F02
WAT
COL
COL
COL
41
125
L. 161
L 161
111
111
!TY 223
233
233
91
»
15
15
»
135
135
41
41
TE A
' A
ITY ?23
111
91
15
ITY 223
ITY 223
550
61
300
500
425
495
495
240
240
240
500
600
600
650
1300
1300
550
550
650
650
495
65
70
800
675
675
T
U
U
U
U
U
U
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
P
I
T
7706
7700
7706
8103
7805
8006
8306
7912
8004
8400
8500
8401
8006
8400
8312
8412
8103
8303
8504
8600
8506
8406
8000
8701
9999
9999
147
-------
Table A. 4. SITED CAPACITY ADDITIONS, 1976 THROUGH 2000: OHIO
UNIT. Ill CO-INDEX NAME
3 BUPI CARDINAL
8 CIGE MIAMI FORT
1 CIGE U. H. ZIMMER
6 COSO CONESVILLE
5 COSO POSTON
6 COSO POSTON
2 DAPO KILLEN STATION
COLU COLUMBUS
1 DAPO KILLEN STATION
1 OHPC RACTNE
VEHP GREENUP
2 CIGE U. H. ZIMMFR
COUNTY MUE STATUS DATE RETIRE FUEL.
81
At
25
31
9
9
1
49
1
105
145
25
615
500
810
403
403
403
600
90
600
40
70
810
U
U
U
U
U
U
U
P
P
P
P
I
7709
7803
7907
7801
8301
8501
8201
8100
8501
7912
8000
9999
COL
COL
IJR
COI
COL
COL
COL
REF
COL
WAT
WAT
IJR
148
-------
Table A.5. SITED CAPACITY ADDITIONS, 1976 THROUGH 2000: PENNSYLVANIA
UNIT_ID CO...INDEX NAME
COUNTY MWE STATUS DATE RETIRE RJEl
1 DULC BEAVER VALLEY
2 DULC BEAVER VALLEY
1 DULC BEAVER VALLEY
2 nULC BEAVER VALLEY
1 DULC SHIPPINGPORT
3 PEEC HOMER CITY
2 PEPC MANSFIELD
3 PEPC MANSFIELD
7 PEEC SEUARD
1 WEPP LOWER ARMSTRONG
2 WEPP LOWER ARMSTRONG
3 WEPP LOWER ARMSTRONG
7
7
7
7
7
63
7
7
43
5
5
5
85
29
800
856
60
693
917
: 917
800
630
630
630
A
A
U
LI
U
U
U
U
P
P
P
P
7803
8404
7704
8205
7710
7712
7710
8010
8405
8303
8403
8503
UR
UR
UR
UR
UR
COL
COL
COL
COL
COI
cm
noi
149
-------
Table A.6. SITED CAPACITY ADDITIONS, 1976 THROUGH 2000: WEST VIRGINIA
UNIT. IP CO..INDEX NAME
1 APPC NEW HAVEN
I MOPC PLEASANTS
2 MOPC PLEASANTS
1 MOPC DAVIS POWER PROJ.
2 MOPC DAVIS POWER PROJ.
3 MOPC DAVIS POWER PROJ.
4 MOPC DAVIS POWER PROJ.
COUNTY MWE STATUS DATF RETIRF FUEL
COL.
cm.
COL
WAI
WAT
WAT
WAT
53
73
73
93
93
93
93
1300
626
626
250
250
250
250
U
U
U
P
P
P
P
8012
7903
8003
8603
8606
8799
8799
150
-------
APPENDIX B
CAPACITY REMOVALS, 1976 THROUGH 2000
Column 1
Column 2
Column 3
Column 4
Column 5
Column 6
Column 7
Column 8
Column 9
UNIT ID
CO_INDEX
NAME
COUNTY
MWE
STATUS
DATE
RETIRE
FUEL
The following tables (B.I through B.6) list capacity removals or retire-
ments by state subdivisions of the ORBES region. Nine separate pieces of in-
formation are given for each generating unit removal:
Unit Identification
Company Index
Unit Name
FIPS County Code
Capacity In MWe
Unit Status
On-line Date
Retirement Date
Primary Fuel
A period (.) in a column indicates that no information was available for that
entry. Further details concerning the interpretation of the coded data are
available in:
Steven D. Jansen, University of Illinois at Chicago Circle,
"Electrical Generating Unit Inventory, 1976-1986: Illinois,
Indiana, Kentucky, Ohio, Pennsylvania, and West Virginia,"
Ohio River Basin Energy Study Phase II, Grant No. EPA R805590
(Washington, D.C., November 1978).
It is assumed that units with retirement dates earlier than the year 2000
will be retired by 2000. Units for which the on-line date and the retirement
date are both unknown are also assumed to retire by 2000. All units that, have
on-line dates earlier than 1967 will be more than 35 years old in 2000 and are
assumed to be retired by 2000. Although most hydroelectric (fuel is WAT) gen-
erating units fit into one of the above categories and are, therefore, listed
in the following tables,they are not considered as retirements for calculating
the number of scenario unit additions.
151
-------
Table B.I. CAPACITY REMOVALS, 1976 THROUGH 2000: ILLINOIS
UNIT-ID CO.INDEX NAME
COUNTY MWE STATUS DATE RETIRE FUEL
1
2
l'
2
3
1
2
3
A
5
7
5
6
7
8
9
10
11
2
3
4
5
6
7
1
3
4
5
A
7
1
3
4
1
2
3
4
1
2
3
1
1
2
3
BETY
BETY
BREE
BRFE1
BREE
BUSH
BUSH
BUSH
BUSH
BUSH
BUSH
CALU
CALU
CALU
CAI.U
CALU
CALU
CALU
CARL.
CARL
CARL
CARL
CARL
CARL
CEIL
CEIL
CEIL
CEII
CEIL
CECL
CEIP
CEIP
CEIP
CEIP
CEIP
CEIP
CEIP
CEIP
CEIP
CFIP
COEC
ELNE
ELNE
ELNE
BETHANY
BETHANY
BREESE
BREESE
BREESE
BUSHNELL
"IG'S'HNELL
BUSHNELL
BUSHNELL
BUSHNELL
BUSHNELL
CARMI
CARMI
CARMI
CARMI
CARMI
CARMI
CARMI
CARLYLE
CARLYLE
CARLYLE
CARLYLE
CARLYLE
CARLYLE
E D EDUARDS
R S UAL LACE
R S UALLACE
R S UALLACE
R S UALLACF
R S UALLACE
COFFEEN
GRAND TOUER
GRAND TOUER
HUTSONVILLE
HUTSONVILLE
HUTSONVILLE
HUTSONVILIE
MEREDOSIA
MEREDOSIA
MEREDOSIA
DRESDEN
JOPPA STEAM
JOPPA STEAM
JOPPA STEAM
139
139
27
27
27
109
109
109
109
109
109
193
193
193
193
193
193
193
27
27
27
27
27
27
143
179
179
179
179
179
135
77
77
33
33
33
33
137
137
137
63
127
127
127
1
1
1
1
2
1
1
2
2
1
1
1
1
1
1
2
2
3
1
3
1
1
1
2
136
25
40
40
86
114
389
81
114
25
25
75
75
58
58
239
209
183
183
183
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
R
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
6000
5200
4800
• 5300
6000
4000
4000
6500
6500
4800
5600
4500
3900
4800
5100
5800
5800
6300
3900
4900
5900
5900
5900
6400
6000
3900
4100
4900
5200
5800
6512
5103
5804
4005 8510
4109 8510
5302
5407
4806
4901
6007
6000
5300
5300
5400 .
UNK
LINK
OIL
OIL
OIL
OIL
OIL
OIL
OIL
0 [1
OIL
OIL
OIL
OIL
OIL
Oil.
on
OIL
COI.
COL
OIL
OIL
OIL
on..
COL
COL
COL
not
COI.
noi
COL
COL
COL
F02
K02
COL
COL
COI.
COL
COI .
UR
COL
COL
COL
(continued)
152
-------
Table B.I. (continued)
I IN IT... ID
4
5
6
2
3
4
1
2
3
4
1
2
3
4
4
5
6
7
1
2
3
4
1
2
5
6
1
2
3
4
5
1
2
3
5
6
1-7
1
2
1
2
1
2
3_INDEX
ELNE
ELNE
ELNF
FACT
FACT
FACT
FMLP
FMLP
FMLP
fMLP
FREE
FREF:
FREE
FREE
GEMU
GEMU
GEMU
GEMU
HIGH
HIGH
HIGH
HIGH
ILPC
ILPC
ILPC
ILPC
TLPC
ILPC
ILPC
ILPC
ILPC
ILPC
ILPC
ILPC
ILPC
ILPC
ILPC
ILPC
ILPP
ILPC
ILPC
ILPC
ILPC
ILPC
NAMF
JOPPA STEAM
JOPPA STEAM
JOPPA STEAM '
FARMER CITY
FARMER CITY
FARMER CITY
FAIRFIELD
FAIRFIELD
FAIRFIELD
FAIRFIELD
FRF E BURG
FREEBURG •'
FREEBURG
FREEBURG
GENESEO
GENESEO
GENESEO
GENESEO
HIGHLAND
HIGHLAND
HIGHLAND
HIGHLAND
BLOOMINGTON
BLOOMINGTON
BLOOMINGTON
BLOOMTNGTON
HAVANA
HAVANA
HAVANA
HAVANA
HAVANA
HENNEPIN
HENNEPIN
HENNEPIN
JACKSONVILLE
JACKSONVIL LE
MARSEILLES
VANDALIA
VANDALIA
VERMILION
VERMILION
WOOD RIVER
WOOD RIVER .
WOOD RIVER
COUNTY
127
127
127
39
39
39
191 .
191
191
191
163
163
163
163
73
73
73
73
119
119
119
119
113
113
113
113
125
125
125
125
125
155
155
155
137
137
99
51
51
183
183
119
119
119
MWE £
183
183
183
1
1
1
2
2
4
5
1
1
1
1
2
1
1
3
2
2
3
6
1
1
2
2
52
52
52
52
52
75
106
125
2
3
2
1
1
75
107
49
50
51
tFA'T
S
S
S
S
S
S
S
S
S
R
P
fc-
R
S
S
S
S
S
S
S
S
S
S
S
S
S
p
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
DATE RETIRE FUEL
5400
5500
5500
6300
4500
5000
4000
4200
4900
5600
4800
4BOO
5300
5900
5700
4900
4700
6100
3600
4800
4700
6100
3200
3200
4900
6000
4700
4700
4800
5000
5000
5300
5900
5900
4900
5200
•
4800
4800
5500
5600
4900
4900
5000
. COL
. COL
. COL
. nii.
. OIL
, OIL
, COL
« COL
, COL
. roi
• OIL
• OIL
• OIL
• OIL
• OIL
• OIL
• OIL
• OIL
• COL
• COL
• COL
• COL
• OIL
• OIL
• OIL
• OIL
• OIL
• OIL
• OTI
• OIL
• OIL
• COL
• COL
• COL
• OIL
• OIL
• MAT
• OIL
• OIL
• COL
• COL
• OIL
• OIL
• OIL
(continued)
153
-------
Table B.I. (continued)
UNIT..ID CO._INPEX NAME
4
5
1
2
3
4
GT1
IC1
IC2
IC3
IC4
Id
IC2
IC3
1
2
3
4
1
2
3
1
2
3
...4
1
2
3
4
2
3
5
6
4
5
1-3
1
2
3
4
5
6
8
ILPC
ILPC
LI OF
MAIL
MAIL
MAIL
MAIL
MCLE
MCLE
MCLE
MCIE
MCLE
MCPD
MCPD
MCPD
MCPU
MCPU
MCPLI
MCPU
NOCH
NOCH
NOCH
PERU
PERU
PERU
PERU
PMIL
PMJL
PHIL
PMIL
REBU
REBU
REBU
REBU
ROOD
ROOD
ROOD
RVLP
RVI.P
RVLP
RVLP
RVIP
RVLP
RVLP
WOOD RIVER
WOOD RIVER-
OTTAWA
MARSHALL
MARSHALL
MARSHALL
MARSHALL
MC LEANSBORO
MC LEANSBORO
MC LEANSBORO
MC LEANSBORO
MC LEANSBORO
MASCOUTAH
MASCOUTAH
MASCOUTAH
MT CARMEL
MT CARMEL
MT CARMEL
MT CARMEL
DAYTON
DAYTON
DAYTON
PERU
PERU
PERU
PERU
PRINCETON
PRINCETON
PRINCETON
PRINCETON
RED BUD
RED BUD
RED BUD
RED BUD
ROODHOUSE
ROODHOUSE
ROODHOUSE
RANTOUL
RANTOUL
RANTOUL
RANTOUL
RANTOUL
RANTOUL
RANTOUL
119
119
99
23
23
23
23
65
65
65
65
65
163
163
163
IBS
185
185
185
99
99
99
99
99
99
99
11
11
11
11
157
157
157
157
61
61
61
19
19
19
19
19
19
19
103
397
32
1
1
1
3
t
1
1
1
2
1
1
1
2
4
8
8
2
1
1
1
3
4
8
3
3
4
4
1
2
1
1
1
1
1
1
1
1
1
t
1
4
5
S
S
s
S
s
s
s
s
s
s
R
S
S
R
S
S
S
S
S
S
S
S
S
s
s
s
s
s
s
s •
s
s
s
s
s
R
s
s
s
s
s
s
s
5400
6400
«
4800
4800
5300
6200
5800
4900
5000
5200
6300
•5100
5 tOO
5800
4100
4900
5200
5700
2500
2500
2500
3600
3800
5000
6000
5300
5800
6500
6500
5900
6500
4800
5300
5700
6400
5000
5100
5100
5300
5400
6400
6400
6400
COUNTY MWE STATUS DATE RET I RE FUEL
COl
COL
IJNK
OTL
OIL.
OIL
OIL
GAS
OIL
OIL
OIL
OIL
OIL
OIL
OTL.
COL.
COL.
COL
COL
WAT
WAT
WAT
OIL
COL
COL
COL
OIL
OTL
OIL
OIL
OIL
OIL
OTL.
OIL
OIL
OIL
OIL
OTL
Oil
OTL
0.iL
on.
on
on.
(continued)
154
-------
Table B.I. (continued)
CO.INDEX NAME
2
3
4
5
6
7
1
2
3
4
5
1
2
3
4
5
1
2
3
1
2
3
4
5
6
7
2
3
4
5
6
7
8
1
2
3
4
5
6
1
2
3
4
5
6
7
UCLP
UCLP
ucu-
WCLP
WCLP
UCLP
UEIL
WEIL
WEIL
WEIL
WEIL
WEIL
WEIL
WEII.
WEIL
WEIL
SOIP
SOIP
SOIP
SPFT
SPFI
SPFI
SPFI
SPFI
SPFI
SPFI
SUII
SUIL
SUII
SUIL
SUIL
SUIL
SUII.
UNEC
UNEC
UNEC
UNEC
UNEC
UNEC
UNIL
UNIL
UNIL
UNIL
UNIL
UNIL
UNIL
WATERI. 00
WATERLOO
WATERLOO
WATERLOO
WATERLOO
WATERLOO
PITTSFIELD
P.ITTSFIELD
PITTSFIELD
PITTSFTELD
PITTSFIELD
WINCHESTER
WINCHESTER
WINCHESTER
WINCHESTER
WINCHESTER
MARION
MARION
MARION
LAKESIDE
LAKESIDE
LAKESIDE
LAKESIDE
LAKESIDE
LAKESIDE
LAKESIDE
SULLIVAN
SULLIVAN
SULLIVAN
SULLIVAN
SULLIVAN
SULLIVAN
SULLIVAN
VENICE NO
VENICE NO
VENICE NO
VENICE NO
VENICE NO
VENICE NO
ABBOTT
ABBOTT
ABBOTT
ABBOTT
ABBOTT
ABBOTT
ABBOTT
COUNTY MWE STATUS DATE RETIRE FUEl
5400 . OIL
4600 . 01L
6400 . OTI
5000 . OIL
5000 . OIL.
5800 . 0[L
4900 . OIL
4900 . OTL
4900 . OIL
5500 . OIL
5500 . OTI
3800 . OIL
3800 . OIL
3800 . OIL
4700 • OIL
4700 . OTI.
6306 . COL
6308 . COL
6309 . COL
3600 8101 COL
3900 8101 COL
4000 . COL
4900 . COL
5300 . COL
6000 , 001.
6500 . COL
6100 . OIL
5600 . OIL
5100 . OIL
4800 . OIL
4600 . OIL
3900 . Oil.
3400 . OIL
4200 . OIL
4200 * OIL
4300 . OIL
4800 . OIL
5000 . COL
5000 , COL
4000 . OIL
4000 , OIL
4800 . OIL
5100 . OIL
5500 . OIL
5900 . OIL
6200 . OIL
133
533
133
133
133
133
D 149
D 149
D 149
D 149
D 149
R 171
R 171
R 171
R 171
R 171
- 199
199
199
167
167
167
167
167
167
167
139
139
139
139
139
, 139
139
. 2 119
. 2 119
. 2 119
.2 119
. 2 119
.2 119
19
19
19
19
19
19
19
1
1
2
1
1
2
1
1
1
3
3
1
1
1
1
1
33
33
33
10
15
15
20
20
38
38
2
2
1
1
1
1
1
40
40
98
98
98
100
3
3
3
3
3
8
8
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
155
-------
Table B.2. CAPACITY REMOVALS, 1976 THROUGH 2000: INDIANA
UNIT.ID C0_INDEX NAME
1
2
3
1
T
3
4
4
5
1
1
2
3
1
2
3
4
5
6
J
1
2
3
4
3
4
5
6
1
2
3
4
5
6
1
2
3
4
5
6
2
3
ALCO
ALCO
ALCO
BLUF
BLUF
BLUF
BLUF
CRAU
CRAW
FOUA
FRAF
FRAF
FRAF
ICIU
INKE
INKE
INKE
INKE
INKE
INKE
INME
INME
INME
INMII
INME
INPl.
INPL
INPL
INPL
INPL
INPL
INPL
INPL
INPL
INPL
INPL
INPL
INPL
INPL
INPL
INPL
INSR
LOSF-
LOSP
UARRICK
WARRICK
UARRICK
BLUFFTON
BLUFFTON
BLUFFTON
BLUFFTON
CRAWFORDSVILLE
CRAWFORDSVILLE
SAINT JOE DAM
FRANKFORT
FRANKFORT
FRANKFORT
CHARLESTOWN
CLIFTY CREEK
CI.IFTY CREEK
CLIFTY CREEK
CLIFTY CREEK
Cl IFTY CREEK
CLIFTY CREEK-
BREED
TANNERS CREEK
TANNERS CREEK-
TANNERS CRFEK
TANNERS CREEK
PERRY K
PERRY K
PERRY K
PERRY K
PRITCHARDf H T
PRITCHARDf H T
PRITCHARDf H T
PRITCHARDf H 1
PRITCHARDf H T
PRITCHARDf H T
STOUT f ELMER U
STOUT f ELMER U
STOUT i ELMER U
STOUT f ELMER U
STOUT f ELMER W
STOUT f ELMER U
HOOSIER
L OGANSPORT
LOGANSF'ORT
173
173
173
179
179
179
179
107
107
3
23
23
23
19
77
77
77
77
77
77
153
29
?9
29
29
97
97
97
97
109
109
109
109
109
109
97
97
97
97
97
97
125
17
17
136
136
136
1
1
3
3
12
13
1
6
10
17
55
225
225
225
225
225
225
496
153
153
215
580
15
15
13
5
46
46
50
69
69
114
37
37
38
43
114
114
234
6
8
S
S
S
S
S
S
S
S
S
S
D
S
S
S
S
S
S
S
S
S
S
S
S
S
S
r;
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
6000
6000
6000
4700
4700
5200
5200
5500
6500
2800
4199
5299
6299
«
5502
5505
5507
5510
5511
5603
6000
5100
5200
5400
6400
2300
2400
3800
3800
4900
5000
5100
5300
5300
560O
3100
3100
4100
4700
5800
6100
•
2900
3900
COUNTY MWE STATUS DATE RETIRE FUKL
. COI.
. coi
. coi
. OTI
. on
. on..
. on.
8699 COL
. COL
. WAT
, COL
. COI
, COL
. IJNK
. COL
. COL
. COL
. COL
. COL
. COL
. COL
. COL.
, COL
. COL
. COI.
. COI
. COL
. COI
. COL
, FO:?
. F02
. COI
. COI..
. COL
. COI.
. F02
. F02
. F02
. K02
. COI
. COL
. COL
. COI.
. COL
(continued)
156
-------
Table B.2. (continued)
UN IT.. ID CO.. INDEX NAME
COUNTY MWF STATUS DATE RETIRE FUEL
4
5
1
2
3
4
1
2
3
1
2
3
6
7
8
66
1
2
3
4
1
2
1
2
3
4
5
4
5
6
7
1
1
1
2
1
2
3
4
5
6
7
1
2
3
4
1
4
LDSF1
I.OSM
NOIF1
NOIF1
NOIP
NOD-
NOIP
NOIP
NOIP
PERT
PERI
PERI
PSIN
PSIN
. PSIN
PSIN
PSIN
1 PSIN
. PSIN
PSIN
PSIN
PSIN
PSIN
.PSIN
PSIN
PSIN
PSIN
RCMP
RCMr
RCMP
RCMP
RIC[
SOIG
: SOIG
SOIG
SOIG
SOIG
SOIG
SOIG
SOIG
SOIG
SOIG
WCIN
WCIN
WCIN
WCIN
RIC[
RTCI
I. OGANSPORT
LOGANSPORT
NORWAY
NORWAY
NORWAY
NORWAY
OAKDALE
OAKDALE
OAKDALE
PERU
PERU
PERU
EDUARD5PORT
EDWARDSPORT
EDWARDSPORT
EDWARBSPORT
GALLAGHER* R
GALLAGHER» R
GALLAGHER, R
GALLAGHER, R
NOBLESVILLE
NOBLESVILLE
WABASH RIVER
WABASH RIVER
UABASH RIVER
WABASM RIVER
WABASH RIVER
RENSSEI.AER
RENSSELAER
RENSSELAER
RENSSELAER
WHITEWATER VALI EY,
CULLEY '
NORTHEAST j
NORTHEAST
OHIO RIVER I
OHIO RIVER |
OHIO RIVER j
OHIO RIVER
OHIO RIVER " '
OHIO RIVER
OHIO RIVER i
WASHINGTON
WASHINGTON
WASHINGTON
WASHINGTON
JOHNSON STREET
JOHNSON STREET
17
17
181
181
181
181
15
15
15
103
103
103
83
83
83
83
43
43
43
43
57
57
167
167
167
167
167
73
73
73
73
177
173
163
163
163
163
163
163
163
163
163
27
27
27
27
177
177
18
25
2
2
2
1
4
3
4
10
5
25
35
40
69
3
150
150
150
150
50
50
113
113
113
113
125
1
A.
3
3
33
50
11
12
8
13
13
20
23
23
23
5
5
3
5
15
15
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
M
M
5800
6400
2305
2305
2305
2305
2511
2511
2511
5000
3300
5900
4400
4900
5100
4400
5900
5800
6000
6100
5000
5000
5300
5300
5400
5400
5600
4099
5099
5799
6499
5500
5500
6300
6400
2900
2900
3600
3800
4500
4900
5100
4700
5700
3800
5700
3400
4000
. COL
, COL
. WAT
. WAT
. WAT
. WAT
, WAT
. WAT
. WAT
. COL
. COL
. COL
8501 F02
8501 BIT
8501 BIT
. F02
. BIT
, BIT
. BIT
. BIT
8501 BIT
850 J Bl. r
. BIT
« K 1 T
, BIT
. BIT
, BIT
B2t? F02
. F02
. F02
. F02
. COL
. COL
. GAS
. GAS
. FO?
. FO?
. F02
. F02
. F02
. F02
. FO:?
. COL
. COL
. COL
, COL
. COL
.. ,COL
157
-------
Table B.3. CAPACITY REMOVALS, 1976 THROUGH 2000: KENTUCKY
UNIT - ID HO...INDEX NAME
COUNTY MWE STATUS DATE RETIRE FUEL
1
2
3
4
5
6
1
.1
2
3
4
1
2
3
4
5
1
2
1
1
2
1
2
3
1
2
3
4
1
2
3
3
1
2
3
1
2
3
4
I
2
3
4
5
CETV
CETV
CETV
CFTV
CETV
CETV
FAKR
F AKR
GAKR
EAKR
FAKR
MEND
HEND
HEND
HEND
HEND
HEND
HEND
KEPC
KEUC
KEUC
KFUC
KEUC
KEUC
KEUC
KEUC
KEUC
KEUC
KEUC
KEUC
KEUC
KEUC
KEUC
KEUC
KEUC
LOGL
LOGE
OGE
OGE
OGE
OGE
OGE
LOGE
LUGE
WOLF CREEK
WOLF CREEK-
WOLF CREEK
WOLF CREEK-
WOLF CREEK-
WOLF CREEK-
COOPER
DALE
PALE
DALE
DALE
POWER STATION ONE
POWER STATION ONE
POWER STATION ONE
POWER STATION ONE
POWER STATION ONE
POWER STATION TWO
POWFR STATION TWO
BIG SANDY
BROWN* E W
BROWNr E W
D.IX DAM
DIX DAM
DIX DAM
GREEN RIVER
GREEN RIVER
GREEN RIVF:R
GREEN RIVER
LOCK *7
LOCK #7
LOCK *7
PINEVILLE
TYRONE
TYRONE
TYRONE
CANE RUN
CANE RUN
CANE RUN
CANE RUN
OHIO FALLS
OHIO FALLS
OHIO FAI LS
OHIO FALLS
OHTO FALLS
207
207
207
207
207
207
199
49
49
49
49
101
101
101
101
101
233
233
127
167
167
79
79
79
177
177
177
177
167
167
167
13
239
239
239
111
111
111
111
111
111
111
111
111
45
45
45
45
45
45
1?1
2 11
22
66
66
I
1
5
5
19
148
148
281
114
180
8
8
8
32
32
75
114
1
1
1
35
31
31
75
113
113
147
163
10
10
10
10
10
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
5100
5100
51.00
5?00
5200
5200
6502
5412
5412
5708
6008
4800
4800
5100
5100
5600
•
•
6301
5705
6306
2599
2599
2599
5003
5001
5404
5907
2799
2799
2799
5107
4710
4806
5307
5400
5600
5800
6200
2700
2700
2700
2700
2700
•
•
«
»
»
*
•
•
*
•
*
*
•
*
*
»
•
•
»
9:->oo
9800
•
•
•
8500
8500
8900
9400
•
•
«
86QO
8200
8300
BHOO
8506
8606
8U06
•
WAT
WAT
WAT
WAT
WAT
WAT
COL
COL
not
COI..
COL
F02
F02
COL
COL
COI
COL
COL
COL
COL
COI.
WAT
WAT
WAT
COL
COL
COL
COI
WAI-
WAT
WAT
COI.
F02
1" 02
COI.
CUL
COL
COI.
COL
WAI
WAT
WAT
WAT
WAT
(continued)
158
-------
Table B.3. (continued)
UNIT_ID CO.INDEX NAME
6
7
8
1
2
3
4
5
6
7
8
1
1
2
4
1
2
3
4
5
1
2
1
2
3
4
5
6
7
8
9
10
LOGE
* LOGE
LOGE
LOGI;:
LOGE
LOGE
LOGE
LOGE
LOGE
LOGE
LOGE
OWFN
OWEN
OWF.N
OWEN
TEVA
TEVA
FEVA
TEVA
TEVA
TEVA
TEVA
TEVA
TEVA
TEVA
TEVA
TEVA
TEVA
TEVA
TEVA
TEVA
TEVA
OHIO FALLS
OHIO FALLS
OHIO FALLS
PADDY'S RUN
PADDY'S RUN
PADDY'S RUN
PADDY'S RUN
PADDY'S RUN
PADDY'S RUN
WATERSIDE
WATERSIDE
ELMER SMITH
OUENSBORO
OWENSBORO
OWENSBORO
KENTUCKY
KENTUCKY
KENTUCKY
KENTUCKY
KENTUCKY
PARADISE
PARADISE
SHAWNEE
SHAWNEE
SHAWNEE
SHAUNEE
SHAWNEE
SHAWNEE
SHAWNEE
SHAWNEE
SHAWNEE
SHAWNEE
COUNTY HUE STATuS DATE
'ill
til
111
111
111
111
111
Itl
111
111
111
59
59
59
59
157
157
157
157
157
\'?7
177
145
145
145
145
145
145
145
145
145
145
10
10
10
25
25
49
45
75
75
20
25
149
7
8
35
37
32
32
32
37
704
704
175
175
175
175
175
175
175
175
175
175
S
S
S
S
S
S
S
S
S
S
S
S
S
S
f)
S
S
S
5
s,
S
s
s
s
s
s
s
s
s
s
s
s
2700
2700
2700
4200
4200
4700
4900
5000
5200
__..64_pO
640
-------
Table B.4. CAPACITY REMOVALS, 1976 THROUGH 2000: OHIO
UN IT..ID CO.... INDEX NAME
COUNTY MWF STATUS DATE RETTRF FUEL
1
2
1
3
4
r;
6
i
2
3
4
5
t
3
6
7
8
1
2
3
3
4
5
1
2
3
4
I
p
3
4
S
7
H
I
2
3
4
5
A
9
ARCA
ARCA
CIOE
CIGE
CIGE
CIGE
CIGE
CIGE
CIGE
CIGE
CIGE
CIGE
COLU
COI.U
COLU
COLU
COLU
COSO
COSO
COSO
COSO
COSO
COSO
COSO
COSO
COGO
COSO
DAPO
DAPO
DAPO
DAPO
DAPO
DAPO
DAPO
DAPO
DAPO
DAPO
DAPO
DAPO
DAPO
DOVE
FITR
GOTR
HAM1
ARCANUM
ARCANUM
DICK'S CRFEK
MIAMI FORT
MIAMI FORT
MIAMI FORT
MIAMI FORT
WALTER C BECKJORD
WALTER C BECKJORD
WALTER C BECKJORD
WALTER C BECKJORD
WALTER C BECKJORD
COLUMBUS
COLUMBUS
COLUMBUS
COLUMBUS
COLUMBUS
CONES VTL.LE
CONESVILLE
CONESVILLE
PICWAY
PICWAY
PTCUAY
POSTON
POS TON
POSTON
POSTON
FRANK M TATT
FRANK M TAIT
FRANK M TAIT
FRANK M TAIT
FRANK M TAIT
FRANK M TAIT
FRANK M TAIT
HUTCHINGS
HUTCHINGS
HUTCHINGS
HUTCHINGS
HUTCHINGS
HUTCHINGS
DOVER
AKRON
AKRON
HAMILTON
37
37
17
61
61
61
61
25
25
25
25
25
49
49
49
49
49
31
31
31
129
129
129
9
9
9
9
113
113
113
113
113
113
113
113
113
113
113
113
113
357
153
153
I/
1
1
120
65
65
too
163
US
113
125
163
245
8
8
13
13
13
125
125
125
30
30
85
40
40
60
60
30
30
35
147
147
30
30
69
69
69
69
69
69
33
58
65
bO
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
5100
4600
6500
3800
4200
4900
6000
5200
5300
5400
5800
6200
•
*
•
•
•
5999
5799
6299
4399
4999
5599
4999
5099
5299
5499
4501
4204
5112
5806
5905
3710
4007
4807
4903
5012
5102
5211
5308
•
•
•
2900
*
•
•
8001
8001
*
»
•
•
t
*
•
7711
/71 1.
7711
'711
80 1 1
»
•
t
no 10
801.0
*
•
»
*
•
•
•
•
•
*
•
*
•
*
«
*
•
*
•
•
«
•
Oil
on
KKR
FO?
ro?
COL
COL
COL
COL
cni..
COL
COI
UNK
UNK
COL
UNK
GAS
COL
COI.
COL
COI
COI
COL
COL
COI
COL
COL
KOL»
K02
h 02
cm
COL
FO:>
FO?
COL
COL
COL
COI
COL
COL
COL
(INK
UNK
r:oi.
(continued)
160
-------
Table B.4. (continued)
UNIT-ID CO_INDEX NAME
4
5
7
GT)
1
1
3
4
S
^
1
2
3
4
5
1-4
6
7
1
2
3
1
2
1
2
3
4
5
6
7
2
3
4
1
2
3
4
5
4
5
6
7
2
HAMI
HAMI
HAMI
HAMI
HAMI
LEOH
LEOH
LEOH
, LEOH
LEOH
MECP"
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHEC
OHPC
OHPC
OHPC
OHVE'
OHVE
OHVF.
OHVH
OHVE
ORRV
PI PC)
PIQU
PIQU
PIQU
PIQU
RESC
RMOI 1
HAMILTON
HAMILTON
HAMILTON
HAMILTON
HAMILTON HYDRO
LEBANON
LEBANON
LEBANON
LEBANON
LEBANON
" CHTLLICOTHE
BURGER r R E
BURGER. R E
BURGER f R E
BURGER f R E
BURGER r R E
EAST PALESTINE
GORGE
GORGE
MAD RIVER
MAD RIVER
MAD RIVER
NILES
NILES
SAMMIS
SAMMIS
SAMMIS
SAMMIS
TORONTO
TORONTO
TORONTO
MUSKINGUM RIVER
MUSKINGUM RIVER
MUSKINGUM RIVER
KYGER CREEK
KYGER CREEK
KYGER CREEK
KYGER CREEK
KYGER CREEK
NORTH VINF ST.
BARBERTON
PIQUA
PIQUA
PIQUA
PIQUA
YOUNGS TOWN
READING
17
17
17
17
17
165
165
165
165
165
'"141
13
13
13
13
13
29
153
153
23
23
23
155
155
8.1
81
81
81
81
81
81
115
115
115
53
53
53
53
53
169
153
109
109
109
109
99
61
9
10
22
11
1
I
1
1
2
3
' 68
66
66
103
16)
161
12
48
48
22
25
24
115
115
188
188
193
193
42
65
65
220
238
238
217
217
217
217
217
89
87
8
I
13
22
28
3
S
S
S
S
S
S
S
S
S
S
S
S
S
R
S
S
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
s
n
s
s
s
s
s
s
s
s
s
s
s
s
s
s
3800
5400
6000
6400
1900
4000
4900
5000
5500
_6100
.
4401
4712
5003
5503
5506
4799
4309
4812
2707
3811
4902
540.1
5406
5908
6007
6107
6211
4010
4908
4911
5406
5712
5305
5502
5506
5509
5511
55.12
•
»
4700
4700
5100
6100
•
4600
COUNTY MWE STATUS DATE RETIRE FUEL
. COL
, COL
. OIL
. OIL
. WAI
. OIL
. Oil.
. Oil.
. OIL
. OIL
. UNK
. BIT
. BIT
. BIT
. B.rr
. BIT
. BIT
, BIT
, BIT
. BIT
. BIT
. BIT
T i "I f'
. BIT
. BIT
. BIT
. Brr
. KIT
. BIT
. BIT
. BIT
. cm.
. COL
. COL
. COI.
. COL
. COL
. CUl
. cm
. COL.
. UNK
. COI
. COI.
. COL
. COL
. UNK
. UNK
(continued)
161
-------
Table B.4. (continued)
UNIT_1D CO..INDFX
4
in
TC?
4
5
1
2
1
?
3
4
5
RMOH
RMOH
RMOH
SHBY
SMML
SMML
SMML
UNCA
UNSS
YOST
OHPC
OHPC
OHPC
OHPC
OHPC
OHPC
OHPC
OHPC
OHPC
OHPC
NAME
READ.TNG
READING
READING
SHELBY
SAINT MARYS
SAINT MARY5J
SAINT MARYS
MARIETTA
YOUNGSTOWN
CAMPBELL
PHILO
PHILO
PHTLO
TIDD
TIDP
UOODCOCK
WOODCOCK
WOODCOCK
UOODCOCK
WOODCOCK
COUNTY
61
61
61
139
11
11
11
167
9
99
.1.19
119
119
81
HI
3
3
3 ""
3
3
MWE
6
?
2
39
1
3
A
160
45
49
85
85
125
105
105
5
5
8
10
JO
STATUS
S
S
S
<•>
S
S
S
S
S
S
M
M
M
M
M
M
M
M
M
M
DATE
5800
6500
6500
«
3900
4600
5700
•
•
*
4110
4206
5708
4599
4899
3800
3800
4100
4700
bOOO
RETIRE
»
*
*
4
t
•
•
•
t
7712
7S06
7JiOH
7505
7610
/A10
7502
/hO'»
750?
7502
7bO?
FUEL
UNK
UNK
(INK
COL
Oil.
COl
COL
COl
UNK
UNK
COl.
COl.
COl
COl.
COl
COl
COl
COl
COL
COl
162
-------
Table B.5. CAPACITY REMOVALS, 1976 THROUGH 2000: PENNSYLVANIA
UNIT_ID CO-INDEX NAME
COUNTY MWE STATUS DATE RETIRE FUEL
1
2
3
4
1
2
3
4
HI
H2
H3
3
4
5
1
?
3
4
IC5
IC6
TC7
1
j>
3
4
5
D
1
1
2
1.
o
3
BESC
DUI.C
DULC
DULC
DULC
DUI.C
DULC
DULC
DULC
JOLS
JOLS
PEEC
PEEC
PEEC
PEEC
PEEC
PFEC
PEEC
PEEC
PF.EC
PEEC
PEEC
PEEC
PEEC
PEPC
PEPC
PEPC
PFPC
PEPC
PEPC
SAJC
SIKO
UNSS
UNSS
UNSS
UNSS
WEPP
WEPP
WEPP
WEPP
WEPP
JOHNSTOWN
ELRAMA
ELRAMA
FLRAMA
L1.RAMA
PHI 11 IPS* F
PHILLIPS* F
PHILLIPS* F
PHILLIPS* F
ALIQUIPPA
PITTSBURG WORKS
PINEY
PINEY
PINEY
SEWARD
SEWARD
SEWARD
SHAUVILLE
SHAWUILLE
SHAWVILI.E
SHAWVILLE
SHAUVILLE
SHAWVILLE
SHAWVILLE
NEW CASTLE
NEW CASTLE
NEW CASTLE
NEW CASTLE
NEW CASTLE
NEW CASTLE
ST. JOSEPH
KOBUTA
CLAIRTON
CLAIRTON
EDGAR THOMSON
HOMESTEAD
ARMSTRONG
ARMSTRONG
MITCHELL
MITCHELL
MITCHELL
21
125
125
1?5
125
3
3
3
3
7
3
31
31
31
63
63
63
33
33
33
33
.33
33
33
73
73
73
73
73
73
7
7
3
3
3
3
5
5
125
125
125
75
99
105
114
176
74
83
8?
148
47
70
9
9
9
35
62
156
133
133
188
188
2
2
2
47
50
115
134
160
6
25
35
49
40
65
68
180
180
89
89
291
S
S
p
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
. .
5206
5303
15411
601.1
4301
4911
5009
5602
. .
. .
2406
2407
2802
4112
5005 9000
5704 9700
5408 9400
5408 9400
591?
6004
6312 .
6312
6312
3900
4700
5200
5800
6400
« t
5902
. «
. •
. «
. .
» •
5800
5900
4800
4900
6300
(INK
COI
COL
COI
COL
COI
COL
COL
COL
UNK
(INK
WAT
WAT
WAT
MUL
MUL
COL
COL
COL
COI.
COL
OIL
OIL
Oil.
COL
COL
COI
COI.
COI
F02
COI
UNK
UNK
UNK
UNK
UNK
COL
COL
T06
F:06
COI
163
-------
Table B.6. CAPACITY REMOVALS, 1976 THROUGH 2000: WEST VIRGINIA
UNTT.TD CO.. INDEX NAME
COUNTY MWE STATUS DATF-. REF[RE HJEl
a
9
i
2
1
2
3
4
5
1
2
3
1
o
3
1
2
3
p
3
1
5
6
1
2
1
2
3
1
2
1-4
APPC
APPC
APPC
APPC
CEOC
CEOC
CEOC
CEOC
CEOC
FOMA
KVPO
KVPO
KVPO
KVPO
Kvro
KVPO
KVHO
KVHO
KVPO
MOPC
MOPC
POEC
MOPC
MOPC
MOPC
MOPC
OHPC
OHPC
OHPC
PIPO
UNCA
IINCA
VIFP
VIEP
WEPP
WESC
CABIN CREEK-
CABIN CREEK-
KAN AWH A K'TVER
KANAWHA RTVFR
SHORN r PHTL
SPORNr PHIL
SPORNr PHTL
\
SPORNf PHIL
SPORN» PHIL
SOUTH CHARLESTON
LONDON
LONDON
L ONDON
MARMET
MARMET
MARMEF
WINFIELD
WINFIELD
WINFIELD
ALBRIGHT
ALBRIGHT
Al. BRIGHT
RTVESVILLE
RTVESVILI.E
WELLOW ISLAND
WILLOW ISLAND
KAMMER
KAMMER
KAMMER
MARTINSVII.LE
ALLOY WORKS
ALLOY WORKS
MT STORM
MT STORM
LAKE LYNN
WEIRTON
39
39
39
39
53
53
53
53
53
39
39
39
39
39
39
39
79
79
79
77
77
77
49
49
73
73
51
51
51
103
19
19
23
23
61
29
85
B5
220
220
153
153
153
153
496
35
5
5
5
5
5
5
5
5
5
76
140
76
48
94
58
188
238
238
238
120
102
123
570
570
52
109
S
S
S
S
S
S
S
5
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
4209
4305
5307
531.2
500.1.
5007
5108
5202
6012
,
3601
3601
3601
3601
3601
3601
3801
3801
3801
5400
5200
5200
4300
S100
4900
6000
5807
5811
5903
.
,
•
6500
6000
4
4
7VIO C()|.
7710 COL
. COL
. COL.
. COL
. COI
. COI.
. COL
. COL
. UNK
. WAT
. WAT
. WAT
. WAT
. WAT
. WAT
, WAT
, WAT
. WAI
, COL
, COL
. COI
. COL.
COI..
COL
COL
COI
COL
COL
UNK
. • WAT
UNK
COL
COL
WAT
UNK
164
-------
APPENDIX C
AIR QUALITY DATA FOR ORBES COUNTIES, 1977
Table C.I. COUNTIES IN ILLINOIS, INDIANA, KENTUCKY, OHIO, PENNSYLVANIA,
WEST VIRGINIA WITH VIOLATIONS OF NAAQS FOR S02 AND/OR
LESS THAN THE FULL PSD INCREMENT AVAILABLE
AT NADB MONITORS IN 1977
Sttite and
County
Illinois
Cook
Du Pag*
Madison
Peoria
iazeweil
Wil liaison
Indiana
Floyd
Jefferson
Lake
Marion
Wayne
Kentucky
Jpffpr?on
Hv,C> aCr'.cn
Ohio
Bclmont
Columbian?
Cuyahoya
Hamilton
Jefferson
Lake
Lora in
Lucas
Number
Violati
3 Hour
0
0
0
0
0
0
0
0
0
0
0
0
C
0
0
1
0
0
0
0
0
of Monitors
ng These
24 Hour
0
0
1
0
1
0
0
0
2
0
0
0
0
0
0
2
0
0
1
1
2
Standards*
Annual
0
0
It
0
0
0
1
0
0
2t
0
3t
A
0
0
1
2(2t)
0
2
1
It
0
(continued)
Number
the Ful
3 Hour
1
0
1
1
1
0
1
1
1
0
1
2
**
0
1
3
0
0
1
1
3
of Monitors with Less than
1 PSD Increment Available**
24 Hour Annual
3
0
0
1
0
0
1
0
0
1
0
1
2
0
1
2
1
1
0
0
1
2t
It
0
0
1
0
0
3t
l(lt)
0
It n i \
(2'!)
/»
v
1 •••
1
4(lt)
1
2(2t)
It
0
1
165
-------
Table C.I. (continued)
State and
County
Ma honing
Monroe
Monto_nicry
Scioio
Stark
Suimri t
Pennsylvania
Allegheny
Philadelphia
West Virginia
Urooke
Hancoc'c
Marshall
Wood
Number of Monitors
Violating These Standards*
3 Hour 24 Hour Annual
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
2
0
0
0
0
1
0
0
0
0
0
5t
4+
0
2t
0
0
Number
the Ful
3 Hour
0
0
0
0
0
0
3
2
0
0
0
0
of Monitors with Less than
1 PSD Increment Available**
24 Hour Annual
0
0
0
0
0
0
3
3
0
1
0
0
2
1
It
0
1
2(4-!)
"• \ /
?•!
2\
rt
l(lt)
1
It
*The violations are defined as two observations -• 1300ug/nr (3 hour), two obser-
vations> 365ug/m-' (24 hour) or one observation > 80V g/m* (annual).
**Thr working definition for "less than the full PSD increment available" in
this table is the measured concentration to which the addition of the Class II
PSD increment equals a violation of the standard. A monitor with a measured
violation is not considered eligible.
tLess than four valid quarters of data were used for averaging the annual mean.
Source: U.S. Environmental Protection Agency. 1978. Air Quality Data - 1977
Annual Statistics Including Summaries with Reference to Standards. EPA
450/2-78-040. Research Triangle Park, N.C. September.
166
-------
Table C.2. COUNTIES IN ILLINOIS, INDIANA, KENTUCKY, OHIO, PENNSYLVANIA
AND WEST VIRGINIA WITH VIOLATIONS OF NAAQS FOR TSP AND/OR LESS THAN
THE FULL PSD INCREMENT AVAILABLE AT NADB MONITORS IN 1977
i ; fiimber of Mcmtsrs VioUlinc
1 The?? F'antJ^rd;*
;State 6 "" 'i_
Ccuntv 'Prirr
Illinois
Adams
Bureau
- -01
3
0
Champaign 0
Cook 10
De Kalb
Ou Page
Effingliam
Jackson
Jefferson
Jo Da vi ess
Kane
Kankakee
Kendall
Knox
Lake
La Salle
MrHenrv
McLean
Ma con
Madison
Massac
Menard
Monroe
Peon a
Rock Island
St. Clair
Sanganon
Tazevtell
Uhitesidc
Will
Williamson
Winnebago
Indiana
Allen
Bdrlholciiiew
Clai k
Uolanare
Dubois
Ell-har:
Moyd
Howard
Grant
Jasper
Jefferson
KM.-
Lake
La Porte
(lad i son
Marion
Monroe
Poi ;er
1
0
0
0
0
0
0
0
0
0
0
T
0
0
0
6
0
0
0
0
0
0
0
0
0
0
0
p
0
0
0
0
0
0
0
0
0
2
0
0
2
0
0
0'
0
0
if
: irr.j?i
I
Number of Monitors «'. :n Less man
the full "SD i-.c-enont rvailabl?" '
2- hour
>«-ccnc--/ Pri-arv ieconcarv Prrncr.-
1
i
0
40
4
d
0
0
1
1
0
1
0
1
1
3
0
1
2
12
1
0
1
2
3
?
1
1
0
3
0
0
0
0
1
0
0
C
0
1
0
2
0
0
14
0
0
7
0
*
0
0
0
20(3T)
K2-)
2
0
IT
0
0
0
0
0
0
0
ST
0
0
2
1K1-)
0
0
0
2
2(r)
1(1")
C
2"
0
3(1")
0
0
0
0
1
0
0
0
0
0
0
1"
0
0
3(6t)
0
0
3
0
0
I"
0
32(6T)
2(3-)
6
0
2"
1
1
0
1
0
1
1
ST
0
1
2
11(2")
r
0
i
<(17)
4(2*)
1(2")
1
2"
1
7(3-)
0
0
0
IT
1
0
1
0
0
}
0
r
G
i
7(15-)
0
0
10(?t)
1
0
0
0
0
3
0
0
0
0
0
1
0
0
0
0
0
1
0
0
1
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
1
Seconcarv
2
0
1
12
1
3
1
0
0
0
1
0
1
0
3
1
0
0
0
1
0
1
0
2
3
1
0
1
1
4
1
2
0
1
0
0
1
0
0
0
1
1
0
1
6
0
3
?
. 1
0
-r.nui i
PriT.i""/1 Seconci' •
2
It
0
8(1")
1
1
It
0
0
0
0
0
0
0
l(l-)
1
0
0
0
1
0
1
0
0
1-
0
0
0
0
2"
0
1
1
0
0
0
0
0
c
0
0
0
0
0
• 4
3
0
1
0
0
0
0
0
12(1")
0
1
0
0
0
0
1
0
IT
0
3(1")
1-
?
0
0
0
0
0
0
0
1
0
1-
0
It
2(1")
1
0
It
0
0
2T
0
1
1
0
1
0
1
0
1(1")
2
K3-)
4
0
4~
(cont Lnued)
167
-------
Table C.2. (continued)
1 1
jStatc i ' ~_i
iCoun:- Tri-j'
St. Joseph 0
Tipp«cenoe 0
Var.dc-rtwrgn 0
Vigo 0
Wayne 0
Kentucky
Ballard 0
Barren 0
Bell 0
boons 0
SourL'On 0
boyd 0
Boyle 0
Bullitt 0
Calculi 0
Callomy 0
Campbell 1
Carlisle C
Carroll 0
Carter 0
Christian 0
Clark 0
Daviess
Fayette 0
Flcyd 0
Franklin 0
Fulton 0
Gallatir. 0
Grays on 0
Grecnui) 0
Nanccck 0
HarcMn 0
Marian 0
hernson 0
Heniicrson 0
Hopkins 0
Jefferson 1
Ken ton 0
Laurel 0
Lawrence 0
Livingston 0
Loaan 0
NcCracken 0
Mac'ison 0
Marsnall 0
Mason 0
Meare 0
Hurlenoury 0
Nelson 0
Ohio 0
Oldham o
Owen 0
* Pencil eton 0
Per-y i
Pul:ski 0
Row?n 0
5h?lby 0
M-,;iiOn 0
Tru.ble 0
x'ar:-nn 0
"'fc'js:-' °
"ucr o- '-lorii lors Yicli-.ing
"nese Standards'
•i.-i;r
'./ ".'.*r.cj~j
0
0
1
3
1
1
0
2
0
0
3
1
1
1
0
1
0
0
1
0
0
4
2
0
0
0
0
0
1
1
0
1
0
6
0
9
0
1
1
0
0
5
1
0
0
0
1
0
1
0
0
0
1
1
0
0
0
1
1
0
0
0
-r-.nu?i
^'--jrv ,S
0
0
0
4t
0
0
0
1(1-)
0
0
2
1
I*
0
0
It
0
0
0
0
0
4
0
0
0
0
0
0
0
0
0
r
0
4
0
0
0
0
0
0
1
0
c
0
0
0
0
0
0
0
1
1
G
0
0
0
1
0
0
0
i
?::rc,jiv
t-
0
2-
6T
2
I-1
1
KK)
l(l-)
0
0
0
IT
1
0
l(l-)
0
0
1
1-
5
0
1
0
1 *
0
1
0
1
0
0
1
6
1
12(27)
1
1
1
0
0
3
1
It
0
0
1
0
0
o
0
0
1
0
0
0
1
1
0
1
1
Numacr of "oniMr; rfitn ,.
2-
P.'V-.dr
0
0
0
0
0
0
0
1
0
0
2
1
0
0
0
0
0
0
1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
o
V
0
0
o
o
0
0
n
u
0
o
0
0
0
i
0
0
0
0
1
0
0
0
hey
•/ =eronc"-/
4
0
1
2
1
1
1
0
3
1
2
0
0
0
0
2
0
1
0
1
2
2
1
1
1
1
1
1
0
1
2
0
1
1
0
5
2
0
0
i
i
0
3
1
0
i
i
1
1
0
0
0
1
0
1
0
0
2
2
1
«nnu
D,..,arv
It
1
1
0
0
o
0
0
1
0
0
0
o
0
o
0
0
1
0
o
1
o
1
o
o
o
o
o
1
1
F
0
0
0
0
0
c
0
0
1 _
1
0
2
1
0
2
1
0
0
0
17
0
|
0
0
o
c
0
535 tflif, ;
el ' *:il»"»
C 1 '
1(4^)
0
0
0
^
0
0
2
1
2(1-)
0
o
o
1
1
1
1
o
o
o
1
3
o
1
o
1
o
o
0
2
0
0
2
0
1(1-)
0
0
1
5
0
*
I
1-
1
0
1
o
0
0
1
o
o
0
y
£
0
(continued)
168
-------
Table C.2. (continued)
1 i
;Siu-,.e £
;Ci. ji t .• '?— .
W;iHify
Ohio
Ac?3TS
Allen
Ashtabula
Athens
Belrnoni
Brown
butler
Carroll
Chairoaion
Clork
Clermonl
Gin-ton
Coiumriana
Cosnocton
Cuyahnga
Oar 1.6
Dofience
Delaware
trie
Franklin
Gal lift
Gcauga
Greene
Guernsey
Kami 1 ton
fiancee*
Ham sen
Henry
Hocking
Jackson
Jefferson
Lake
Lawre-ice
Lacking
Logan
Lorain
Lucas
Nahoning
Marion
Medina
Keigs
Miami
Monroe
Montgomery
tfuskinajm
Noble "
Percy
Portage
Richlarid
Ross
SandusLv
Scioto
Seneca
Shelby
Stark
Sunnit
Trumbull
1 us cam w.i s
Union
NuT.oer
i- • i •„
.-.:r \
0
n
1
r,
i
0
i
0
0
0
0
c
1
c
9
0
2
C
0
1
0
0
0
0
c
G
0
0
0
0
5
1
£
0
0
"l
0
6
0
0
0
C
0
0
0
0
0
0
0
0
5
1
G
0
0
0
0
0
n
of Mom
Tnpse st
r
:.cr.'.c.r.
1
0
2
2
0
3
1
2
2
1
O
i
i
5
0
20
0
5
1
1
6
0
1
*i
0
7
1
1
4
0
1
10
9
7
0
1
6
2
9
0
2
1
2
2
5
0
1
0
0
7
1
n
c
1
1
11
6
6
1
?
tors Violating
irdarcs*
•irr.ua!
J"i-er/ 1
1
0
0
0
0
2
K17)
1
0
0
0
1
0
3
0
15(2f)
It
IT
0
0
3
I*
0
0
0
e(it)
0
0
1
0
c
7(1")
3
3(lT)
0
1
2
0
6(3t)
0
IT
1
1
1
<(!T)
0
0
0
0
K2t)
0
5(17)
2(lt)
0
0
5
2
3
0
0
i
•
e^crcar/ ?r
0
0
2
4
0
3
1(2-)
5
1
1
2
2
2
7
0
2' (2-J
Hl7)
K3-)
1
2
6(27)
It
1
1
0
19(1-)
1
1
2
It
1
8(1")
8
6(1-)
0
1
6
7(3t)
6(37)
0
2(27)
1
Kl7)
6(3")
1
1
0
2
5(3-)
1
7(lt)
6(17)
1
It
not)
9(4")
7
0
2
Number of Monitors
»>• -"til! B?C Ire re-
e.-
1 ::t r
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
2
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
1
0
0
1
1
0
1
0
0
0
0
1
1
0
0
0
0
0
G
3
1
0
0
0
0
1
0
0
.-cur
v Secc'icar- ;
n
1
0
5
1
1
0
4
0
0
0
3
1
2
1
7
2
2
0
1
5
2
2
2
1
22
0
9
2
1
0
1
4
1
1
0
2
10
2
1
2
0
1
0
7
2
0
1
2
2
2
1
3
i
]
5
9
1
1
1
with Less
•sit ivaila
. "Ol'l 1
>r,.-ar, <,<••:
o
it
0
i
i
i
0
0
0
o
0
1
0
0
0
3
0
1-
0
0
1
0
0
1-
0
4
0
1
0
0
0
1
1
1
0
0
0
0
0
0
0
0
1
1
0
1
0
0
0
1-
0
0
0
1
1
5
2
0
0
tiar.
.'"-f.rv
o
2
0
Ul7)
o
o
0
1
2
o
0
2
0
0
1
6(5-)
0
2r
0
0
3
1
2
1
1
6(lt)
0
1
1
0
0
0
4
IT
1
0
4(1")
4(lt)
1
1
0
0
0
1
2(2T)
0
0
1
0
2-
3
5
Kl")
l(l-)
0
1
l(l-)
0
0
1
(continued)
169
-------
Table C.2. (continued)
.'•u.TEitr of Xomtors Violating
The;? S'.anoarc:*'
Ctita 5 '> •
Caimt ' P'n-ir
Warren 0
Kcsirington 0
Wayne 0
Wooa 0
Wyar.dct 2
Pennsylvania
Alleonsny 13
beaver 3
Berks 0
Blair 1
Bucks 0
Cambria <
Chester 0
Cumber l;nc 0
Dauphin 0
Delaware 0
Erie 2
T;.-:*.t: 0
Harrissurg 0
lackamanna 0
Lzncaster 0
Lawrence 2
Lenigh 0
Luzerne 1
Lycoming 0
Marcer 1
.'lontaoTcry 0
NortnnafiiptGiO
Philadelphia
Washington 0
Westnorelzodl
York 1
West Virginia
Berkeley 0
Lrooke 0
Cabeil 0
F*yette 0
Hancock 0
Harrison 0
Kt>nav>ha 0
Lewis 0
Mjrton 1
Mar snail 0
Mineral 0
Monongal 13 C
Ohio 0
Putnam 0
Raleigh 0
Wood 0
-£••„ r :
.' ^e.oncir/
1
0
1
0
3
19
8
2
2
0
5
2
0
1
0
3
1
0
1
2
2
0
3
1
2
1
2
7
2
1
4
1
2
1
1
3
1
6
1
2
1
0
0
2
0
1
1
f nnuf i
D'inar/ <•*
1
0
0
0
2
20t
8L
li-
lt
0
5-1
1-
0
It
0
37
C
0
It
1-
2-
0
2-
0
2t
0
It
7-
0
1-
4-
0
1
0
0
0
0
1
0
1
1
0
0
2-
0
0
0
:cr,aai-v
1
0
H2t)
0
3
23-
8t
5r
2-
0
5-
2-
!•••
I*
0
3-
It
1"
2t
4t
2"1
0
4T
2~
2~
3T
47
10J-
2r
1 A
5-
0
1
1
0
2(1 1)
0
2
0
1
1
0
0
«?
0
0
0
: Number c
:he full
: 2£ .ucur
.'.Prirarv Sec
0
0
0
0
0
1
2
0
0
0
0
0
0
1
o •
1
0
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
0
0
1
0
0
1
0
0
0
0
0
0
1
0
'" Monitors
rncarv • ?
2
1
3
2
0
3
0
3
0
0
0
2
1
1
0
1
C
1
2
2
0
0
1
1
0
2
1
3
2
0
2
0
0
1
0
0
0
0
0
0
0
2
1
1
1
0
0
witn Less tnar i
ont iva liable" '
-nnua
riir.arv
0
2
0
0
0
0
0
0
0
0
0
2-
0
17
!•"•
0
0
0
1-
1-
0
0
0
0
0
0
0
2-
1*
0
lr
0
0
0
0
0
0
1
0
p
0
0
0
0
0
0
0
,
^ec^ncer/
2
0
1
3
0
0
0
2"
0
3*
1*
G
lx
2"
2-1
2*
C
0
0
2-
0
2-
4
0
0
It
2+
It
1-
0
I1
0
1
0
0
0
1
2
1
0
0
0
0
1
1
0
2
*The violations are defined as two observations greater tnan 260 ug/n" {p-irrar> 24
hour), two observations create- than 150 jg/r,J (secondary 24 hour), one onservation
equal to 75 ug/n3 (primary annual) 0" one ooservation esual to 60 ug/irj (seconaary
annual). / measured violation of the prir.ary standard is includec as a violation
of the secondary standard.
"The working definition given to "less than the full PSD increment available" is the
measured concentration to whici the addition of the Class :i PSD increment equals
a violation of f,a standards, j.g., 22-' + 37 > 250, 114 - 37 > 150, *1 + 19 - CO
and i.6 + 19 - 75. A moi.ilor *\u\ a ncasjred ..ulaLion is not eliyiLi: fci r3;
Increment corsioe-'ition with tne exre3f.cn of ? violation of the secondary 24 hour
standard ( > 1?0 us/1"3) that leaves less tnan ;ne full PSD increTent toward the
primary 2J hour stanaard, e.g., 22* ix 126C.
tLess than four valid quarters of oata were us"d for averaging the annual mean.
Sourjce. U.S. Environmental Protection Atjjncy. 1978. Air Quality Oats - 1977 Annual
Statistics Inducing Surn->rieS v«itn Reference to Standards. -rA-ISO/2-78-0-0.
Research Triangle Park. S C Septe-nbcr
170
-------
APPENDIX D
COUNTIES EXCLUDED AS SITES FOR COAL-FIRED SCENARIO UNIT ADDITIONS,
BASE CASE ENVIRONMENTAL CONTROLS
Table D.I. COUNTIES EXCLUDED AS SITES FOR COAL-FIRED SCENARIO UNIT ADDITIONS,
BASE CASE ENVIRONMENTAL CONTROLS
State
Illinois
Indiana
Kentucky
Ohio
Pennsylvania
West Virginia
FIPS
fnHp
V>(JUC
079
167
013
019
061
089
111
131
145
147
177
233
027
091
105
139
141
003
039
083
093
Air Quality
Nonattainment3 PSD Public
T.anH«
TSP SO Class lb
Marion
Vigo
Bell
Boyd Boyd
Edmonson
Greenup
Jefferson Jefferson Leslie
McCracken McCracken
McCreary McCreary
Muhlenberg Muhlenberg
Webster
Clinton
Logan
Meigs
Richland
Ross
Allegheny Allegheny
Kanawha
Randolph
Tucker
"County designated nonattainment area, primary standards.
County contains mandatory Class I area.
CA11 of county in public lands, actual ownership.
171
-------
APPENDIX E
COUNTIES EXCLUDED AS SITES FOR COAL-FIRED SCENARIO UNIT ADDITIONS,
STRICT ENVIRONMENTAL CONTROLS
Table E.I. COUNTIES EXCLUDED AS SITES FOR COAL-FIRED SCENARIO UNIT ADDITIONS,
STRICT ENVIRONMENTAL CONTROLS
State FIPS
Code
Illinois 001
003
Oil
069
081
091
095
099
113
115
119
127
129
133
143
151
155
163
167
179
181
199
Air Quality
Nonattainment3 PSD Public
Landsc
TSP SO Class lb
2
Adams
Alexander
Bureau
Hardin
Jefferson
Kankakee
Knox
LaSalle
McLean
Macon
Madison
Massac Massac
Menard
Monroe
Peoria Peoria
Pope Pope
Putnam
St. Clair
Sangamon
Tazewell Tazewell
Union
Williamson
(continued)
172
-------
Table E.I. (continued)
State IMPS
Code
Indiana 019
025
029
037
067
097
101
117
123
163
167
177
Kentucky 013
019
029
037
051
059
061
065
089
101
109
111
127
129
131
145
147
151
157
165
177
Air Quality
Nonattainment3
TSP SO
2
Clark
Dearborn
Dubois
Howard
Marion Marion
Vanderburgh
Vigo Vigo
Wayne Wayne
Bell
Boyd Boyd
Bullitt
Campbell
Daviess Daviess
Greenup
Henderson Henderson
Jefferson Jefferson
Lawrence
McCracken McCracken
Madison
Marshall
Muhlenbere Muhlenbere
PSD Public
La nd 3 c
Class lb
Crawford
Martin
Orange
Perry
Clay
Edmonson
Estill
Jackson
Lee
Leslie
McCreary McCreary
Men! fee
(continued)
173
-------
Table E.I. (continued)
State FIPS
Code
Kentucky 193
]95
197
205
233
235
Ohio 003
009
013
017
019
021
023
025
027
029
031
037
049
053
057
061
079
081
087
091
099
101
103
105
109
111
113
115
JJ9
129
133
135
139
Air Quality
Nonattainment3 PSD Public
TSP
Perry
Pike
Whitley
Allen
Belmont
Butler
Carroll
Champaign
Clark
Clennont
Clinton
Columbiana
Darke
Franklin
Gallia
Greene
Hamilton
Jackson
Jefferson
Lawrence
Logan
Mahoning
Medina
Meigs
Miami
Monroe
Montgomery
Muskingum
Portage
Preble
Richland
Land?
SO, Class lb
2
Powe L 1
Rowan
Webster
Whitley
A.1 len
Athens
Clermont
Columbiana
Coshocton
Franklin
Gallia
Greene
Hamilton
Jefferson
Lawrence
Mahoning
Marion.
Medina
•
Montgomery
Morgan
Muskingum
Pickaway
(continued)
174
-------
Table E.I. (continued)
State FIPS
f*nr1 A
VfUUl.
Ohio U1
U5
149
151
153
155
157
J63
167
169
175
Pennsylvania 003
005
007
021
051
033
073
085
125
129
West Virginia 009
029
049
051
069
075
083
093
107
Air Quality
Nonattainment3 PSD
TSP
Ross
Scioto
Shelby
Starke
Summit
Trumbull
Tuscarawas
Washington
Wayne
Wyandot
Allegheny
•
Beaver
Cambria
Fayette
•
Lawrence
Mercer
Washington
Westmoreland
Brooke
Hancock
Marion
Marshall
Ohio
Wood
SO Class lb
2
Starke
Summit
Trumbull
Washington
Allegheny
Armstrong
Fayette
Washington
Westmoreland
Brooke
Hancock
Randolph
Tucker
Public
Lands0
Scioto
Vinton
Forest
Pocahontas
Randolph
Tucker
aCounty contains nonattainment area, primary and secondary standards.
'County contains mandatory Class 1 area.
Slajority of county in public lands; total area, including designated purchase
area.
175
-------
APPENDIX F
COUNTIES EXCLUDED AS SITES FOR NUCLEAR-FUELED SCENARIO UNIT ADDITIONS
Table F.I. COUNTIES EXCLUDED AS SITES FOR NUCLEAR-FUELED
SCENARIO UNIT ADDITIONS
FIPS
State Code
Illinois 003
033
047
055
059
065
069
077
081
083
087
101
119
127
133
145
151
153
157
159
163
165
181
185
189
191
193
199
Seismic Population
Suitability3 Densityb
Alexander
Crawford
Edwards
Franklin
Gallatin
Hamilton
Hard in
Jackson
Jefferson
Jersey
Johnson
Lawrence
Madison
Massac
Monroe
Perry
Pope
Pulaski
Randolph
Richland
St. Clair
Saline
Union
Wabash
Washington
Wayne
White
Williamson
(continued)
176
Public
Lands0
Alexander
Hardin
Pope
1
Union
-------
Table F.I. (continued)
State
FIPS
Code
Seismic
Suitability3
Population
Density"
Public
Landsc
Indiana
Kentucky
Ohio
025
051
083
097
101
117
123
129
163
007
003
037
039
051
055
065
067
075
083
101
105
109
111
117
131
139
143
145
148
157
165
197
205
225
233
235
017
049
061
Gibson
Knox
Marion
Posey
Vanderburgh
Ballard
Galloway
Carlisle
Crittenden
Fulton
Graves
Henderson
Hickman
Crawford
Martin
Orange
Perry
Vanderburgh
Campbell
Clay
Estill
Fayette
Jefferson
Kenton
Livingston
Lyon
McCracken
Marshall
Union
Webster
Jackson
Leslie
McCreary
Menifee
Powell
Rowan
Whitley
Butler
Franklin
Hamilton
(continued)
177
-------
Table F.I. (continued)
State
Pennsylvania
West Virginia
FIPS
Code
087
099
113
145
151
153
163
003
053
069
075
083
093
Seismic Population
Suitability3 Density1*
Mahoning
Montgomery
Stark
Summit
Allegheny
Ohio
Public
Lands0
Lawrence
Scioto
Vinton
Forest
Pocahontas
Randolph
Tucker
aCounty within relative seismic suitability zone III.
County population density _> 500 persons per square mile.
°Majority of county in public lands; total area, including designated purchase
area.
178
-------
APPENDIX G
ECAR REGION SITE INVENTORY
The East Central Area Reliability (ECAR) council maintains an inventory
of major thermal electric power plants that are located in the service areas
of the council's member utilities. The inventory is based on the utilities
reporting sites they own or have substantial holdings with options to buy.*
Potential sites that companies may be evaluating for their suitability, or that
have been indicated as potential alternate sites as required by a state siting
agency or other authority, are not included. The ECAR site inventory includes
all of the ORBES region except the Illinois state subregion, which is in the
Mid-America Interpool Network (MAIN). No region site inventory is available
for MAIN.
The ECAR region site inventory consists of a map of sites and a listing
of selected site information. The site information includes the site name; the
utility that reports the site; fuel types of generating units; and information
about the size (in MWe) and number of units that are located at the site, or are
under construction or planned. Those sites that can physically accomodate ad-
ditional generation are also identified, although the inventory does not speci-
fy the magnitude of generation that could be added while meeting current air
and water quality regulations and other certification requirements. These "ex-
pandable" sites that are located in the ORBES region are listed in Table C-l.
*Correspondence form Mr. Owen A. Lentz, Executive Manager, ECAR,
dated July 20, 1979 and August 20, 1979.
179
-------
Table C.I.
INVENTORY OF ELECTRIC UTILITY SITES IN THE ORBES PORTION OF INDIANA, KENTUCKY, OHIO. PENNSYLVANIA
AND WEST VIRGINIA THAT ARE CAPABLE OF ACCOMODATING CAPACITY ADDITIONS
00
O
State
Subregion
INDIANA
KENTUCKY
OHIO
County
Jasper
Jefferson
Morgan
Parke
Pike
Posey
Spencer
Sullivan
Switzerland
Boone
Davies
Hancock
Henderson
Lewis
Mason
Pulaski
Trimble
Webster
Adams
Athens
Clermont
Jefferson
Lawrence
Meigs
Morgan
FIPS
Code
18073
077
109
121
125
129
147
153
155
21015
059
091
101
135
161
199
223
223
39001
009
025
081
087
105
115
Site Name
R. M. Schahfer
Marble Hill
Paragon
Cayuga
Frank E. Ratts
A. B. Brown
Reckport
Breed
Patriot
East Bend
Elmer Smith
Coleman
Henderson
St. Paul
Project 2602
H. L. Spurlock
J. S. Cooper
Trimble County
Unnamed
Reid/Henderson 02
Klllen
Sandy Springs
Poston
Z inane r
Rayland
Hanging Rock
Great Bend
Muskingum Mine
Company
Acronym
Coal
NIPS •
PSI
I PL
PSI •
HED •
SIGE •
AEP •
AEP •
IPL •
CG&E •
OMU •
BIRI •
AEP
AEP
AEP •
EK •
EK •
LG&E •
AEP
BIRI •
DPL •
AEP
CSOE •
CG&E
CEI
CSOE
AEP
Fuel Type Size (MWe)
and Number
Under
Oil Nucl Future Present Construction
477(1)
•
•
• 1024(6)
244(2)
400(1)
•
399(2)
455(3)
•
•
•
300(1)
354(2)
•
• 455(3)
•
• 250(5)
682(2)
2260(2)
250(1)
2600(2)
1200(2)
500(1)
495(1)
1200(2)
375(1)
807(1)
of Units
Planned
7500
7300
(continued)
-------
Table C.I. (continued)
QO
Sub region
PENNSYLVANIA
WEST
VIRGINIA
Councy
Armstrong
Beaver
Mason
Pleasants
PIPS
Code
42005
007
54053
073
Site Name
Lower Armstrong
Mansfield
Apple Grove
Mountaineer
Pleasants
Company Fuel Type
Acronym
Coal Oil Nucl Future
APS •
OE •
AEP •
AEP •
APS •
Size (MUe) and Number
Under
Present Construction
1260(2)
825(1)
2600(2)
1252(2)
of Units
Planned
SOURCE: ECAR Region Site Inventory. 1979.
------- |