April 1971
mplemenlatioD Planning Program
Environmental Protection Agency
Air Pollution Control Office
Washington, D.C.
Contract No. PH 22-68-60
TRW>
-------
SOME APPLICATIONS OF THE
IMPLEMENTATION PLANNING PROGRAM
D. H. Lewis, S. E. Plotkin,
K. R. Woodcock
April 1971
Prepared for
Environmental Protection Agency
Air Pollution Control Office
Washington, D.C.
TRW SYSTEMS GROUP
7600 Colshire Drive,
-------
The work upon which this
publication is based was performed
pursuant to Contract No. PH 22-68-60
with the Air Pollution Control Office,
Environmental Protection Agency.
-------
TABLE OF CONTENTS
Page
1.0 EXECUTIVE SUMMARY 1
1.1 THE IMPLEMENTATION PLANNING PROGRAM 1
1.2 STUDY OBJECTIVES 1
1.3 STUDY ASSUMPTIONS 2
1.4 ESTIMATING NATIONAL LEVEL CONTROL MEASURE DEMAND 3
1.4.1 Aggregation Approach 5
1.4.2 Extrapolation Approach 16
1.4.3 305(a) Approach 20
1.4.4 Conclusions and Recommendations 24
1.5 305(a) EMISSION STANDARDS EVALUATION 27
1.6 USE OF RAPID SURVEY AS IPP INPUT 29
2.0 INTRODUCTION 37
3.0 EXTRAPOLATION APPROACH 41
3.1 RESULTS OF POLLUTANT CONTROL STRATEGIES - INDUSTRY COSTS,
EMISSIONS, AND AIR QUALITY 42
3.1.1 Cincinnati AQCR 42
3.1.2 St. Louis AQCR 61
3.1.3 Washington, D. C. AQCR , 62
3.2 COST OF SIMULTANEOUS PARTICULATE AND S02 CONTROL 62
3.3 RESULTS OF POLLUTANT CONTROL STRATEGIES - DEVICE DEMAND. . 65
3.3.1 Cincinnati AQCR 65
3.3.2 St. Louis AQCR 65
3.3.3 Washington, D. C. AQCR 69
4.0 AGGREGATION APPROACH 71
4.1 RAPID INVENTORIES 72
4.2 CONSTRUCTION OF THE EMISSION SOURCE FILE 74
4.2.1 Exhaust Gas Volume 77
4.2.2 Fuel Data - Price, Sulfur Level, Heat Content ... 83
4.2.3 Stack Exit Temperature 85
4.2.4 Rated Capacity 85
4.2.5 Operating Hours 86
-------
TABLE OF CONTENTS (Continued)
Page
4.2.6 Source Identification ............... 88
4.2.7 Existing Control Equipment ............ 89
4.2.8 Source Location .................. 91
4.2.9 Stack Height ................... 91
4.2.10 Emission Sources with Multiple Stacks ....... ^3
4.3 IMPACT OF 3 05 (a) EMISSION CONTROL STRATEGY ........ 95
4.4 COMPARISON OF RAPID SURVEYS AND DETAILED INVENTORIES. . . 102
4.4.1 Emission Inventory Comparison ........... 102
4.4.2 Air Quality Estimates ............... 105
4.4.2.1 Rapid Survey versus Detailed Inventory -
Direct Comparison ............ 105
4.4.2.2 Comparison of Rapid Survey and Detailed
Inventory Calibration .......... 113
4.4.3 Comparison of 305 (a) Strategy Impacts - Detailed
versus Rapid Survey Inventory ........... 124
5.0 THE 305 (a) APPROACH ...................... 127
5.1 INTRODUCTION ....................... 127
5.2 DESCRIPTION OF THE 305 (a) APPROACH AND PRESENTATION OF
REGIONAL DEMAND ESTIMATES ................ 130
5.2.1 Introduction ................... 130
5.2.2 Scope of the Economics of Clean Air ........ 132
5.2.3 The Economics of Clean Air Control Cost
Estimating Procedure ...............
5.2.4 305(a) Approach - Step #1 ............. 142
5.2.5 305(a) Approach - Step //2 ............. 177
5.2.5.1 Source Category and Process ....... 187
5.2.5.2 Typical Control System Type ....... 188
5.2.5.3 Number of Systems Required ........ 188
5.2.5.4 Substitute Fuels Required ........ 188
5.2.5.5 Full Additional Control Investment. . . . 189
5.2.5.6 Typical Gas Volumes ........... 189
5.3.0 Limitations of the 305 (a) Approach ........ 190
5.3.1 Comprehensiveness of Source Identification .... 192
5.3.2 Completeness of the Identification of Source
Characteristics .................. 192
-------
TABLE OF CONTENTS (Continued)
Page
5.3.3 Appropriateness of the Selected Control System. . 193
5.3.4 Accuracy of Control Cost Estimates 194
6.0 REFERENCES 197
APPENDICES
A IMPLEMENTATION PLANNING PROGRAM 199
B ESTIMATING CONTROL COSTS 202
C AREA SOURCE SCALE FACTORS 210
-------
TABLES
Page
1-1 Emission Standards Used in Comparison 4
1-2 Impact of Steam Electric Powerplants 6
1-3 Aggregation Approach - Comparison of Rapid and Detailed
Emission Inventories/Regionwide 8
1-4 Aggregation Approach - Emission Inventory Comparison
Cincinnati Major Point Sources 10
1-5 Aggregation Approach - Location of St. Louis Powerplants ... 11
1-6 Aggregation Approach - Comparison of 305(a) Strategy
Impact on Point Sources 15
1-7 Extrapolation Approach - Control Costs -$/Ton Removed 18
1-8 Extrapolation Approach - Steam Electric Powerplants 19
1-9 305(a) Approach - Demand/Cost Predictions - Powerplants. ... 22
1-10 305(a) Approach - Demand/Cost Predictions - Powerplants -
IPP Using 305 (a) Control Measures 23
1-11 305(a) Approach - Demand/Cost Predictions - Industrial
Boilers 25
1-12 Achieving Air Qualtiy Standards Using 305(a) Emission
Standards 28
1-13 Major Point Sources in the Ohio Portion of MCIAQCR 34
3-1 Effects of Regulations; Cincinnati 305(a) S02 Strategy
(Detailed Inventory) 43
3-2 Effects of Regulations; Cincinnati 305(a) Particulate
Strategy (Detailed Inventory) 44
3-3 Effects of Regulstions; St. Louis 305(a) S02 Strategy
(Detailed Inventory) 45
3-4 Effects of Regulations; St. Louis 305(a) Particulate
Strategy (Detailed Inventory) 46
3-5 Effects of Regulations; St. Louis 305(a)/4 Particulate
Strategy (Detailed Inventory) 47
3-6 Effects of Reulgations; Washington, D.C. 305(a) S02 Strategy
(Detailed Inventory) 48
-------
TABLES (Continued)
Page
3-7 Effects of Regulations; Washington, D.C. 305(a)
Particulate Strategy (Detailed Inventory) 49
3-8 Effects of Regulations; Washington, D.C. 305(a)/2 S02
Strategy (Detailed Inventory) 50
3-9 Effects of Regulations; Washington, D.C. 305(a)/2
Particulate Strategy (Detailed Inventory) 51
3-10 Cost of Control for Three Cities, 305(a) S02/Particulate
Strategy 64
3-11 Cincinnati Device Demand and Cost, 305(a) Strategy
(Detailed Inventory) 66
3-12 St. Louis Device Demand and Cost, 305(a) Strategy
(Detailed Inventory) ... 67
3-13 Washington, D.C. Device Demand and Cost, 305(a) Strategy
(Detailed Inventory) 68
4-1 Implementation Planning Program Point Source Variables .... 75
4-2 Gas Volume Factors 79
4-3 Measured Gas Volume Factors 81
4-4 Operating Hours 87
4-5 Particulate Emission Factors for Coal Combustion
(Without Control Equipment) 90
4-6 Location of St. Louis Power Plants 92
4-7 Champion Paper Company Characteristics 94
4-8 Effects of Regulations; Cincinnati 305(a) S02 Strategy
(Rapid Survey) 96
4-9 Effects of Regulations; Cincinnati 305(a) Particulate
Strategy (Rapid Survey) 97
4-10 Effects of Regulations; St. Louis 305(a) S02 Strategy
(Rapid Survey) 98
4-11 Effects of Regulations; St. Louis 305(a) Particulate
Strategy (Rapid Survey) 99
4-12 Cincinnati Device Demand and Cost, 305(a) Strategy
(Rapid Survey) 100
-------
TABLES (Continued)
Page
4-13 St. Louis Device Demand and Cost, 305(a) Strategy
(Rapid Survey) 101
4-14 Emission Inventory Comparison, Cincinnati Major Point
Sources 103
4-15 Emission Inventory Comparison, St. Louis Major Point
Sources 104
4-16 Comparison of Rapid and Detailed Emission Inventories/
Regionwide 106
4-17 Correlation Between Rapid Survey and Detailed Inventory
Air Quality Predictions 108
4-18 Rapid Survey and Detailed Inventory Calibration Runs 114
4-19 Comparison of 305(a) Strategy Impact on Point Sources .... 125
4-20 Comparison of Major Device Demands of 305(a) Strategy .... 126
5-1 Scope of 1971 Economics of Clean Air Report 133
5-2 Solid Waste Disposal 144
5-3 Steam-Electric Power 146
5-4 Other Fuel Combustion Sources 147
5-5 Industrial Processes 148
5-6 Regional Emission Summary by Source Category 178
5-7 Particulate & Sulfur Dioxide Control Cost Summary for
Select Air Quality Control Regions 182
5-8 Control System Demand for the Cincinnati AQCR 183
5-9 Control System Demand for the Washington, D.C. AQCR 184
5-10 Control System Demand for the St. Louis AQCR 185
5-11 Control System Demand for the Philadelphia AQCR 186
B-l Applicable Control Devices 204
C-l Composition of Cincinnati Area Sources 211
C-2 305 (a) Standards for Area Sources 212
-------
TABLES (Continued)
Page
C-3 Area Source Scaling Factors 216
D-l Allowable Rate of Particulate Emission Based on
Process Weight Rate 218
-------
FIGURES
Page
1-1 AGGREGATION APPROACH - Cincinnati Particulate Calibration
Detailed Inventory ....................... 13
1-2 AGGREGATION APPROACH - Cincinnati Particulate Calibration
Rapid Survey Inventory .................... 14
1-3 Impact. of 305 (a) Emission Standards on Air Quality
Standards - Cincinnati .................... 30
1-4 Impact of 305 (a) Emission Standards on Air Quality
Standards - St. Louis , ................... 31
1-5 Impact of 305 (a) Emission Standards on Air Quality
Standards - Washington, D. C ................. 32
3-1 Annual S02 Ground Level Concentrations, 305 (a) Strategy,
Cincinnati AQCR ....................... 52
3-2 Annual Particulate Ground Level Concentrations, 305 (a)
Strategy, Cincinnati AQCR .................. 53
3-3 Annual S02 Ground Level Concentrations, 305 (a) Strategy,
St. Louis AQCR ........................ 54
3-4 Annual Particulate Ground Level Concentrations, 305 (a)
Strategy, St. Louis AQCR ................... 55
3-5 Annual Particulate Ground Level Concentrations, 305(a)/4
Strategy, St. Louis AQCR ........... ........ 56
3-6 Annual S02 Ground Level Concentrations, 305 (a) Strategy,
Washington, D. C. AQCR .................... 57
3-7 Annual Particulate Ground Level Concentrations, 305 (a)
Strategy, Washington, D. C. AQCR ............... 58
3-8 Annual S02 Ground Level Concentrations, 305 (a)/ 2 Strategy,
Washington, D. C. AQCR .................... 59
3-9 Annual Particulate Ground Level Concentrations, 305(a)/2
Strategy, Washington, D. C. AQCR ............... 60
4-1 Relationship between Type of Fuel Burned, Excess Air, and
Resulting Volume of Combustion Products ........... 80
4-2 Scatter Diagram and Regression Line of Pollutant Concentration
Estimates, Rapid Survey versus Detailed Inventory, Cincinnati
-------
FIGURES (Continued)
4-3 Scatter Diagram and Regression Line of Pollutant
Concentration Estimates, Rapid Survey versus Detailed
Inventory, Cincinnati Particulates 110
4-4 Scatter Diagram and Regression Line of Pollutant
Concentration Estimates, Rapid Survey versus Detailed
Inventory, St. Louis S02 Ill
4-5 Scatter Diagram and Regression Line of Pollutant
Concentration Estimates, Rapid Survey versus Detailed
Inventory, St. Louis Particulates 112
4-6 Diffusion Model Calibration Plot Cincinnati S02
(Detailed Inventory) 116
4-7 Diffusion Model Calibration Plot Cincinnati S02
(Rapid Survey, adjusted) 117
4-8 Diffusion Model Calibration Plot Cincinnati Particulate
(Detailed Inventory) 118
4-9 Diffusion Model Calibration Plot Cincinnati Particulate
(Rapid Survey) 119
4-10 Diffusion Model Calibration Plot St. Louis S02
(Detailed Inventory) 120
4-11 Diffusion Model Calibration Plot St. Louis S02
(Rapid Survey) 121
4-12 Diffusion Model Calibration Plot St. Louis Particulate
(Detailed Inventory) 122
4-13 Diffusion Model Calibration Plot St. Louis Particulate
(Rapid Survey) 123
5-1 Procedure for Developing Control Measure Demand Estimates
with the 305 (a) Approach 129
5-2 The Development of Decision-Making Information from the
Economics of Clean Air Control Cost Estimates 131
5-3 The Scope of the Demand Study 136
5-4 Control Cost Estimating Procedure in the Cost of Clean
Air Report 138
5-5 Detailed View of EGA Cost Estimating Procedure 141
A-l Flow Chart 200
-------
FIGURES (Continued)
Page
D-l Maryland Particulate Emission Standards for Fuel
Burning Installations 219
D-2 New York State Particulate Emission Regulation
For Refuse Burning Equipment 219
-------
1.0 EXECUTIVE SUMMARY
This section is intended to provide an overview of the study
objectives and a summary of the conclusions and recommendations. Tech-
nical backup data and detailed descriptions of the study procedures are
presented in Sections 2.0 through 5.0.
1.1 THE IMPLEMENTATION PLANNING PROGRAM
This report describes a study of certain applications of the
Implementation Planning Program (IPP) , undertaken as a part of Phase
III of the Regional Air Pollution Analysis Project. The Implementation
Planning Program is an air resource management planning tool which has
been developed by TRW over a period of two years under contract to the
Air Pollution Control Office (APCO). IPP includes mathematical models
representing the atmospheric diffusion of pollutants, the cost and
effectiveness of pollution control measures, and the potential air
quality impact of air pollution control strategies upon which control
legislation may be based. IPP thus provides an objective means for
analyzing both the extent of a region's air pollution problem and the
adequacy of potential solutions.
IPP is currently being used in the preparation of implementation
plans by a number of States and has been used by TRW in the preparation
of Implementation Plans for Cincinnati and Washington, D.C. In this
report some other applications of IPP-will be evaluated.
1.2 STUDY OBJECTIVES
Now that the Implementation Planning Program is completed, it is of
interest to look at several questions relating to the operational use of
the program. The primary question to be answered by this study is
whether or not a regionally oriented program such as IPP can be used to
answer national-level questions in the air pollution control area. A
second question to be answered has to do with the validity of the assump-
tion made in the Cost of Clean Air Report (submitted by APCO to the Con-
gress) relating emission standards to air quality standards. The diffu-
sion model segment of IPP makes it possible to determine whether or not
-------
quality standards. The third question concerns the input data require-
ments for IPP in an operational environment, i.e., whether or not existing
emission inventory data (collected using a modified Rapid Survey technique;
see Reference 3) is adequate to run the program.
There are a number of national level questions which might be asked
regarding the air pollution control problem. The one selected for analy-
sis here is that of estimating the demand for control measures generated
by the need to meet the requirements of the Clean Air Act, as amended.
Of particular concern is the demand for low sulfur fuels and the relative
trade-off between fuel use and flue-gas desulfurization devices. Since
the Nation's fossil fuel sources are not unlimited, an estimate of the
national demand for fossil fuels, both in terms of magnitude and mix
(fuel types arid grades) is of great importance. Most of the resources of
the study will be devoted to trying to answer this question using IPP.
The second objective is to look at the emission standards which were
used in the Cost of Clean Air Report (Section 305(a) of the Clean Air Act,
as amended) and to determine whether or not the assumption that these
emission standards will achieve air quality standards is valid. Since
the 305(a) report is designed to present a comprehensive study of the
economic impact of the Act, it is essential that the estimated costs
reflect what is actually required to achieve air quality standards.
Because of the tremendous number of regions included in the 305(a) report,
it is not feasible to do an air diffusion modeling study for every region;
consequently, some uniform emission standards must be assumed for applica-
tion throughout the United States. The approach here will be'to look at
the air quality impact of these assumed emission standards in selected
regions where there is sufficient data to do a detailed diffusion modeling
study.
The final objective is to look at the operational data requirements
of an IPP type analysis, to determine whether or not these requirements
may be relaxed and the corresponding data collection problems reduced.
1.3 STUDY ASSUMPTIONS
Detailed emission inventories for Cincinnati, St. Louis and
-------
Cincinnati and Washington, these are the same inventories used in prepar-
ing Implementation Plans. When evaluating the Rapid Survey data (avail-
able only for Cincinnati and St. Louis),the detailed inventories will be
assumed to be correct.
To allow comparison between the three control measure estimation
techniques, the same emission standards will be used in all three
approaches. The implicit assumption is made that these emission stan-
dards are adequate to achieve air quality in each of the regions under
study. The specific emission standards used are presented in Table 1-1.
All control cost outputs will be compared in terms of annualized costs
(operations and maintenance plus annualized capital costs) in 1967-68
dollars. Instantaneous compliance with the Clean Air Act is assumed,
i.e., there is no finite time period allocated for achieving compliance
with the Act. The control costs computed will be those for controlling
particulates and sulfur dioxides as emitted by point sources. The 305(a)
background data will be that to be used in the 1971 Economics of Clean
Air report.
1.4 ESTIMATING NATIONAL LEVEL CONTROL MEASURE DEMAND
The primary question to be answered is whether or not the IPP can
be used to estimate the national level demand for control measure of all
types, especially fuel substitution. The estimation procedure involves
simply running as many regions as possible through IPP and adding up the
total demand generated by complying with the emission standards in each
region. Theoretically this approach is valid; the question is whether or
not the quality of the data which is available at the present time for a
large number of regions is adequate for use in the IPP. Because this
technique is basically a summation over regional results, it has been
labeled the Aggregation Approach.
Two other procedures which have been suggested for estimating nation-
al level control measure demand were also evaluated; each of them requires
considerably less work than the Aggregation Approach described above. One
of these techniques, which has been labeled the Extrapolation Approach,
involves the intensive study of three regions and an evaluation of the
-------
PARTI CULATES
FC: MARYLAND COMBUSTION REGULATION
IP: SAN FRANCISCO PROCESS WEIGHT CURVE
SW: NEW YORK STATE INCINERATOR CURVE
FC: i*46 IBS / io6 (EQUIVALENT TO 1% SULFUR COAL)
IP! EXHAUST GAS CONCENTRATION LIMIT, 500 PPM
Note: These emission standards are defined in
detail in Appendix D.
-------
other technique, labeled the 305(a) Approach (the third of three candidates
considered in this study), relies on the use of existing 305(a) background
cost data and looks at the possibility of tabulating the control measure
demand implied by these cost data.
In order to limit the number of source categories which must be
considered in the initial comparison of the three demand estimation
approaches, it is of interest to look at the three regions to be used in
this study to see if the intuitive feeling that powerplant emissions
dominate the total regional emission picture is, in fact, correct. In
Table 1-2, the steam electric powerplant contribution to each of the three
regions is presented in terms of particulate and SO., emissions in tons per
day. In the left-hand column, the source category breakdown is presented
with the two major fuel combustion source types included, i.e., powerplants
and industrial boilers. The numbers in parentheses indicate the relative
contribution of these particular fuel combustion sources to the total fuel
combustion emissions in the region. This table shows that the powerplant
contribution varies from a low of 42% of the particulate emissions in St.
Louis to a high of 58% of the particulate emissions in Cincinnati. The
powerplant contribution to sulfur dioxide emissions is even greater,
varying from a low of 63% in St. Louis to a high of 91% in Washington.
The total contribution of fuel combustion sources to the regional emis-
sions, (as indicated in the last line of the table) never falls below 70%
and reaches a high of 100% for sulfur dioxide emissions in Washington.
Clearly, fuel combustion sources (especially powerplants) are the dominant
sources of emissions of these two pollutants and should be the first
source category considered in evaluating the three candidate approaches to
control measure demand estimation.
1.4.1 Aggregation Approach
There is no doubt that this approach is theoretically correct, i.e.,
if enough regions are run through IPP with adequate input data, summing
over the regional results will give a good estimate of national control
measure demand (assuming an iteration or two to allow fuel prices to ad-
just to increased demand). Thus, the question is whether or not the
emission inventory data now available is adequate to produce meaningful
-------
REGIONAL EMISSIONS. TONS/DAY
FUEL COMBUSTION
POWERPLANTS
INDUSTRIAL BOILERS
INDUSTRIAL PROCESS
SOLID WASTE
CINCINNATI
PART.
296
(224)
(70)
68
22
so2
962
(876)
.(82)
24
0
ST. LOUIS
PART.
309
(188)
(116)
. 130
7.
so2
1307
(1030)
(263)
325
0
WASHINGTON
PART.
62
(36)
(-)
10
14
so2
542
(492)
(3)
-
-
o\
387 986
447 1632
86
542
POWERPLANT CONTRIBUTION
TO EXISTING EMISSIONS
FUEL COMBUSTION
TO EXISTING EMISSIONS
58% 89%
77% 98%
42%
69%
63%
80%
43%
91%
72% 100%
-------
or Rapid Survey emission inventories have been collected for some 40
regions, and the data is currently on file in APCO's Division of Air
Quality and Emission data. The first task in evaluating this Aggregation
Approach is to examine these Rapid Survey emission inventories to deter-
mine if they would be adequate. Two of the regions for which detailed
inventories are available are also covered by the Rapid Survey inventories:
Cincinnati and St. Louis. The Rapid Survey inventory differs from the
detailed inventory in the number of point sources (10 to 50 per region
rather than several hundred) and the amount of time expended in compiling
the inventory. The detailed inventory requires on the order of one man
year or more to assemble, while the Rapid Survey inventory can be com-
pleted in three to six weeks. The data which exists in AQED files, and
which is to be the subject of this study, was not collected using the
complete Rapid Survey technique as described in the publication by
Ozolins and Smith. It was collected using somewhat less manpower (1 to
2 weeks rather than 3 to 6 weeks), and was intended primarily to serve as
the data base for air quality control region designation work.
On completion of the IPP runs using the Rapid Survey data, the
results were compared with those obtained using the detailed inventory
from the following standpoints:
Accuracy of emission inventory
Air quality prediction
Control device demand prediction
Resources required per region for analysis
In comparing the Rapid Survey and detailed emission inventories, it
is important to keep in mind that less detail is expected, i.e., there
will be fewer point sources and, generally, the major point sources won't
be reported on a stack-by-stack basis. However, to be sure that the same
region is being described, some characteristics must agree with those in
the detailed inventory, i.e., the overall (or "net") descriptions of the
large major sources and the total regional emissions must agree.
In Table 1-3, a comparison of the Rapid Survey and detailed emis-
sion inventories is presented for Cincinnati and St. Louis on a regionwide
basis. As can be seen by examining the column in the center of the table
-------
REGIONAL EMISSIONS (TONS/DAY)
NUMBER OF SOURCES
00
CINCINNATI
PARTICULATES
DETAILED INVENTORY
RAPID SURVEY
S°2
DETAILED INVENTORY
RAPID SURVEY
ST. LOUIS
PARTICULATES
DETAILED INVENTORY
RAPID SURVEY
^2
DETAILED INVENTORY
RAPID SURVEY
POINT
386.7
211.1
986.1
1,054.7
446.8
385.9
1,631.7
1,532.4
AREA
212.6
158.8
369.4
104.3
108.3
92.2
46.2
157.7
FC
296.0
199.2
962.5
1,052.3
309.5
245.4
1,306.7
1,204.0
IP
68.3
5.4
23.5
2.4
130.3
134.4
324.9
328.4
SW
22.4
.6.4
0
0
7.3
5.1
0
0
TOTAL
599.3
369.9
1,355. '5
1,159.0
555.1
478.1
1,677.9
1,690.1
FC
71
44
71
43
47
44
' 46
44
IP
34
8
6
1
60
39
13
8
SW
20
6
0
0
3
2
0
0
TOTAL
125
58
77
44
110
85
59
52
-------
most cases, it is less than 15%. This one exception is for the Cincinnati
region where the particulate emissions from industrial processes are
essentially totally absent and the particulate emissions from area sources
are underestimated by some 60 tons per day. In general, however, the
regionwide figures agree reasonably well.
In Table 1-4 and 1-5, these two regions are presented again, but
individual sources are described in greater detail. In Table 1-4, major
point sources for the Cincinnati region are presented. The specific
sources being considered are listed in the left-hand column. The S0? and
particulate emissions as described by the detailed and Rapid Survey
inventories are listed in the four columns on the right-hand side. This
table shows clearly that there are sizeable errors, especially in the
powerplant emission rates, which are both positive and negative. For ex-
ample, the Rapid Survey inventory includes an overestimate for the Beckjord
S0» emission rate of 150 tons per day. Because of the magnitude of the
emissions from powerplants, errors such as these mean that the level of
accuracy in the remainder of the inventory is of little consequence.
Looking at particulate emissions in this region, essentially the same
situation is found and, in fact, in the case of the Miami powerplant, the
Rapid Survey is in error by a factor of 10.
The situation is much better for St. Louis, at least insofar as the
magnitudes of the emission rates are concerned. There is another diffi-
culty, however, with the Rapid Survey inventories which is described in
Table 1-5. In this table, the locational errors which were found in St.
Louis are listed.
The differences in the X and Y coordinates (third and sixth columns
in the table) should each be constant, i.e., a bias reflecting the differ-
ent coordinate system origins used. The variation about the average X
coordinate bias is about +_ 1 km, assuming the large discrepancy between
Portage Des Sioux and the remaining plants is due to a blunder. The Y
coordinate bias is much closer to being constant. It is difficult to
evaluate exactly what a 2 km error means in terms of ground-level concen-
trations. Since the Diffusion Model must be calibrated using air quality
measurements taken at specific points in the region, and these locational
-------
Power Plants
Tanners Creek
Beckjord
Miami
Reading
Hamilton
Sorg Paper (Boiler)
/
Champion Paper (Boiler)
Proctor and Gamble (Boiler)
Dupont Sulfuric Acid
Philip Carey (Boiler)
General Electric (Boiler)
Interlake Steel (Electric
Arc)
Armco Steel (Boiler)
ACTP.CO Steel (Open Hearth)
SO- Emissions
(Tons/Day)
Detailed
Inventory
961.33
339.30
452.52 .
162.20
1.10
6.21
4.03
15.73
6.08
14.20
2.38
3.64
0
11.77
0
Rapid
Survey
905.23
437.00
308.60
137.58
.50
21.55
3.70
10.40
8.92
-
3.13
3.12
0
82.20
-
Particulate Emissions
(Tons /Day)
Detailed
Inventory
225.28
64.75
29.62
125.99
.09
4.83
12.50
3.37
4.72
0
7.35
9.22
13.00
5.48
32.87
Rapid
Survey
50.24
15.90
20.31
12.07
.07
1.89
12.40
10.50
6.20
-
7.89
1.57
1.50
68.50
-
TABLE 1-4. AGGREGATION APPROACH - EMISSION INVENTORY COMPARISON
CINCINNATI MAJOR POINT SOURCES
-------
PLANT
MERAMEC
ASHLEY
CAHOKIA
VENICE
PORTAGE
DES SIOUX
V
138.00
151.21
150.6
151.5
141.5
XR
77.29
88.94
88.86
89.37
75.43
w
60.71
62.27
61.74
62.13
66.07
*D
192.6
218.6
215.0
222.0
249.5
y
41.94
68.20
64.46
71.31
99.09
Y -Y
D R
150.66
150.40
150.54
150.69
150.41
X_ « detailed inventory X coordinate, kilometers
X_ » rapid survey X coordinate, kilometers
Jx
Xn-Xp « "conversion factor" between two coordinate systems
TABLE 1-5. AGGREGATION APPROACH - LOCATION OF ST. LOUIS POWERPLANTS
-------
measured air quality, a poorer calibration is likely to result. There
were other inaccuracies in the Rapid Survey inventories which affect air
quality, e.g., stack height data. These inaccuracies are described in
greater detail in Section 4.0 of this report.
The air quality implications of these inaccuracies in the Rapid
Survey emission inventory are best revealed through calibration against
measured data. As has been stated, the total regional emissions are
about the same, but there are mislocated sources, and there are errors in
the emission rates (both positive and negative) which could distort the
regional air quality picture. Figures 1-1 and 1-2 display the regression
lines which were obtained in calibrating the model predictions of partic-
ulate ground-level concentrations for the Cincinnati region. Figure 1-1
is the regression line using the detailed inventory; Figure 1-2 is that
using the Rapid Survey inventory. As can be seen, the scatter when using
the Rapid Survey inventory is much greater and the resulting lower regres-
sion coefficient gives much less confidence in the calibrated model re-
sults, i.e., 58% of the data variation is explained by the detailed
2
regression line (R = 0.58) as compared with 8% for the Rapid Survey
2
regression line (R = .08). Although this is the worst scatter obtained
on any of the Rapid Survey runs, none of the regression lines came close
to achieving the 95% confidence level.
The total regional control costs in terms of dollars per year and
in terms of control cost effectiveness in dollars per ton of pollutant
removed are presented in Table 1-6. In the first column, the existing
regional emissions given agree very well, as has been pointed out
previously. In addition, the percentage reduction of emissions tend to
agree between IPP results based on the detailed and Rapid Survey
emissions inventories. The fact that the number of applicable and con-
trolled sources is lower for the Rapid Survey is to be expected because
of the nature of the emission inventory. However, the regional cost
figures should be comparable, and as is shown in this table, they are not.
In fact, there are differences of as much as a factor of 3 between control
costs as predicted by the detailed inventory and those predicted by the
Rapid Survey inventory.
-------
wo
120
D
_rs
00
100
80
o
A.
40
20
75 100 125 150 175 200 225 250
Pollutant Concentration, vg/m^ (Calculated)
Regression Coefficient, R = .762
For 95% Confidence Level, R g5 = .335
-------
_S
DO
c
a
u
C
o
-------
CINCINNATI
PARTICULATES
DETAILED INVENTORY
RAPID SURVEY
so2
DETAILED INVENTORY-
RAPID SURVEY
ST. LOUIS
PARTICULATES
DETAILED INVENTORY
RAPID SURVEY
so2
DETAILED INVENTORY
RAPID SURVEY
EXISTING
REGIONAL
EMISSIONS
(TONS/DAY)
POINT SOURCES
386.7
211.1
986.1
1,054.7
446.8
385.9
1,631.7
1,532.4
CONTROLLED
REGIONAL
EMISSIONS
(TONS /DAY)
POINT SOURCES
42.9
31.1
306.5
353.9
75.6
55.9
361.2
550.8
PERCENT
REDUCTION
OF
EMISSIONS
89
85
69
66
83
86
78
64
REGIONAL
COST
,$/YEAR
3,338,000
2,344,000
3.969,000
7,344,900
12,693,000
4,526,100
20,187,100
22,688,000
NO. SOURCES
APPL
125
58
77
44
110
85
59
52
CON-
TROLLED
83
50
30
9
85
63
46
41
COST
EFFECTIVENESS
$/TON
OF POLLUTANT
REMOVED
26.6
35.7
16.0
28.7
93.7
37.6
43.5
64.71
-------
Because the Rapid Survey data on file at AQED was not originally
intended to support IPP modeling, considerable effort was required to
augment this data to the point where it could be used in the program.
Approximately 1.5 man months was required for the data preparation for
each of these two regions. In addition, some data which is essential
for IPP was either not collected, or, because of the intended use of this
data, was not recorded. Examples of specific data types that were diffi-
cult to recover are the following:
A. Fuel data: sulfur and ash content are both required,
as well as the price for each fuel grade.
B. Plant Stack Data: including physical stack height
and stack exit conditions.
C.. Existing Control Equipment; particularly the
efficiency for both S0_ and particulates.
1.4.2 Extrapolation Approach
This procedure is based on the intensive study of a few regions to
derive normalized control cost figures, e.g., dollars/kilowatt for power-
plants, which could then be applied to national level output by industrial
sector (e.g., installed capacity for powerplants). An evaluation of this
procedure consisting of the following steps was conducted:
A. Tabulate control measures and costs for each region.
The tabulation can be in terms of control costs per
ton of pollutant removed. This tabulation is then
-examined for region-to-region and intra-regional
differences and an attempt is made to isolate the
causes for these differences.
B. Look at the components in the cost equation, i.e.,
the inputs that are needed to run the cost model,
and determine how they vary from region to region,
e.g., fuel costs vary considerably, perhaps a
factor of two or three for a given fuel type across
the regions. Other examples of cost input variations
are device manufacturer's prices which don't change
appreciably from region to region, and labor rates
which can change by as much as 50%.
-------
C. Determine the sensitivity of the computed control
costs to variations in these input parameters.
Data from the IPP computer runs based on the detailed emission inventory
(prepared to serve as the standard of comparison in evaluating the aggre-
gation approach) were used, i.e., no special runs were required to evaluate
the extrapolation approach.
Although the primary concern is with the estimation of control
measure demand, control costs must be considered because of the strong
interrelationship between demand and cost introduced by the IPP least-
cost-alternative approach to control measure selection.
Control costs in terms of dollars per ton of pollutant removed
for Cincinnati, St. Louis and Washington are presented in Table 1-7 for
representative sources in each major source category. The fuel combus-
tion category is represented by powerplants and industrial boilers,
integrated iron and steel and petroleum refineries are included as
examples of industrial process sources, and finally, municipal incinera-
tors are included as representatives of solid waste disposal sources. In
looking at the first row in this table, it is important to keep in mind
that only electric powerplants have a uniform representation in all
regions, i.e., all are relatively large regions, with total installed
generating capacities on the order of 3,000 megawatts. Reading across
the top row in this table, note that the costs for controlling particulates
can vary by a factor of 8. Apparently, steam electric powerplants are a
likely candidate for the first look at the details of the control cost
calculation.
The IPP Cost Model output shows that most of the fuel combustion
control measures consist of fuel substitutions, either coal-to-oil or
coal-to-coal switches. Because of the number of fuel substitutions, an
important variable in the calculation of the regional control is the cost
of the substitute fuel. The variation in fuel prices (in terms of
dollars per 10 BTU) found across the three regions studied here is pre-
sented in Table 1-8. The prices included in this table are for fuels
which meet the emission standard, i.e., 1% sulfur coal or 1.36% sulfur
oil. As can be seen from this table, for a given fuel type, the price
can vary by a factor of 2 from region to region; in addition, within a
-------
CINCINNATI
PART. SO,
ST. LOUIS
PART. SO,
WASHINGTON
PART. SO
FUEL COMBUSTION
STEAM ELECTRIC
INDUSTRIAL BOILERS
21 16
37 5
34
51
46
82
164
18
INDUSTRIAL PROCESS
00
IRON & STEEL
PETROLEUM REFINERIES
32
191
293
SOLID WASTE
36
INCINERATORS
30
32
-------
LOW SULFUR FUEL PRICES - $/T(TBTU -
COAL
OIL
CINCINNATI
.24
.45
ST. LOUIS
.46
.49
WASHINGTON
.29
.50
-------
given region, a factor of 2 variation is possible in going from one fuel
type (coal) to another (oil). The fuel prices tabulated here were ini-
tially derived from Fuels for 50 Cities and were updated using, in the
case of Cincinnati and Washington, information from local control offi-
cials and APCO. It is important to keep in mind that these represent
prices which obtain at the current level of demand, and that they could
be considerably in error in the presence of any widespread switching to
low sulfur fuels. Reliance on static fuel prices is, incidently, a
shortcoming which is common to all three of the techniques which are
being studied here.
Given these variations in the prices for low sulfur fuels, the next
step is to determine the sensitivity of the final computed control costs
to these changes. For a typical 1,000 megawatt powerplant (e.g., Tanner's
Creek, Cincinnati) a 10% change in the price of a substitute fuel results
in a $930,000 change in the control cost for this one plant (an increase
of 50%). This very large multiplier effect occurs because of the impact
of incremental fuel costs, i.e., the additional cost of replacing the
original fuel with a substitute fuel of lower sulfur content. For a
region like Cincinnati, with close to 3,000 megawatts of installed capa-
city, a 10% fuel price change would then imply a 2 million dollar change
in predicted control costs. Because the nearly constant costs associated
with wet limestone injection, relative price changes of this magnitude
can cause a significant shift in the control measure demand. Thus, this
very high sensitivity when coupled with the variation in fuel prices dis-
played in Table 1-7, indicates that powerplants must be treated individu-
ally, i.e., region-to-region and intra-regional variation in fuel prices
must be considered.
1.4.3 305(a) Approach
In coming up with the national level control cost estimates presented
in the 305(a) report, background data was collected which should allow the
tabulation of the corresponding control measure demand, i.e., the types of
control measures assumed for each industry category are known and the costs
incurred are known; consequently, it should be possible to compute the num-
ber of devices or the quantity of low sulfur fuel which would be required.
A procedure for this "back-calculation" was developed and applied to the
-------
three regions for which the detailed emission inventories are available.
The control measure demand and cost estimates were then evaluated by com-
paring the 305(a) results with those obtained using detailed emission
inventories. The comparison was made on the basis of the predictions of
control measure demand and cost and the structure of the cost equations.
Because the importance of major fuel combustion sources has already
been noted (Section 1.3 above), these sources were emphasized in this
comparison. A detailed comparison of costs for controlling powerplants in
Cincinnati is presented in Table 1-9. According to the IPP results (first
row of the table) 3.9 million dollars are required for powerplant control.
Note that the IPP program is set up to select the least-cost alternative
available to each source subject to control. In the second line of this
table, the 1970 305(a) report prediction that 31 million dollars would be
required to control Cincinnati powerplants is listed; here it is assumed
that all powerplants will switch to low sulfur oil. The last line shows
that the 1971 305(a) report forecasts a 8.4 million dollar control cost,
with all powerplants going to wet limestone injection. All of these
control cost projections are based on essentially the same fuel price
information; consequently, there must be another reason for the tremendous
difference. In Table 1-10 the most likely cause of the difference, i.e.,
the assumptions made in selecting control measures" is evaluated by using
the IPP Cost Model and applying the 305(a) 1970 and 1971 assumptions, i.e.,
switch to low sulfur oil and installation of wet limestone injection,
respectively. As shown in this table, IPP predicts a cost of 35.7 million
for the switch to low sulfur oil. This is compared with an estimate from
the 305(a) report of 31 million. For the 1971 assumption IPP predicts a
switch-to-wet-limestone control cost of 9.6 million dollars, as compared
with 8.4 million in the 305(a) report, indicating the importance of the
assumption that is made regarding control measure selection. The region-
to-region variation in fuel price has been shown to be important and is
already being considered in the 305(a) approach, at least for industrial
boilers. For powerplants, however, the availability of lower cost fuel
substitution control measures is not being considered, i.e., every power-
plant is assumed to switch to wet limestone injection, even though it
might be possible to use much cheaper low sulfur coal. The 3.9 million
projection by IPP (Table 1-9) consisted primarily of switches to low
sulfur coal. For this particular region, low sulfur coal is not only
-------
POWERPLANTS - CINCINNATI
SOURCE OF PREDICTION
IPP
305(A) REPORT - 1970
305(A) REPORT - 1971
CONTROL MEASURE
LEAST COST TO
INDIVIDUAL SOURCE
SWITCH TO LOW SULFUR OIL
WET LIMESTONE
-POWERPLANT CONTROL COST (IP6 $)
3.9
31.0
8.4
-------
APPLY 305(A) RULES FOR CONTROL MEASURE SELECTION TO IPP COST MODEL RESULTS
SOURCE OF PREDICTION CONTROL MEASURE POWERPLANT CONTROL COST (TO6 $)
IPP
305(A) 1970
SWITCH TO LOW SULFUR OIL
SWITCH TO LOW SULFUR OIL
35.7
31.0
IPP
305(A) 1971
WET LIMESTONE
WET LIMESTONE
9.6
8.4
-------
much cheaper than residual oil, it is probably the only fuel that will be
available. Residual fuel oil would have to be imported on the Ohio River,
generating an additional increment in barge traffic which would push the
total traffic level beyond that which the river can accommodate.
Industrial boilers are another significant source of particulate
and SO emissions. For the Cincinnati region, this category of fuel com-
X
bustion source was examined to determine if the same phenomenon as has
been found in the powerplant estimating procedure existed. In Table 1-11,
the first line shows the IPP prediction of $800,000 for control of indus-
trial boilers in the Cincinnati region. Here, the least cost alternative
for each boiler, for which the original fuel sulfur content is greater
than 1 percent, is selected. The second line gives the 305(a) report
estimate of 8 million dollars with all sources switching to low sulfur
fuel. Again, if the 305(a) control measure assumption is applied to the
IPP Cost Model results, the resulting IPP estimate of 8.5 million dollars
is very close to that in the 305(a) report. The problem inherent in
allocation of national fuel use figures to individual regions on the
basis of population (as is done in the 305(a) report) shows up in this
table as well. Most of the large industrial boilers in Cincinnati already
burn 1 percent sulfur fuel and would not, in fact, have to switch fuels.
1.4.4 Conclusions and Recommendations
The Aggregation Approach to estimating national level control
measure demand, which was known at the outset to be theoretically suitable,
is not recommended at present because of inadequacies in the existing
Rapid Survey data. This data was collected to serve as input to a diffu-
sion model in the designation of air quality control regions, and does
not give adequate accuracy even when augmented to meet the more compre-
hensive input requirements of IPP.
The Extrapolation Approach looks unattractive because it can't
account for region-to-region variations, primarily those in fuel prices.
The 305(a) Approach includes variations in fuel prices, but since
all combustion sources are assumed to select only one control measure
(local least-cost alternatives are not available), it does not give
sufficient resolution of fossil fuel demand. The assumption that all
powerplants are going to use wet limestone injection may give an indica-
tion of what control costs are going to be, but it is not adequate when
-------
to
Ul
INDUSTRIAL BOILERS - CINCINNATI
SOURCE OF PREDICTION CONTROL MEASURE
IPP
305(A) REPORT,1971
APPLY 305(A) RULES:
SOURCE OF PREDICTION
IPP
305(A) REPORT,1971
IPP
LEAST COST TO
INDIVIDUAL SOURCE
ALL SWITCH TO LOW
SULFUR OIL
CONTROL MEASURE
ALL SWITCH TO LOW
SULFUR OIL
ALL SWITCH TO LOW
SULFUR OIL
THOSE >U S SWITCH
TO OIL
INDUSTRIAL BOILER CONTROL COST (TO6 $)
0.8
7.9
INDUSTRIAL BOILER CONTROL COST (IP6 $)
8,5
7.9
1.9
-------
trying to estimate fuel demand, since many powerplants will surely select
low sulfur coal or residual fuel oil, if either of these options is avail-
able and is cheaper than flue gas desulfurization.
The recommended approach for estimating control measure demand,
given the data base which is currently available, consists of combining a
limited IFF modeling effort on powerplants and major industrial boilers,
with data generated using the 305(a) Approach for all other source
categories.
It is recommended that the IPP Control-Cost and Strategy Models be
run on the individual powerplants using the 305(a) emission standards and
local fuel prices, so that the least-cost alternative will be selected at
each site. The IPP Atmospheric Transport Model would not be used (because
of lack of data), so that it would have to be assumed that the emission
standards used would achieve the air quality standards, except for some
major AQCR's where the actual local emission standards could be used (if
they were from an approved Implementation Plan).
For powerplants, various fuel switches should be considered, along
with wet limestone injection, as control measure alternatives. More
detail will be needed on fuel prices, and some technique for accounting
for the effects of increased fuel demand on prices will be required.
The two control cost estimates in the Cincinnati powerplant example used
above probably bound the variation in control cost due to demand generated
changes in fuel prices. The cost associated with many switches to low
sulfur coal is a lower bound, since static fuel prices are used which would
be unrealistic in the event of widespread switching to low sulfur fuel on
the national level. The 305(a) estimate (assumes most select wet limestone
injection) provides an upper bound, since if fuel prices exceeded this level
there would be switching to the wet limestone devices. In the fuels analy-
sis, it is also important to include some consideration of the availability
of fuels, especially low sulfur coal.
The Federal Power Commission has collected detailed information on
every powerplant in the nation (Form 67), which could be used to update
and augment the existing emission inventory.
-------
It won't be possible to determine the level of control required on
powerplants in each individual region, since this would require diffusion
modeling of every region in sufficient detail that the relative impact,
and, consequently, the degree of control required, could be identified
It has already been pointed out that the existing emission inventory data
is inadequate for this task.
The improvement which would give the next best return on resources
invested would be a detailed emission inventory on major industrial
boilers. These boilers could then be analyzed in the same manner as
described above for powerplants.
Control measure demand predictions for the remaining industrial
process and solid waste disposal sources could be derived using the
305 (a) Approach.
1.5 305 (a) EMISSION STANDARDS EVALUATION
There is an important assumption implicit in all three of the
control demand estimation techniques which must be evaluated, i.e., that
the emission standards that have been imposed (Table 1-1) will be adequate
to meet local air quality standards. In Table 1-12, for each of the three
regions, an indication is given where the particulate and SO,, air quality
standards have been met. A standard of 65 micrograms per cubic meter was
selected for both pollutants for this evaluation. This is probably a good
figure for particulates, but may tend to be somewhat higher than the air
quality standards being adopted for sulfur dioxide.
Of the total of six pollutant/region combinations studied, air
quality equal to the standard was achieved in only three, where meeting
the standard was interpreted to mean that no point in the region may ex-
ceed the standard. For each pollutant/region combination in this table,
the margin by which the standard is met is indicated in parentheses
3
(in yg/m ); a negative sign means the standard was exceeded.
In order to determine the incremental cost incurred in achieving
the air quality standards, additional runs were made for those regions
where air quality was not achieved, i.e., St. Louis and Washington. The
emission standards described in Table 1-1 were applied at increasing
levels of stringency until the air quality standards were met. Increased
-------
10
00
Cincinnati
St. Louis
Washington
Particulates*
YES (5)**
NO (-50)
NO (-15)
so2*
YES (<1)
YES (5)
NO (-35)
* - Air Quality Standard of 65 pg/m Assumed
3
** - Margin by which Standard is Met, yg/m (negative sign means standard exceeded)
-------
stringency was achieved by dividing the 305(a) allowable emissions by 2,
i.e., predictions of air quality and control costs were made using
305(a)/2, 305(a)/4, etc. In St. Louis, meeting air quality standards
required an increment in costs of 50% over the 305(a) control costs,
corresponding to allowable emissions which were scaled to 305(a)/4.
Washington, B.C. where neither air quality standard was achieved, required
1000% and 180% cost increments to meet the standards for particulates and
SO,,, respectively; allowable emission rates for both pollutants were
scaled to 305(a)/2. The very large increment in control costs for
Washington particulates is due to the fact that the unsealed 305(a)
emission standards had virtually no effect, and, consequently, a very
low control cost. Ignoring this one data point, it appears that meeting
air quality standards, rather than simply complying with uniformly applied
emission standards, can have a significant effect on predicted control
costs (50-200%).
To further evaluate the impact of the 305(a) emission standards on
air quality, Figures 1-3, 4 and 5 were prepared. Increasing levels of
air quality standards are plotted on the horizontal axis in these figures;
the number of receptors which exceed the air quality standard is plotted
on the vertical axis. This allows evaluation of air quality standards
3
other than the 65 Mg/m level used in this study. Although the receptor
grid spacing is not uniform, the number of receptors which exceed a
standard is roughly proportional to the area over which the standard is
exceeded.
1.6 USE OF RAPID SURVEY AS IPP INPUT
Although the existing Rapid Survey data is inadequate to support the
Aggregation Approach, some modifications to the way in which the technique
is applied should make the data useful for IPP modeling.
For many regions, 6 to 12 large sources emit 75% or more of the
regional totals for particulates and SO . This group, which will usually
include powerplants, integrated iron and steel plants, and perhaps a few
others, will prove to be substantially more important, in terms of total
emissions, than the many small fuel combustion, industrial process, and
solid waste disposal sources which make up the remainder of the point
source emission inventory. An illustration of the importance of these
-------
100 '
90 .
CO
c
2 80
00
a
H
T3
0)
0)
O
X
w
05
70 '
60 -
o
4-1
« 50
a
O
)-<
1)
30
20
10
CINCINNATI AQCR
318 RECEPTORS
Particulates
65
Air Quality Standard, pg/m"
FIGURE 1-3. IMPACT OF 305(a) EMISSION STANDARDS
ON AIR QUALITY STANDARDS - CINCINNATI
-------
100 -I
Tl 90 H
rt
13
c
cfl
CO
c
H
T3
OJ
0)
a
o
.u
P.
a)
o
01
80 H
70 -I
60 -I
50
40
ST. LOUIS AQCR
509 RECEPTORS
Particulate results
were incorrect due
to source file error.
30
20
10
45
50
i
55
\
60
65
Air Quality Standard, yg/m"
FIGURE 1-4. IMPACT OF 305(a) EMISSION STANDARDS
ON AIR QUALITY STANDARDS - ST. LOUIS
-------
240 '
220
200 '
180
160 -
140 -
120 -
100 -
80
40
20
\
\
so.
WASHINGTON, D.C. AQCR
261 RECEPTORS
Particulates
\
45
50
55
60
65
Air Quality Standard, vig/nT
FIGURE 1-5. IMPACT OF 305(a) EMISSION STANDARDS ON
AIR QUALITY STANDARDS - WASHINGTON, D.C.
-------
major point sources in the Ohio portion of the Metropolitan Cincinnati
Interstate Air Quality Control Region is presented in Table 1-13. An in-
spection of this table reveals that a total of 7 sources - 2 powerplants,
2 paper companies, a steel plant, a chemical plant, and a manufacturing
plant - provide 85% of the total S0_ emissions and 68% of the total
particulate emissions in the area. A small additional effort at in-
creasing the accuracy and level of detail for these few sources will
have a substantial effect on improving the accuracy of IPP results
derived from a Rapid Survey emissions inventory.
Because the existing Rapid Survey data was collected for another
purpose, values for many important variables were either not collected,
or were not recorded because they were only intermediate quantities in
terms of the original purpose for collecting the data. For an IPP data
base, however, documentation of sources of information, assumptions,
and methods of calculation is very important.
Finally, a handbook which would give values for exit conditions,
stack heights, process rates, etc., which reflect industry practice, would
allow any gaps in the emission inventory to be filled. If this data, used
on smaller sources, is combined with the more intensive investigation of.
the large sources mentioned above, an emission inventory adequate for IPP
modeling should result.
-------
TABLE 1-13. MAJOR POINT SOURCES IN
OF MCIAQCR (Reference
THE OHIO PORTION
8)
OHIO
Total Emissions,
All Sources
Andersonville Dump
Armco Steel Corp.
Center Hill
Incinerator
Champion Paper Co.
Sulfur Oxides
(tons/day)
670.55
0.17
11.93
0.58
15.73
Cincinnati G&E Co.
(Beckjord Station) 368.59
Cincinnati G&E Co.
(Miami St. Station) 162.20
Colerain Dump
Container Corp. of
America
Crystal Tissue Co.
0.17
1.90
1.56
Diamond National Corp.
(Lockland Plant) 5.42
Diamond National Corp.
(Middle town Plant) 4.42
Dunbar Incinerator
0.14
E.I.DuPont de Nemours
Co., Inc. 14.20
Emery Industries Inc. 3.50
Ford Motor Co.
(Sharonville Plant) 1.30
Formica Corp.
Fox Paper Co.
3.67
1.28
General Electric Co. 3.64
Gulf Oil Corp.
Hamilton Municipal
Power Plant
Hilton Davis. Chem.
International Chern
2.90
6.21
Co. 2.53
. Co. 2.59
% of Total
100.00
.03
1.78
.09
2.35
54.97
24.19
.03
.28
.23
.81
.07
.02
2.12
.52
.19
.55
.19
.54
.43
.93
.38
.39
Particulates
(tons/day)
333.85
6.85
50.62
1.62
3.37
28.12
125.99
6.85
1.50
2.63
1.21
1.64
1.24
2.80
3.30
1.43
9.22
0.24
4.99
0.11
P»^
% of Total
100.00
2.05
15.16
.49
1.01
8.42
37.74
2.05
.45
.79
.36
.49
.37
.84
.99
.43
2.76
.07
1.49
.03
-------
TABLE 1-13. MAJOR POINT SOURCES IN THE OHIO PORTION
OF MCIAQCR (Reference 8) (Continued)
National Lead Co.
of Ohio
Philip Carey Corp.
Procter and Gamble
Company
Reading Power Plant
Sorg Paper Co.
Stearns and Foster Co
West Fork Incinerator
Sulfur Oxides
(tons/day)
1.56
2.38
6.08
t 1.10
4.03
Co. 1.87
tor 0.30
% of Total
.23
.35
.91
.16
.60
.28
.04
Particulates
(tons/day)
1.05
7.35
5.22
0.09
12.50
1.11
1.84
% of Total
.31
2.20
1.56
.03
3.74
.33
.55
-------
2.0 INTRODUCTION
The Implementation Planning Program is now completed and has been
installed on the APCO computer in Durham, North Carolina. The primary
purpose in developing this program was to provide a tool which could be
used in solving the regional air resource management problem. IPP has
been used extensively for this purpose by States, local control agencies,
universities and private contractors throughout the country. The purpose
of this study is to determine if IPP can be used to answer some more-or-
less interrelated questions which are not a part of the air resource
management problem. Specifically, the study had the following three
objectives:
A. To evaluate three approaches to the estimation
of the national demand for air pollution control
measures generated in response to the Air Quality
Act.
B. To investigate the air quality implications of
the enforcement of those emission standards
outlined in the 305(a) Cost of Clean Air Report
to Congress.
C. To determine the feasibility of upgrading
currently available Rapid Survey emission
inventories so that they may be utilized in
the Implementation Planning Program computer
model.
The three demand estimating approaches evaluated under objective A are:
1) Extrapolation Approach. The extensive study
of control measure demand in a small number
of regions and extrapolation results to the
national level.
2) Aggregation Approach. The generation of
control measure demand in a large number of
regions (representing a large proportion of
the total nation) and aggregation of the
results to the national level.
-------
3) 305(a) Approach. The utilization of existing
Economics of Clean Air (305 (a)) background cost
data to derive national control measure demands.
Most of the resources of the study were committed to objective A and more
specifically to the evaluation of the Aggregation Approach; that is, the
primary objective of the study is to determine whether or not running all
the regions now covered by emission inventory data in the Division of Air
Qualtiy and Emissions Data (AQED) files through the IPP is worthwhile.
Thus, the main activity in the study was basically a data evaluation exer-
cise, since it was known at the outset that (theoretically) the Aggrega-
tion Approach should work. By properly structuring the work required to
answer this question, the questions associated with objectives A 1), B and
C can also be answered with very little additional effort. For example,
it is necessary to run a regional analysis using the best available emis-
sions inventory data to provide a standard of comparison for the evalua-
tion of the Rapid Survey data. By using the emission standards outlined
in the 305(a) report in running these regional analyses, it is possible
to answer the question regarding the air quality impact of the 305(a)
emission standards at the same time. Similarily, the feasibility of using
the Rapid Survey technique to provide a data base for the Implementation
Planning Program can be examined as part of the process of evaluating the
existing Rapid Survey data for use in IPP.
Objectives B and C are not addressed specifically in separate
sections in this report. Rather, they are discussed at the point at
which they naturally occurred in the course of the analysis. The air
quality question (objective B) is answered in Section 3.0 where the des-
cription of the detailed inventories and the IPP results based on these
inventories is presented. The feasibility of using the Rapid Survey
technique, which is a logical outgrowth of the work done in upgrading the
existing Rapid Survey data to run IPP, is discussed in Section 4. A brief
description of the content of each section is presented below.
In Section 3.0, Extrapolation Approach, the discussion of the de-
tailed inventory results is presented. The original intent in running
these regional analyses was to cover four regions: Cincinnati, Philadelphia,
St. Louis and Washington. Problems with computer tapes generated in running
-------
the Philadelphia model, however, made future processing for this region
impossible. Consequently, this analysis proceeded with just the three
regions. In this section the 305(a) strategy, which is simply the collec-
tion of emission standards outlined in the 305 (a) report, is introduced
and its impact in terms of applicable sources, controlled sources, cost,
cost-effectiveness and device demand is presented. The discussion of the
ability of the 305(a) emission standards to achieve air quality standards
is also included in this section. In addition, the effects of more strin-
gent levels of 305(a) emission standards where these are required to meet
air quality standards are presented.
In Section 4, Aggregation Approach, the discussion of the Rapid
Survey results is presented. Of the three regions which remained in the
detailed inventory regional analyses group, only two (Cincinnati and St.
Louis) were covered by the Rapid Survey data in the AQED files, so at this
point the study sample was reduced to two regions. The description of the
Rapid Survey technique, the development of the IPP data file by augmenting
the Rapid Survey data, and the discussion of errors in the Rapid Survey
data is presented in this Section. The impact of the 305(a) strategy is
again described; in this case, as predicted by running IPP using the Rapid
Survey data base. Considerable detailed discussion is presented in com-
paring the results of the control group (i.e., the detailed inventory runs)
and the Rapid Survey results. This discussion is presented in terms of a
direct emission inventory comparison, an air quality estimate comparison
and a direct comparison of the 305 (a) strategy impact.
The 305(a) Approach is described in Section 5. A description of the
non-modeling procedure is presented, and its relationship to the other two
approaches (which both involve modeling) is described. The basic data
used in the 305(a) Approach is presented for all four regions in this sec-
tion, along with the prediction of control measure demand implied by this
cost data.
More data has been presented in this report than was used in the
evaluation of control measure estimation alternatives. Conclusions have
been drawn and recommendations have been made (see Section 1.0) largely
on the basis of the results when each of these three candidate approaches
is used on large fuel combustion sources. However, all of the technical
-------
backup data has been included in this report, since it can provide a data
base for additional analyses.
-------
3.0 EXTRAPOLATION APPROACH
The Extrapolation Approach to the prediction of national pollution
control costs and control measure demand involves a detailed analysis,
on an industry-by-industry basis, of the effects of a particular abate-
ment strategy on a small number of regions. Extrapolation to national
level predictions is achieved by scaling with the ratio of such variables
as total emissions, total production, kilowatt-hours of power generated
(in the case of steam/electric plants), and others. An overly simplified
example of this type of extrapolation is the following:
Total National Cost of Control for Steam/Electric Plants
~ m <- in <- n- *.-\ * National KW-Hrs.
= (.Control Cost in Cincinnati) * : :trT, TT
Cincinnati KW-Hrs.
Since linearity is not likely to hold in these extrapolations, a
minimum of three points is necessary to produce an accurate estimate. In
fact, the probability of fluctuations in control costs due to factors
difficult to account for in this type of analysisfor instance, the
presence of one very large plant with special control problems causing
industry costs to be high in relation to the value of the assigned vari-
abledemands that additional points be added as they become available,
i.e., as additional regions are run on IPP. An additional problem with
the use of only these three regions is that many industry groups are
either not represented or else are represented by only one or two sources,
which is certainly an insufficient number for extrapolation purposes.
It should be noted that the above difficulties with this approach
were well known before modeling was begun. However, this technique has
been suggested on a number of occasions, and an investigation into the
magnitude of these problems was deemed to be useful. No additional model
runs were necessary to implement this analysis, however, since the de-
tailed runs were necessary to provide a comparison with the Rapid Survey
runs described in Section 4, and also to provide an estimate of the
adequacy of the 305(a) regulations with respect to achieving air quality
standards.
Three detailed inventoriesthose of Washington, B.C., Cincinnati,
and St. Louiswere available for the study. As described in Section 2,
-------
these regions were run through the IPP model using the Cost of Clean Air
o
305(a) emission standards; where compliance with a 65 ug/m air quality
standard was not achieved, the standards were made uniformly more strict
(by lowering the allowable emissions of the three standards by some factor)
Sections 3.1 through 3.3 describe the results of these strategy runs.
3.1 RESULTS OF POLLUTANT CONTROL STRATEGIES - INDUSTRY COSTS, EMISSIONS,
AND AIR QUALITY
Tables 3-1 through 3-9 describe the effects of the various emission
control strategies applied to the three AQCR's investigated in this study.
Figures 3-1 through 3-9 are plots of "isopleths," or lines of constant
air quality for each of the nine strategies. These plots indicate the
extent to which each of the strategies satisfy the specified air quality
3
goal of 65 pg/m for SO- and particulates.
The 305(a) strategy for SO- and particulates is defined in Section 2
of this report. The designation "305(a)/2" defines a strategy where the
allowable emissions are one-half those specified in Section 2. During the
course of this study, additional strategies based on the 305(a) standards
were investigated, but the results would not significantly add to those
presented here and are therefore not discussed.
3.1.1 Cincinnati AQCR
The yearly cost* of implementing the 305 (a) emission strategy in the
Cincinnati AQCR is $3,969,000 for S02 and $3,338,000 for particulate con-
trol. By far the greatest component of this cost, especially in the case
of S09, is the combustion of fuels.
Steam/electric powerplants require the major proportion of these
funds and, in addition, account for 89 percent of the SO- and 58 percent
of the particulate point source emissions in the region. After controls
have been installed, these percentages drop to 79 and 48 percent of the
new (controlled) total, respectively, mainly because:
The powerplants were using lower grade fuel than
the other combustion sources in the region.
*As discussed in Appendix B, these costs exclude agency and small source
costs.
-------
TABLE
EFFECTS OF REGULATIONS; CINCINNATI 305(a)
S02 STRATEGY(DETAILED INVENTORY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/Institutiona]
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
17
12
42
0
2
0
0
3
1
0
0
0
0
0
71
6
0
77
TOTAL
CONTROLLED
SOURCES
13
1
13
0
0
0
0
3
0
0
0
0
0
0
27
3
0
30
COST
1000 's OF
$ PER YR.
3,794
-3
52
0
0
0
0
126
0
0
0
0
0
0
3,843
126
0
3,969
EXISTING
EMISSIONS
TONS /DAY
876.4
4.6
81.5
0
.2
0
0
20.
2.9
0
0
0
0
0
962.5
23.5
0
986.0
ALLOWABLE
EMISSIONS
TONS /DAY
303.6
5.1
145.7
0
2.0
0
0
1.5
1.5
0
0
0
0
0
454.4
5.0
0
459.
CONTROLLED
EMISSIONS
TONS /DAY
242.8
4.6
52.0
0
.2
0
0
4.1
2.9
0
0
0
0
0
299.4
7.
0
306.5
COST
EFFECTIVENESS
$ PER TON
REMOVED
16.4
-133.
4.
-
-
-
_
21.1
-
-
-
-
-
-
15.9
21.1
_
-------
TABLE 3-2
EFFECTS OF REGULATIONS; CINCINNATI 305(a) PARTICIPATE STRATEGY(DETAILED INVENTORY)
SOURCE CATEGORY
o Steam/Electric
o Conunercial/InstitutionaJ
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
0 Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
17
12
42
0
12
6
3
1
1
4
4
0
2
1
71
34
20
125
TOTAL
CONTROLLED
SOURCES
14
10
38
0
8
0
0
1
0
4
0
0
0
0
62
13
8
83
COST
1000' s OF
$ PER YR.
1,577
37
795
0
671
0
0
15
0
13
0
0
0
0
2,409
699
231
3,338
EXISTING
EMISSIONS
TONS /DAY
223.9
2. 1
70.0
0
63.3
2.0
.4
.2
.2
1.2
.8
0
.5
.3
296.0
68.3
22.4
386.7
ALLOWABLE
EMISSIONS
TONS /DAY
37.0
1.3
16.4
0
6.1
.3
.1
.1
.7
.1
1.1
0
.3
.1
54.8
9.0
4.6
64.3
CONTROLLED
EMISSIONS
TONS /DAY
20.4
1.0
10.7
0
5.7
2.0
.4
.0
.2
.1
.8
0
.5
.3
32.1
9.4
1.4
42.9
COST
EFFECTIVENESS
$ PER TON
REMOVED
21.2
100.2
36.7
-
31.9
-
212.0
_
33.4
_
_
_
25.0
32.5
30.2
-------
TABLE 3-3
EFFECTS OF REGULATIONS; ST. LOUIS 305(a) S02 STRATEGY (DETAILED INVENTORY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/Institutiona]
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
0 Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
10
4
32
0
1
0
1
6
4
0
0
0
0
1
46
13
0
59
TOTAL
CONTROLLED
SOURCES
10
4
27
0
0
0
1
4
0
0
0
0
0
0
41
5
0
46
COST
1000 's OF
$ PER YR.
13,717
315
5,886
0
0
0
*
270
0
0
0
0
0
0
19,917
270
0
20,187
EXISTING
EMISSIONS
TONS /DAY
1,030.4
13.6
262.8
0
17.3
0
232.0
35.7
39.8
0
0
0
0
.1
1,306.8
324.9
0
1 ,631 -7
ALLOWABLE
EMISSIONS
TONS /DAY
346
3.7
126.1
0
2.7
0
10.5
80.4
85.1
0
0
0
0
19.7
475.8
198.3
n
674 5
CONTROLLED
EMISSIONS
TONS /DAY
223.7
0
66.6
0
17.3
0
5.8
8.0
39.8
0
0
0
0
.1
290.3
70.9
o
361 2
COST
EFFECTIVENESS
$ PER TON
REMOVED
45.6
63.4
82.2
-
-
-
*
26.6
-
-
-
-
-
-
53.7
2.9 *
43.5
-------
TABLE 3-4
EFFECTS OF REGULATIONS; ST. LOUIS 305(a) PARTICULATE STRATEGY (DETAILED INVENTORY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/Institutional
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
0 Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
10
4
33
0
7
4
6
5
5
0
4
15
0
14
47
60
3
110
TOTAL
CONTROLLED
SOURCES
8
4
28
0
4
4
3
5
4
0
2
15
0
6
40
43
2
85
COST
1000's OF
$ PER YR.
2,118
124
1,976
0
555
27
520
293
946
0
153
4,469
0
1,428
4,218
8,390
86
12,693
EXISTING
EMISSIONS
TONS /DAY
188.0.
5.2
116.2
0
11.9
2.6
2.2
5.1
10.1
0
10.0
50.4
0
10.1
309.4
102.5
7.3
419.1
ALLOWABLE
EMISSIONS
TONS /DAY
35.2
.8
21.1
0
3.9
.3
1.0
1.5
3.9
0
1.6
2.4
0
3.9
57.1
18.5
2.1
77.7
CONTROLLED
EMISSIONS
TONS /DAY
18.5
.4
9.9
0
3.9
0
.6
.6
1.3
0
5.7
.6
0
5.5
28.8
18.2
.8
48.4
COST
EFFECTIVENESS
$ PER TON
REMOVED
34.2
71.2
50.7
-
190.7
29.2
1,381.8
178.1
292.5
-
96.6
245.7
-
837.7
41.2
274.5
36.5
-------
TABLE 3-5
EFFECTS OF REGULATIONS; ST. LOUIS 305(a)/4 PARTICULATE STRATEGY (DETAILED INVENTORY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/Institutiona]
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
o Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
10
4
33
0
7
4
6
5
5
0
4
15
0
14
47
60
3
110
TOTAL
CONTROLLED
SOURCES
10
4
32
0
5
4
5
5
4
0
2
15
0
7
46
47
2
95
COST
1000's OF
$ PER YR. .
6,634
163
2,797
0
983
27
757
404
949
0
153
4,471
0
1,473
9,594
9,217
145
18,956
EXISTING
EMISSIONS
TONS /DAY
188.0
5.2
116.2
0
11.9
2.6
2.2
5.1
10.1
0
10.0
50.4
0
10.1
309.4
102.5
7.3
419.1
ALLOWABLE
EMISSIONS
TONS /DAY
8.8
.2
5.3
0
1.0
.1
.4
1.0
0
0
.4
.6
0
1.0
14.2
4.7
.5
19.5
CONTROLLED
EMISSIONS
TONS/ DAY
4.3
.0
2.6
0
3.9
0
.5
.2
1.2
0
5.7
.5
0
5.0
6.9
17.2
.3
24.3
COST
EFFECTIVENESS
$ PER TON
REMOVED
98.9
86.4
67.4
-
364.7
29.2
1,249.6
224.4
291.3
-
96.6
245.4
-
793.1
86.9
297.8
56.6
-------
TABLE 3-6
EFFECTS OF REGULATIONS; WASHINGTON, D. C.
305(a) SO STRATEGY(DETAILED INVENTORY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/Institutiona]
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
o Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
22
120
9
0
0
0
0
0
0
0
0
0
0
0
151
0
0
151
TOTAL
CONTROLLED
SOURCES
8
1
0
0
0
0
0
0
0
0
0
0
0
0
9
0
0
9
COST
1000' s OF
$ PER YR.
1,732
22
0
0
0
0
0
0
0
0
0
0
0
0
1,754
0
0
1,754
EXISTING
EMISSIONS
TONS /DAY
492.2
47.2
2.6
0
0
0
0
0
0
0
0
0
0
0
542
0
0
542
ALLOWABLE
EMISSIONS
TONS /DAY
328.9
66.9
3.6
0
0
0
0
0
0
0
0
0
0
0
399.5
0
n
399.5
CONTROLLED
EMISSIONS
TONS /DAY
234.1
46,9
2.6
0
0
0
0
0
0
0
0
0
0
0
283.6
0
n
283. 6_
COST
EFFECTIVENESS
$ PER TON
REMOVED
18.4
243.3
-
-
-
-
.
-
-
-
-
-
-
-
18.6
-
-------
TABLE 3-7
EFFECTS OF REGULATIONS; WASHINGTON, D. C. 305(a) PARTICIPATE STRATEGY(DETAILED INVENTORY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/Institutional
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
o Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
22
120
9
0
0
0
0
0
7
0
0
0
1
151
8
21
180
TOTAL
CONTROLLED
SOURCES
2
12
0
0
0
0
0
0
0
1
0
0
0
0
16
1
6
23
COST
1000' s OF
$ PER YR.
642
276
0
0
0
0
0
0
0
10
0
0
0
0
918
10
140
1,069
EXISTING
EMISSIONS
TONS /DAY
36.4
23.8
.4
0
0
0
0
0
0
9.7
0
0
0
.7
61.6
10.4
13.5
85.5
ALLOWABLE
EMISSIONS
TONS /DAY
40.5
13.4
.8
0
0
0
0
0
0
1.3
0
0
0
0
54.8
1.3
3.9
59.9
CONTROLLED
EMISSIONS
TONS /DAY
25.7
7.1
.4
0
0
0
0
0
0
7.2
0
0
0
.7
33.2
7.9
3.8
44.9
COST
EFFECTIVENESS
$ PER TON
REMOVED
164.3
45.3
-
-
-
-
-
_
11.0
-
-
-
-
88.5
11.0
46.3
-------
TABLE 3-8
EFFECTS OF REGULATIONS; WASHINGTON, D.C. 305(a)/2 SO STRATEGY (DETAILED INVENTORY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/Institutiona]
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
0 Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
22
120
9
0
0
0
0
0
0
0
0
0
0
0
151
0
0
151
TOTAL
CONTROLLED
SOURCES
22
108
9
0
0
0
0
0
0
0
0
0
0
0
139
0
0
139
COST
1000' s OF
$ PER YR.
15,245
8,931
325
8
0
0
0
0
0
0
0
0
0
0
24,502
0
0
24,502
EXISTING
EMISSIONS
TONS /DAY
492.2
47.2
2.6
0
0
0
0
0
0
0
0
0
0
0
542.0
0
0
542.0
ALLOWABLE
EMISSIONS
TONS /DAY
164.5
33.5
1.8
0
0
0
0
0
0
0
0
0
0
0
199.8
0
0
199.8
CONTROLLED
EMISSIONS
TONS /DAY
146.2
25.4
1.3
0
0
0
0
0
0
0
0
0
0
0
172.9
0
0
172.9
COST
EFFECTIVENESS
$ PER TON
REMOVED
120.7
1,123.5
702.0
-
-
-
-
-
-
-
-
-
-
181.9
-
-------
TABLE 3-9
EFFECTS OF REGULATIONS; WASHINGTON, B.C. 305(a)/2 PARTICULATE STRATEGY (DETAILED INVENTORY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/InstitutionaJ
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
0 Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
21
120
9
0
0
0
0
0
0
7
0
0
0
1
150
8
21
179
TOTAL
CONTROLLED
SOURCES
13
26
0
0
0
0
0
0
0
3
0
0
0
0
39
3
7
49
COST
1000' s OF
$ PER YR.
2,026
747
0
0
0
0
0
0
0
24
0
0
0
0
2,773
24
156
2,953
EXISTING
EMISSIONS
TONS /DAY
36.4
23.8
.4
0
0
0
0
0
0
9.8
0
0
0
.7
60.6
10.6
13.5
84.7
ALLOWABLE
EMISSIONS
TONS /DAY
21.8
6.9
.4
0
0
0
0
0
0
.6
0
0
0
0
29.0
.6
3.7.
33.4
CONTROLLED
EMISSIONS
TONS /DAY
13.5
3.7
.4
0
0
0
0
0
0
7.2
0
0
0
.7
17.6
7.9
3.7
29.3
COST
EFFECTIVENESS
$ PER TON
REMOVED
242.4
102.6
_
-
-
-
-
-
24.5
-
-
-
-
176.8
24.5
43.6
-------
Kilometers
100
Figure 3*4.-. Annual SO Ground Level Concentrations,
305(a) Strategy, Cincinnati AQCR
-------
KENTON CAMPBELL
Kilometers
100
Figure 3-2. Annual Particulate Ground Level Concentrations,
305(a) Strategy, Cincinnati AQCR
-------
Kilometer*
S°
Figure 3-3. Annual SO.. Ground Level Concentrations,
-------
Kilometers
Particulate
Figure 3-4. Annual Particulate Ground Level Concentrations,
305(a) Strategy, St. Louis AQCR
-------
Kilometers
Particulates
100
Figure 3-5. Annual Particulate Ground Level Concentrations,
305(a)/4 Strategy, St. Louis AQCR
-------
S02 " I5o
Figure 3-6. Annual S02 Ground Level Concentrations,
305(a) Strategy, Washington, D.C. AQCR
-------
Particulates ~ - ug/m
Figure 3-7. Annual Particulate Ground Level Concentrations,
305(a) Strategy, Washington, D.C. AQCR
-------
S°2 "
Figure 3-8. Annual SO. Ground Level Concentrations,
305(a)/2 Strategy, Washington, D.C. AQCR
-------
Particulates ~
Figure 3-9. Annual Particulate Ground Level Concentrations,
305(a)/2 Strategy, Washington, D.C. AQCR
-------
The particulate standards become stricter as plant
size increases, so powerplant emission reductions
are the greatest.
3
The desired 65 yg/m air quality is achieved for both SCL and
particulates using the base 305(a) emission standards (Figures 3-1 and
3-2).
3.1.2 St. Louis AQCR
The annual cost of applying the 305(a) emission standards to the
point sources of the St. Louis AQCR is $20,187,000 for SO control and
$12,693,000 for particulate control. These costs are significantly
greater than those incurred in Washington, D.C. and Cincinnati. In
terms of cost effectiveness, the three regions compare as follows:
Cost Effectiveness, $/Ton Removed
Region S0_ Particulate
Cincinnati 16.00 26.60
St. Louis 43.53 93.68
Washington, D.C. 18.60 76.89
One reason for the very high SO- costs is that according to the
available data, low sulfur fuel is quite expensive in St. Louis. For
instance, the price that powerplants pay for 1.25% sulfur content coal
is $9.20 per ton in St. Louis, and only $6.25 per ton in Cincinnati.
In addition, the variation of price with sulfur level is high, so that
a fuel switch from 2.3% sulfur content coal to 1.25% sulfur content coal
costs $3.20 per ton additional for powerplants. A switch to 1% sulfur
content coal costs $6.40 per ton, which is extremely high.
As in Cincinnati, powerplants produce a major share of the total
emissions in the region both before and after control. However, the
larger industrial base in St. Louis contributes to a more balanced emis-
sion inventory. Powerplants account for 63 percent of the SO and 42
percent of the particulate point source emissions prior to implementation
of the 305(a) strategy, and 62 and 25 percent, respectively, of the post-
strategy emissions.
According to the isopleths in Figures 3-3 and 3-4, the 305(a) SO
3
strategy is successful in achieving a 65 yg/m level throughout the region,
-------
whereas the particulate strategy is notably unsuccessful. Particulate
strategies of twice and four times the strictness of the 305(a) standards
were unsuccessfully tested; the results of the latter are illustrated in
Table 3-5 and Figure 3-5. Unfortunately, the isopleths in Figures 3-4 an
3-5 include the effects of an error in the source file discovered after
the completion of the runs. A coking plant had been assigned a particulate
emission of 30.8 tons/day, whereas its actual emission is 3.08 tons/day.
Since the original 305(a) particulate standards achieve a 77 percent reduc-
tion in emissions, they should probably be considered as satisfactory for
the major point sources of the region.
3.1.3 Washington, D.C. AQCR
The annual cost of implementing the 305(a) emission control strategy
in the Washington, D.C. AQCR is $1,754,000 for S02 and $1,069,000 for par-
ticulate control. A striking feature of the Washington control cost and
emissions "picture" is the lack of industrial emissions. There are no
major industrial SC- emission sources in the AQCR, and only a few particu-
late sources. Powerplants supply 91 percent of the S0_ and 43 percent of
the existing point source emissions, and 83 and 57 percent, respectively,
of the post-control emissions.
Despite the lack of industrial polluters, the 305(a) standards do
not achieve the desired air quality in the AQCR. The heavy concentration
of small residential emission sources and the area's powerplants and
incinerators require stricter standards. According to Figures 3-10 and
3-11, the application of standards twice as strict as the 305(a) base
standards cause a notable improvement in air quality, and it may be ex-
pected that a slightly stricter set of standards will indeed achieve the
desired level. The additional annual cost to the region of the 305(a)/2
strategies is $22,748,000 for SC>2 and $1,813,000 for particulate control,
above those predicted for the 305(a) strategies.
3.2 COST OF SIMULTANEOUS PARTICULATE AND S02 CONTROL
The costs presented in Tables 3-1 through 3-9 represent the costs
of applying a one-pollutant control program independently of any other
programs. The interaction between pollutant strategies is ignored.
-------
It must be recognized that this separation of pollutants does not
apply in the real world. Many control devices are effective in controlling
more than one pollutant; when these devices are utilized, a link between
separate pollutant control strategies is established and should not be
ignored. The "cost" of instituting simultaneous particulate and S0_
strategies cannot be considered as merely the sum of the costs of each
taken separately. There is considerable "double-counting" and other errors
inherent in such a simple calculation. For instance, a fuel switch to
natural gas can be used for both particulate and SO control. When this
switch appears for a single source in both a particulate strategy run and
an SO run, the analyst must subtract the cost of a single switch from the
combined total to avoid counting this cost twice. The analyst must also
reconcile the two strategies when the following situation occurs:
For a single source;
Particulate strategy chooses device XI at cost Yl.
S0~ strategy chooses device X2 at cost Y2.
A third device X3 exists which will satisfy both
the particulate and the SO standards, and whose
cost Y3 is less than the combined total (Yl + Y2)
but greater than either Yl or Y2.
, In a least-cost solution, device X3 would be the proper choice. In
a proper summation procedure, Y3 would be substituted for (Yl + Y2).
Table 3-10 presents the combined costs of SO^ and particulate runs
for the three AQCR's. The summation procedure is as described above. It
is interesting to calculate the difference between the reconciled total
cost and the simple sum of the two costs. The simple sum of the costs in
the Cincinnati AQCR yields a total cost of 7.307 million dollars per year
to control the major sources in the region. Reconciling the two strategies
yields a reduction of 675 thousand dollars, about a 9 percent change.
While a change of this magnitude is probably well within a reasonable mar-
gin of error for this type of calculation, it can be expected that a sig-
nificantly larger difference will appear under altered circumstances, for
instance, where widespread switching from coal to low sulfur oil or natural
gas is used for both SO- and particulate control.
-------
TABLE 3-10. COST OF CONTROL FOR THREE CITIES,
305(a) SO/PARTICULATE STRATEGY
SOURCE CATEGORY
'COST , $/YEAR
Steam/Electric
Commercial/
Institutional Boilers
Industrial Boilers
Kraft
Iron and Steel
Grey Iron
Non-Ferrous Metals
Sulfuric Acid
Oil Refineries
Asphalt Batching
Cement
Grain
. Varnish
Other Process Sources
Solid Waste Disposal
St. Louis
14,173,363
314,518
6,105,726
0
554,654
27,248
519,500
562,578
945,928
0
152,643
4,469,265
0
1,947,441
85,921
Cincinnati
4,717,571
37,325
824,732
0
670,805
0
0
140,852
0
13,273
0
0
0
13,273
231,000
Washington, D.C.
2,373
344
0
0
0
0
0
0
0
9
0
0
0
0
164
,590
,225
,923
,000
-------
3.3 RESULTS OF POLLUTANT CONTROL STRATEGIES - DEVICE DEMAND
Tables 3-11 through 3-13 show the control device demands and costs
generated by the 305(a) emission standards. The selection of these de-
vices is based on the assumed policy of selecting the least expensive
device which will satisfy the required allowable emissions specified by
the emission standards. The "total" columns compensate for the double-
counting described in Section 3.2.
3.3.1 Cincinnati AQCR
The bulk of the air pollution control required by the 305(a) stan-
dards in Cincinnati is handled by Wet Limestone Injection and Fuel
Switching, for S0_ control, and by Fabric Filters and Electrostatic
Precipitators for particulate control. The fuel switching to .8% sulfur
coal has some beneficial effect on particulate emissions because the coal
has a lower ash content than the high sulfur coal being used. The nearly
3 million dollar cost of the switching represents 1,370,000 tons of low
sulfur coal, of which 1,250,000 tons are consumed by powerplants.
Other significant features are:
The Wet Limestone Injection process is represented
by only one device, but it removes 160 tons of S09
daily from the emissions of one of the AQCR's major
powerplants.
Centrifugal collectors (cyclones), while not
accounting for a substantial portion of the
costs, are nevertheless represented by 23 devices.
3.3.2 St. Louis AQCR
The control device demands generated in the St. Louis AQCR by the
305(a) standards are similar to those in Cincinnati but are of a higher
magnitude. Significant features are:
The Wet Limestone Injection process is represented
by 8 large devices costing a total of nearly 11
million dollars.
Switching to low sulfur oil and natural gas is
extremely substantial, representing an annual
-------
TABLE 3-11
CINCINNATI DEVICE DEMAND AND COST, 305(a) STRATEGY (DETAILED INVENTORY)
DEVICE NAME
1. Wet Scrubber - Hi Eff
3. Wet Scrubber - Lo Eff
7. Cyclone - Hi Eff
8. Cyclone - Med Eff
10. Elect. Precipitator - Hi Eff
11. Elect. Precipitator - Med. Eff
12. Elect. Precipitator - Lo Eff
13. Gas Scrubber
15. Mist Eliminator - Lo Vel
16. Fabric Filter - Hi Temp
17. Fabric Filter - Med Temp
18. Fabric Filter - Lo Temp
28. Fuel Subst., .8% S. Coal
29. Fuel Subst., .3% S. Coal
30. Fuel Subst., 1.25% S. Oil
42. Wet Limestone Injection
so2
NO. SOURCES
0
0
0
0
0
0
0
3
0
0
0
0
26
0
0
1
COST,$/YR.
0
0
0
0
0
0
0
126,100
0
0
0
0
2,630,000
0
0
1,212,900
PARTICULATES
NO. SO URGES
1
7
12
11
5
2
10
0
1
10
1
3
14
4
1
1
COST
67,200
201,000
136,800
93,500
329,400
143,900
700,500
0
14,700
853,400
4,900
672,200
343,200
10,300
16,600
1,212,900
"TOTAL" (reconciled)
NO. SOURCES
1
7
12
11
5
5
8
3
1
10
1
3
31
10
0
1
COST
67,200
201,000
136,800
93,500
329,400
143,900
368,500
126,100
14,700
853,400
4,900
672,200
2,633,000
10,300
0
-------
TABLE 3-12.
ST. LOUIS DEVICE DEMAND AND COST, 305(a) STRATEGY (DETAILED INVENTORY)
DEVICE NAME
2. Wet Scrubber - Med Eff
3. Wet Scrubber - Lo Eff
7. Cyclone - Hi Eff
8. Cyclone - Med Eff
10. Elect. Precipitator - Hi Eff
11. Elect. Precipitator - Med Eff
12. Elect. Precipitator - Lo Eff
13. Gas Scrubber
14. Mist Eliminator - Hi Velocity
15. Mist Eliminator - Lo Velocity
16. Fabric Filter - Hi Temp
18. Fabric Filter - Lo Temp
20. Catalytic Afterburner
32. Fuel Subst., 1% Sulfur Oil
33. Fuel Subst., Natural Gas
42. Wet Limestone Injection
43. Sulfuric Acid Plant Contact Proc.
S02 PARTICULATES
NO. SOURCES
0
0
0
0
0
0
0
4
0
0
0
0
0
12
21
8
1
COST, $/YR.
/
0
0
0
0
0
0
0
270,100
0
0
0
0
0
6,967,800
1,395,000
10,897,900
*
NO. SOURCES
2
3
5
6
10
6
9
0
1
4
21
8
1
1
7
0
COST
85,900
146,000
126,900
391,700
2,176,500
1,392,900
563,700
0
19,000
273,400
4,763,200]
489,200
1,228,200
174,700
442,600
0
"TOTAL" (reconciled)
NO. SOURCES
2
3
1
4
7
1
0
4
1
4
15
8
1
12
23
8
1
COST
85,900
146,000
2,800
357,800
1,273,800
456,700
0
270,100
19,000
273,400
4,410,400
489,200
1,228,200
6,967,800
1,519,000
10,897,900
*
-------
TABLE 3-13. WASHINGTON, B.C. DEVICE DEMAND AND COST, 305(a) STRATEGY (DETAILED INVENTORY)
DEVICE NAME
2. Wet Scrubber - Med. Eff .
3. Wet Scrubber - Lo Eff.
9. Cyclone - Lo Eff.
12. Elect. Precipitator - Lo Eff.
16. Fabric Filter - Hi Temp
18. Fabric Filter - Lo Temp
28. Fuel Subst. .7% S Coal
30. Fuel Subst. 1.0% S Oil
so2
No . Sources
0
0
0
0
0
0
8
1
Cost,$/Yr.
0
0
0
0
0
0
1,732,000
22,200
P ARTICULATES
No. Sources
2
4
1
5
8
1
0
1
Cost
52,800
87,300
8,200
666,700
194,900
9,900
0
26,200
TOTAL (Reconciled)
No. Sources
2
4
1
5
8
1
8
2
Cost
52,800
87,300
8,200
666,700
194,900
9,900
1,732,000
48,500
-------
outlay of over 8 million dollars. The
requirement for 1% sulfur content oil in
the AQCR exceeds 4.6 million barrels (over
190 million gallons) annually,
Fabric filters are used to control very
large feed and grain operations at a cost
exceeding 4 million dollars.
An interesting feature is the use of a
sulfuric acid plant (contact process) which
actually shows a net profit through the sale
of the collected and transformed SO.. The
controlled plant is a lead smelter.
3.3.3 Washington. B.C. AQCR
The major control devices utilized in the Washington, B.C. AQCR
are electrostatic precipitators for particulate and fuel switching to
low sulfur coal for S02 control. A demand for 2,775,000 tons of .7%
sulfur content coal, at an additional cost of $1,732,000 is generated
by the 305(a) S02 standards.
-------
4.0 AGGREGATION APPROACH
The Aggregation Approach to the prediction of national pollution
control costs and control device demands involves the processing of a
large number of regions through the IPP model, generating estimates of
regional and industry group control costs and device demand, and simply
adding (aggregating) the results to provide a "national-level" estimate.
The approach is attractive because Rapid Survey emission inventories
already exist for upwards of 40 regions, and thus an expensive data
collection effort is avoided if these surveys can provide a usable and
accurate "emission source file" for use in the model.
The Cincinnati and St. Louis Rapid Surveys were chosen as test cases
to evaluate the approach, primarily because of the availability of detailed
emission inventories for these regions which can be used for comparative
purposes. These detailed inventories represent a substantial step upwards
in the care and effort with which they were assembled. They normally re-
quire a man-year of labor or more, whereas a Rapid Survey inventory can be
completed in three to six man-weeks. It should be noted, however, that
the key difference between the Rapid Survey, in its ideal form, and the
detailed inventory is the number of variables collected and the care with
which they are obtained; the inaccuracies in the Rapid Surveys which are
due solely to the need to estimate rather than measurefor instance, with
respect to annual emission ratesare to a certain extent present in the
detailed inventory also. The detailed inventory provides a good but far
from perfect standard against which to measure the Rapid Survey.
Since the variables collected by the Rapid Surveys are insufficient
for running the IPP model, the survey data must be expanded using industry-
wide or region-wide averages, and empirical or theoretical relationships
between various emission/control variables. The expanded inventories are
then utilized in the model to measure the effects of a pollution control
strategy consisting of the emission standards of the 1970 Cost of Clean
Air (305(a)) Report to Congress; the results are directly comparable to
detailed inventory IPP results based on the same strategy.
*
The utility of the Aggregation Approach may be judged on the basis
of the answers to the following questions:
-------
Does the (built-up) Rapid Survey emission
inventory provide a reasonably accurate
measure of control device demand and cost?
Is the effort to "build-up" these inventories
and run them on IPP moderate enough to justify
repeating the process for 40 additional regions?
To what extent do the 40 or so regions represent
the nation in terms of population, industrial
output, electric power generated, etc.?
4.1 RAPID INVENTORIES
The Rapid Survey technique of compiling community air pollution
inventories is centered about estimating air pollution emissions without
incurring the expense of direct stack sampling and detailed and extensive
plant surveys. The technique is designed to produce reasonably accurate
estimates of:
1. The annual total and seasonal weight of
emissions of selected pollutants.
2. The relative importance of various fuels
and types of sources in producing the
emissions.
3. The relative amounts of pollutants emitted
in various geographic sub-areas of the
community (Reference 3). The technique is
not meant as a substitute for detailed surveys
but instead is designed to "provide reasonable
information for making intelligent decisions
in air conservation programs in many communities
at an earlier time than might otherwise be
possible"(Reference 3). Thus, it is to be
expected that the accuracy and level of detail
of the Rapid Surveys will be significantly lower
than that of the Detailed inventories.
The pollution "sources" in a community include both the large indus-
trial and commercial plants and also such small but important (in the
aggregate) sources as automobiles, backyard burning, apartment house
-------
incinerators, etc. As the first step in the Rapid Survey technique, the
sources in the community are divided into two categories: point sources,
which are those sources which satisfy some predetermined minimum size
measure (a minimum emission in tons/year, minimum plant capacity, or some
other measure), and area sources, which include all other sources in the
community.
Point sources are handled individually in the survey; one plant
equals one pollution source. In an average sized community, a Rapid
Survey will include 10 to 50 point sources, although this number may
expand considerably for a large city. The point source inventory normally
includes:
all steam-electric power plants
municipal incinerators
large manufacturing plants (both as fuel
combustion sources and process sources)
. large institutions such as hospitals
Point sources are individually surveyed for the following information
Name
Number of employees
Industrial Classification (SIC code)
Location
Fuel use, by fuel (if plant is a fuel combustion source)
Total
Process needs
Space heating
Fuel sulfur and ash content
Equipment, by fuelfor each boiler
Equipment type
Rated capacity
Fuel burned
Control equipment type
Control efficiency
If the plant is a "process source," i.e., if the pollution is directly
caused by a manufacturing process rather than by combustion of fuel in an
indirect heat exchanger, the survey would be altered to collect data on
production rather than fuel use.
-------
Given the information listed above, a reasonably accurate estimate
of air pollution emission may be made by making use of "emission factors"
which relate emission to either fuel use (combustion sources) or production
(process sources). For instance, from Reference 3, sulfur dioxide emissions
from the burning of 1 ton of coal are approximately equal to 38 pounds
times the sulfur content (in percent by weight) of the coal. Particulate
(ash) emission from the burning of coal may be measured in a similar
manner, although the emission factors are here strongly dependent on the
type of boiler used.
An area source represents an aggregation of many small sources which
is treated as a single source. The magnitude of emissions from each com-
ponent of an area source (if the source is composed of more than one type
of emitter) is calculated by various methods described in detail in
Reference 3. For instance, emissions from residential areas (space
heating) may be estimated by the use of an empirical formula using degree
days, number of households using a fuel, and an average fuel requirement
per household per degree day. Data on degree days is available from the
weather service, while the number of households is available from local
fuel distributors.
4.2 CONSTRUCTION OF THE EMISSION SOURCE FILE
Table'4-1 presents the point source variables used by the IPP model.
As noted, several of them are unnecessary for operation of the model, and
several others may be assumed to be equal to region-wide or industry-wide
averages, probably without significant loss of accuracy. Since at this
time no detailed study has been conducted of the precise role that each
variable plays in determining the accuracy of the model results, the
process of "building up" a Rapid Inventory to the point where it may be
used in IPP will consist of calculating as many of the variables in
Table 4-1 as possible.
It should be recognized at the onset that the nature of the goals of
this task...to gauge the usefulness of the Rapid Survey Technique in imple-
plementation modeling, and to judge whether the Aggregation Approach should
be used to estimate national contVol device demand and costs...requires
that each technique selected to calculate a variable should, ideally, be
-------
Table 4-1. IMPLEMENTATION PLANNING PROGRAM POINT SOURCE VARIABLES
Source Variable Comments
Source ID 1
X Location 1
Y Location 1
Political Jurisdiction 2
Ownership 2
Source Type* 1
Stack Height (M) 3
Stack Diameter (M) 5
Exit Velocity (M/Sec) 5
Exit Temperature (°K) 3
Existing Control Device 4
Existing SO Control Efficiency 4
Existing Particulate Control Efficiency 4
Exhaust Gas Volume, ACFM, from stack 1
Rated Capacity of Plant (BTU/hr) 1
Operating Time (Hr/Yr) 3
Annual Shifts (Shifts/Day) 3
Original Coal Cost ($/Ton) 3
Ori-ginal Oil Cost ($/Gallon) 3
3
Original Gas Cost ($/Ft. ) 3
Original Coal Heat Content (BTU/Ton) 3
Original Oil Heat Content (BTU/Gal) 3
3
Original Gas Heat Content (BTU/Ft. ) 3
Electricity Cost ($/kwh) 3
3
Control Device Fuel Cost ($/Ft. ) 3
Disposal Factor, S02 ($/Ton)** 3
Disposal Factor Particulates ($/Ton) 3 '
Liquid Cost ($/Gal) 3
Manufacturing Process Unit 1
Process Rate (Process Units/Day)*** 1
Coal Use (Tons/Day) 1
Oil Use (Gal/Day) 1
Gas Use (Ft. /Day) 1
-------
Table 4-1. IMPLEMENTATION PLANNING PROGRAM POINT SOURCE VARIABLES
(Continued)
Source Variable Comments
Original Coal Sulfur Content (%) 6
Original Oil Sulfur Content(%) 6
Emission Rate SO (Tons/Day) 1
Emission Rate Particulates (Ton/Day) 1
*Fuel Combustion Source, Process Source, or Solid Waste Disposal Source.
**Disposal Factor is the price that the plant can get for the material
captured by the control device. For instance, certain SO^ control
devices produce marketable sulfuric acid.
***The Process Rate is defined as the rate of raw material consumed by
the plant in a manufacturing process.
Comments:.
1. These variables are absolutely essential for operation of the model
and must be calculated on an individual basis, i.e., industry-wide
or region-wide averages are insufficient.
2. These variables are not strictly needed for a region-wide analysis.
3, These variables are desirable but may be estimated using either
region-wide or industry-wide averages, or else estimated as a
function of SIC code.
4. These variables fit into comment 3, except that they assume much
greater importance when emission rates have been calculated, using
emission factors, rather than measured.
5. These variables are desirable but may be left out. An error is
then introduced into the air quality calculations, however.
6. These variables are extremely desirable, but are often unavailable
and must be assumed to be equal to region-wide averages.
-------
a "best" technique in terms of both accuracy and ease of calculation. Of
course, due to the limits of time, this demand is not always realized.
The variables collected in the "ideal" Rapid Survey are described in
Section 4.1. The surveys investigated did not follow the precise format
outlined in Reference 3, because of a lack of manpower resources and a
stress on diffusion modeling, which does not require precise boiler-by-
boiler information. As a result, some variables were entirely absent, and
others were present in only a haphazard fashion (the St. Louis and Cincinnati
Rapid Surveys were of course examined in great detail; however, several
others, including those of Denver and Seattle, werp also examined.)
In general, the following variables were available:
Location, X and Y coordinates
Source type
Manufacturing process (source I.D.) but often in
an overly general way, i.e., Chemical Plant,
Rubber Plant, etc.
Fuel use, if a fuel combustion source
Production or process rate, if a process source
(occasionally missing in Cincinnati Survey)
S0« and particulate emission rates (obviously
calculated from fuel rates or production figures)
Variables which are occasionally available include:
Control device and/or efficiency
Operating days per year
Fuel sulfur levels
Equipment types, i.e., spreader stoker, pulverized
wet bottom, etc.
The following sections describe the process of estimating the missing
variables.
4.2.1 Exhaust Gas Volume
Exhaust gas volume is a key variable in determining the cost of
control devices. It is calculated for process sources by utilizing "gas
volume factors" which relate gas volume to other key plant variables.
Table 4-2 lists some key gas volume factors collected by the Research
Triangle Institute (Reference 4). Note that many industries are not
-------
represented. Gas volumes for these industries were filled in, where
possible, from the available detailed file. It is possible that the
missing gas volume factors could be obtained with additional systems
studies of the industries involved.
Exhaust gas volume for fuel combustion sources can be calculated
by the derivation of factors similar to those in Table 4-2. The HEW
document Atmospheric Emissions from Coal Combustion (Reference 5)
contains a chart (reproduced as Figure 4-1) which allows the calculation
of gas volume in ambient cubic feet per minute (acfm) given the fuel
type, percent of excess air in the burning process, BTU content of the
fuel and gas temperature.
As a sample calculation, assume:
Fuel is LOW VOLATILE BITUMINOUS COAL
Coal heat content«26 x 10 BTU/ton
Gas exit temperaturea300° F
3 4
From Figure 4-1, gas volumes 150 ft, dry flue gas per 10 BTU
fired. (Corrected to 60° F, 30" Hg).
Correcting to 300°F to get acfm,
o
Gas Volume "= 9.5 x 10 acfm per ton/hr of coal fired
Reference 6 gives us a useful comparison with actual coal fired
plants (Table 4-3). The comparison with the derived factor seems quite
reasonable. For coal with a different heat content, multiply the
derived factor by
Heat Content, Millions of BTU's per Ton
26
For plants burning oil and gas, the following factors apply:
OIL
Residual fuel oil, 35 percent excess air, 300°F gas exit temperature
Gas Volume = 48>^ acfn* x Heat Content. 1Q5 BTU/gal
gal/hr of oil fired f~5
-------
TABLE 4-2. GAS VOLUME FACTORS* (Reference 4)
PROCESS
Kraft Processes
GAS VOLUME PRODUCTION
Recovery furnace 25000 acfm/100 tons per day air-dried pulp
Lime kiln 3200 acfm/100 tons per day air-dried pulp
Smelt-dissolving tank 3100 acfm/100 tons per day air-dried pulp
Bark boiler 8000 acfm/100 tons per day air-dried pulp
Aluminum Industry
Prebaked cells
Soderberg cells
7.15*10 acfm/1000 tons of annual capacity
1.14*10 acfm/1000 tons of annual capacity
Phosphate Fertilizer
750 acfm (at 150°F) per ton-per-day plant
capacity
Oil Refinery Total acfm
(Catalyst Regenerator)
x
Cement Industry
Wet process
Dry process
2830 acfm
1000 bbl/day
feed rate (1000 bbl/day)
Asphalt Batching 150 acfm per ton of mix per hour
40*10 acfm per 1000 bbls/day
48*103 acfm per 1000 bbls/day
Lime
Rotary kiln
Vertical kiln
5500 acfm per ton/hour
3200 acfm per ton/hour
*Based on average conditions and various assumptions described in
Reference 4.
-------
Figure 4-1. Relationship between Type of Fuel Burned, Excess Air, and
Resulting Volume of Combustion Products. (Reference 5)
100-
HIGH VOLATILE _
BITUMINOUS
- ANTHRACITE
-------
TABLE 4-3. MEASURED GAS VOLUME FACTORS (Reference 6)
Type of Boiler Firing
Coal Rate,
Ton/Hour
ACFM
10 ACFM
Tons/Hour
Vertical
Corner
Front-wall
Spreader-stoker
Cyclone
Horizontally opposed
65.6
56.1
52.2
9.2
64.4
9.6
549000
518000
453000
91900
792300
92700
8.37
9.23
9.99
8.67
12.30
9.66
-------
GAS
Natural gas, 3 percent excess air, 300°F gas exit temperature
r .. , .256 acfm Heat Content. 1Q3 BTU/ft3
Gas Volume = r x :~r
ft /hr of gas fired
In conclusion, the exhaust gas volume in acfm for fuel combustion
sources may be calculated from an accurate estimate of the fuel rate
with some reasonable degree of validity. It is also possible that acfm
may be estimated accurately for most industrial processes given consider-
able details of the exact process and an accurate estimate of the process
rate. However, the required level of detail and accuracy is not available
in most Rapid Surveys. Given the close relationship (see Section 3.4.1)
between exhaust gas volume and control device capacity (and cost), the
importance of upgrading the variables which figure in the calculation
of acfm is evident.
-------
4.2.2 Fuel Data - Price, Sulfur Level, Heat Content
The accuracy of fuel data is a weak link in the modeling process.
Users and dealers often do not know what the sulfur level and heat con-
tent of their fuel is, and obtaining accurate data is quite difficult.
Also, because strict S0? regulations often demand widespread switching
to low sulfur fuels, prices of such fuels are subject to fluctuation
which cannot be accounted for in the IPP model.
Fuel heat content and price was not available in the Rapid Survey
data, and sulfur content was only occasionally given. Therefore,
region-wide averages were used in place of the missing data. Average
values were obtained from The Fuel of Fifty Cities (Reference 7) and
from the detailed files.
The gathering of this average data is not a straightforward process.
Reference 7 rarely has only one price for a given fuel (since there are
several sources of data and often several grades of the same fuel).
Furthermore, there are four prices for most fuelsDomestic, Commercial,
Industrial, and Power Plant - and the price difference between the four is
considerable. For instance, Reference 7 gives the following prices for
coal in the St. Louis area:
Domestic $24-26 per ton
Commercial $21-24 per ton
Industrial $7.15 per ton
Power Plants $4.45 per ton
In general, a point source may be expected to pay "industrial" prices at
the most; the names refer more to the level of fuel use rather than the
actual business the source is in. However, depending on the size limita-
tion placed on point sources in the inventory, it is conceivable that
the smallest sources may be charged at rates higher than "industrial";
-------
in addition, extremely large industrial sources might pay rates approaching
"power plant" rather than "industrial" rates.
Another potentially confusing point is the difference in price
between "interruptible" and "firm" gas prices. "The essential difference
between the two classes is that the commercial user can be supplied without
straining the system, while an industrial user will usually have some
restriction on demand, for example, interruptible or dual fuel service,
maximum daily limits, or the stipulation that the user be located on a
major gas main," (Reference 7).
For point sources, "firm" prices are used when only gas is burned
(it is assumed that a substitute fuel is impractical); when the plant
burns several fuels, "interruptible" gas prices are assumed. In St. Louis,
the "firm" price is $.055 per therm, while "interruptible" services cost
$.033 per therm.
Although unavoidable, the use of region-wide averages is quite
dangerous here. The effect that such use has on the results for individual
sources ("MICROSCOPIC" result) is well known; if a plant is paying for fuel
at a rate significantly different from the region-wide average, or else
using a significantly lower or higher grade fuel than the average, cost
results for control of the source can be considerably in error. However,
it might be assumed that regional ("MACROSCOPIC") results will not suffer
from the use of averages because these errors will balance out, i.e., some
plants will be "charged" too little and some too much. This is not
necessarily the case. A simple example will illustrate the possibility of
a very significant error in the overall results:
Assume region-wide average sulfur level of fuel is one
percent
Assume this average actually consists of:
50 percent of the fuel contains .5 percent sulfur
50 percent of the fuel contains 1.5 percent sulfur
If a maximum one percent fuel sulfur limitation is enforced in the
region, the model will predict ZERO CONTROL COSTS and no improvement in
the air quality of the region.
-------
The scenario that is enacted here has the users of the high grade
.5 percent sulfur fuel switching to a lower grade one percent fuel, which
is highly unlikely. A more likely result might be a general maintenance
of the higher grade usage among this group, coupled with a shift to one
o
percent coal among the previous "1.5 percent" group, generating a very
substantial cost of control and air quality improvement.
4.2.3 Stack Exit Temperature
Although exit temperature is not given in the Rapid Survey, it is
an important variable:
Given a calculated "standard gas volume flow" it
determines actual flow rate according to the gas law.
It can be a determining factor in control device
selection. For instance, fabric filters are rated
as to their heat tolerance. In the cost section of
IPP, three selections of fabric filters are available
according to exit temperature of the source.
It is a factor in calculating the "plume rise" of the
exit gas, and therefore plays a role in predicting
air quality.
An attempt was made to derive an empirical relationship between
exit temperature and gas volume (which is a measure of plant size) for
combustion sources, using the available detailed inventories for data.
It was expected that larger plants would be more efficient and thus have
lower exit temperature. While some correlation seemed to be present in
the Cincinnati inventory, further investigation of the Philadelphia and
St. Louis inventories did not prove as fruitful. It was decided to assume
a constant exit temperature of 300°F (422° K) for all combustion sources.
Process source temperatures were obtained, where possible, as industry-wide
averages from Reference 4. Judging from the values in the detailed files,
deviation from these "averages" is high.
4.2.4 Rated Capacity
The particulate fuel combustion emission standard used in the report
is based on the rated capacity, in BTU/hour, of the applicable plant.
Although boiler capacity is listed in Section 4.1 as a variable to be
-------
collected, neither of the two rapid inventories contained rated capacity.
Actual heat intake, in terms of the heat content of the fuel used, was
therefore substituted in place of the correct values. This calculation is
not likely to be very accurate, since few plants actually operate at their
rated capacity. In addition, the value calculated will be an average heat
input over the year whereas an estimate of the maximum heat input would be
of more use for an estimate of rated capacity. The accuracy of this
variable could be improved by the incorporation of an accurate "use factor"
into the calculations.
An idea of the impact of an error in rated capacity can be obtained
by inspecting the emission standard in Figure D-l and noting that the rate
of allowable emissions varies between .6 pounds per million BTU and .12
pounds per million BTU as the capacity rating of the emission source
increases. A 100 percent error in rated capacity, which is not really
a remarkable error when considering the difficulties of calculation, will
result in "an error in the allowable emission rate on the order of 30
percent or so.
4.2.5 Operating Hours
Most of the 305(a) (Cost of Clean Air) regulations are stated in
terms of rates PER HOUR, whereas plant variables are normally given in
the rapid inventory in terms of rates PER YEAR. An accurate measure of
operating hours per year is therefore necessary to properly apply the
regulations. Note that the calculation of rated capacity above is
directly dependent on an accurate estimate of operating hours.
An inspection of the detailed emission source files of Philadelphia
and Cincinnati revealed that there exists a pattern of operating hours/year
for most industry groups. Although quite a bit of "scatter" exists, it
seemed reasonable to construct a table of operating hours per year versus
industry groups for use in constructing the source file for the Rapid
Survey data (see Table 4-4).
Emission rates for air pollution source inventories are calculated
on an "average annual" basis: Emission in tons/year is simply divided by
*
365. In this manner, a summation (over all sources) of "average annual"
daily emissions will give a reasonable picture of the average emissions on
-------
TABLE 4-4. OPERATING HOURS
SIC Code (two digit)
20
25
26
27
28
29
32
33
34
35
37
4911
4953
80
82
99
11 oil refineries
51 asphalt batching
non-foundry
foundries
except hospitals
hospitals
Operating Hours Per Year
8760
5840
8760
5840
8760
8760
4380
8760
8760
4380
5840
5840
8760
8760
7300
5840
8760
5840
7300
-------
a given day. A summation of emissions calculated on an "emission/operating
day" basis will be larger than that of an actual average day, because on
any given day some plants will be shut down. (The use of "average annual"
emissions is an approximation because plant shutdown is not a random affair;
usually more plants shut down on weekends, or during certain seasons, so
that an average Sunday will have less air pollution than an average
Wednesday.) During the course of constructing the source files, it was
discovered that several discrepancies existed between fuel sulfur content,
fuel use, and S09 emissions in the Rapid Survey data. The three factors
are linearly dependent; for instance, in coal combustion,
SO emissions, Tons/Year = .019 S x (coal rate, Tons/Year)
where S = percent by weight of sulfur in coal
It was found that daily "average annual" S0_ emission rates did not agree
with the formula. Apparently, yearly emission rates had been correctly
calculated, but had then been divided by actual operating days rather
than 365 to obtain daily rates. Extensive corrections were then made,
especially in the St. Louis file; the result is that both the St. Louis
and Cincinnati emission inventories used in this study no longer agree
with those originally found in the Rapid Surveys.
4.2.6 Source Identification
The eight digit Source I.D. used to identify each emission source
in the IPP source inventory consists of the following:
a four digit Standard Industrial Classification
(SIC) code number
a three digit site number
a one digit process-detail code
In many cases, the plant descriptions in the Rapid Surveys are
of insufficient detail to precisely identify the proper SIC code or
process-detail code (the latter code identifies the processfuel
combustion, open hearth furnace, sinterer, electric arc, etc., etc.
given the SIC code). The result is that a proper choice of control
devices cannot be made. It is expected that some of these omissions are
*
the result of data handling problems rather than a lack of collecting
-------
these data in the first place. The original calculation of annual
emissions must have been based on a knowledge of process details.
The lack of precise details of the process type (or, in the case of
fuel combustion sources, of the equipment used) can lead to serious errors
in the emission estimates. Table 4-5 reproduces table F6 in Reference 3.
The table explains why the Rapid Survey and Detailed Inventory particulate
estimates in Tables 5-1 and 5-2 are so often not in agreement; a wrong
guess about combustion equipment can result in a substantial error in the
emission calculation.
4.2.7 Existing Control Equipment
The Rapid Surveys contain only very sketchy details of existing
particulate and S0_ control devices. Where any data appears, it is
invariably in the form of either a control efficiency or a device name.
The latter is undoubtedly the more reliable of the two. It is probable
that a considerable number of air pollution control device users have no
way of telling just how efficient their devices are, especially after a
few years of use and the accompanying degradation of efficiency due to
imperfect maintenance, wear and tear, etc. In the cases where only a
device name was given, an attempt was made to find an average value for
that device's efficiency; this value was used in the source inventory.
Reference 3 contains some useful information on this subject.
A knowledge of the present level of control is particulately
important in the application of the 305(a) standards to incinerators.
The particulate solid waste disposal standard is a curve of potential
emission versus allowable emission. The model uses present emissions and
control efficiency to calculate potential emissions:
. , . present emissions
potential emissxons = ;r ..... .-
1 - control efficiency
Neither the control efficiency nor the control device name, if any, is
given for the incinerators listed in the St. Louis and Cincinnati Rapid
Survey data. Control efficiency is therefore assumed to be zero for
these sources.
-------
TABLE 4-5. PARTICULATE EMISSION FACTORS FOR COAL COMBUSTION
(WITHOUT CONTROL EQUIPMENT); (Reference 3)
Type of Unit
Pulverized - General
Dry Bottom
Wet Bottom
Without Reinjection
With Reinjection
Cyclone
Spreader Stoker
Without Reinjection
With Reinjection
All Other Stokers
Hand-Fired Equipment
Particulate Emission
Lb/Ton of Coal
Burned
16A*
17A
13A
24A
2A
13A
20A
5A
20A
*A equals percent ash in coal (by weight).
-------
4.2.8 Source Location
The identification of accurate emission source locations is an
important factor in obtaining an accurate regional air quality pattern.
A comparison of rapid and detailed source inventories reveals that the
measurement of plant locations has been inaccurate in one or the other
(or both). For example, since both the St. Louis rapid inventory and
detailed inventory have been set up in Cartesian coordinates (with
different origins), the difference between the two X coordinates (or Y
coordinates) of any point should be a constant throughout the region
.....since this is a linear measurement. Table 4-6 compares the locations
of several power plants and indicates the "constant" conversion factor to
switch between coordinate systems.
Ignoring the discrepancy between X coordinates for the Portage Des
Sioux plant, which is possibly due to a data handling error, the locational
error in the X coordinate can be as high as 2 kilometers, a severe error.
The Y coordinates seem to be an order-of-magnitude more accurate.
Although it is not possible to state with confidence precisely what
order of effect this error will have on predicted air quality patterns,
this type of error contains the potential for creating high pollutant
concentration "hot" spots where they do not exist (or hiding them when
they do).
4.2.9 Stack Height
An accurate estimate of stack height is necessary for proper
computation of air quality levels, since this parameter enters directly
into the diffusion equation. The Rapid Surveys did not include this data;
in most cases, point source stack heights were assumed to be 75 meters
(with a very few obvious exceptions such as quarries). Area source stack
heights were assumed to be zero. Although the inaccuracy of this data
will hurt the analysis of St. Louis and Cincinnati, stack height should
not be a difficult variable to obtain. Upgrading of other Rapid Surveys
to include this variable should be a simple matter.
-------
TABLE 4-6. LOCATION OF ST. LOUIS POWER PLANTS
PLANT
MERAMEC
ASHLEY
CAHOKIA
VENICE
PORTAGE
DBS SIOUX
V
138.00
151.21
150.6
151.5
141.5
XR
77.29
88.94
88.86
89.37
75.43
-------
4.2.10 Emission Sources with Multiple Stacks
Reference 3 suggests that a Rapid Survey should account for each
stack in a plant separately. This procedure was not followed in the
St. Louis and Cincinnati Rapid Surveys, and it is not often possible with
the limited time and resources usually allotted to such surveys. On the
other hand, the more exacting detailed surveys usually separate sources
on a stack-by-stack basis.
Although this lack of detail will not adversely affect the accuracy
of the existing air quality patterns calculated in the Rapid Survey, it
will make a substantial difference in all calculations concerning the
effects of the 305(a) standards. The particulate emissions standard for
industrial process sources assigns a variable allowable emission rate to
plants of varying capacity. The larger the plant, the more stringent the
emission standard becomes. It is profitable, therefore, for the plant
to have each stack considered separately, because the sum of allowable
emissions calculated by this method is somewhat greater than the allowable
emission of the plant if it were considered as a single source.
To test the variation in allowable emissions caused by the two
interpretations of the standards (or by the two levels of detail in
the source file) , a sample calculation was made for the Champion Paper
Company in Cincinnati. In the detailed inventory, the plant is considered
as five separate sources. In the rapid inventory, it is considered to be
one. To be certain we were calculating only the difference between the
number of sources and that no extraneous differences crept in, the five
sources in the detailed inventory were actually compared with a single
source computed by combining the five. The standard applied is the
Maryland Particulate Emission Standard (see Figure D-l). The relevant
details of the five sources are shown in Table 4-7.
The allowable emission for the individual stacks sum to a total of
1.81. The allowable of the plant considered as a single source is 1.48
tons/day, an 18 percent difference.
Given the lack of detail, it is tempting to divide the source into
five stacks of equal capacity and equal heat input as a first approximation
to the detailed case. In fact, however, this approximation will produce
-------
TABLE 4-7. CHAMPION PAPER COMPANY CHARACTERISTICS .
STACK
NUMBER
1
2
3
4
5
ALLOWABLE
EMISSIONS,
TONS /DAY
1.28
.26
.19
.04
.04
RATED
CAPACITY
BTU/HR
573 x 106
131 x 106
94 x 106
100 x 106
100 x 106
COAL
USE
TONS /DAY
423
60.3
40
10
10
COAL HEAT
CONTENT
BTU/TON
26
26.6
26.6
25.4
24.5
-------
a maximum allowable emission. In the present case, the allowable based on
this assumption is 2.12 tons/day, which represents approximately the same
level of error, but now in the opposite direction.
An interesting by-product of this multiple source error is that the
air quality effects of the more stringent interpretation of the emission
regulation (the computation of the allowable emissions on a "total plant"
basis) are better simulated by a source file which lacks boiler-by-boiler
details, since IFF as it is presently constructed has no provision to
"combine" multiple sources for the purpose of calculating allowable emis-
sions. This lack of detail is, of course, a severe disadvantage in calcu-
lating control costs, because the Control Cost Model will proceed to apply
a single stack device to a source which actually has several stacks. A
substantial error would be present only if the cost of control for the
device varied linearly with rated capacity and heat input and if the cost
equation contained no constant term; this is probably never true.
4.3 IMPACT OF 305(a) EMISSION CONTROL STRATEGY
Tables 4-8 through 13 provide a summary of the effects of the 305(a)
emission standards on the Cincinnati and St. Louis AQCR's (as represented
by their Rapid Survey emission files). Tables 4-8 through 4-11 provide
an industry by industry breakdown of emissions and costs; Tables 4-12
and 4-13 indicate the control device demands and costs for each device
category. These tables are directly comparable to their detailed
inventory counterparts in Section 3.
As in the detailed inventory results, the cost and emission
breakdowns highlight the importance of a relatively small number of
powerplants to the overall pollution control effort in the two AQCR's.
In Cincinnati, powerplants produce 9 out of every 10 tons of SO- emission
contributed by major plants, and 1 out of every 4 tons of particulates
(this latter figure occurs only because powerplants already exert a
degree of control over their particulate emissions). In St. Louis, the
appropriate proportions are 6 out of every 10 tons of SO and 3 out of
every 10 tons of particulate matter. Powerplants will also spend a large
share of the total regional cost of control, although in three of the four
cases the cost effectiveness of the control devices is better than the
regional average.
-------
TABLE 4-8.
EFFECTS OF REGULATIONS; CINCINNATI 305(a) SO STRATEGY (RAPID SURVEY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/Institutiona]
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
*
o Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
5
9
29
0
0
0
0
1
0
0
0
0
0
0
43
1
0
44
TOTAL
CONTROLLED
SOURCES
4
2
2
0
0
0
0
1
0
0
0
0
0
0
8
1
0
9
COST
1000' s OF
$ PER YR.
6,638
19
664
0
0
0
0
24
0
0
0
0
0
0
7,321
24
0
7, "US
EXISTING
EMISSIONS
TONS /DAY
905.2
9.2
137.8
0
0
0
0
2.4
0
0
0
0
0
0
1,052.3
2.4
0
1.054.7
ALLOWABLE
EMISSIONS
TONS /DAY
295.1
6.6
143.3
0
0
0
0
.2
0
0
0
0
0
0
445.0
.2
0
445.2
CONTROLLED
EMISSIONS
TONS /DAY
231.1
5.8
116.5
0
0
0
0
.5
0
0
0
0
0
0
353.4
.5
0
353.9
COST
EFFECTIVENESS
$ PER TON
REMOVED
27.0
15.1
85.4
-
-
-
-
34.5
-
-
-
-
-
-
28.7
34.5
_
-------
TABLE 4-9. EFFECTS OF REGULATIONS; CINCINNATI 305(a) PARTICIPATE STRATEGY .(RAPID SURVEY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/Institutional
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
o Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
6
10
28
0
2
5
1
0
0
0
0
0
0
0
44
8
6
58
TOTAL
CONTROLLED
SOURCES
4
10
27
0
2
0
1
0
0
0
0
0
0
0
41
3
6
50
COST
1000's OF
$ PER YR.
1,178
102
894
0
77
0
29
0
0
0
0
0
0
0
2,175
106
63
2,344
EXISTING
EMISSIONS
TONS /DAY
50.4
7.5
141.3
0
5.3
.2
.2
0
0
0
0
0
0
0
199.2
5.4
6.4
211.1
ALLOWABLE
EMISSIONS
TONS /DAY
30.6
1.8
20.7
0
.8
.2
.0
0
0
0
0
0
0
0
53.1
1.1
1.6
55.7
CONTROLLED
EMISSIONS
TONS/ DAY
13.0
1.2
15.4
0
.1
.2
0
0
0
0
0
0
0
0
29.6
.2
1.4
31.1
COST
EFFECTIVENESS
$ PER TON
REMOVED
86.3
44.3
19.5
-
40.6
-
470.6
-
-
-
-
-
-
35.1
56.0
34.2
-------
TABLE 4-10. EFFECTS OF REGULATIONS; ST. LOUIS 305(a)
S02 STRATEGY (RAPID SURVEY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/Institutional
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Grey Iron
o Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
7
8
29
0
0
0
2
4
0
0
0
0
0
2
44
8
0
52
TOTAL
CONTROLLED
SOURCES
6
8
22
0
0
0
0
4
0
0
0
0
0
1
36
5
0
41
COST
1000's OF.
$ PER YR.
8,832
1,737
11,748
0
0
0
0
353
0
0
0
0
0
19
22,316
372
0
22,688
EXISTING
EMISSIONS
TONS /DAY
892.6
44.5
266.9
0
0
0
237.6
69.8
0
0
0
0
0
21.0
1,204.0
307.4
0
1,532.4
ALLOWABLE
EMISSIONS
TONS /DAY
271.1
14.1
276.5
0
0
0
442.8
4.6
0
0
0
0
0
2.4
561.6
447.4
0
1,009.0
CONTROLLED
EMISSIONS
TONS/ DAY
178.0
4.7
116.5
0
0
0
237.6
14.0
0
0
0
0
0
19.4
299.3
251.5
0
550.8
COST
EFFECTIVENESS
$ PER TON
REMOVED
33.9
119.8
214.0
_
-
_
-
17.3
-
-
-
'
-
31.9
67.6
18.2
-
-------
TABLE 4-11. EFFECTS OF REGULATIONS; ST. LOUIS 305(a) PARTICULATE STRATEGY (RAPID SURVEY)
SOURCE CATEGORY
o Steam/Electric
o Commercial/Institutiona]
Boilers
o Industrial Boilers
o Kraft Industry
o Iron and Steel
o Gr,ey Iron
o Non-Ferrous Metals
o Sulfuric Acid
o Oil Refineries
o Asphalt Batching
o Cement
o Grain
o Varnish
o Other
o Fuel Combustion
o Industrial Process
o Solid Waste Disposal
o TOTAL
TOTAL
APPLICABLE
SOURCES
7
8
29
0
5
0
2
5
3
0
2
10
I
11
44
39
2
85
TOTAL
CONTROLLED
SOURCES
5
8
23
0
2
0
1
2
0
0
2
10
0
8
36
25
2
63
COST
1000' s OF
$ PER YR:
1,066
241
1,310
0
550
0
173
87
0
0
148
502
0*
394
2,617
1,854]
56
4,526
EXISTING
EMISSIONS
TONS /DAY
123.2
25.1
97.2
0
3.7
0
5.7
24.4
.3
0
3.9
27.5
2.4
67.5
245.4
135.4
5.1 .
385.9
ALLOWABLE
EMISSIONS
TONS /DAY
28.9
3.3
33.8
0
2.3
0
.8
2.1
2.5
0
.2
2.2
.1
3.7
66.0
13.9
1.1
81.0
CONTROLLED
EMISSIONS
TONS/ DAY
15.8
1.7
15.9
0
2.1
0
5.1
1.1
.3
0
.1
.3
2.4
10.9
33.4
22.3
.2
55.9
COST
EFFECTIVENESS
$ PER TON
REMOVED
27.2
28.3
44.1
-
912.7
-
847.9
10.2
-
-
107.3
50.6
-
19.1
33.8
44.9
30.9
37.6
vO
The only control for this source is a catalytic afterburner costing nearly 5 million dollars.
judged impractical, as cost/ton ~ $6,000.
-------
TABLE 4-12. CINCINNATI DEVICE DEMAND AND COST, 305(a) STRATEGY (RAPID SURVEY)
DEVICE NAME
3 Wet Scrubber - Lo Eff.
8 Cyclone - Med. Eff.
9 Cyclone - Lo Eff.
10 -Elect. Precipitator - Hi Eff.
12 Elect. Precipitator - Lo Eff.
13 . Gas Scrubber
16 Fabric Filter - Hi Temp.
18 Fabric Filter - Lo Temp.
28 Fuel Subst., .8% S Coal
so2
NO . SOURCES
0
0
0
0
0
1
0
0
8
COST, $/YR.
0
0
0
0
0
24,300
0
0
7,311,400
PARTICULATES
NO. SOURCES
6
15
10
1
3
0
5
3
0
COST
63,200
142,500
115,500
29,200
1,394,100
0
187,000
97,200
0
TOTAL (Reconciled)
NO. SOURCES
6
15
10
1
3
1
5
3
8
COST
63,200
142,500
115,500
29,200
1,394,100
24,300
187,000
97,200
7,311,400
o
-------
TABLE 4-13. ST. LOUIS DEVICE DEMAND AND COST, 305(a) STRATEGY (RAPID SURVEY)
DEVICE NAME
7 Cyclone - Hi Eff.
8 Cyclone - Med. Eff.
10 Elect. Precipitator - Hi Eff.
11 Elect. Precipitator - Med. Eff.
*
12 Elect. Precipitator - Lo Eff.
13 Gas Scrubber
14 Mist Eliminator - Hi Velocity
15 Mist Eliminator - Lo Velocity
16 Fabric Filter - Hi Temp.
18 Fabric Filter - Lo Temp
32 Fuel Subst., 1% Sulfur Oil
33 Fuel Subst., Natural Gas
40 Alkalized Aluminum
42 Wet Limestone Injection
SO PARTICIPATES TOTAL (Reconciled)
NO. SOURCES
0
0
0
0
0
5
0
0
0
0
13
18
1
4
COST, $/YR.
0
0
0
0
0
371,600
0
0
0
0
11,172,500
2,447,300
2,616,000
6,080,600
NO. SOURCES
4
2
7
3
7
1
1
1
16
18
1
0
0
0
COST
t65,200
135,500
1,395,000
552,900
671,700
18,600
63,400
23,300
442,000
948,800
2,700
0
0
0
NO. SOURCES
6
6
1
1
1
18
13
18
1
4
COST
855,200
390,100
63,400
23,300
85,600
948,800
11,172,500
2,447,300
2,616,000
-------
An interesting contrast between the two AQCR's is the importance of
industrial sources of S0? emissions in the St. Louis AQCR as compared to
their relative lack of importance in Cincinnati. The disparity is due
mainly to the existence of several process sources of SO in St. Louis
as well as the use of low sulfur (1 percent) coal in Cincinnati's
industrial boilers.
4.4 COMPARISON OF RAPID SURVEYS AND DETAILED INVENTORIES
In order to judge the quality of the results of the Rapid Survey IPP
runs, they may be compared to those of the detailed inventory. Such a
comparison should answer the following questions:
How well does the Rapid Survey duplicate the
detailed emission inventory?
How well do the air quality estimates generated
by the Rapid Survey diffusion model agree with
those generated by the detailed inventory?
How similar are the 305(a) impacts predicted by
the Rapid Survey and detailed inventory?
4.4.1 Emission Inventory Comparison
Tables 4-14 and 4-15 compare the emissions of the major point sources
in Cincinnati and St. Louis as recorded by the Rapid Surveys and the de-
tailed inventories. An inspection of the tables reveals that the S0_ es-
timates tend to be in reasonable agreement, with a few notable exceptions
(Armco Steel in Cincinnati, for instance), whereas the particulate esti-
mates display a wide variation. These results are not surprising, assuming
a fairly good estimate of plant fuel use and fuel sulfur content. As noted
in Section 4.2, particulate emissions from fuel combustion sources are
strongly dependent upon the precise type of boiler used (see Table 4-5) ,
while SO- emissions are solely a function of fuel use and sulfur content.
A comparison of the other important variables in the "source files"
would be quite tedious here. The major problem areas were pointed out in
Section 4.2 to be:
Operating hours. *
Fuel datasulfur and ash contents, prices.
Stack data.
-------
TABLE 4-14. EMISSION INVENTORY COMPARISON, CINCINNATI MAJOR POINT SOURCES*
Power Plants
Tanners Creek
Beckjord
Miami
Reading
Hamilton
Sorg Paper (Boiler)
Champion Paper (Boiler)
Proctor and Gamble (Boiler)
Dupont Sulfuric Acid
Philip Carey (Boiler)
General Electric (Boiler)
Interlake Steel (Electric
Arc)
Armco Steel (Boiler)
Armco Steel (Open Hearth)
SO Emissions
(Tons/Day)
Detailed
Inventory
961.33
339.30
452.52
162.20
1.10
6.21
4.03
15.73
6.08
14.20
2.38
3.64
0
11.77
0
Rapid
Survey
905.23
437.00
308.60
137.58
.50
21.55
3.70
10.40
8.92
-
3.13
3.12
0
82.20
-
Particulate Emissions
(Tons /Day)
Detailed
Inventory
225.28
64.75
29.62
125.99
.09
4.83
12.50
3.37
4.72
0
7.35
9.22
13.00
5.48
32.87
Rapid
Survey
50.24
15.90
20.31
12.07
.07
1.89
12.40
10.50
6.20
-
7.89
1.57
1.50
68.50
-
*(Not a complete list.)
-------
TABLE 4-15. EMISSION INVENTORY COMPARISON, ST. LOUIS MAJOR POINT SOURCES*
Power Plants
Ashley
Wood River
Mound
Meramec
Cahokia
Venice
Highland
Portage Des Sioux
St. Jo Lead Smelter
Coking Plant
Brewery
CKW Chemical Plant (Boiler)
National Lead (Boiler)
Monsanto (Boiler)
Alpha Cement
Chrysler (Boiler)
SO,- Emissions
(Tons/Day)
Detailed
Inventory
1030.41
46.20
249.73
-
358.00
40.30
115.00
4.18
217.00
232.00
17.30
-
8.56
22.00
32.80
0
5.80
Rapid
Survey
1009.44
38.89
248.80
.25
377.10
31.90
116.80
3.60
192.10
232.23
-
17.90
7.18
20.26
31.69
0
4.74
Particulate Emissions
(Tons/Day)
Detailed
Inventory
189.00
11.45
120.42
-
9.76
4.68
36.61
.48
5.60
.61
30.80
-
14.67
7.30
10.70
1.09
2.52
Rapid
Survey
137.78
4.98
95.30
.02
12.00
4.00
14.91
4.60
1.97
.59
-
2.29
9.17
3.90
19.26
1.90
6.05
*(Not a complete list.)
-------
Gas exit volumes, temperature, and velocity.
Plant rated capacity.
Process and equipment details.
Table 4-16 compares the recorded regional emissions of the emission
source files for Cincinnati and St. Louis. The results for St. Louis are
excellent; emission estimates agree quite closely in every emission cate-
gory. Note that a comparison between the number of sources in each source
file must take into account the greater detail of the detailed inventory,
which counts every stack as a separate source. For instance, most of the
powerplants are registered as single sources in the Rapid Survey and as
multiple sources in the detailed survey. (Comparisons in Tables 4-14 and
4-15 take this into account by adding up the separate stack emissions.)
The results for Cincinnati are quite poor. The Rapid Survey missed
most of the industrial process and solid waste disposal sources, and
apparently severely underestimated the particulate emissions of the major
sources (see Table 4-14). The data for Cincinnati contained far less
equipment detail than did the St. Louis data, and thus was subject to the
variance in particulate emissions rates noted above.
4.4.2 Air Quality Estimates
In order to test the validity of diffusion model results generated
by the Rapid Survey emission estimates, comparisons of the Rapid Survey-
based and detailed inventory-based results were made in two different ways.
4.4.2.1 Rapid Survey versus Detailed Inventory - Direct Comparison
A direct comparison of the two sets of results was conducted. Using
uncalibrated results (i.e., air quality estimates prior to being adjusted
or "calibrated" by actual physical measurements of pollutant concentrations),
plots were made using the Rapid Survey concentrations as the dependent vari-
able (y) and the corresponding detailed inventory concentration as the inde-
pendent variable (x). Each point on the plot then represents the Rapid
Survey-based and detailed inventory-based pollutant concentrations at a
single receptor. If the detailed inventory and Rapid Survey are identical,
the plot will simply be a straight, line through the origin with a slope
of 1. If some systematic linear bias between the two exists, a positive or
negative y-intercept and/or a slope other than 1 will result; however, this
-------
TABLE 4-16. COMPARISON OF RAPID AND DETAILED EMISSION INVENTORIES /REGIONWIDE
REGIONAL EMISSIONS (TONS/DAY)
NUMBER OF SOURCES
CINCINNATI
PARTICULATES
DETAILED INVENTORY
RAPID SURVEY
*
so2
DETAILED INVENTORY
RAPID SURVEY
ST. LOUIS
PARTICULATES
DETAILED INVENTORY
RAPID SURVEY
so2
DETAILED INVENTORY
RAPID SURVEY
POINT
386.7
211.1
986.1
1,054.7
446.8
385.9
1,631.7
1,532.4
AREA
212.6
158.8
369.4
104.3
108.3
92.2
46.2
157.7
FC
296.0
199.2
962.5
1,052.3
309.5
245.4
1,306.7
1,204.0
' IP
68.3
5.4
23.5
2.4
130.3
134.4
324.9
328.4
SW
22.4
6.4
0
0
7.3
5.1
0
0
TOTAL
599.3
369.9
1,355.5
1,159.0
555.1
478.1
1,677.9
1,690.1
FC
71
44
71
43
47
44
46
44
IP
34
8
6
1
60
39
13
8
SW
20
6
0
0
3
2
0
0
TOTAL
125
58
77
44
110
85
59
-------
bias will disappear after calibration and the two models will predict air
quality in an identical manner. If the data points are scattered around a
line, the final air quality predictions of the two models will differ.
In this analysis, lines were fitted through the plots using a least
squares regression technique (identical to that used in calibrating the
diffusion model) and the correlation coefficients and standard errors
of estimate* were calculated. In addition, the values of the correlation
coefficient representing the 95 percent confidence level for correlation
were calculated for each case. Table 4-17 presents the results of the
above computations. The magnitude of the Cincinnati correlation coeffi-
cients can be considered as the more significant since they are based on
more data points. Figures 4-2 through 4-5 present the plots of the points
and the regression lines. The values in Table 4-17 and the amount of
scatter evident in the figures suggests that the Rapid Survey predictions
of air quality are likely to be of limited accuracy. Although 3 out of 4
correlation coefficients are higher than those required for 95 percent
confidence of correlation, the magnitude of the coefficients is quite low
and the standard error of estimate is very high.
* The (correlation coefficient)2 predicts the fraction of the variation
of the dependent variable y that is accounted for by the regression on the
independent variable x. For instance, R2=.99 means that the regression
of y on x accounts for 99 percent of the variance of y. The remaining 1
percent is due either to other variables affecting y which are not included
in the analysis or else to a nonlinearity in the function y=y(x) not
accounted for in the linear regression. In physical terms, R2=.99 means
that the other variables and/or the nonlinear term in y(x) do not have
much real effect and can be ignored in all but the most sensitive applica-
tions where small errors can be critical. Note that the R measured in
this analysis is only an approximation of the true value of the correla-
tion coefficient; the greater the number of data points used in the
regression, the closer the measured R is likely to be to the true value.
The standard error of estimate measures the scatter in the vertical
(y) direction of the data points about the regression line.
-------
TABLE 4-17. CORRELATION BETWEEN RAPID SURVEY AND DETAILED
INVENTORY AIR QUALITY PREDICTIONS
R Standard Error
Run //Points R(.95) (Measured) of Estimate
Cincinnati S02 44 .294 .467 75.0
Cincinnati Particulate 44 .294 .641 34.6
St. Louis S02 22 .418 .531 102.8
St. Louis Particulate 14 .514 .281 123.1
-------
o
\o
Survey
Rap
m
ug
Pollutant Concentration
260
N5
-P-
O
S3
N3
O
NJ
O
O
M
OO
O
M
.p-
O
M
t-0
O
(-1
O
O
00
O
O^
O
-P-
O
20
0
20
40
60
80 100 120 140
Pollutant Concentration ug/m (Detailed Inventory)
Figure 4-2. Scatter Diagram and Regression Line of Pollutant Concentration Estimates,
-------
260
3
Pollutant Concentration, yg/m (Rapid Survey)
M M M M r-1 N3 KJ N3 t
»\ OOONJJ>OOOON5*~C
O OOOOOOOOOC
^x
j
/
y
x
. /
^
^
1
* ,
.
/
1
/
/
/
80
100
120 140 160 180 200 220 240 260
Pollutant Concentration, yg/m , (Detailed Inventory)
Figure 4-3. Scatter Diagram and Regression Line of Pollutant Concentration Estimates,
-------
01
T3
H
CX
o
H
J-J
n)
i-i
w
c
0)
o
C
o
o
td
4J
o
0*
500
400
300
200
100
50
100
150
200
250
300
Pollutant Concentration, ug/m (Detailed Inventory)
Figure 4-4. Scatter Diagram and Regression Line of Pollutant Concentration
-------
600
500
0)
>
M
3
CX
cfl
_B
00
C
o
H
4J
rt
M
O
c
o
u
c
iJ
3
o
cu
400
300
200
100
50
100
150
200
250
Pollutant Concentration, ug/m (Detailed Inventory)
Figure 4-5. Scatter Diagram and Regression Line of Pollutant Concentration Estimates,
-------
4.4.2.2 Comparison of Rapid Survey and Detailed Inventory Calibration
The calibration procedure for the diffusion model, where the results
of the theoretical model are adjusted by comparing them to actual pollutant
concentration measurements, provides a good test of the validity of the
input data and the model itself. If very low correlation exists between
the calculated air quality and the measured concentrations, then the
diffusion procedure is invalid. Assuming accurate physical measurements,
a low correlation can be an indication of serious errors in the original
emission estimates.
An obvious test of the efficacy of the Rapid Survey's capabilities
for air quality prediction in comparison to that of the detailed inventory
is to compare the results of such calibration runs for both data sets.
The appropriate calibration runs were conducted, and the correlation
coefficients and standard errors of estimate are given in Table 4-18. The
clear superiority of the detailed inventory air quality estimates are
evident from the considerable differences in correlation between the two
data sets, and the lower scatter as evidenced by the values of the standard
errors of estimate.* Of particulate importance is the fact that none of
the Rapid Survey results achieve the .95 confidence level of correlation,
which is usually the minimum level required for continuing with the
modeling studies.
* It should be noted that the difference between the relative magnitudes
of the standard errors of estimate in Table 4-17 and those in Table 4-18 is
caused not by different degrees of "fit" of regression lines but merely bv
different "dependent variables." In Table 4-17, the dependent variable is
the Rapid Survey data, whose range of values is high, whereas the dependent
variable in Table 4-18 is the measured data, whose range is smaller. The
standard error of estimate measures the vertical differences between the
data points and the regression line; if the axes were rotated in the second
analysis, the order of magnitude of the standard errors of estimate for the
two tables would have been similar.
-------
TABLE 4-18. RAPID SURVEY AND DETAILED INVENTORY CALIBRATION RUNS
Standard Error
Run //Points R(.95) R* of Estimate
Cincinnati SO /Detailed
Cincinnati SO /Rapid
Cincinnati SO /Rapid, Adjusted**
Cincinnati Particulate/Detailed
Cincinnati Particulate/Rapid
St. Louis SO /Detailed
St. Louis SO /Rapid
St. Louis Particulate/Detailed
St. Louis Particulate/Rapid
8
8
7
34
34
23
22
15
14
.666
.666
.707
.335
.335
.404
.413
.497
.514
.644
.241
.371
.762
.282
.747
.548
.818
.282
9.5
12.8
13.9
14.6
21.6
20.6
25.7
23.1
41.8
*R(.95) = Correlation coefficient of .95 confidence level
R = Correlation coefficient
** = One data point was dropped because of extreme
disagreement between measured and calculated
concentration.
-------
Figures 4-6 through 4-13 present the plots of the data points and
corresponding regression lines for 8 of the 9 cases presented in Table
4-18. Three of the four regression lines corresponding to the Rapid Sur-
veys are nearly horizontal; this indicates that the diffusion models are
not accounting for the variation in air quality observed in the Air
Quality Control Regions and are therefore relatively worthless. This
result has strong implications regarding the utility of the Rapid Surveys
for implementation planning; the aim of such planning is the achievement
of air quality standards, and, in the case of St. Louis and Cincinnati,
the Rapid Survey-based diffusion models could not have accurately pre-
dicted when such standards would be met.
There is a possibility of a bias in favor of the detailed inventories
with regards to the diffusion model calibration runs. In the early cali-
bration runs of the detailed inventories, attempts were made to increase
the correlation of the results. Certain air quality measurements were
thrown out of the calibration, because of suspicion of severe inaccuracies;
measurements in extreme disagreement with the calculated values were ob-
viously tempting targets for such removal. The Rapid Survey calibrations
used the identical final measurement data as the detailed inventory, and
did not have the benefit of such selective removal of data points. It is
felt that the results of the calibration comparison are so firm that such
favorable treatment would not have altered them significantly....in fact,
the Cincinnati SO- calibration did drop one extremely unfavorable point
(Figure 4-7), with no significant change in the results.
In addition, the primary cause for throwing out receptor measure-
ments was the actual discovery of measurement problems with the instrumen-
tation, and not merely the existence of a disagreement between calculated
and measured ground level pollutant concentrations. However, the existence
of this possible bias in the comparison should be recognized.
-------
90
o
41
80
c
5
70
o
e
o
u
60
501-
25'
50
75
100
125
150
Pollutant Concentration, ug/m (Calculated)
-------
100
125
150
175
200
225
250
Pollutant Concentration, jig/m (Calculated)
-------
00
'140
I 120
I
% 100
c
80
g
3
tH
Q 60
40
20
75
100
125
150
175
200
225
250
Pollutant Concentration, yg/m (Calculated)
-------
VO
140
120
S 100
g
2
o
u
40
75
100
125
150
175
200
225
250
275
Pollutant Concentration, |ig/m (Calculated)
-------
150
125
100
o
01
u
01
J3
O
2 75
o
c
o
CJ
50
25
50
100
150
200
250
300
Pollutant Concentration, wg/n (Calculated)
-------
150
125
« 100
o
_B
en
c
o
H
a
3
50
25
c
u
3
tH
tH
O
100
150
200
250
300
350
400 450
500
Pollutant Concentration, vig/m (Calculated)
-------
NJ
NJ
in o m
01 o r-*
CM
-------
225
5 200
I
01
s
n
fl,
175
150
4J
c
100
75
150
200
250
300
350
400
450
500
550
600
Pollutant Concentration, jig/m (Calculated)
-------
A.4.3 Comparison of 305(a) Strategy Impacts - Detailed versus Rapid
Survey Inventory
The key to the usefulness of the Rapid Surveys is their ability to
predict the effects of various emission control strategies on the demand
and cost of control devices and on air quality.
Accuracy in prediction of air quality effects can be inferred from:
the level of accuracy of the pre-control air
quality predictions (discussed in 4.5), and
the accuracy of post-control emission predictions.
Table 4-19 indicates the post-control regional point source emissions
and the percent reduction of emissions (caused by implementation of the
305(a) Strategy) for both the Rapid Survey and detailed inventory IPP
model runs 'for St. Louis and Cincinnati. The level of agreement between
the results of the two data collection methods is quite good, with the
exception of the St. Louis SO- run.
Table 4-19 also compares the regional costs of the 305(a) strategies.
There is no consistent pattern of accuracy apparent in the results. The
only truly accurate cost prediction is that for S0« control in St. Louis,
and the indicators of cost effectiveness and emission reduction for that
case are relatively inaccurate. The extreme disagreement between the
St. Louis particulate results is of particular significance, because the
emission data of the Rapid Survey and detailed inventory are in reasonable
agreement. The only possible source of error lies in the remaining source
file variables (gas volume rate, previous level of emission control, etc.,
etc.).
Table 4-20 compares the device demands and costs generated by the
305(a) Strategy using the two data sources. Judging from the table, the
Rapid Survey predicts the existence of major demands when they involve a
number of sources, but is prone to miss expensive devices when they
involve only one source. The Rapid Survey cost predictions are not
generally very accurate (assuming that the detailed inventory predictions
are), but they do show the major trends of device demands. This may be
quite satisfactory for some purposes.
-------
TABLE 4-19. COMPARISON OF 305(a) STRATEGY IMPACT ON POINT SOURCES
to
Cn
CINCINNATI
PARTICULATES
DETAILED INVENTORY
RAPID SURVEY
*
so2
DETAILED INVENTORY
RAPID SURVEY
ST. LOUIS
PARTICULATES
DETAILED INVENTORY
RAPID SURVEY
so2
DETAILED INVENTORY
RAPID SURVEY
EXISTING
REGIONAL
EMISSIONS
(TONS/DAY)
POINT SOURCES
386.7
211.1
986.1
1,054.7
419.1
385.9
1,631.7
1,532.4
CONTROLLED
REGIONAL
EMISSIONS
(TONS /DAY)
POINT SOURCES
42.9
31.1
306.5
353.9
47.9
55.9
361.2
550.8
. PERCENT
REDUCTION
OF
EMISSIONS
89
85
69
66
89
86
78
64
REGIONAL
COST
$/YEAR
3,338,000
2,344,000
3,969,000
7,344,900
12,693,000
4,526,100
20,187,100
22,688,000
NO.
APPL
125
58
77
44
110
85
59
52
SOURCES
CON-
TROLLED
83
50
30
9
85
63
46
41
COST
EFFECTIVENESS
$/TON
OF POLLUTANT
REMOVED
26.6
35.7
16.0
28.7
93.7
37.6
43.5
-------
TABLE 4-20. COMPARISON OF MAJOR DEVICE DEMANDS OF 305(a) STRATEGY
NJ
ST. LOUIS
WET LIMESTONE INJECTION
ALKALIZED ALUMINUM
SWITCH TO NATURAL GAS
SWITCH TO 1% SULFUR OIL
. FABRIC FILTERS
ELECTROSTATIC PRECIPITATORS
CATALYTIC AFTERBURNER
CINCINNATI
WET LIMESTONE INJECTION
SWITCH TO .8% SULFUR COAL
FABRIC FILTER
ELECTROSTATIC PRECIPITATORS
RAP IE
# SOURCES
4
1
18
13
19
8
0
0
8
8
4
SURVEY DETAILED INVENTORY
COST, $/YR
6,000,000
2,600,000
2,500,000
11,200,000
1,000,000
900,000
0
0
7,300,000
300,000
1,400,000
# SOURCES
8
0
23
12
23
8
1
1
31
14
15
COST, $/YR.
10,900,000
0
1,500,000
7,000,000
4,900,000
1,700,000
1,200,000
1,200,000
2,600,000
1,500,000
-------
5.0 THE 305(a) APPROACH
5.1 INTRODUCTION
The third method examined in this study for estimating control
measure demand is called the "305(a) Approach." This name has been
selected since the demand estimating procedure utilizes engineering data
and control cost estimates from the Economics of Clean Air (EGA) Report*,
Can annual economic report from the Administrator of the Environmental
Protection Agency to Congress, required by Section 305 (a) of the Clean
Air Act of 1967).
The 305(a) Approach is unlike the Aggregation and Extrapolation
approaches in that the IPP model is not used for predictive purposes.
None of the models developed by TRW under this contract are used in the
Approach. The method was selected because it is simple and uses data
which is currently available. Also, the method provides a basis for com-
parison with the two other approaches which are based on the utilization
of the IPP model. Finally, by analyzing the data in the Report at the
regional level, better insight into the methods employed in making national
control cost estimates is made possible. In other words, this analysis
provides the reader the opportunity to observe the pertinent factors which
may contribute to deficiencies in the ECA Report control cost estimates.
To prevent confusion, it is important that a distinction be made
between the goals of the ECA Report and the objectives of this report.
The ECA Report evaluates the costs resulting from the implementation of
emission control regulations stimulated by the Clean Air Act, as amended.
In addition, the impact of the control costs on specific industries is
analyzed. The analysis includes the effect of the control costs on prices
in the economy. Finally, expected price changes in specific industries
are analyzed with the use of an input-output model, and price level in-
creases are predicted for the automobile and construction industries.
* - The Economics of Clean Air. . Report of the Administrator of the
Environmental Protection Agency to the Con ress of the United States,
U. S. Government Printing Office, March 1971.
-------
The goal of this study is to more closely examine the specific
resources demanded by such legislation by identifying and categorizing
the specific control systems and fuel substitutions. New legislation
for the control of air pollution stimulates a demand for a complex mix
of resources. This is unlike some legislation which stimulates or de-
presses the market for only one specific good. The promulgation of
emission control regulations requires a combination of resources which
include specific types of fuel, control hardware, construction equipment,
labor, electric power, and numerous others.
The 305(a) Approach has been developed to provide decision-makers
with estimates of the resources demanded by air pollution control re-
quirements. The steps involved in producing the demand estimates for
the decision-maker are illustrated in Figure 5-1. As shown, engineering-
economic data and assumptions on sources of air pollution and air pollution
control systems are the inputs to a control cost estimating procedure. The
estimating procedure (which is described in detail in Section 5.2) varies
for each industry according to the quality of the input data for that indus-
try. Specifically, industries in which plants are identified and charac-
terized individually have control costs developed on a plant-by-plant basis.
On the other hand, in cases where only a number of plants in an industry
can be identified (and not specific plant characteristics such as process
size), model plants, which are selected to be representative of plants in
the industry, must be determined from nationally available statistics.
Control costs are then based on the model plant characteristics.
As indicated in Figure 5-1, this data bank of control costs is used
to prepare the annual EGA Report to Congress. This same data bank is the
basis for the "305(a) Approach" demand estimating technique. There are
two steps involved in preparing the demand estimate. First, the data from
the EGA Report data bank are structured by source category to identify the
additional control systems and quality and quantity of fuel required for
the control of emissions from each process. Also, data on the investment
and annual control costs in addition to data on the quantity of controlled
and uncontrolled particulate and sulfur dioxide emissions are identified
on a process-by-process basis (where possible).
-------
ENGINEERING-
ECONOMIC
DATA
K)
VO
ENGINEERING-
ECONOMIC
ASSUMPTIONS
305(a) APPROACH
STEP #1
EGA REPORT
COST
ESTIMATING
PROCEDURE
ECA
CONTROL
COST
ESTIMATES
STEP #2
PLANT-BY-
PLANT
DEMAND
DATA
REGIONAL OR
NATIONAL
DEMAND
ESTIMATE
BASIS FOR
ECONOMICS OF
CLEAN AIR
REPORT
-------
The second step of the 305(a) Approach requires an aggregation of the
data at the regional or national level for demand estimation presentations.
The 305(a) Approach is a simple technique for converting the control
cost data into useful regional and national control measure demand esti-
mates. The procedure requires some engineering calculations which are
needed to determine low-sulfur fuel-use estimates. No other engineering
or economic analyses are required.
If the demand estimates which result from the 305(a) Approach are
good approximations of the estimated demand, few resources will be required
to develop the EGA Report data bank into useful demand estimates. The pur-
pose of this report is to evaluate the reliability of the demand estimates
and recommend whether or not this approach has merit. If it does, then
another potential use for the EGA Report control cost data bank has been
identified (see Figure 5-2).
To illustrate the 305(a) Approach, data from the third report to
Congress (i.e., the March 1971 version) have been analyzed for four air
quality control regions (AQCRs), By selecting the same AQCRs that were
used in demonstrating the Aggregation and Extrapolation approaches, com-
parisons of the resulting demand estimates are made possible. The compari-
son of the three demand estimating techniques are presented in Section 1.0.
5.2 DESCRIPTION OF THE 305(a) APPROACH AND PRESENTATION OF REGIONAL
DEMAND ESTIMATES.
5.2.1 Introduction
This section describes the two steps of the 305 (a) Approach. Because
the quality of the demand estimates produced by the 305(a) Approach is
directly dependent upon the data and procedures used in developing the EGA
Report control cost estimates, the EGA Report procedures will be reviewed
also. In addition, estimates of control system and control costs on a
plant-by-plant basis will be presented for four AQCRs. Finally, control
measure demand estimates will be presented for each of the four AQCRs.
No national demand estimate will be presented in this report although it
is recognized that such as estimate is feasible. The national estimate
can be determined by applying the*305(a) Approach to control cost estimates
for the entire nation. At this time such an estimate is not possible since
-------
FIGURE 5-2
THE DEVELOPMENT OF DECISION-WKING INTORWTION FROM THE ECONOMICS OF CLEAN AIR
COML COST ESTIMATES
ENGINEERING-ECONOMIC
DATA BANK
DECISION-f-lAKING INFORMATION
AIR POLLUTION
CONTROL COST
ESTIMATES
ESTIMATES OF FULL ADDITIONAL CONTROL COSTS
FOR THE NATION BY SOURCE CATEGORY
ECONOMIC IMPACT OF CONTROL COSTS ON
EMISSIONS SOURCES
IMPACT OF CONTROL COSTS ON PRICE LEVELS
WITHIN THE NATIONAL ECONOMY
* NATIONAL DEMAND FOR AIR POLLUTION CONTROL
SYSTEMS AND LOW SULFUR FUELS
-------
the 1971 EGA Report provides control cost estimates for sources in selected
industries in only 298 of the nation's metropolitan areas. The 1972 EGA
Report will comprehensively evaluate all sources in the nation which are
part of the selected industries.
Before describing the steps of the 305(a) Approach and the resulting
control measure demand estimates, the control cost estimating procedures
of the EGA Report are reviewed. This provides %the necessary background
for a clear understanding of the two steps of the 305(a) Approach.
5.2.2 Scope of the Economics of Clean Air
The Economics of Clean Air is required by the Clean Air Act, as
amended. The section of the Act which calls for the Report is as follows:
Section 305. (a) In order to provide the basis for
evaluating programs authorized by this Act and the develop-
ment of new programs and to furnish the Congress with the
information necessary for authorization of appropriations
by fiscal years beginning after June 30, 1969, the Secretary,
in cooperation with State, interstate, and local air pollution
control agencies, shall make a detailed estimate of the cost
of carrying out the provisions of this Act; a comprehensive
study of the cost of program implementation by affected units
of government; and a comprehensive study of the economic im-
pact of air quality standards on the Nation's industries,
communities, and other contributing sources of pollution,
including an analysis of the national requirements for, and
the cost of, controlling emissions to attain such standards
of air quality as may be established pursuant to this Act
or applicable State law. The Secretary shall submit such
detailed estimates and the results of such comprehensive
study of cost for the five-year period beginning July 1,
1969, and the results of such other studies, to the Congress
not later than January 10, 1969, and shall submit a re-
evaluation of such estimate and studies annually thereafter.
To respond to this requirement, APCO has developed, over the past
three years, control cost estimating procedures and a sizable engineering-
economic data bank based on industrial statistics, national fuel-use es-
timates, research studies of air pollution control systems and costs, air
pollution engineering factors, and numerous engineering assumptions.
Table 5-1 presents the pollutants which were studied in the 1971 EGA
Report. The listed pollutants are those for which criteria and control
technology documents have been or soon will be published. The source
-------
TABLE 5-1
SCOPE OF 1971 ECONOMICS OF CLEAN AIR REPORT
POLLUTANTS
PARTI CULATES
SULFUR DIOXIDE
HYDROCARBONS
CARBON MONOXIDE
FLUORIDES
LEAD
STATIONARY SOURCE CATEGORIES
SOLID'WASTE DISPOSAL
COMMERCIAL-INSTITUTIONAL HEATING
INDUSTRIAL BOILERS
RESIDENTIAL HEATING
STEAM-ELECTRIC POWER
ASPHALT BATCHING
BRICK AND TILE
COAL CLEANING
CEMENT
ELEMENTAL PHOSPHORUS AND PHOSPHATE FERTILIZER
GRAIN MILLING AND HANDLING
GREY IRON FOUNDRIES
IRON AND STEEL
KRAFT PULP
LIME
PETROLEUM REFINING AND STORAGE
PRIMARY AND SECONDARY NON-FERROUS METALLURGY
RUBBER (TIRES)
SULFUR1C ACID
VARNISH
-------
categories listed are the stationary sources which are the major contribu-
tors of the identified pollutants.
The EGA Report evaluates the cost of controlling emissions from
processes which are specifically affected by the emission control regula-
tions required by the Clean Air Act. Not all processeswhether large or
small, clean or dirtyare identified in the Report. Only specific pro-
cesses in the selected industries (which covers nearly all of the
identifiable stationary sources of emissions) are included in the analysis.
This means that the EGA Report is not totally comprehensive and, yet,
consequently, the control measure demand estimates of the 305(a) Approach
will not be totally comprehensive.
Gas cleaning equipment is used for numerous purposes in our economy,
that is, for purposes other than meeting emission control regulations re-
quired by the Clean Air Act. Thus, it is important to identify "air
pollution control equipment" as used in this report. The terms "air
pollution equipment" or "control measures" will be used to identify the
gas cleaning equipment specifically demanded by the requirements of emis-
sion control regulations for the pollutants and processes identified in
Table 5-1. A further discussion of this follows.
An "air pollution control system" is defined as equipment or fuel
switching which is installed for the primary purpose of reducing air
pollutant emissions to comply with emission standards. The term "gas-
cleaning equipment" is used to describe all air pollution control systems
as well as equipment used for other process purposes. Air pollution
economics draw the fine line between what is and what is not an "air
pollution control system." The distinction is based on the utility of
the equipment as a profitable or "break-even" investment. If gas-cleaning
equipment is used for cost-cutting (i.e., profitable) purposes (in addi-
tion to the goal of air pollution control) in a manufacturing process and
the investment is recovered over the depreciable life of the equipment,
then the gas-cleaning equipment is not considered an "air pollution
control system."
When used to capture valuable material before entering the atmos-
phere, gas-cleaning equipment (which pays for itself over the normal life
of the equipment) is considered "process equipment" rather than an air
-------
pollution control system (such as electrostatic precipitator on a basic
oxygen furnace). Nor is gas-cleaning equipment considered an air pollu-
tion control system when it is used internally in a manufacturing process.
Further, emission control systems which are justified on the basis of
reduced plant maintenance (such as the avoidance of excess wear in
bearings or the reduction in cost to clean the plant environment from
emitted dust) are not defined as "air pollution control systems" (econo-
mists consider these costs internal to the firm. Air pollution, as de-
fined here, is an externality. Emissions are controlled when receptors
external to the firm are in need of protection).
The EGA Report accounts for the entire emission control investment,
including the cost of control hardware, auxiliary equipment, and installa-
tion, but does not generally include the cost of the pollutant disposal
system. The incremental costs of low-sulfur fuels and of boiler conver-
sions are also included as air pollution control costs.
It is also important to acknowledge that the EGA Report does not
include other resources demanded by air pollution control legislation,
i.e., the demand for surveillance instrumentation and staff, other control
agency resources, and other equipment and manpower requirements of the
private sector.
Figure 5-3 illustrates the sector of the gas-cleaning market that is
analyzed in this study. As shown, the demand study examines the control
systems needed to control emissions from the major polluting sources
affected by emission standards (indicated by the shaded area. The rela-
tive size of the shaded area in no way indicates the relative size of the
market for the resources. It is believed that the costs for the systems
identified in the EGA Report are a sizable portion of the total gas
cleaning market). In-plant and process gas-cleaning systems are not
considered. Also, minor sources of atmospheric emissions (such as small
lumber yards, construction and demolition activities, small fabricating
plants, etc.) have not been analyzed. Finally, minor polluting processes
or major industries which are generally controlled because of localized
"plant nuisances" are not analyzed since, in general, these processes are
not affected by air pollution control regulations but are controlled
because of "good company practice."
-------
FIGURE 5-3
THE SCOPE OF THE DEMAN) STUDY
CONTROL SYSTEMS
MAJOR EMISSION
RCE CATEGORIES
CONTROL SYSTEMS FOR
MAJOR POLLUTING
PROCESSES
CONTROL SYSTEMS
FOR MINOR POLLUTING
PROCESSES
CONTROL SYSTEMS
FOR MINOR EMISSION
SOURCE CATEGORIES
IN-PLANT AND PROCESS
GAS CLEANING SYSTEMS
ATMOSPHERIC
GAS CLEANING
SYSTEMS
-------
5.2.3 The Economics of Clean Air Control Cost Estimating Procedure
Procedures have been developed by APCO's Division of Economic
Effects Research (DEER) for estimating control costs for source categories.
The analytical procedures combine the best available engineering and cost
data with assumptions as necessary to arrive at the desired control cost
estimates. A general cost estimating procedure has been identified as the
basis for the development of all control cost estimates. Because varying
amounts of input data are available for each source category under study,
modifications to the procedure have been created to compensate for data
inadequacies. First, the general procedure is presented, followed by a
description of the specific methods used for specific source categories.
A flow chart which illustrates the general procedure is shown in Figure
5-4.
The data and assumptions which are inputs to the estimating proce-
dure are listed along the left-hand side of the flow chart. A general
description of each input category follows:
Identification of Emission Sources - Data which
identify all major and minor sources of specific
pollutants in each selected source category.
Source Engineering Information - Data on the type
of process, process production rate or capacity,
type and quantity of fuel burned, and other
source characteristics.
Emission Factors - Engineering factors which
relate the potential rate of process emissions
to process size.
Assumed Current Level of Control - Data or
assumptions on the current emission control
practices of a source or source category.
Emissions Standard - An emission control
regulation for each process category assumed
adequately stringent to comply with commonly
accepted air quality standards.
Control System Technical Characteristics and
Availability - Data on the alternative control
systems for a given source including the
-------
FIGURE 54
GONTOL COST ESTIMATING PROCEDURE IN THE COST OF CLEAN AIR REPORT
IDENTIFICATION
OF EMISSION
SOURCES
SOURCE
ENGINEERING
INFORMATION
EMISSION
FACTORS
ASSUMED
CURRENT
LEVEL OF
CONTROL
ESTIMATE OF
EXISTING EMISSIONS
EMISSION
STANDARD
REQUIRED
CONTROL
EFFICIENCY
ESTIMATE OF
CONTROLLED EMISSIONS
CONTROL SYSTEM
TECHNICAL
CHARACTERISTICS
AND AVAILABILITY
SELECTION Of
CONTROL
SYSTEM
CONTROL SYSTEM
COST FUNCTIONS
I
CONTROL
COST
ESTIMATE
-------
effectiveness of emission control and the
availability of the system or fuel for
solving the control problem.
Control System Cost Functions - Engineering
and cost data which relate the cost of emission
control to the capacity or production rate of a
given source type.
The top four input categories in Figure 5-4 are required to deter-
mine an estimate of the existing emissions from sources within a source
category. The emission standard determines the level of control efficien-
cy required to obtain a desirable air quality. A selection of the most
desirable control system (for the purpose of control cost estimates) is
made following an examination of the technical characteristics of alter-
native control systems, their availability, current industrial practices,
and other engineering assumptions. Finally, a control system cost func-
tion for the selected control alternative is used to determine the control
cost estimate.
As previously indicated, the statistics for identifying emission
sources and the descriptive data for such sources differ considerably for
the industries studied. Inconsistencies in the quality of input data re-
quire modifications to the basic cost-estimating procedure. Thus alter-
native estimating methods have been developed to cope with the problem of
data inconsistencies among industries.
For convenience in identifying the different methods, the following
terminology will be used:
Method Situation
I For multi-process plants where process
capacities are known.
II For single-process plants where the
plant capacity is known.
Ill A A "model plant" is used for a multi-
process plant (individual plant
capacities not known).
Ill B A "model plant" is used for a single-
process plant (individual plant
capacities not known).
-------
Method Situation
IV Gross regional characteristics are
used as a basis for estimation.
Figure 5-5 facilitates a better understanding of the specific
methods. As shown, air quality control regions contain some sources
which have been identified by name and/or location. Regions also contain
sources which have not been identified individually. In general, data
are available for industrial processes as well as steam-electric power-
plants and municipal incinerators. Domestic and commercial solid waste
disposal units as well as other stationary combustion sources (i.e., ex-
cluding powerplants) are not generally identified on a source-by-source
basis.
In general, the two types of sources which are identified by name
and location are (1) plants which have only one major process type which
is susceptible to emission control regulations, and (2) plants which have
more than one process. For a single process plant, there may or may not
be source engineering information on plant capacity. When such informa-
tion is available, control cost estimates are determined by the direct
application of a control cost function to the known plant capacity
(Method II). When plant capacities for an industry are not known, a
model plant is determined through engineering analysis. The model plant
is representative of the "typical" plant found in the industry. The
assumption is made that each plant within the industry has the character-
istics of the model plant and control cost estimates are based on that
assumption (Method III B).
For some industries with multi-process plants, statistics are avail-
able on the capacity of each polluting process. In that case, control
cost functions for these processes are directly applied (as a function of
the known process capacity) to determine control cost estimates (Method I).
When process capacities are not known for multi-process plants, a model
plant is determined. The model plant characteristics are consistent with
the plant design most commonly found within the industry. Control cost
estimates for such plants are based on the assumed mix of processes
(Method III A).
-------
FIGURE 5-5 DETAILED VIEW OF EGA COST ESTIMATING PROCEDURE
AIR QUALITY
CONTROL REGION
SOURCES IDENTIFIED BY
NAME OR LOCATION
SOURCES NOT !
INDIVIDUALLY IDENTIFIED I
\
MULTI-PROCESS
PLANTS
SINGLE PROCESS
PLANTS
HAVE WOWLEDGE OF
CAPACITY OF
INDIVIDUAL PROCESSES
HAVE KNOWLEDGE
OF PLANT CAPACITY
USE MODEL PLANT
CONTROL COST FACTORS
APPLIED AS A FUNCTION
OP oopULATiPr, PUEL USE/
'-'ASTF. OEM7ATED, r
IMVT DATA AfD RESULTING ESTIfWTES
COMPANY 5» COMPANY OR NUMBER OF PLANTS
TYPE PROCESS = TYPE PROCESS
PROCESS CAPACITY |« MODEL PLANT CAPACITY
ASSUMED OR KNOWN CONTROLS
(TYPE AND EFF.)
ANNUAL PART, ATE) SCI.
EMISSIONS *
(CONT. AND UNCONT,)
CONTROL INVESTMENT?!)
ANNUAL CONTROL COST($)
CONTROL SYSTEM
(TYPE AND EFF,)
TYPE CONTROL REGULATION
ASSUMED
I
ASSUMED CONTROLS
(TYPE AND EFF,)
ANNUAL PART. AND SOL
EMISSIONS
(CONT, AND UNCONT.)
CONTROL INVESTMBtfW
ANNUAL CONTROL COSTW
CONTROL SYSTEM
(TYPE AND EFF.)
TYPE CONTROL REGULATION
ASSUMED
CCMPAW
TYPE PROCESS
PLANT CAPACITY
ASSUMED OR KNOWN CONTROLS
(TYPE AND EFF.)
ANNUAL PART, AND SOY
EMISSIONS *
(COOT, AND UNCONT,)
CONTROL INVESTMENT^)
ANNUAL CONTROL COST($)
CONTROL SYSTEM
(TYPE AND EFF.)
TYPE COFOROL REGULATION
ASSUMED
1
i
POPULATION DATA
QUALITY OF SW GENERATED
* CONTROL SYSTEM
(TYPE AND EFF,)
CONTROL INVESTMENT^
ANNUAL CONTROL COST($)
ANNUAL PART, AND SO^
(CONT, AND UNCONT,)
NUMBER OF BOILERS:
RESIDENTIAL HEATING
COW, AND INSTIT,
INDUSTRIAL/ AND
QUANTITY OF FUELS BURNED
COAL (s carrerr)
OIL (s CONTENT)
NG
COWERSIOfJ COSTS
TYPE & QUAfrriTY OF FUEL SWITCH
SCURCE CATEGORIES
IRON AND STEEL i
PETROL, REFINING i
!
i
]
MUNICIPAL INCINERATORS COAL CLEAN IMG f OTHER SOLID WASTE DISPOSAL
» STEAM-ELECTRIC POWER = GRAIN MILLING i OTHER STATIONARY COMBUSTION
PRIMARY HONFERROUS SMELTERS s GRAIN HANDLING S o INDUSTRIAL BOILERS
SULFURIC ACID = PHOSPHATE FERTILIZER(PARTIALs «> COfl'icPCIAL-INSTITUTIONAL
PHOSPHATE FERTILIZER(PARTIAL) s ASPHALT BATCHING s HEATING PLANTS
jj» CB1ENT \ LIME S ° RESIDENTIAL HEATING
FGREY IRON FOUNDRIES SECONDARY MONFERROUS METAL 5 PLANTS
-------
When the identification of individual sources is not economically
justified for analytical purposes, regional characteristics, such as
population, fuel use, or quantity of solid waste generated, are used in
still another control-cost-estimating procedure. No sources are treated
individually. Here, the emission contribution from each source alone is
small even though the emission from the total aggregate of sources may be
substantial. Cost factors are applied to the regional characteristics to
determine the cost of control (Method IV).
The detailed cost and engineering factors utilized in the five
methods are not reported in the EGA Report. The relevant data and
assumptions are presented in the back-up report to the 1970 Cost of
Clean Air Report (Reference 4). The 1971 Report is also supported by
a technical back-up document (Reference 9).
5.2.4 305(a) Approach - Step #1
The procedures described in the previous section have been used to
develop the DEER control cost data bank. The first step in the 305(a)
Approach uses this data and structures it by source category according
to a desired format. The following data elements have been aggregated
for this report:
Air quality control region
Company or plant name
Process description
- type
- capacity
- number of units
Required controls
- type of system
- system efficiency
- alternative fuel
-------
- fuel type
- fuel quantity
Particulate and sulfur dioxide emission data
- existing quantity of emissions
- controlled quantity of emissions
Full additional control costs
- investment
- annual cost
This information is considered essential to characterize the
resources that will be demanded as a result of the emission control
regulations
The data for stationary source control of particulates and sulfur
dioxide emissions are presented for four AQCRs in Table 5-2 through 5-5.
The following source categories have been identified:
Solid waste disposal
Steam-electric power generating
Other fuel combustion sources
Industrial processes.
Footnotes at the base of each table indicate the specific cost-
estimating procedure utilized in the determination of the control cost
and emission estimates.
Notice that the'following source categories can serve as examples
where the cost estimating methods described in the previous section have
been applied:
Method Example Source Category
I Petroleum Refining
II Sulfuric Acid
III A Coal Cleaning
III B Asphalt Batching
IV Other Fuel Combustion Sources
Also identified in some of the tables are descriptions of the type
and efficiency of the control systems assumed to be currently on each
source. The assumptions are required for calculating the quantity of
existing emissions as well as in determining the adequacy of the existing
-------
Table 5-2. SOLID WASTE DISPOSAL
PAGE 1 OF 2
AQCR
Cinti
Wash.
PROCESS
TYPE
Municipal
ii
M
1 1
it
M
ii
M
II
II
11
M
II
II
II
" (3!
Dom. & Conun.
Municipal
M
tl
II
II
"
It
II
Dom. & Commsr
CAPACITY
(TON/DAY)
100
50
50
200
180
50
300
19
50
500
60
150
300
500
150
120
125
425
300
500
750
500
1,050
300
YEAR BUILT
1929
1930
1930
1931
1931
193A
1935
1937
1949
1954
1957
1960
1962
1965
1965
1932
1932
1949
1955
1957
1961
1965
1967
ADDITIONAL
CONTROLS REQUIRED
TYPE
Wet Scrubber
"
M
11
II
II
II
M
"
II
II
It
II
II
It
II
It
Wet Scrubber
11
II
II
II
It
M
ir
"
EFFICIENCY
(%)
85
85
85
85
85
85
85
85
85
90
85
85
85
90
85
85
85-95
85
87.5
85
90
95
90
95
85
85-95
FULL ADDITIONAL (1)
CONTROL COST
INVESTMENT
($1000)
50
25
25
100
90
o =;
t.~i
15C
10
25
250
30
75
120
100
30
24
1,542
63
213
150
250
375
200
210
60
4,039
ANNUAL
($1000/YR.;
36
18
18
72
65
18
108
7
18
180
22
54
99
155
47
37
683
45
153
108
180
270
165
325
43
1,831
PARTICIPATE EMISSIONS
(TONS/YR.)
(2)
EXISTING
63.8
31.9
31.9
127.6
114.8
31.9
191.4
121.2
31.9
319.0
38.3
95.7
191.4
319.0
95.7
76.6
14,813.0
79.8
271.2
191.4
319.0
478.5
319.0
669.9
191.4
33,791.0 -
CONTROLLED
38.2
19.1
19.1
76.4
68.8
19.1
114.6
7.3
19.1
127.5
23.0
47.3
114.6
127.5
47.3
45.9
1,899.0
47.7
134.2
114.6
127.5
95.5
127.5
134.0
114.6
3,584.0
(1) Control costs for municipal incinerators based on known capacity and "year built."
(2) Assumed existing incinerators controlled by inertial separator (75% efficient).
(3) Domestic and commercial category includes costs and emissions of new incinerators required for disposal of some uncollected and collected
-------
Table 5-2. SOLID WASTE DISPOSAL
' PAGE 2 OF 2
AQCR
St. Louis
Phila.
PROCESS
TYPE
Municipal
(3)
Dom, & Comm.
Municipal
ii
it
it
M
ii
ii
ii
ii
ii
ii
ii
ii
it
it
ii
ii
ii
ii
ii
ii
ii
(3)
Dom. & Comm.
CAPACITY
(TON/DAY)
500
500
40
25
97
200
300
600
100
50
450
200
200
600
100
300
600
300
500
500
500
300
300
250
600
.__
YEAR BUILT
1950
1958
1920
1927
1938
1950
1950
1951
1954
1954
1955
1955
1955
1956
1958
1959
1960
1961
1961
1961
1962
1962
1964
1965
1965
~~"
ADDITIONAL
CONTROLS REQUIRED
TYPE
Wet Scrubber
it
ii
ii
it
it
ii
ii
IT
11
II
II
II
II
It
II
II
II
II
II
II
II
II
II
II
Tl
II
EFFICIENCY
(%)
90
90
85-95
85
85
85
85
85
90
85
85
90
85
85
90
85
85
90
85
90
90
90
85
85
85
90
85-95
FULL ADDITIONAL (1)
CONTROL COST
INVESTMENT
($1000)
250
250
1,739
20
13
49
100
150
300
50
25
225
100
100
300
50
150
300
120
200
200
200
120
120
50
120
7,984
ANNUAL
($1000/YR.)
180
180
1,094
14
9
35
72
108
216
36
18
162
72
72
216
36
108
216
99
165
165
165
99
99
78
186
3,200
PARTICULATE EMISSIONS
(TONS/YR.)
EXISTING^
319.0
319.0
22,340.0
25.5
16.0
61.9
127.6
191.4
382.8
63.8
31.9
287.1
127.6
.127.6
382.8
63.8
191.4
382.8
191.4
319.0
319.0
319.0
191.4
191.4
159.5
382.8
53,855.0
CONTROLLED
127.5
127.5
1,968.0
15.3
9.5
37.1
76.4
114.6
153.0
38.2
19.1
115.0
76.4
76.4
153.0
38.2
114.6
153.0
114.6
127.5
127.5
127.5
114.6
114.6
95.5
153.0
5,790.0
-------
Table 5-3. STEAM-ELECTRIC POWER
AOCR
Cinti.
Wash.
St. Louis
Phila.
Plant
W.C. Beckjord
West End
Tanners Creek
Miami Fort
Benning Road
Buzzard Point
Possum Point
Potomac River
Dickerson
Chalk Point
Wood River
Cahokia
Venice It 2
Meramec
Ashley
Edgemoor
Chester
Cromby
Delaware
Eddystone
Richmond
Schuylkill
Southwark
Deepwater
Burlington
Barbadoes
Delaware City
Capacity
(Megawatts)
760.5
219.3
1,098.0
263.8
270.0
491.0
650.1
300.0
500.0
800.0
389.8
256.0
417.5
499.3
707.2
474.8
325.4
345.0
Fuel Sulfur
Content
COal (%)
1.50
1.50
2.80
1.50
1.50
1.50
1.00
1.50
1.80
1.80
3.30
3.27
3.27
2.68
2.68
2.75
2.17
2.17
2.17
2.17
2.17
2.17
2.17
2.75
3.00
2.17
0.00
Oil (%)
1.36
1.70
1.36
1.36
1.36
0.50
2.17
2.17
1.70
1.70
0.00
1.70
1.70
1.48
2.75
2.75
2.75
2.75
2.75
2.75
2.75
2.50
2.75
2.75
1.48
Additional
Controls
Required
Wet Limestone
Scrubbing
it
ii
H
ii
it
it
n
it
ii
ii
M
n
n
it
"
II
11
It
II
II
It
II
II
II
II
tl
Emission Data (Ton/Yr . ) (D (2)
Particulates
Existing
17,275
1,287
8,656
7,562
3,871
1,814
7,208
8,148
744
10
14,178
1,087
6,678
21,665
190
7,233
506
6,796
7,302
13,911
5,316
9,030
10,939
4,049
178
2,189
13
Required
1,233
92
261
540
277
130
514
582
53
1
1,013
1,087
476
1,547
190
517
36
485
522
994
380
645
781
289
13
2,189
13
Sulfur Dioxide
Existing
48,034
3,578
111,313
21,027
13,381
6,258
17,814
30,212
2,888
372
84,797
6,455
39,577
102,947
903
42,138
3,912
27,973
31,058
57,257
24,734
37,166
45,022
24,983
5,610
9,219
332
Required
4,803
358
11,131
2,102
1,338
626
4,454
3,024
289
37
8,480
1,971
3,958
10,295
337
4,214
391
2,797
3,106
5,726
2,473
3,717
4,502
2,498
561
4,162
51
Incremental. Control Cost
Investment .
($1000)
7,490(3)
2,518
11,479
5,650
6,394
3,938
6,303
5,944
5,784
6,907
7,504
0
6,111
10,189
0
4,725
Annual
($1000/Yr.)
2,410(3)
504
3,700 '
1,801
1,669
770
1,217
1,772
349
303
2,516
100
1,666 l
3,191
15
1,473
3,004 855
4,289
6,168
6,822
6,558
4,951
5,424
5,192
5,993
0
0
1,307
1,869
2,112
2,030
1,523
1,656
1,838
1,883
311
338
(1) Particulate emission approximately 80% controlled for all sources.
(2) Sulfur oxides are essentially uncontrolled.
-------
Table 5-4. OTHER FUEL COMBUSTION SOURCES
AQCR
Cinti.
Wash.
St. Louis
Phila.
COMBUSTION
TYPE
Industrial
Comm. & Inst,
Residential
Industrial
Conm. & Inst
Residential
Industrial
Comm. & Inst
Residential
Industrial
Comm. & Inst.
Residential
UNITS
NUMBER
1,670
7,733
485,383
704
13,419
800,034
3,213
11,348
708,200
10,127
29,768
1,529,512
EXISTING FUEL COMBUSTION
NAT. GAS
QUANTITY
(106cf.)
0
16,755
34,288
0
8,677
28,436
0
22,604
36,789
0
44,367
63,630
j
COAL
QUANTITY
(103tons)
840 (1)
65.3(2)
267.8(2)
. Neg.
50.0(2)
484.6(2)
2,050 (1)
0.0(2)
646.4(2)
3,700 (1)
124.0(2)
459.4(2)
SULPHUR
CONTENT (%)
2.0
1.0
1.1
1.6
0.6
3.5
1.0
2.5
1.8
0.6
OIL
QUANTITY
(106gal.)
15
30.9
67.8
Neg.
190.2
197.2
111
49.3
100.9
911
340.8
758.2
SULPHUR
CONTENT (%)
1.75
.8
.5
1.75
.8
.5
1.75
.8
.5
1.75
.8
.5
REQUIRED FUEL CONSUMPTION
NAT. GAS
QUANTITY
(106cf.)
0
310
4,980
0
5
29
0
0
10,410
0
1,960
3,950
OIL
QUANTITY
106gal.)
138
9
9
0
580
4,620
337
0
26
608
7
43
SULPHUR
CONTENT (Z)
1.0
0.2-0.75
0.2-0.5
1.0
0.2-0.75
0.2-0.5
1.0
0.2-0.75
0.2-0.5
1.0
0.2-0.75
0.2-0.5
EMISSION DATA
PARTICIPATES
EXISTING
(TONS)
37,100
854
2,936
Neg.
1,975
5,410
87,600
334
6,990
174,000
3,540
8,710
REQUIRED
(TONS)
1,600
264
420
Neg.
1,309
910
9,800
334
665
15,800
2,350
4,400
SULFUR DIOXIDE
EXISTING
(TONS)
33,600
3,100
8,600
Neg.
13,030
12,600
145,000
2,820
18,800
300,000
23,930
35,050
REQUIRED
(TONS)
12,200
2,280
3,000
Neg.
11,090
9,000
40,000
2,820
5,050
122,000
20,200
31,000
FULL ADDITIONAL
CONTROL COST (4)
INVESTMENT
($1000)
8,826
726
(3)
Neg.
671
(3)
8,894
(3)
15,609
1,426
(3)
ANNUAL
($1000)
7,962
- 410
(3)
Neg.
1
(3)
16,052
(3)
10,445
38
(3)
(1) Ash content - 8%.
(2) Emission factor for particulates = 200/ton of coal.
(3) The transition from residential burning of coal to oil, natural gas and electricity is not directly a result of
concern over air pollution. No control costs have been allocated for the switch from coal to other fuel because
of the long observed trend away from coal.
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: IRON AND STEEL
INDUSTRY CODE:
06
PACK 2 OF 3
AQCR
Phila.
||
If
It
tt
II
II
If
If
tl
M
II
It
COMPANY
U. S. Steel
tl
Luken Steel
it
ii
Phoenix Steel
Allanwood St.
tt
ii
M
II
M
PROCESS
TYPE
>pen Hearth
Sintering
)pen Hearth
Slectric
ii
)pen Hearth
ii
Sleet ric
Electric
it
it
-
Sintering
CAPACITY
(Ton/Melt)
395
6,000.00
145
145
100
150
140
18
45
40
11
23
NUMBER
9-
)
6
1
2
5
9
1
1
1
1
1
2,OQO.OO|)
ADDITIONAL
CONTROLS REQUIRED
TYPE
40" w.g.
Venturi
20" w.g.
Venturi
50" w.g.
Venturi
H.T. Fabric
Filter
It
40" w.g.
Venturi
"
H.T. Fabric
Filter
ii
ii
20" w.g.
Vpnturi
EFFICIENCY
96.4
98.9
92.3
92.0
69.0
92.3
92.3
84.0
87.0
87.0
84.0
84.0
97.0
FULL ADDITIONAL (3)
CONTROL COST
INVESTMENT
(S1000)
16,517
2,879
4,748
4,693
\
5,694
/
1,204
ANNUAL
(S1000/YR.)
6,656
2,551
2,167
1,947
' 2,367
968
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(4)
12,917
20,279
5,852
3,797
4,578
6,759
REQUIRED
619
344
914
386
476
300
SULFUR DIOXIDE
EXISTING
(2)
(2)
(2)
(2)
(2)
(2)
REQUIRED
(2)
(2)
(2)
(2)
(2)
(2)
oo
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY:
IRON AND STEEL
INDUSTRY CODE:
06
PACK 3 OF 3
AQCR
Phila.
II
II
II
II
II
II
COMPANY
Midvale-
Heppenstall
it
»
"
"
Phoenix Steel
PROCESS
TYPE
Electric
ii
"
"
it
it
CAPACITY
(Ton/Melt)
30
50
100
25
5
150
145
NUMBER
1
1
1
1
2
6
1
ADDITIONAL
CONTROLS REQUIRED
TYPE
H.T. Fabric
Filter
II
II
11
II
"
EFFICIENCY
87.0
87.0
89.0
84.0
84.0
92.0
92.0
FULL ADDITIONAL (3)
CONTROL COST
INVESTMENT
($1000)
\
596
/
( 3,396
ANNUAL
(S1000/YR.)
388
1,396
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(4)
1,521
2,725
REQUIRED
455
277
SULFUR DIOXIDE
EXISTING
(2)
(2)
REQUIRED
(2)
(2)
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: IRON AND STEEL
INDUSTRY CODE:
06
PACK
OF
AQCR
Cinti.
ii
IT
II
II
tl
It
St. Louis
tt
ii
ii
it
tt
COMPANY
Interlake Steel
ii
it
Armco Steel
ii
it
Am. Compressed
Steel
Granite City
tt
u
tt
u
u
PROCESS
TYPE
Open Heartl
u
Electric
Open Heartl
u
it
Electric
Electric
u
u
u
u
Sintering
CAPACITY
(Ton/Melt)
78
97
75
180
190
300 .
6
250
300
400
500
600
3,238.6(1:
NUMBER
2
5
3
3
3
6
1
2
2
1
1
1
ADDITIONAL
CONTROLS REQUIRED
TYPE
40" w.g.
Venturi
u
H.T. Fabric
Filters
40" w.g.
Venturi
ti
u
H.T.' Fabric
Filter
H.T. Fabric
Filter
II
tt
tl
It
20" w.g.
Venturi
EFFICIENCY
CO
90.0
90.0
89.0
94.0
94.0
95.6
84.0
95.0
95.0
95.0
95.0
95.0
98.0
FULL ADDITIONAL (3)
CONTROL COST
INVESTMENT
($1000)
3,492
5,060
33
\
12,138
i
1,785
ANNUAL
(S1000/YR.)
1,575
2,086
22
4,892
1,503
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(4)
4,084
4,033
46
\
»
/
9,447
10,947
REQUIRED
740
346
21
459
308
SULFUR DIOXIDE
RXISTING
(2)
(2)
(2)
(2)
(2)
REQUIRED
(2)
(2)
(2)
(2)
(2)
(1) Capacity for sintering reported in tons per day.
(2) Sulfur dioxide emissions considered negligible relative to applicable emissions standards.
(3) Control cost estimates based on known capacity of processes. Total Production Average Control
(A) All basic oxygen furnaces assumed controlled to 95%; also, the following were assumed: Process Controlled (%) Ef ficiency(%)
Open Hearth 27 90
Electric 61 90
Sintering (windbox) 90 75
Sintering (discharge) 0 0
Ln
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: GREY IRON FOUNDRY
INDUSTRY CODE:
07
PACK 1 OF 5
AQCR
Cinti.
11
11
II
M
II
II
. II
fl
II
II
COMPANY
Star Foundry
Klaene Foundry
Independence
Foundry
Reliance Fdy.
Chris Erhart
Foundry
Cinti. Milling
Standard
Castings
Buckeye Fdy.
Oberhelman -
Ritter
Tri-State Fdy.
Black
Clawson (1)
H.P. Deuscher
PROCESS
TYPE
Cupola
11
H
ii
ii
ii
ii
it
it
ti
M
CAPACITY
(Ton/Hr.)
(1)
tt
.11
11
II
II
tt
M
II
II
It
II
NUMBER
(D-
II
II
M
II
II
II
II
II
II
II
It
ADDITIONAL
L CONTROLS REOUTRF.D
TYPE
16-35" w.g.
Venturi
Scrubbers
M
H
M
H
ii
ii
it
ii
H
11
u
EFFICIENCY
(%)
85-95%
11
II
II
tt
"
II
"
11
"
It
II
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
48
62
48
75
73
120
68
105
103
91
126
95
ANNUAL
(S1000/YR.)
14
18
14
22
22
36
20
31
31
27
38
29
EMISSION' DATA (TON/YEAR)
PARTICULATES
EXISTING
(2)
30
35.9
3.0
67.9
62.8
221.1
51.1
135.2
127.7
84.5
209.3
99.5
REQUIRED
0.3
4.1
0.3
7.7
7.1
61.4
5.8
15.4
14.5
9.6
23.8
11.3
SULFUR DIOXIDE
EXISTING
Neg.
U
II
It
II
II
II
M
"
"
II
REQUIRED
Neg.
11
tl
tt
II
tl
"
II
"
M
tl
II
(1) Capacity per plant not known; cost and emission estimates based on known "value of shipments" and "number of cupolas" for each plant.
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: GREY IRON FOUNDRY
INDUSTRY CODF: °7
PACK 2 OK
AQCR
Cinti.
n
M
>t. .Louis
it
M
ti
ii
it
M
COMPANY
Hamilton Fdy.
Campbell-
Hausfeld
Black -
Clawson (2)
Spuck Iron and
Foundry
Laclede Stoker
Banner Iron
Works (1)
Liberty Fdy.
Didion & Sons
Banner Iron
Works (2)
ABEX
PROCrSS
TYPE
Cupola
ii
ii
ii
ii
H
ii
H
H
ii
CAPACITY
(Ton/Hr.)
(1)
II
II
II
II
II
II
II
II
II
NUMBER
(1.)
II
tl
M
II
II
II
M
II
II
ADDITIONAL
CONTROLS REQUIRED
TYPE
16-35" w.g.
Venturi
Scrubbers
ii
H
16-35" w.g.
Venturi
Scrubbers
n
M
M
II
II
II
EFFICIENCY
(%)
85-95%
IT
It
85-95%
n
n
ti
M
n
ti
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
320
99
92
63
90
0
0
70
0
82
ANNUAL
CSIOOO/YP.)
96
30
28
19
27
0
0
21
0
25
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(2)
421.0
128.2
111.5
38.7
106.4
19.2
18.4
54.1
19.2
84.5
REQUIRED
51.3
14.6
12.7
4.4
12.1
19.2
18.4
6.1
19.2
9.6
SULFUR DIOXIDE
EXISTING
Neg.
||
II
II
II
II
tl
II
II
M
REQUIRED
Neg.
tl
tl
II
It
IT
II
II
II
11
Ul
N)
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: GREY IRON FOUNDRY
INDUSTRY CODE: °7
PACK 4 OK
AQCR
Phila.
tl
II
It
tl
II
tl
II
II
II
II
COMPANY
Bonair Fdy.
Schneider
Bowman
Glode
Enterprises
H.G. Enderlein
Perserverance
Iron
Pottstown Mach.
March Brownbaci
Keystone Grey
Iron Foundry
Buffalo Pipe &
Foundry
Royersford Fdy.
Richmond Ring
PROCESS
TYPE
Cupola
it
ti
it
ii
ii
it
it
it
it
CAPACITY
(Ton/Hr.)
(1)
II
II
II
II
II
II
M
II
II
II
NUMBER
(1)
II
II
II
II
II
II
II
II
II
It
ADDITIONAL
CONTROLS RKOUIRFD
TYPE
16-35" w.g.
Venturi
Scrubber
ii
M
ti
ti
ii
n
ii
n
M
it
EFFICIENCY
CO
85-95%
It
II
II
II
II
It
II
It
tl
It
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
64
81
70
72
55
58
75
64
88
62
40
ANNUAL
(S1000/YR.)
19
24
21
22
16
17
23
19
26
19
14
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(2)
40.7
82.0
10.9
59.3
17.7
24.7
67.6
39.9
101.4
34.7
1.0
REQUIRED
4.6
9.3
1.2
6.7
2.0
2.1
7.7
4.5
11.5
3.9
0.1
SULFUR DIOXIDE
EXISTING
Neg.
"
II
II
II
II
||
It
"
11
II
REQUIRED
Neg.
"
It
II
M
It
It
tl
tl
U
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: GREY IRON FOUNDRY
INDUSTRY CODF.:
07
PACK 5 OF 5
AQCR
Phila.
1)
II
COMPANY
Spring City
Foundry
Stanley G Flag
ABEX
PROCESS
TYPE
Cupola
ii
it
CAPACITY
(Ton/Hr.)
(1)
II
II
NUMBER
(1)
"
11
ADDITIONAL
CONTROLS REQUIRED
TYPE
16-35" w.g.
Venturi
Scrubber
ii
M
EFFICIENCY
(%)
85-95%
1 1
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
63
181
91
ANNUAL
(S1000/YR.)
19
55
27
EMISSION' DATA (TON/YEAR)
PARTICULATES
EXISTING
(2)
36.5
405.1
107.5
REQUIRED
4.1
46.1
12.2
SULFUR DIOXIDE
FX I STINT,
Neg.
"
II
REQUIRED
Neg.
u
Ln
-------
INDUSTRY: ftEFY TROW
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY CODE: QT_
PACK 3 OK 5
AQCR
St. Louis
it
ii
II
II
Phila.
If
II
M
II
It
COMPANY
Excelsior Fdy.
Buckler Fdy.
East St. Louis
Castings
Modern Fdy. &
Mfg.
Autocrat
Burlington
Beloit
Howmet
Seamless
Modena
King Founda-
ries
Link-Belt
PROCESS
TYPE
Cupola
ii
"
"
II
II
II
II
II
II
It
CAPACITY
(Ton/Hr.)
(1)
"
ii
M
II
t|
tt
II
II
II
II
NUMBER
(1)
||
"
II
It
II
II
II
tt
"
II
ADDITIONAL
CONTROLS REQUIRED
TYPE
16-35" w.g.
Venturi
Scrubber
ii
"
tl
II
16-35" w.g.
Venturi
Scrubber
tt
ti
ti
M
"
EFFICIENCY
(%)
85-95%
II
II
"
11
85-95%
II
tl
"
II
"
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
68
53
138
67
67
50
102
102
74
56
0
ANNUAL
(S1000/YP.)
21
16
42
20
20
15
30
30
22
17
0
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(2)
50.7
12.6
250.5
47.3
47.0
5.1
135.2
135.2
24.0
19.1
15.4
REQUIRED
5.8
1.4
28.5
5.4
5.3
0.6
15.4
15.4
2.7
2.2
15.4
SULFUR DIOXIDE
EXISTING
Neg.
tl
tl
tl
II
II
ri
fi
ti
M
II
REQUIRED
Neg.
II
II
tl
II
.1
Tl
"
tl
"
"
Ul
Ul
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: PRIMARY NONFERROUS METALS
INDUSTRY CODF: 08
PACK 1 OF1
AQCR
St. Louis
"
tl
COMPANY
Lewis Mathes
Co.
St. Joseph
Lead
Am. Zinc of
111.
PROCESS
TYPE
Copper
Lead
Zinc
CAPACITY
(TON/YR. )
42,500
200,000
69,000
NUMBER
1
1
1
ADDITIONAL
CONTROLS REQUIRED
TYPE
(1)
ir
ii
EFFICIENCY
(%)
__
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
ANNUAL
(S1000/YR.)
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
38
1,140
166
REQUIRED
38
1.140
166
SULKIR DIOXIDE
EXISTING
Neg.
4,500
1,450
REQUIRED
Neg.
4,500
1,450
Ul
(1) Current controls for copper, lead and zinc smelters indicated assumed 95% efficient for particulate and 49.9 efficient for sulfur dioxides,
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY:
SULFURIC ACID
INDUSTRY CODK:
09
PACK l Of l
AQCR
Cinti.
II
"
t
St. Louis
it
it
Phila.
11
tl
COMPANY
American
Cyanamide
E. I. Dupont
International
Min. & Chem.
Mobile Chem.
Charles Pfizer
American Zinc
National Lead
Atlantic-
Richfield
E. I. Dupont
Rohm & Haas
PROCESS
TYPE
Contact
n
it
it
n
ti
n
"
It
II
CAPACITY
(Ton/Day)
90
175
45
25
5
146
500
140
70
88
NUMBER
1
4
II
It
II
II
It
II
»
II
ADDITIONAL
CONTROLS REQUIRED
TYPE
Secondary
Absorption
Tower and
Demister
M
n
it
ti
it
it
M
it
it
EFFICIENCY
(%)
(3)
It
It
II
II
It
II
II
II
II
FULL ADDITIONAL
CONTROL COST (1)
INVESTMENT
($1000)
501
765
329
229
58
675
1,469
656
435
495
ANNUAL (2)
(SIOOO/Y?.)
109
168
71
49
14
148
343
144
95
108
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(4)
174
338
87
48
10
282
967
271
135
170
REQUIRED
106
207
53
29
6
172
591
165
82
104
SULFUR DIOXIDE
EXISTING
(4)
1,780
3,461
890
494
99
2,088
9,890
2,769
1,384
1,740
REQUIRED
249
484
124
69
14
404
1,384
387
193
243
(1) Control cost and emission estimates based on known capacity of plants.
(2) Annual cost does not include credit for additional sulfuric acid recovered.
(3) Required sulfur dioxide control of 86%; particulate control range from 43% to 93% according to plant size.
-------
- Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: PHOSPHATE FERTILIZER
INDUSTRY CODE: 10
PACK 1 OK 1
AQCR
Cintl.
"
II
St. Louis
II
Phila.
COMPANY
Inter. Minerals
& Chemicals
Mobil Chem.Co.
Tennessee Corp
FS Services, In
National
Stockyard
Kerr-McGee
Chemical
PROCr.SS
TYPE
N.S.(2)
N.S.
N.S.
c N.S.
N.S.
N.S.
CAPACITY
;iOOO Ton/Yr
35e (2)
35e
100
60
35e
35e
NUMBER
1.
1
1
1
1
1
ADDITIONAL
CONTROLS REQUIRED
TYPE
(3)
M
II
11
II
II
EFFICIENCY
(%)
(3)
"
II
II
II
II
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
(3)
II
II
"
II
II
ANNUAL
(S1000/YP.)
(3)
M
it
ti
it
M
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
3,150
3,150
9,000
5,400
3,150
3,150
REQUIRED
3,150
3,150
9,000
5,400
3,150
3,150
SULKL'R DIOXIDE
rx IST INC
Neg.
II
It
II
"
II
REQUIRED
Neg.
11
it
it
"
M
(1) Normal superphosphate = N.S.
(2) "f" indicates plant capacity not known, thus, model plant of 35,000 Ton/Yr. was assumed for emission and control cost estimates;
otherwise, estimates based on known plant capacity.
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: PETROLEUM REFINING
INDUSTRY CODE:
11
PACK
OF
AQCR
Cinti.
M
II
St. Louis
i
i.
tl
"
If
M
11
COMPANY
Chevron Asphalt
Gulf Oil Corp
it
Amer.Oil Co.
ii
Clark Oil & Ref
II
Mobile Oil Co.
II
Shell Oil Co.
PROCESS
TYPE
Refinery
Refinery
Cat
Cracking
Refinery
Cat
Cracking
Refinery
Cat
Cracking
Refinery
Cat
Cracking
Refinery
CAPACITY
(b/cd)
11 , 000
40,700
27,000(1)
86,500
65,000(1)
35,000
14,000(1)
47,000
31,000(1)
194,000
NUMBER
1
1
1-
1
1
1
1
1
1
1
ADDITIONAL
CONTROLS REQUIRED
TYPE
S.R.P. (2)
S.R.P. (3)
H.E.E.P. (3)
S.R.P.
H.E.E.P.
S.R.P.
H.E.E.P.
S.R.P.
H.E.E.P.
S.R.P.
EFFICIENCY
(%)
50
50
82
50
82
82
50
82
50
FULL ADDITIONAL
CONTROL COST (4)
INVESTMENT
($1000)
130
130
260
240
120
100
190
450
ANNUAL
(S1000/YP.)
25
44
20
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(5)
Neg.
Neg.
30
Neg.
70
Neg.
14
Neg.
3
Neg.
REQUIRED
Neg.
Neg.
6
Neg.
12
Neg.
2
Neg.
Neg.
Neg.
SULFUR DIOXIDE
EXISTING
(5)
1,800
6,000
Neg.
14,400
Neg.
6,000
Neg.
10,000
Neg.
32,000
REQUIRED
1,800
3,000
Neg.
7,200
Neg.
3,000
Neg.
5,000
Neg.
16,000
VO
(1)
(2)
(3)
(4)
(5)
Capacity reported in b/sd (barrels per stream day).
= Sulfur recovery plant for control of sulfur dioxide emissions.
= High efficiency electrostatic precipitator for control of particulate emissions.
Control cost and emission estimates based on known process size.
Assumed average control of particulates from catalyst regenerator of 67%; refining existing controls of 67% particulate emission
-------
INDUSTRY: PETROLEUM REFINING
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY CODE: U
PACK 2 OF 2
AQCR
Phila.
it
11
1 11
it
ti
it
ti
COMPANY
Atlantic
Richfield
II
Sun Oil Co.
II
Sinclair Ref.
II
Sun Oil Co.
II
PROCESS
TYPE
Refinery
Cat
Cracking
Refinery
Cat
Cracking
Refinery
Cat
Cracking
Refinery
Cat
Cracking
CAPACITY
(b/cd)
155,000
40.000C
158,300
78,500 C
104,000
44,000 (
158,000
90,000 (:
NUMBER
1
) 1
1
) 1
1
) 1
1
) 1
ADDITIONAL
CONTROLS REQUIRED
TYPE
(6)
H.E.E.P. (3)
(6)
H.E.E.P.
(6)
H.E.E.P.
(6)
N.E.E.P.
EFFICIENCY
(%)
82
82
82
82
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
180
260
200
160
ANNUAL
CS1000/YP.)
36
56
40
32
EMISSION' DATA (TON/YEAR)
PARTICULATES
EXISTING
Neg.
44
Neg.
86
Neg.
56
Neg.
38
REQUIRED
Neg.
8
Neg.
14
Neg.
10
Neg.
7
SULFUR DIOXIDE
EX 1ST INC
13,000
Neg.
14,000
Neg.
8,600
Neg.
13,200
Neg.
REQUIRED
13,000
Neg.
14,000
Neg.
8,600
Neg.
13,200
Neg.
See Page 1 of 2 for footnotes (1), (2), (3), (4) and (5).
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: ASPHALT BATCHING
INDUSTRY CODE:
12
PACK OF
AQCR
Cinti
Wash.
St. Louis
Phila.
COMPANY
Model Plant
(1)
If
II
II
PROCESS
TYPE
Asphaltic
Concrete
it
it
it
ii
CAPACITY
(Ton/Yr . )
510,000
10,200,000
7,650,000
3,060,000
15,810,000
NUMBER
1
20
15
6
31
ADDITIONAL
CONTROLS REQUIRED
TYPE
10" w.g. (2)
Venturi
Scrubber
ii
n
ii
n
EFFICIENCY
00
98.2
II
II
tt
II
FULL ADDITIONAL
CONTROL COST (3)
INVESTMENT
($1000)
14.0
280
210
84
434
ANNUAL
($1000/YR.)
11.3
226
170
68
350
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(2)
870
17,400
13,050
5,220
26,970
REQUIRED
57
1,140
855
342
1,767
SULFUR DIOXIDE
EXISTING
(2)
Neg.
Neg.
ii
ii
it
REQUIRED
Neg.
Neg.
"
II
II
(1) A list of the asphalt batching plants which operated in the selected AQCR's in 1967 may be found on the following page.
(2) All asphaltic concreteplants were assumed to have process cyclones, reducing particulate emissions by 80%
-------
ASPMLT BATCHING
AQCR
PLANT NAME
Phila.
D. C.
Union Paving Co.
Eastern Asphalt Co. (2 plants)
Flintkote Co.
Philgite Co., Inc.
Warner Co.
Dover Equipment and Machine
Asphalt Industries, Inc.
Miller Quarries, Inc. (2 plants)
Interstate Amiesite Corp.
Ackworth Materials Corp.
Ackworth Materials Corp.
Asphalt Paving and Supply Co.
Newport Paving Supply Co.
JDM Materials Co., Inc.
Highway Materials, Inc.
Newport Paving Supply Co.
Bucks Co. Asphalt Co.
Bucks Co. Asphalt Co.
Highway Materials Co.
Highway Materials Co.
Union Paving Co.
M & M Stone.Co., Inc.
Eureka Stone Quarry, Inc.
Union Paving Co.
Barrett Paving Mixtures
Dover Equipment and Machine Co.
Saienni Bros.
Warren Brow. Co., Inc.
Petrillo Bros., Inc.
Fairfax Asphalt Co., Inc.
Kensington Bituminous Corp.
F. 0. Day Bituminous Co.
Arlington Asphalt Co.
Arlington Asphalt Co.
Prince Georges Bituminous Co.
-------
ASPHALT BATCHING (CONTINUED)
AQCR
PLANT NAME
(D. C.)
Cincinnati
St. Louis.
Prince Georges Bituminous Co.
Washington Contractors, Inc.
National Asphalt Paving Corp.
National Asphalt Paving Corp.
Arlington Asphalt Co., Inc.
Newton Asphalt Co., Inc.
Troxler Asphalt Co., Inc.
Bituminous Construction Co.
Plasma Chem. Systems, Inc.
Northern Ky. Asphalt Co.
Armrel Construction Co,, Inc.
Eaton Asphalt Paving Co., Inc
Byrnes-Conway Co. (2 plants)
Eaton Asphalt Paving Co., Inc.
Paz-0-Products, Inc.
Chevron Asphalt Co.
Brewer Co.
Brewer Co.
Brewer Co.
Brewer Co.
Bitucote Products Co.
Northern Ky. Asphalt Co.
Valley Asphalt Corp.
Allied Chemical Corp. (2 plants)
Triasco Paving Corp.
Hotmix, Inc.
Great Lakes Asphalt, Inc.
Barrett
Bitucote Products Co.
Asphalt Material Co., Inc.
Trumbell Asphalt Co. of Delaware
Asphaltic Concrete Corp.
Asphaltic Concrete Corp.
Ken-Mar Venetian Blind Co.
Bitumix, Inc.
Source: Dun & Bradstreet Computer Printout.
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: CEMENT
INDUSTRY CODE:
13
PACK
OF
AQCR
St, Louis
it
it
COMPANY
River Cement
Co.
Alpha Portlant
Missouri
Portland
PROCrSS
TYPE
Dry
Wet
Wet
CAPACITY
f106bbl/vr.')
3.0
2.6
5.0
NUMBER
1 '
1
1
ADDITIONAL
CONTROLS REOIITREn
TYPE
Fabric Filtei
E.S.P. (1)
E.S.P.
EFFICIENCY
(%)
>99
>99
>99
FULL ADDITIONAL
CONTROL COST (2)
INVESTMENT
($1000)
665
839
1,395
ANNUAL
(S1000/YR.)
189
186
312
EMISSION' DATA (TON/ YEAR)
PARTICULATES
EXISTING
(3)
2,256
1,683
3,238
REQUIRED
45]
112
32
SULFUR DIOXIDE
rx i STING
Neg.
ii
"
REQUIRED
Neg.
11
II
(1) E.S.P. = electrostatic precipitators.
(2) Control cost and emission estimates based on known plant size.
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY:
LIME
INDUSTRY CODE:
14
PACK
OF
AQCR
Phila.
ii
t
it
COMPANY
G & W. H.
Carson
Amos K.
Stoltzfus
Everett V.
Moser
PROCESS
TYPE
(1)
(1)
(1)
CAPACITY
(Ton/Day)
(1)
(1)
(1)
NUMBER
1
1
1
ADDITIONAL
CONTROLS REQUIRED
TYPE
Venturi
Scrubber
it
it
EFFICIENCY
(%)
97
II
M
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
50
50
50
ANNUAL
(S1000/YR.)
69
69
69
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(2)
1,209
1,209
1,209
REQUIRED
97
97
97
SULFUR DIOXIDE
EXIST INC;
Neg.
"
"
REQUIRED
Neg.
"
"
Ol
(1) Control cost and emission estimates based on a model plant of 500 ton/day - 80% of capacity in a single rotary kiln, the remainder in a vertical kiln.
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY:
COAL CLEANING
INDUSTRY CODF:
15
PACK OF
AQCR
St. Louis
tt
ii
M
COMPANY
Belle-Valley
Coal
Peabody Coal
Peabody Coal
PROCESS
TYPE
(1)
II
II
CAPACITY
Clfinn Trm/Yr
120
3,304
869
NUMBER
))
1
1
1
ADDITIONAL
CONTROLS REQUIRED
TYPE
CD
ii
.u
EFFICIENCY
(%)
(1)
II
II
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
57
1,582
416
ANNUAL
(S1000/YR.)
10
271
71
EMISSION DATA (TON/ YEAR)
PARTICULATES
EXISTING
(2)
23.8
129.5
58.5
REQUIRED
9.0
64.9
23.5
SULFUR DIOXIDE
EXISTING
Neg.
11
It
REQUIRED
Neg.
It
II
O\
(1)
(2)
Control costs and emission estimates based on known capacity and model plant (which identifies the process elements in model plant). Model plant
is comprised of the following unit processes and percent of total plant capacity: pneumatic cleaners (7.1%); fluidized-bed driers (8.4%); and
flash driers (56.5%)- percentage of production not indicated is assumed air dried with negligible emissions. Wet scrubbers were required control.
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY:
fIRATN MTT.T.TMr:
INDUSTRY CODE: 16
PACK OF
AQCR
Cinti.
Wash.
St. Louis
Phila.
COMPANY
Model Plant
(1)
II
ft
II
PROCESS
TYPE
Animal
Feed
Mill
II
II
II
II
CAPACITY
(Ton/Yr.)
200,000
T/Yr.
1,400,000
600,000
3,600,000
2,800,000
NUMBER
1.
7
3
18
14
ADDITIONAL
CONTROLS .REQUIRED
TYPE
Fabric
Filter
II
U
II
II
EFFICIENCY
(%)
99
II
II
II
II
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
20
140
60
360
280
ANNUAL
(S1000/YP.)
7
49
21
126
98
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(2)
350
2,450
1,050
6,300
4,900
REQUIRED
54
378
162
972
756
SULFUR DIOXIDE
EXISTING
Neg.
Neg.
II
II
II
REQUIRED
Neg.
Neg.
II
II
II
(1) Control cost and emission estimates based on model plant indicated. A list of animal feed mills which operated in the selected AQCR's in 1967 may
be found on the following page.
-------
GRAIN MILLING
AQCR
PLANT NAME
D. C.
Phila.
St. Louis
Bowman Bros. Inc.
Chesapeake Feed Co.
Binger Bros. Ind.
Atlas Canine Products, Inc.
Pet Foods, Inc.
Bailsman Pet Supplies
Seaboard Supply Co.
Thrivo Co., Inc.
Crisfield Dehydrating Co., Inc.
Rosenberg Ben, Inc.
Amburgo Manufacturing Co., Inc.
R. F. French Co.
Deer Run Packing Co., Inc.
Greater Valley Feed Co., Inc.
Wilson Phrm. and Chemical
Wafer Mills Co., Inc.
American Pet Food Corp.
Ralston Purina Co.
Textron Inc. Del.
Cardinal Food Products, Inc.
Campbell and Co., Inc.
Ultra-Life Laboratories, Inc.
Dixie Mills Co.
Consumers Products Co., Inc.
Allied Mills, Inc.
Joggerst Milling Co.
Chesterfield Elev. & Sup. Co.
Agway, Inc.
Gordon Foods Div.
Van Camp Div.
International Multifoods Corp.
Nutreana Mills Div.
Professional Feeds
Agway, Inc.
Fenton Feed Mill & Welding Co.
-------
GRAIN MILLING (CONTINUED)
AQCR
PLANT NAME
Cincinnati
Cooperative Mills, Inc.
Kentucky Chemicals Industry
Dearborn Mills Co.
Clark Selma Mis. J.E.
Midwest Feeds Co., Inc.
Ralston Purina Co.
Thorobred Co., Inc.
Source: Dun & Eradstreet Computer Printout.
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: GRAIN HANDLING
INDUSTRY CODE: 17
PACK j OF 2
AQCR
Cinti
II
II
II
II
II
COMPANY
Model Plants
II
II
II
II
II
II
II
||
II
(2)
"
II
*l II
"
It
PROCESS
TYPE
Country
ii
ii
ii
ii
Terminal
it
ii
it
it
Country
it
-
Terminal
"
II
CAPACITY
(1000 Bushe:
0-25 (1)
25-75
75-150
150-250
250-500
500-1000
1000-2500
2500-5000
5000-10,000
10,000 +
25-75
150-250
250-500
500-1000
1000-2500
2500-5000
NUMBER
s)
1
1
1
1
1
1
1
1
1
1
1
3
4
6
4
3
ADDITIONAL
CONTROLS REQUIRED
TYPE
Fabric
Filters
II
"
II
II
II
II
II
II
II
II
II
II
"
ii
11
EFFICIENCY
(%)
99
M
It
tl
It
"
M
ii
it
ii
n
it
n
ti
11
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
2.7
5.4
12.4
22.5
52.0
84.0
195.0
420.0
840.0
1,110.0
5.4
67.5
208.0
504.0
780.0
1,260.0
ANNUAL
(S1000/YR.)
1.0
1.9
4.4
7.9
18.2
29.4
68.2
147.0
294.0
388.0
1.9
23.7
72.8
176.4
272.8
441.0
EMISSION DATA (TON/YEAR)
PARTICULATES
EXISTING
(3)
1.6
6.3
14.1
25.0
46.8
390.0
910.0
1,950.0
3,900.0
7,800.0
6.3
75.0
187.2
2,340.0
3,640.0
5,850.0
REQUIRKD
Neg.
0.1
0.1
0.3
0.5
6.0
14.0
30.0
60.0
120.0
0.1
0.9
2.0
36.0
56.0
90.0
SULFUR DIOXIDE
EXISTING
Neg.
II
"
II
It
II
II
"
II
II
II
It
It
II
It
REQUIRED
Neg.
n
n
n
n
n
n
n
n
n
H
it
"
"
(1) Range of processes sizes.
(2) Control cost and emission estimates based on model plants indicated. No grain handling plants were identified by name.
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: GRAIN HANDLING
IN'DUSTRY CODE: 17
PACK 2 OK 2
AQCR
St. Louis
ti
it
ii
it
ii
ii
ii
it
Phila.
II
COMPANY
(2)
tl
II
II
||
II
II
II
II
II
II
pROcrss
TYPE
Country
ii
ii
it
ii
Terminal
ii
"
M
Terminal
ii
CAPACITY
[1000 Bushel
0-25
25-75
75-150
150-250
250-500
500-1000
1000-2500
2500-5000
5000-10,000
500-1000
100-2500
NUMBER
0
1 .
3
7
3
7
6
5
8
2
1
1
ADDITIONAL
CONTROLS REQUIRED
TYPE
Fabric
Filters
M
tl
tl
It
II
It
II
tl
II
It
EFFICIENCY
(%)
>99
M
M
"
M
"
tl
II
11
II
II
FULL ADDITIONAL
CONTROL COST
INVESTMENT
(Siono)
2.7
16.2
86.8
67.5
364.0
504.0
975.0
3,360.0
1,680.0
84.0
195.0
ANNUAL
($1000/YR.)
1.0
5.7
30.8
23.7
127.4
176.4
341.0
1,176.0
588.0
29.4
68.2
EMISSION' DATA (TON/YK.AR)
PARTICULATES
EXISTING
(3)
1.6
18.9
98.7
75.0
327.6
2,340.0
4,550.0
15,600.0
7,800.0
390.0
910.0
REQUIRED
Neg.
0.3
0.7
0.9
3.5
36.0
70.0
240.0
120.0
6.0
14.0
SUL1TR DIOXIDE
i::asTiNc;
Neg.
"
It
II
M
tl
"
II
It
II
It
RK.QUIRKD
Neg.
"
"
"
"
11
"
it
M
"
M
-------
SECONDARY NDNFERROUS METALLURGICAL
AQCR
PLANT NAME
METAL TYPE
D. C.
Phila.
Bladensburg Metals
Hyman Viener & Sons
Atlantic Metals Corporation
General Smelting Company
Metallurgical Products Company
George Sail Metals Company
North American Smelting Company
Joseph Rosenthal's Sons, Inc.
Bers & Company, Inc.
Electric Storage Battery Company
Franklin Smelting & Refining Company
Thomas F. Lukens
National Lead Company
Joseph Rosenthal's" Sons, Inc.
U. S. Smelting Works
Ajax Metal, Division of H. Kramer
L. Goldstein's Sons, Inc.
Metal Bank of America
Thermal Reduction Corporation
Acme Alloys, Inc.
Bers & Company, Inc.]
Halpern Metals Company
National Lead Company
Secondary Lead, Tin, Babbitt, Solder,
Aluminum, Lead, Brass, Copper, Sec. Lead
Lead, Tin, Babbitt, Solder, Zinc.
Aluminum
Aluminum, Zinc.
Aluminum, Brass, Copper
Aluminum, Lead, Brass, Copper, Zinc.
Aluminum, Brass, Copper, Sec. Lead, Tin,
Babbitt, Solder
Lead
Lead
Lead
Lead, Brass, Copper, Zinc
Lead, Sec. Lead, Tin, Babbitt, Solder
Lead
Lead, Brass, Copper
Lead, Sec. Lead, Tin, Babbitt, Solder
Brass, Copper
Brass, Copper
Brass, Copper
Brass, Copper
Lead, Tin, Babbitt, Solder
Lead, Tin, Babbitt, Solder
Lead, Tin, Babbitt, Solder
-------
SECONDARY NONFERROUS METALLURGICAL (CONTINUED)
AQCR
PLANT NAME
METAL TYPE
St. Louis
A. Perez & Son, Div. of Abrams Metal Co.
Besco Metals Corporation
Superior Zinc Corporation
Superior Zinc
Deleware Valley Smelting Corp.
Lefton Iron & Metal Company
Aluminum Service Corp.
Grossman Smelting Company
M. Holtzman Metal Company
Lewin Mathes Co., Div. of Cerro de
Pasco Corporation
Lewin Mathes Metals Div., Cerro de
Pasco Corporation
American Zinc, Lead, & Smelting Co.
Federated Div. of Amer. Smelting and
Refining Company
National Lead Company
American Zinc, Lead & Smelting Co.
Theo. Hiertz Metal Company
National Lead Company
Shanfield Brothers Metal Company
American Zinc Co. of 111.
American Zinc, Lead and Smelting Co.
Monsanto Chemical Company
Lead, Tin, Babbitt, Solder
Zinc
Zinc
Zinc
Zinc
Aluminum
Aluminum
Aluminum, Brass, Copper, Zinc
Aluminum, Brass, Copper, Lead,
Tin, Babbitt, Solder, Zinc
Aluminum, Lead, Brass, Copper, Secondary
Lead, Babbitt, Solder, Zinc
Lead, Brass, Copper, Zinc
Lead, Tin, Babbitt, Solder, Zinc
Aluminum
Lead
Lead
Lead, Tin, Babbitt, Solder
Lead, Tin, Babbitt, Solder
Lead, Tin, Babbitt, Solder, Zinc
Zinc.
Lead, Tin, Babbitt, Solder, Zinc.
-------
SECONDARY ^DNFERROUS METALLURGICAL (CONTINUED)
AQCR
PLANT NAME
METAL TYPE
Cincinnati
American Compressed Steel Corp.
G. A. Avril Company
Certified Metals Mfg. Company
Eagle-Picher Company
National Lead Company
Moskowitz Brothers
National Lead Company
Aluminum, Lead, Secondary Lead,
Tin, Babbitt, Solder, Zinc
Aluminum, Lead, Secondary, Lead,
Tin, Babbitt, Solder, Zinc
Aluminum, Lead, Secondary Lead,
Tin, Babbitt, Solder, Zinc.
Lead, Secondary Lead, Tin, Babbitt,
Solder
Lead, Tin, Babbitt, Solder
Lead, Tin, Babbitt, Solder
Lead, Secondary Lead, Tin, Babbitt,
Solder
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: SECONDARY NON-FERROUS METALS
INDUSTRY CODE: 18
PAH. OK
AQCR
;
Cinti.
Iff
ft
Ha»h.
II
1*
tl
St. Louis
tt
COMPANY
Model Plant
M
It
II
(1)
it
ii
it
n
M
M
it
n
PROCESS
TYPE
Alum.
Copper,
Brass, &
Bronze
Lead
Zinc
Alum.
Lead
Zinc
Alum.
Copper
Lead
Zinc
Alum.
Copper
CAPACITY
CTon/Yr.'*
4,300
5,500
1,200
240
12,900
8,400
720
4,300
6,500
480
240
25,800
26,000
NUMBER
4 '
6
1
3
7
3
1
1
2
1
6
4
ADDITIONAL
rflMTRf)T.<; PFnTTTRFD
TYPE
Wet Scrubber
Fabric
Filter
Fabric
Filter
Fabric
Filter
Wet Scrubber
Fab. Filter
n
Wet Scrubber
Fab. Filter
n
n
Wet Scrubber
Fab. Filter
EFFICIENCY
(%)
95
95
95
95
95
it
n
n
ii
ii
n
n
n
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
82
140
15
5
246
105
15
82
140
30
5
492
560
ANNUAL
($1000/YR.)
26
54
5
1
78
35
3
26
54
10
1
156
216
EMISSION DATA (TOX/YKAR)
PARTICULATES
EXISTING
(2)
8.7
65.0
34.0
1.6
26.1
238.0
4.8
8.7
65.0
68.0
1.6
52.2
260.0
REQUIRED
1.0
6.5
3.4
0.2
3.0
23.8
0.6
1.0
6.5
6.8
0.2
6.0
26.0
SULITR DIOXIDE
DUSTING
Neg.
II
II
||
II
tl
II
II
||
II
II
II
It
REQUIRED
Neg.
It
1.
II
11
II
tl
II
II
II
It
tl
It
(1) Control costs and emission estimates based on model plants indicated. A list of plant names are included on the following page.
-------
Table 5-5. INDUSTRIAL PROCESSES
INDUSTRY: SECONDARY NON-FERROUS METALS
INDUSTRY CODK:
18
7 ?
HACK OF
AQCR
St. Louis
(Con't.)
II
1
Phila.
it
it
ii
COMPANY
(1)
it
it
ii
ii
ii
PROCHSS
TYPE
Lead
Zinc
Alum.
Copper
Lead
Zinc
CAPACITY
(Ton/Yr.1
12,000
2,160
17,200
58,500
18,000
1,680
NUMBER
10
9
4
9
15
7
ADDITIONAL
CONTROLS REQUIRED
TYPE
Fab. Filter
"
rfet Scrubber
Fab. Filter
M
II
EFFICIENCY
(%)
95
II
II
M
II
ft
FULL ADDITIONAL
CONTROL COST
INVESTMENT
($1000)
150
45
328
1,260
225
35
ANNUAL
(S1000/YR.)
50
9
104
486
75
7
EM1SS::1 VTA (TOX/YKAR)
PARTICULARS
EXISTING
(2)
340.0
14.4
34.8
585.0
510.0
11.2
REQUIRK1)
34.0
1.8
4.0
58.5
51.0
1.4
SUL1TR DIOXIDE
I:XISTINC;
Neg.
II
II
M
II
M
RKQUIRKD
Neg.
II
II
II
II
II
i
-------
emissions as well as in determining the adequacy of the existing control
system to meet emissions standards. Finally, the "year built" (as found
with municipal incinerators) is identified since the control-cost-estima-
ting procedure is dependent upon the age of the facility.
The data observed in Tables 5-2 through 5-5 have been presented to
facilitate comparisons between the data used in the 305(a) Approach and
the data and results obtained through the Extrapolation and Aggregation
Approaches. To further simplify the comparisons, summaries of regional
emission estimates and control costs are presented in Tables 5-6 and 5-7.
The efforts involved in Step //I of the 305 (a) Approach are nearly
all data handling tasks, that is, accounting efforts. The data bank
supporting the EGA Report is partially computerized, and part of the data
is stored in files and in other locations. The Step #1 task was one of
obtaining and organizing the data.
There was some engineering analysis required in Step #1. Control
efficiencies for specific processes were necessary on a process-by-process
basis and were easily determined (along with the quantity of controlled
and uncontrolled emissions). Also, thermal balance calculations were re-
quired to determine the quantities of low-sulfur oil and natural gas.
Such calculations were required because the analysis presented here is con-
ducted on a regional level while the EGA Report procedures are designed
for calculating national control cost estimates.
5.2.5 305(a) Approach - Step #2
Step #2 requires that the data organized in Step #1 be presented in
a form most useful to the decision-maker. In this report, the data has
been aggregated for each AQCR investigated. If the relevant data was
available for the entire nation, such demand estimates could be presented
for entire states, multi-state regions and/or the nation.
The demand estimates are presented in Table 5-8 through 5-11 for the
four AQCRs. In addition to the basic data developed in Step #1, data on
"typical gas volumes" has been included to present rough indication of the
size control systems that are required for each process type. The data on
gas volumes was obtained from various sources in the literature.
-------
Table 5-6. REGIONAL EMISSION SUMMARY BY SOURCE CATEGORY
Page 1 of 4
AIR QUALITY CONTROL REGION: ' ' CINCINNATI
cniTT? r*T?
oUUKC.ll
CATEGORY
Solid waste
Steam-electric Power
Industrial Boilers
Commercial and Instit. Heat.
Residential Heating
Iron and Steel
Gray Iron Foundry
Primary Nonferrous Metals
Sulfuric Acid
Phosphate Fertilizers
Petroleum Refining
Asphalt
Cement
Lime
Coal Cleaning
Grain Milling
Grain Handling
Secondary Nonferrous Metals
Regional Total
1967 EMISSION BY SOURCE CATEGORY (Ton/Yr.)
PARTICULATES
Uncontrolled
16,695
34,780
37,100
854
2,936
8,163
1,762
-
647
15,300
30
17,400
-
-
-
2,450
12,099
269
150,485
Controlled
2,814
2,126
1,600
264
420
1,107
240
-
395
15,300
6
1,140
-
-
-
378
185
27
26,002
SULFUR DIOXIDE
Uncontrolled
183,952
33,600
3,100
8,600
-
-
-
6,625
-
7,800
-
-
.
-
-
-
243,677
Controlled
18,394
12,200
2,280
3,000
-
-
-
926
-
4,800
-
-
-
-
-
-
41,600
-------
Table 5-6. REGIONAL EMISSION SUMMARY BY SOURCE CATEGORY (Con't.)
Page 2 of 4
AIR QUALITY CONTROL REGION: WASHINGTON, D.C.
COTTT?/"1 17
OUUKL.EJ
CATEGORY
Solid Waste
Steam-electric Power
Industrial Boilers
Commercial and Instit. Heating
Residential Heating
Iron and Steel
Gray Iron Foundry
Primary Nonferrous Metals
Sulfuric Acid
Phosphate Fertilizers
Petroleum Refining
Asphalt
Cement
Lime
Coal Cleaning
Grain Milling
Grain Handling
Secondary Nonferrous Metals
Regional Total
1967 EMISSION BY SOURCE CATEGORY (Ton/Yr . )
PARTICULATES
Uncontrolled
33,311
21,831
-
1,975
5,410
-
-
-
-
-
-
13,050
-
-
-
1,050
-
143
76,760
Controlled
4,482
1,557
-
1,309
910
-
-
-
-
-
-
855
-
-
-
162
-
15
9,290
SULFUR DIOXIDE
Uncontrolled
_
70,925
-
13,030
12,600
-
-
-
-
-
-
-
-
-
-
-
-
96,555
Controlled
9,768
11,090
9,000
-
-
-
-
-
-
-
-
-
-
-
-
29,858
-------
Table 5-6. REGIONAL EMISSION SUMMARY BY SOURCE CATEGORY (Con't.)
Page 3 of 4
AIR QUALITY CONTROL REGION: ' ST- LOUIS
CnTTDPT?
oUUKLhi
CATEGORY
Solid Waste
Steam-electric Power
Industrial Boilers
Commercial and INstit. Heating
Residential Heating
Iron and Steel
Gray Iron Foundry
Primary Nonferrous Metals
Sulfuric Acid
Phosphate Fertilizers
Petroleum Refining
Asphalt
Cement
Lime
Coal Cleaning
Grain Milling
Grain Handling
Secondary Nonferrous Metals
Regional Total
1967 EMISSION BY SOURCE CATEGORY (Ton/Yr.)
PARTICIPATES
Uncontrolled
22,978
43,798.
87,600
334
6,990
20,394
749
1,344
1,259
8,550
187
5,220
7,177
-
212
6,300
30,812
667
244,571
Controlled
2,223
4,313
9,800
334
665
767
135
1,344
769
8,550
32
342
189
-
97
972
471
68
31,071
SULFUR DIOXIDE
Uncontrolled
_
234,679
145,000
2,820
18,800
-
-
-
12,077
-
62,400
-
-
'
-
-
-
475,776
Controlled
-?*
25,041
40,000
2,820
5,050
-
-
-
1,802
-
31,200
-
-
-
-
-
-
105,913
00
-------
Table 5-6. REGIONAL EMISSION SUMMARY BY SOURCE CATEGORY (Con't.)
Page 4 of 4
AIR QUALITY CONTROL REGION: ' PHILADELPHIA
cnm?riT7
oUUKLi £j
CATEGORY
Solid Waste
Steam-electric Power
Industrial Boilers
Commercial and Instit. Heating
Residential Heating
Iron and Steel
Gray Iron Foundry
Primary Nonferrous Metals
Sulfuric Acid
Phosphate Fertilizers
Petroleum Refining
Asphalt
Cement
Lime
C*'al Cleaning
Grain Milling
Grain Handling
Secondary Nonferrous Metals
. Regional Total
1967 EMISSION BY SOURCE CATEGORY (Ton/Yr.)
PARTICULATES
Uncontrolled
58,293
67,462
174,000
3,540
8,710
58,428
1,363
-
576
3,150
224
26,970
-
3,627
-
4,900
1,200
1,141
413,584
Controlled
7,954
6,882
15,800
2,350
4,400
3,771
168
-
351
3,150
39
1,767
-
291
-
756
20
115
47,814
SULFUR DIOXIDE
Uncontrolled
309,404
300,000
23,930
35,050
-
-
-
5,893
-
48,800
-
-
.
-
-
-
723,077
Controlled
34,198
122,000
20,200
31,000
-
-
-
823
-
48,800
-
-
-
-
-
-
257,021
-------
Table 5-7. PARTICULATE & SULFUR DIOXIDE CONTROL COST SUMMARY FOR SELECT AIR QUALITY CONTROL REGIONS
C ATTT?PT?
oUUKL-li
CATEGORY
Solid Waste
Steam-electric Power
Industrial Boilers
Comm.& Inst. Heating
Residential Heating
Iron and Steel
Sray Iron Foundry
Prim.Nonfer. Metals
Sulfuric Acid
Phosphate Fert.
Petroleum Refining
Asphalt
Cement
..ime
Coal Cleaning
Srain Milling
Srain Handling
secondary Non.Met.
Regional Total
REGIONAL CONTROL COST PER SOURCE CATEGORY ($1000) FOR 1967
CINCINNATI WASHINGTON
INV.
2,671
27,137
8,826
726
-
8,585
1,525
-
1,824
-
260
280
-
-
-
140
2,825
366
55,165
ANNUAL
1,637
8,415
7,962
410
-
3,683
456
-
397
-
25
226
-
-
-
49
999
116
23,555
INV.
5,560
35,270
-
671
-
-
-
-
-
-
-
210
-
-
-
60
-
257
42,028
ANNUAL
3,120
6,080
-
1
-
-
-
-
-
-
-
170
-
-
-
21
-
91
9,482
ST. LOUIS
INV.
2,239
23,804
8,894
-
-
13,923
698
-
2,202
-
1,670
84
2,899
-
2,055
360
7,056
1,247
67,131
ANNUAL
1,454
7,488
16,052
-
-
6,395
211
-
505
-
124
68
687
-
352
126
2,470
431
36,363
PHILADELPHIA
INV.
11,046
53,126
15,609
1,426
-
39,727
1,348
-
1,586
-
800
434
150
-
-
280
279
1,848
127,659
ANNUAL
5,646
17,195
10,445
38
-
18,440
435
-
347
-
164
350
207
-
-
98
98
672
i
-------
TABlf 5-8. CONTROL SYSTH1 D0WD R» THE CINCIflWI /OCR
SOllRCr.
CODE
01
02
01
D4
05
06
07
08
09
10
11
12
13
14
15
16
17
18
SOURCE CATEGORY AND PROCESS
!;ul.l 1) WASfL DISPOSAL
Municipal Incinerators (Existing)
e Domestic and Commercial Incinerators
STEAM-KLECTRK POWER GENERATION
e Boiler Flue Gas
INDUSTRIAL BOILERS
e Boiler
COMMERCIAL- INSTITUTIONAL HkATING PLANTS
e Boi lers
RESIDES'TIAI HEATING PLANTS
e Home Heating Units
IRON AND STEEL
e Open Hearth
e Electric Furnace
e Sintering
GREY IRON FOUNDRY
e Cupola
PRIMARY NON-FURIOUS METALS
e Copper
e Smelter
a Leed
e Refinery
e Zinc
e Smelter
SULFURIC ACID
e Contact Proceae
PHOSPHATE FERTILIZER
e Normal Superphoephace
PETROLEUM REFINING
e Refinery
e Catalyst Rageneretora
ASPHALT BATCHING
CEKEN t
e Dry Kiln
e Wet Klin
LIKE
e Vertical Kiln
e Rotary Kiln
COAL CLEANING
e Fluldlzed-Bed Drier
e Flash Drier
e Pneumatic Cleaner
GRAIN MILLING
e Animdl l-'eed Hilling
GRAIN HANDLING
e Country Elavetora
e Terminal Elevator*
SECONDARY .'INKERROUS MKTALS
e Aluminum Melting
e Bras*, and Bronse Helting
e Rpverberatory
e Lead Refining
a cupola
e Zinc Melting
e G.ilv.mi ting Kettlee
REGIONAL TOTAL
TYPICAL CONTROL SYSTEM TYPE
e Wet 'Jcruhbcr
'..'el Scrubber
a U'et Limestone Scrubbing
e Switch to Oil with Sulfur
e Switch to lov Sulfur Oil and
Natural Gas
e Conversion to Distillate
Oil and Natural Gas
60" w.g, Vancurl Scrubber
High Teaperacure Fabric
Filter
a 20" w.g. Vet Scrubber
a 16-35" w.g. Uet Scrubber
e Acid riant with Lla* Scrubber
a Acid Plant
a Acid Flant
Secondary Absorption Tower
a Fabric Filter
a Sulfur Recovery Planet
a High Efficiency Electrostatic
Pracipi tators
a Fabric Filter
e Electrostatic Freclpltator
e Cyclonic Scrubber
e Ver.turf Scrubber
e 15" w.g. Venturi Scrubber
a 10" w.g. Venturi Scrubber
e 10" w.g. Venturi Scrubber
a Fabric Filter
a Fabric Filter
e Febric Filter
a Fabric Filter
a Fabric Filter
a Fabric Filter
NUMBER 01-
REQUIRED
16
(4)
(5)
N/A
N/A
N/A
19
4
0
15
0
0
0
6
0
2
1
20
0
0
0
0
0
0
0
7
8
13
3
0
7
3
SUBS, . :'
Ol'AHTlTY
(10 <',xl)
138
9
9
156
t rra.s
SULFUR
aiN'TI-INT
<;:>
EijlMHL'U
NAT. CAS
QUANTITY
t
(10 cf)
0
310
4,980
5,290
i n.i
ADUlllnNU.
CON! Rnl.
INVESTMENT
(S1000)
1,129
1,542
27,137
8,826
726
(1)
)
> 8,585
7
1,525
1,824
130
130
280
140
281
2,544
246
105
15
55,165
IV!1: ' AI
! 1 1 1 > i :
>ri i
!)h'\l
(.')
-')
C)
(2)
N/'A
N/A
N/A
N/A
-94
~96
~93
-86
N/A
~62
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
S/A
N/A
N/A
N/A
N/A
N/A
'. : (Jl
:t l, - , )
i-AX-m-
-,-.
Sb-'J >
{ n
' 1)
( n
89.5-97.5
84-95
95-98.8
85-95
~95
~95
~95
~67
0-80
~90
~90
97. 8-99.9
97.6-99.6
sn-io
97.^-99. >
~ > 7 . fl
~<)i.J
~94.S
~99
~9»
-99
~.r.
~95
~*;
...
. '>l
AS
':''. yjs
' ! ' '" "
]>
-.>«)»
; xi
S/A
N.'A
10-650
30-400
10-400
5-150
10-500
15-300
10-500
6. 5-90
20-80
100-1500
20-JOO
10-60
'.''.-100
60-650
5-15
50-125
10-150
10-150
10-130
10- JO
9-50
10-50
10-50
Natural transition from coal to oil. gn and clactrlc heating, thus no control costs.
Allowable emission rate of 1.46 pounds of sulfur dioxide per million BTU Input (contro
dependent on sulfur content of existing fuel),
Allowable emission rate determined by combustion regulation of State of Maryland
e Figure E-? In Appendix K).
:,umber of control system* for domestic and coavicrcial Incinerators and new municipal
Incinerators not identified.
Number of power plants Is Indicated (not tho number of boilers).
-------
TABLE 5-9, CONTROL SY5TB1 BEWE FOR THE WASHINGTON, D.C. AQCR
sm RCIC
CODE
01
02
01
04
05
06
07
08
09
10
11
12
n
14
15
16
17
18
SOURCE CATEGORY AND PROCESS
Sill. ID WASTE DISPOSAL
Municipal Incinerators (Exlatlng)
Domestic and Commercial Incinerators
STtAM-ELHCIRIC POWER CF.NKRATION
Boiler flue Gas
INDUSTRIAL BOILERS
Boiler
COMMERCIAL-INSTITUTIONAL HEATING PLANTS
Boilers
RESIDENTIAL HEATING PLANTS
Home Heating Units
IRON AND STEEL
Open Hearth
Electric Furnace
Sintering
GREY IRON FOUNDRY
Cupo U
PRIMARY NON-FERgOUS METALS
Copper
Smelter
Lead
Refinery
Zinc
Smelter
SULPURIC ACID
Contact Proeeae
PHOSPHATE FERTILIZER
Normal SuparphoealMte
PETROLEUM REFINIK
e Refinery
Catalyst Regenerator*
ASPHALT BATCHING
Asphalt Batching Kroceaa
CEMEN1
e Dry Kiln
Wet Kiln
LIME
Vertical Kiln
e Kotary Kiln
COAL CLEANING
e Fluidlzed-Bed Drier
a Flash Drier
Pneumatic Cleaner
GRAIN MILLING
e Animal Feed Hilling
GRAIN HANDLING
e Country Elevatore
Terminal Elevator*
SECOSDARY ..UNtEWOUS MCTALS
e Aluminum Meltln*
RevprSeratory Furnace
Brags wild Bronte Melting
Rev«rberatory
Lead Ke fining
Cupola
e Zinc Melting
e c.i Iv.mlilng Kettlee
REGIONAL TOTAL
TYPICAL CONTROL SYSTEM TYPE
Wet Scrubber
'..'et Scrubber
Uet Limestone Scrubbing
Switch to Oil with Sulfur
Content lass than 1.38X
Switch to low Sulfur Oil and
Natural Gas
Conversion to Dletlllate
Oil and Natural Gaa
40" w.g. Venturl Scrubber
High Temperature Fabric
Filter
20" w.g. Wet Scrubber
16-35" w.g. Wat Scrubber
Acid riant with Lima Scrubber
Acid Plant
e Acid Plant
e Secondary Abaorption Tower
e Fabric Filter
e Sulfur Recovery Planta
High Efficiency Electroatatic
Precipitate ra
a 10" w.g. Wet Scrubber
e Fabric Filter
e Electrostatic Precipitate*
a Cyclonic Scrubber
e Vecturl Scrubber
a 15" w.g. Venturl Scrubber
e 10" w.g. Venturl Scrubber
e 10" w.g. Venturi Scrubber
a Fabric Filter
e Fabric Filter
e Fabric Filter
e Wet Scrubber
e Fabric Filter
a Fabric Filter
Fabric Filter
NUMBER (IF
SYSTEMS
REQUIRED
8
(4)
6(5>
N/A
N/A
N/A
15
3
1
1
2
1
SHi-. ! 1 1 in: i TI:I.S ui:ijnni:!j
' i.
QUAjrl ITY
do" Cal)
0
5
29
34
Sfl.FTK
CONTENT
(Z)
NA1. (.AS
QUANTITY
(Id6 cl)
{)
580
4,620
-
5.200
1 I'M
AUDI [ In:.'AJ.
CUNT Kill.
INVi:SrMJ-NT
(S1000)
I.52J
4,039
35,270
0
671
(i)
-..
-
210
60
82
140
30
5
42,028
i N ! ' .1. :]
n > n i:
SI'I 1 1 l<
UlilAlM!
'.' ' \
<'.'>
(2)
(2)
N/A
N/A
N/A
N/A
~94
-96
~93
~86
N/A
~62
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
.-CM , i 1
I'.Md [1 -
MAM :
H'
B'j-'l'>
<:.
i I)
(i)
89.5-97.5
84-95
95-98.8
85-95
-95
~95
-95
~f>7
0-SO
-90
-90
79-98.2
97.8-99.9
97.6-99.6
50-90
97.H-99. )
- '1 7 . 8
- <> ) . 2
-94.5
-99
-------
TABLE 5-10. ONTO. SYST01E0WB FOR TrE ST. LOUIS AQCR
soijRci;
CODE
01
02
01
04
05
06
0?
oe
09
10
11
12
13
14
15
16
17
Ifl
SOURCE CATEGORY AND PROCESS
Sill. ID WASTE DISPOSAL
P ( g
S TEAM-KLtCTRIC POWER flF.NKRATION
Boiler Flue Gas
INDUSTRIAL BOILERS
Boiler
COMMKRC1 A!.- INSTITUTIONAL HEATING PLANTS
Boilers
RESIDENTIAL HEATING PLANTS
Home Heating Unite
IRON AND STEEL
Open Hearth
Electric Furnace
Sintering
GREY IRON FOUNDH
Cupo la
PRIMARY SON- PERIODS METALS
Copper
S«elter
L«»d
Refinery
Zinc
Smelter
SULFUR1C ACID
Contact Proeeee
PHOSPHATE FERTILIZER
riormel Su| aiafcupluta
PETROLEUM REFINING
Refln«ry
Catalyst Regeneratore
ASPHALT BATCHING
Asphalt Batching Proceaa
CEMENT
a Dry K41n
Wet Kiln
LIME
a Vertical Kiln
a Rotary Kiln
COAL CLEANING
a Fluidliod-iad Drier
a Flanh Drier
a Pneumatic Cleaner
GRAIN MILLING
a Animal Kaed Milling
ORAIN HANDLING
a Country Elevetore
a Terminal Klavatora
6ECOWDARV .^INFEUOUS METALS
Alunlnura Malting
Kevvrheratory Furnece
a BraaN And Ironia Malting
Reverberecory
a Lead Refining
a Cupola
Zinc Melting
a c.i limiting Kattlaa
REGIONAL TOTAL
TYPICAL CONTROL SYSTEM TYPE
a
a Wet Limestone Scrubbing,
a Switch to Oil with Sulfur
Content leas than 1 383:
Switch to low Sulfur Oil and
Natural Gas
Converalon to Distillate
Oil and Natural Gaa
a 40" w.g. Venturl Scrubber
High teoperatura Fabric
Filter
20" v.g. Wet Scrubber
16-35" w.g. Wat Scrubber
Acid Plant with lie* Scrubber
Acid Plant
Acid Plant
a Secondary Abaorptton Towar
Fabric filter
Sulfur Recovery Planta
High Efficiency F.lectrostatic
Preclpitatora
a 10" w.g. Wet Scrubber
Electrostatic Preclpltator
a Cyclonic Scrubber
Verturl Scrubber
15" w.g. Venturl Scrubber
10" w.g. Venturl Scrubber
a 10" w.g. Venturl Scrubber
e Fabric Filter
a Fabric Filter
a Fabric Filter
e Wat Scrubber
e Fabric Filter
e Fabric Filter
e Fabric Filter
NUMBER OF
REQUIRED
2
5(5)
N/A
N/A
N/A
0
7
1
12
0
0
0
3
0
4
4
6
1
2
0
0
3
18
21
21
6
4
10
9
si'Bsi ii1;
i
QUANTITY
do6 can
337
0
26
363
i: ITL-.I.S i
L
SULFUR
TONTKNT
(7)
...
...
i:i)um:ij
NAT. CAS
QUANTITY
(I06 cf)
0
0
10,410
...
10,410
ITU.
ADUI rin:;.u
CONTROL
INVESTMENT
(S1000)
500
I 739
23,804
0
0
)
(21
(2)
N/A
N/A
N/A
N/A
~94
-96
-93
-86
N/A
~62
N/A
N/A
N/A
N/A
S/A
N/A
N/A
N/A
N/A
N/A
N/A
S/A
N/A
N/A
N/A
N/A
!\v:i k II
:i:n:s (. .)
p.AKTir-
K ' ' '
85-4 :
( II
( 1)
n>
89.5-97.5
IW-95
95-98.8
85-95
-95
-95
-95
67
0-Rt)
QO
-90
79-98.2
9 7 . 8-9'! , Q
97.6-99.6
50-40
97..S-99. i
~
-------
TABLE 5-11. 03NTPDL SfSTBI DBWO FOR M PHILADELPHIA AQCR
SOL'HCE
CODE
01
02
01
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
SOURCE CATEGORY AND PROCESS
Stil.ID WASTE DISPOSAL
a Domestic and Commercial Incinerators
STEAM-ELECTRIC POWER GENERATION
a Boiler Flue Gas
INDUSTRIAL BOILERS
a Builer
COMMERCIAL-INSTITUTIONAL HEATING PLANTS
a Boilers
RESIDENTIAL HEATING PLANTS
a Home Heating Units
IRON AND STEEL
e Open Hearth
a Electric Furnace
a Sintering
GREY IRON FOUNDRY
a Cupola
PRIMARY NON-FERaOUS METALS
a Copper
a Saaltar
a Lead
a Refinery
a Zinc
a Snaltar
SULFURIC ACID
a Contact Proeaaa
PHOSPHATE FERTILIZER '
e riornal Superahoevhata
PETROLEUM REFINING
a Refinery
a Catalyat Rageneratore
ASPHALT BATCHING
e Asphalt Batching Proeaaa
CEMENT
a Dry Kiln
e Uat .Kiln
LIME
a Vertical Kiln
a Rotary Kiln
COAL CLEANING
a Fluldlzed-Bed Drier
a Flash Drier
a Pneumatic Cleaner
GRAIN MILLING
e Animal Feed Milling
('.RAIN HANDLING
e Country Elavators
e Terminal Elavatora
SECONDARY ..INFERHOUS METALS
a Aluminum Malting
a Braan And gronaa Malting
a Reverberatory
a Lead Refining
e Cupola
a Zinc Halting
a Colv.inlslna, let t lea
REGIONAL TOTAL
TYPICAL CONTROL SYSTEM TYPE
a '..'el Scrubber
a u'et Limestone Scrubbing
a switch to Oil with Sulfur
Content laaa than 1.38X
a Switch to low Sulfur Oil and
Natural Gas
a Converalon to Distillate
Oil And Natural Gaa
a 40" w.g. Vanturl Scrubbar
a High Taaaiaratura Fabric
Ftltar
a 20" w.g. Wat Scrubbar
a 16-li" w.g. Uat Scrubbar
a Acid Plant with Lla* Scrubbar
a Acid Plant
Acid riant
a Secondary Abeoratlon Tovar
a Fabric Filter
a Sulfur Recovery Planta
High Efficiency EUctroetatlc
Pfaclaltatore
10" w.g. Wat Scrubbar
a Fabric Filter
a Electroatatlc Preclpltater
a Cyclonic Scrubber
a Varturl Scrubber
a 15" w.g. Venturl Scrubber
a 10" w.g. Venturl Scrubbar
a 10" w.g. Venturl Scrubber
a Fabric Filter
a Fabric Filter
a Fabric Filter
a Wet Scrubbar
Fabric Filter
a Fabric Filter
a Fabric Filter
NUMBER OF
SYSTEMS
REQUIRED
23
(4)
12(5)
N/A
N/A
N/A
29
21
2
20
0
0
0
3
0
4
4
31
0
0
0
3
0
0
0
14
0
2
4
9
15
7
SUBS1IIITE EL'ELS HICIJIjlREU
OIL
QUANTITY
6
(10 Gal)
608
7
43
658
bULKUK
CONTENT
m
_
NAT. GAS
QUANTITY
£
(10° cf)
0
1,960
3,950
5,910
I'UI.I.
ADDITIONAL
CONTROL
INVESTMENT
(S1000)
3 062
7,984
53,126
15,609
1,426
(1)
)
> 35,644
4,083
1,348
1,586
800
434
0
150
280
279
328
1,260
225
35
127,659
lYI'KJA. CO'!! KOI
I'.KFICIESCII.S ( ',)
SULFl'R
DIOXIDE
".' ' \
(2)
<1)
C.')
C)
N/A
N/A
N/A
N/A
-94
-96
~93
86
N/A
-62
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
...
PARTIC-
ULATES
H5-')1)
85-9^
(J)
'l>
1 1)
(1)
89.5-97.5
84-95
95-98.8
85-95
-95
-95
-95
-67
0-80
90
90
79-9S.2
97. 8-99.9
97.6-99.6
SO-'tO
o.'.s-9-.l
~ nc i m
10- 1 00
l-i'l
.'9-2600
I- H.O
:'.>
N/A
10-650
30-400
10-400
5-150
10-500
15-300
10-500
6.5-90
20-80
100-1500
20-300
10-60
40-300
60-650
5-15
50-125
10-150
10-150
30-130
10-50
9-50
10-50
10-50
(1) Natural transition from coal to oil. gas and electric hMCing* thus no control tottt.
(2) Allowable emission rate of 1.46 pound* of sulfur dioxide per Billion BTU Input (control efflcien
dependent on sulfur content of existing fuel).
(3) Allowable emission rate determined by combustion regulation of State of Maryland
(see Figure E-2 in Appendix E).
(A) Number of control systems for domestic and commercial incinerators and new municipal
Incinerators not identified.
(5) Number of power plants is indicated (not the number of boilers).
-------
The data presented in Tables 5-8 through 5-11 include:
Source Code - a code used in this study
for accounting purposes.
Source Category and Process - an identification
of the major polluting source types and processes
affected by emission regulations.
Typical Control System Type - the control technology
most likely to be applied to a majority of the
specific processes identified.
Number of Systems Required - the number of control
systems required for compliance with the assumed
emission standards.
Substitute Fuels Required - an estimate of the type
and quantity of fuels which will be substituted for
presently used fuels.
Full Additional Control Investment - the estimated
cost of the installed control systems for each
specific process type in the region.
Typical Control Efficiencies - a range of control
efficiencies (in percent) that are required for the
specific process type for the two pollutants indicated.
Typical Gas Volumes - the range of gas volume rate
typically found for the specific process types
indicated.
The following discussion will examine demand estimates in greater
detail.
5.2.5.1 Source Category and Process
The major source categories and processes identified are the major
contributors of particulate and sulfur dioxide emissions in the model
nation. These are sources that will be affected by emission control regu-
lations. The processes identified include those primary processes with
any plant, as well as secondary processes. That is to say, more than just
the major process within a plant has been included in the analysis. Secon-
dary processes, such as a grey iron cupola in a waffle iron manufacturing
plant, have been identified and included in the analysis. Another example
-------
of a secondary process is a sulfuric acid plant in a fertilizer complex.
Resources such as Dun & Bradstreet, literature from manufacturing associa-
tions, Department of Commerce reports, Federal Power Commission surveys,
and others, have been used to identify the plants and processes.
5.2.5.2 Typical Control System Type
The "typical control system type" is the specific control technology
which has been selected for cost estimating purposes in the EGA Report.
The control technology identified is one that will "most likely" be applied
to the process. It is, necessarily, a technology which will control the
pollutants to a level consistent with the emission standards. In many
cases, a process can be controlled with more than one type of control sys-
tem. In reality, the selection of the control system by a plant is based
on the criteria of competitive prices, regional environmental parameters,
and other technical conditions.
The main criterion used in selecting the "typical" control system
was the capability of attaining the required control efficiency. Technical
considerations regarding the gas stream, the pollutant, currently accepted
practices in each industry, and the trends in equipment selection were
considered in the evaluation process. No supply considerations were
examined.
5.2.5.3 Number of Systems Required
The "number of systems required" is an estimate of the number of
processes within the model nation which are identified as not in compliance
with the emission standards, which have been assumed. It is assumed that
all plants will comply with the emission standards by installing externally
applied control systems and alternative fuels. No plants will modify pro-
cesses or shut down the process. Where possible, processes which currently
meet assumed emission standards have been identified and have not been in-
cluded as candidates for control.
5.2.5.4 Substitute Fuels Required
The fuel substitution calculations are heavily dependent on a number
of assumptions used in the EGA Report. The major assumptions are:
The currently available supply of low sulfur fuel
is inadequate to fulfill the needs of steam-electric
-------
powerplants in the nation. Therefore, wet
limestone scrubbing will be used by most
plants in this category.
Other stationary fuel combustion sources
now burning coal will switch to either low
sulfur oil or natural gas.
The residential heating market is now
undergoing a natural transition from the
use of coal to oil, natural gas or
electricity.
Because the EGA Report estimating procedure for fuel substitution
was dependent on one assumption for any source type, the estimates of
expected shifts in fuel demand are of limited value. Certainly, a mix of
flue gas desulfurization methods and fuel switching will be used by power-
plants.
The estimated quantity of fuels demanded was determined by a BTU
balance for the sources identified in the AQCRs. The control investment
indicates the conversion costs for switching to low sulfur fuels.
5.2.5.5 Full Additional Control Investment
The "required control efficiency" for processes has been calculated
by relating the emission.limitation of the assumed emission standards to
the emission potential of the processes. A range of control efficiencies
is presented since many or most of the emission standards base the allow-
able quantity of emissions on process size. The "required control effi-
ciency" estimates have been determined for a range of process sizes which
are typically operated in the nation. The efficiencies reported are not
merely representative of the process sizes found in the four selected AQCRs.
5.2.5.6 Typical Gas Volumes
A range of gas volumes is presented as an indicator of control system
size required for any given process. The range is an estimate of the gas
volumes found on the nation's most typical processes. The control volumes
reported are not merely representative of sources in the four AQCRs.
-------
5.3.0 Limitations of the 305(a) Approach
The 305(a) Approach and example control measure demand estimates have
been presented. Estimates have been determined for a selected group of
major polluting industries. The control systems demanded are the resources
needed to limit air pollutant emissions from processes not now in compliance
with newly proposed emission standards stimulated by the Clean Air Act.
An evaluation of the limitations of the 305(a) Approach is needed.
This evaluation focuses on the expected accuracy of the demand estimates.
Other aspects of the Approach have been previously evaluated, such as the
availability of data and the simplicity of the calculations.
Because the quality of the demand estimates from the 305(a) Approach
are directly dependent upon the EGA Report control cost estimates, the EGA
Report cost estimating procedures are, in essence, being examined. Theo-
retically, the examination should be a comparison between the actual con-
trol systems and costs that are demanded (i.e., a "perfect" assessment)
and control systems and costs estimated by the EGA Report. A "perfect"
demand study requires a source-by-source engineering evaluation of current
emissions and the least-cost control system to meet the desirable air qua-
lity. Such a study requires the following tasks:
(i) The identification of all sources of pollution that
need control
(ii) The development of information on the process type,
size, and other plant characteristics
(iii) The documentation of the current level of emissions
control
(Jv) An analysis of the specific control technology that
will reduce emissions at least cost to the level
required by the appropriate emission standard
(v) The determination of a detailed estimate of the
installation and annual cost of operating the system.
Yet, such a "perfect" study is not in the interest of society because
of the excessive time and the resources needed to obtain such detailed
decision-making information. The procedure developed for the EGA Report
approximately predicts the costs of control and accomplishes this task with
limited resources,
-------
This Section identifies the gaps between the "perfect" demand
estimates that might result from a detailed engineering evaluation of
each and every source and the results obtained from the EGA Report control
cost estimating-procedures and the 305(a) Approach. Conclusions on the
merits of the 305(a) Approach are not made until the comparison of the
three approaches have been conducted in the summary in Section 1.0.
As previously identified, the EGA Report control cost estimating
procedure is comprised of five methods. A comprehensive evaluation of
each of the methods is desirable but beyond the scope of this report.
Such an effort should evaluate the strengths and weaknesses of each
method relative to specific criteria. The evaluations should shed light
on the expected accuracy of the EGA Report estimates and identify the
areas of the EGA Report cost estimating procedure that could be most
profitably strengthened (i.e., areas where additional efforts will bring
the greatest improvement in the desired results).
Although such an evaluation is not possible here, some limitations
have been identified and are presented in order that the three proposed
approaches for estimating control measure demand can be reasonably com-
pared in Section 1.0.
The four criteria used in the evaluation are:
(i) Comprehensiveness of source identification
(ii) Completeness of the source characteristics
(iii) Appropriateness of the selected control system
(iv) Accuracy of control cost estimates
It is important to understand how these criteria are used. First,
the EGA Report cost estimating procedures are investigated to see if all
affected sources have been identified. Second, it is assumed that all
sources have been identified and only the characteristics of the identified
source, such as size and current level of control, are to be evaluated.
Third, it is assumed that sources have been precisely characterized and the
selection of the most appropriate control systems are to be investigated.
Finally, sources with the most appropriate control system are assumed and
the accuracy of the control cost estimates are questioned. The evaluation
is conducted in the following manner: first, the evaluation criteria are
-------
stated; second, important questions regarding the criteria are cited; and,
third, specific merits or deficiencies of the EGA Report data and procedures
are stated and alternatives expressed.
5.3.1 Comprehensiveness of Source Identification
The question under consideration is whether or not all sources
affected by the control regulations have been identified. All sources in
the source categories analyzed by Methods I and II have been accurately lo-
cated (including the identification of individual processes). Specific
plants have also been identified for some industries analyzed by Methods
III A and III B.
Individual sources in the grain handling industry (Method III B) have
not been identified. Also, other fuel combustion sources (commerical, in-
stitutional and residential), small incinerators, and industrial boilers
have not been individually identified (Method IV).
5.3.2 Completeness of the Identification of Source Characteristics
The questions under consideration are: Have the size (and other
relevant characteristics) of each source been identified? and, Is the
current level of emissions control known?
(i) The size of processes and plants are known where
methods I and II are used in estimating control costs.
(ii) The size of sources is not known in cases where
Methods III A and B and IV are used. This means
that data on the distribution of control measure
size is not available for these source categories.
Also, lack of information on fuel combustion source
size prevents detailed estimates of alternative
control practices which may be applied (for example,
it is likely that large fuel combustion sources may
use wet scrubbing devices, whereas, small fuel combustion
sources may be controlled by fuel switching).
(iii) Information on the current level of control is inadequate
for nearly all source categories (the primary copper
smelting industry may be an exception). This factor
prevents the analyst from distinguishing between sources
-------
which merely require an upgrading of the
control system and those which require totally
new systems. Also, specific sources which are
currently using low sulfur fuel (or fuel with a
low ash content) are not always clearly identified.
This information is very important for predicting
control system demand and costs for the control of
powerplants as well as other stationary fuel
combustion sources.
5.3.3 Appropriateness of the Selected Control System
The questions under consideration are: Are the assumed emission
standards (which stipulate the required emission control efficiency)
appropriate for the air quality standards of the political jurisdiction
where sources are located? Have the control systems with the greatest
likelihood of application been identified for each source? Are all of
the demanded fuels and control systems technologically feasible?
(i) The EGA Report procedure applies uniform emission
standards to all sources. The standards are
relatively stringent, that is, adequate to meet
air quality standards in the nation's major
metropolitan areas (see Section 1.0 in this study).
A more accurate technique would vary the emission
standards in each AQCR according to the existing
air quality levels. Such an approach may be
practical after the submittal of implementation
p la ns.
(ii) The EGA Report assumes the application of one
control technology for all processes of a similar
type. Although this procedure is entirely adequate
for cost estimating purposes, it is inadequate for
a demand study. Currently, alternative (i.e.,
competitive) control techniques exist for most
processes. Certain system designs are more cost-
effective at some locations or for some size
-------
processes than others. Factors, such as the
availability of water, access to markets, or,
climate, influence the selection of the control
system. Because the control measure demand
estimate for a process is limited to only one
control technology, the alternative control
systems and the expected frequency of utilization
is not predicted (for example, it is not possible
to predict the proportion of gray iron foundries
that will use wet scrubbers, fabric filters,
electrostatic precipitators or switch to electric
induction melting).
(iii) The supply of economically reasonable and technically
feasible control technologies has a direct effect on
control measure demand estimates. Today, there are
many questions regarding the availability of control
systems, especially for the control of gaseous
pollutants. The 305(a) Approach estimates the demand
for the "most-likely" candidate control system. In
cases where no economically reasonable control
technologies have been demonstrated (and where
promising alternatives are now under evaluation),
the alternatives and, if possible, the expected
probability of success of such alternatives,
should be cited.
5.3.4 Accuracy of Control Cost Estimates
The questions under consideration are: How accurate are the control
cost estimates for fuel substitution and control hardware? Have all the
relevant costs been included in the control cost estimates?
(i) Projected supply and demand relationships have not
been factored into the control cost calculations
for low sulfur fuel substitutions. Economic studies
of the necessary relationships are needed to truly
reflect the substitution costs (as current fuel
prices wildly fluctuate).
-------
(ii) Some costs, such as the cost for shipping control
systems (i.e., transportation costs), have been
deleted from the control cost estimates. A
demand study should reflect these costs.
(iii) The control costs for some industrial categories
are based on linear and non-linear control cost
functions developed for the control technique
documents. Specific engineering studies of such
source categories will improve the quality of
the control cost estimates.
In summary, the EGA Report data bank is a useful source of informa-
tion for control measure demand estimates. Limitations in the data have
been identified. In general, the estimates for industrial process sources
where Methods I and II are used are more accurate (and informative) than
those estimates determined by Methods III A and B. The fuel use and flue
gas desulfurization demand estimates are the least useful because of:
(i) Lack of detailed information;
(ii) The current uncertainties in the supply-demand
relationships for low sulfur fuels; and,
(iii) The potential impact of flue gas desulfurization
techniques.
Demand studies using the 305(a) Approach can be significantly im-
proved if current source data (1971) is used. Naturally, forecasts of
sources and emissions are even more desirable. Demand studies should also
be longitudinal, that is, the study should predict demand as a function of
time. Control system demand should be related to the expected rate of en-
forcement of emission control regulations.
To conclude, a study which merges an up-graded version of the current
data, bank with accurate predictions of new control technology developments,
the growth of industrial activities, and the rate of air pollution control
enforcement will determine the desirable decision-making information.
-------
6.0 REFERENCES
1. The Cost of Clean Air, Report to the 91st Congress, Document No. 91-65,
U. S. Government Printing Office, Washington, D. C., March 1970.
2. Control Techniques for Particulate Air Pollutants, National Air
Pollution Control Administration Publication No. AP-51, Washington,
D. C., January 1969.
3. A Rapid Survey Technique for Estimating Community Air Pollution
Emissions, G. Ozolins and R. Smith, Public Health Service Publication
No. 999-AP-29, October 1966.
4. Comprehensive Economic Cost Study of Air Pollution Control Costs for
Selected Industries and Selected Regions, by M. E. Fogel, et. al. ,
Research Triangle Institute Final Report R-OU-455, February 1970.
5. Atmospheric Emissions from Coal Combustion, W. Smith and C. Gruber,
National Air Pollution Control Administration Publication No. AP-24
(Clearinghouse PB 170851), 1966.
6. Emissions from Coal-Fired Power Plants, National Air Pollution Control
Administration Publication No. AP-35 (Clearinghouse PB 174708), 1967.
7. The Fuel of Fifty Cities, Ernst and Ernst, Washington, D. C., November
1968.
8. Proposed Implementation Plan for the Control of Particulates and
'Sulfur Oxides for the State of Ohio Portion of the Metropolitan
Cincinnati Interstate AirQualtiy Control Region, TRW Systems Group
Document SN 14838.000, September 1970.
9. Comprehensive Study of Specified Air Pollution Sources to Assess the
Economic Effects of Air Quality Standards, D. A. LeSourd, et. al.,
Research Triangle Institute, Research Triangle Park, North Carolina,
December 1970.
-------
APPENDIX A
IMPLEMENTATION PLANNING PROGRAM
The Implementation Planning Program (IPP) is a series of models
developed by TRW under contract to APCO which help in the evaluation of
alternative strategies for control of air pollution sources. A flow
chart of the program is illustrated in Figure A-l.
The heart of IPP is the atmospheric diffusion model, which predicts
expected regional ambient concentrations of pollutants by mathematically
simulating the dispersion of these pollutants throughout the region. The
inputs to this model are a detailed emission inventory, various meteoro-
logical data, and measured pollutant concentration data. The emission
inventory lists individually the major sources of pollutants (powerplants,
incinerators, etc.) and describes in detail those parameters which
characterize the sources and their emissions. The inventory also charac-
terizes those emission sources which are too small to be identified indi-
vidually by aggregating them to form "area sources." The meteorological
data includes wind speed and direction, mixing depth, and other phenomena
which describe the transport mechanism which carries the pollutants from
the .sources throughout the region. The measured concentration data is
used to calibrate the theoretical model, in order to account for inaccu-
racies in the diffusion equation, inaccurate source emissions and meteoro-
logical data, irregularities in the area's topography (the diffusion
equation assumes a flat plain), and other errors.
Besides predicting present air quality conditions, IPP can predict
the effects of a pollutant control "strategy" (a series of emission con-
trol standards which apply to all major sources in the area) on the area's
air quality and measure the resulting pollution control device demand and
cost. This is accomplished by a control cost, a control standards, and a
control strategy model.
The control cost model assigns to each major source all those control
devices which may reasonably be used for reducing emissions. Devices for
the control of particulates are generally applied to the outlets of the
polluting process (usually the stacks). Sulfur dioxide control is usually
-------
Figure A-l. Flow Chart
INPUTS
Air Quality Data
Meteorological Data
Emission Inventory
Emission Standards
CONTROL COST MODEL
CONTROL STANDARDS MODEL
CONTROL
STRATEGY
MODEL
FOR CALIBRATION
ATMOSPHERIC DIFFUSION MODEL
EXISTING AI
QUALITY
DISPLAY
POST-CONTROE
AIR QUALITY DISPLAY
POINT SOURCE
CONTROL COSTS
-------
accomplished by switching to low sulfur fuels or by the use of flue gas
desulfurization techniques. The model's output consists of lists of
device names, their efficiencies and costs, and their effects on pollutant
emissions, for each major emission source.
The control standards model applies a series of emission standards
to the three categories of emission sources: fuel combustion, industrial
process, and solid waste disposal sources. Output consists of a list, for
each major source, of the applicable standards, their prescribed allowable
emissions, suitable control devices selected on the basis of effectiveness
and least cost (one device for each standard), and the cost and effect on
emission of the devices (obtained from the control cost model).
The control strategy model calculates the effects of applying
selected sets of three emission standards (two in the case of S0_) to
every political jurisdiction in the control area. Selecting the applicable
results from the control standards model, the model develops a picture of
the change'in emissions (and costs) resulting from a realizable pollution
control alternative. Using the output of the atmospheric diffusion model,
the strategy model recomputes the pollutant concentration distribution
resulting from the new emission pattern, thus allowing a decision to be
made on the effectiveness of the strategy based on both the resulting
cost-s (on a regional, industry-by-industry, political jurisdiction-by-
political jurisdiction, or source-by-source basis) and the actual air
quality produced by implementation of the standards.
-------
APPENDIX B
ESTIMATING CONTROL COSTS
The industry-wide and regional costs of air pollution control pre-
sented in this report are obtained by a detailed source-by-source cost
analysis conducted with the Implementation Planning Program described in
Appendix A. These costs represent the cost to a major industry group or
region of buying, installing, and operating selected control devices in
order to control emissions to the degree required by an air pollution
standard. It should be noted that pollution sources which are smaller
than some predetermined size are aggregated and included in the analysis
as area sources (the large individual sources are called point sources).
These area sources play a role in the air quality predictions output by
the model but are not included in the cost calculations (regional control
agency costs are also excluded). Typically, the components of area
sources are automobiles, apartment house incinerators, small industries
and boilers, etc.
The basis used in this report for estimating the cost of control of
an individual point source is as follows:
A. The source is defined in some detail. Process
details, fuel consumption, stack exit conditions
and size, plant capacity, and other variables
are specified.
B. Those control devices which can be applied to
this class of source are selected; the total
annual cost of control and the device efficiency
are calculated for each selection. Costs are
based on source parameters, known device parameters,
and certain regional information such as labor costs,
interest rates, etc.
C. An emission standard is specified; from this, a
required collection efficiency is calculated.
D. The lowest-priced device of those that can equal or
exceed the desired efficiency is selected, with the
-------
proviso that if none are sufficient, the most
effective device will be chosen regardless of
cost. This method of selecting control devices,
based on individual source characteristics,
contrasts with that employed in the Cost of
Clean Air report. The methodology used there
involves the application of a single device
type to all the plants in a specified industry
group. Device selection is based on analysis
of a typical or model plant. The flexibility
of the IPP approach is illustrated with an
example from the Cincinnati 305(a) strategy.
Application of the San Francisco Process
Weight regulation for industrial process sources
to asphalt batching plants in the Kentucky juris-
diction dictates the selection of three device
types for the four applicable sourcesa medium
temperature fabric filter with a 99 percent
control efficiency, a medium efficiency cyclone
with a 75 percent efficiency, and two high
efficiency cyclones with 85 percent efficiencies.
Step B in the selection process deserves considerably more explana-
tion. The applicability-of a control device to a particular source de-
pends upon a large number of engineering factors related to both source
and device characteristics. For instance, some devices cannot handle the
high temperature corrosive gas streams associated with certain source
types. Other devices are designed for control of gaseous pollutants only,
or particulates only; a fabric filter is obviously unsuitable for control
of an SO source. The program handles the selection problem by identifying
sources using the four digit Standard Industrial Classification (SIC) and
specifying for each classification a list of applicable devices. A fifth
digit is added to the classification when more detail is necessary (for
instance, distinguishing between an electric arc and open hearth furnace).
Table B-l presents the range of applicability of the control devices
considered by the model for each source type. Note that all combustion
sources are subject to devices specified in the first row.
-------
TABLE B-l. APPLICABLE CONTROL DEVICES
souncE
TYPE
COMBUSTION
SIC Z04O Feed and Grain
SIC 2819 rmnl Sulfuric Acid
SIC 2819 nm»2 Nitric Acid
SIC 2819 iwr»3 Ammonium Nitrate
SIC 2822 Synthetic Rubber
SIC 2851 Paint and Varnish
SIC 2671 Fertilizers
SIC 3210 Flat Glass
SIC 3241 Cement Manufacturing
SIC 2951 Asphalt Batching
SIC325O Structural Clay Prod.
SIC 3295 Rockj sand & Gravel
SIC 3312 nnnl iron & steel/Blast Furn
SIC 3312 nrw2 Basic Oxveen Furnace
SIC33l2iwn3 Sintering
SIC 3312 nm4 Coking Operations
SIC 3312 nnn5 Electric Arc Furnace
SIC3312rmn6i Open Hearth Furnace
SIC 3323 Steel Foundries
SIC 3332 nnnl Lead Smelters/Sinterine
SIC 3332 nm2 Blast Furnar.e
SIC 3332 mn3 Reverbatorv Furnace
SIC 4953 Solid Waste Disposal
SIC 2911 Petroleum Refi- .-v
SIC 2800 Cheaical & Allied Prod.
m
0
0
o
0
0
0
0
0
0
0
0
0
fY>
o
0
0
o
0
0
0
0
0
0
o
0
0
0
0
0
0
0
rwi
n*>
nc
("17
0
0
o
0
0
0
0
o
0
o
rtt
0
0
0
0
0
0
0
0
0
0
no
0
O
o
0
0
0
0
0
o
in
0
0
0
0
0
0
0
o
0
0
0
o
0
0
o
0
0
0
0
0
0
0
0
0
f>
0
o
0
0
0
0
o
1 -I
o
o
0
0
14
0
0
I'S
0
COI
16
0
0
o
o
0
0
0
0
0
0
0
0
ITR
17
0
0
0
0
0
0
0
0
0
0
0
0
01
IB
0
o
0
0
0
0
0
0
0
0
0
0
DEV
10
0
ICE
20
0
n
21
0
PE
22
0
27
0
28
0
o
Yl
0
It
0
T>
o
V*
o
^
-------
Table B-l. APPLICABLE CONTROL DEVICES (Continued)
S3
o
Ol
POLLUTION REDUCTION DEVICES OR METHODS
001 Wet Scrubber - High Efficiency
002 Wet Scrubber - Medium Efficiency
003 Wet Scrubber - Low Efficiency
004 Gravity Collector - High Efficiency
005 Gravity Collector - Medium Efficiency
006 Gravity Collector - Low Efficiency
007 Centrifugal Collector - High Efficiency
008 Centrifugal Collector - Medium Efficiency
009 Centrifugal Collector - Low Efficiency
010 Electrostatic Precipitator - High Efficiency
Oil Electrostatic Precipitator - Medium Eff.
012 Electrostatic Precipitator - Low Efficiency
013 Gas Scrubber
014 Mist Eliminator - High Velocity
015 Mist Eliminator - Low Velocity
016 Fabric Filter - High Temperature
017 Fabric Filter - Medium Temperature
018 Fabric Filter - Low Temperature
019 -Catalytic Afterburner
020 Catalytic. Afterburner with Heat Exchanger
021 Direct Flame Afterburner
022 Direct Flame Afterburner with Heat Exchanger
027 Fuel Substitution to Low Coal A
028 Fuel Substitution to Low Coal B
029 Fuel Substitution to Low Coal C
030 Fuel Substitution to Low Sulfur Oil A
031 Fuel Substitution to Low Sulfur Oil B
032 Fuel Substitution to Low Sulfur Oil C
033 Fuel Substitution to Natural Gas
039 Catalytic Oxidation - Flue Gas Desulfurization
040 Alkalized Alumina
041 Dry Limestone Injection
042 Wet Limestone Injection
043 Sulfuric Acid Plant - Contact Process
044 Sulfuric Acid Plant - Double Contact Process
-------
The philosophy of calculating the actual cost of control given the
source and device type closely follows that expressed in the APCO "Control
Techniques for Particulate Air Pollutants" document. Total cost is based
on:
manufacturer's price
installation costs
annual capital charges
operation and maintenance costs.
These factors are in turn based on device characteristics, detailed source
characteristics (such as gas stream velocity and temperature, stack sizes,
fuel use, plant capacity, etc.), desired efficiency, and locational fac-
tors (utilities prices, labor charges, etc.). The model considered "point"
devices, i.e., devices with a specified design and efficiency. When a
range of efficiency or other design characteristics are necessary in order
to truly represent a class of devices, a "high, medium, and low" design
are utilized. Costs for the specified device are then dependent solely on
source and locational characteristics. Although a collection efficiency
is specified for each device, it is recognized that there are different
degrees of efficiency associated with the use of a device on different
kinds of sources, and efficiency correction factors have been installed in
the model for each, source type/device type combination.
Manufacturer's Price
The general form of the equation used to calculate manufacturer's
price is
2
y = a + bx + ex
where
y = cost in thousands of dollars
x = some measure of source size (Examples:
exhaust gas volume, steam electric power
plant, generating capacity in Mw, amount
of sulfur emitted in tons/day, etc.)
a,b,c = coefficients dependent upon device
characteristics.
Accuracy of the equations is enhanced by specifying different coefficients
according to varying ranges of a source variable. For instance, as the
-------
percentage P of SO- in the exhaust volume of a lead smelting plant varies,
the coefficient "a" of the cost equation for device 43 is allowed to vary
as follows:
010.0 a = 1.3449
Installation Cost
The cost of installing a control device on a particulate source is
calculated by applying a factor expressed as a percentage of the manufac-
turer's price. For devices which require extensive modification of the
source, installation cost is included in the manufacturer's price calcula-
tion.
Annual Capital Charge
The capital charge is based on the rated life of the control device
and the prevailing (regional) interest rates.
Operating and Maintenance Costs
A number of factors must be considered in determining the operating
and maintenance cost of a particular source-device combination: The
amount of power (calculated as electrical power) necessary to force the
effluent gas stream through the control device, the quantity of labor
required and its cost per hour, the cost of liquid or additional fuel used
by the device, the cost or credit resulting from disposal of the collected
pollutant, the replacement cost of certain device components (such as
filters), and others. For most devices, an equation of the form
0 & M = Z (factor unit cost) * (factor rate)
(Example: factor = electricity, kwhr/yr; rate = $/kwhr) can be used.
Other sources utilize equations based on manufacturer's specifications
and operating experience reported in the literature, and are of slightly
different form.
-------
Fuel Substitution
One "device" which is not included in the above discussion is
fuel substitution. Briefly, substitution costs are calculated by assuming
that the old fuel is replaced by a quantity of fuel which will yield an
equivalent heat energy output, taking into account the efficiency of the
boiler (which varies with fuel type) and the BTU content of both old and
new fuel. "Cost" of this control method is simply the difference between
the cost of the new fuel and that of the old.
Sources of Error
It should be recognized that details which are ignored in the cal-
culations can be quite important in the total cost for a single source
and can cause the calculation for that source to be substantially in error.
A partial list of possible sources of error is as follows:
Gas corrosive level or temperature which is well above
average for a particular source category.
Inaccessibility of site causing transportation costs to
be above average.
Special structural problems in the plant requiring expensive
construction techniques, structural modifications, etc.
Sudden high demand for device or fuel substitute following
passage of regulation causing prices to rise and/or shortages
to develop.
Zoning and other regulations requiring special noise
abatement procedures.
Possible differences in company maintenance procedures.
Possible need to expand electrical and waste treatment
facilities, water supply, etc., to accommodate device.
Scheduling problems requiring extensive overtime and/or
specialized labor.
Differences in device base prices due to degree of instru-
mentation, quality of workmanship and materials, competitive
market or lack of it, etc., etc.
-------
Despite these and other factors, the cost estimates for an entire industry
group, averaged over several sources, should be reasonable and quite ade-
quate to serve as a basis for decision-making with regard to selecting
control measures and regulations and for estimating total control costs
for a region. However, great care should be taken in interpreting the IPP
results for specific sources.
-------
APPENDIX C
AREA SOURCE SCALE FACTORS
The IPP model does not apply emission standards to area sources.
However, it is possible to manually input scale factors which will alter
the emission from these sources, thus simulating the effect on air
quality of applying new standards.
The effort involved in individually scaling every area source in
a region is very great. A more tractable approach is to assume that the
distribution of the various area source componentstransportation, resi-
dential space heating, etc.is fairly uniform; two scale factors (one
for each pollutant), applied to all area sources, will then suffice.
Using the Cincinnati AQCR as an illustration, area source scale
factors may be derived as follows.
Using data supplied by APCO's Division of Abatement, the composition
of the Cincinnati AQCR area sources can be defined in Table C-l. According
to the data, the area sources contain no industrial process emissions.
The emission standards to be applied to these sources can be obtained
by inspection from Appendix D. Note that all component sources of an area
source are, by definition, SMALL SOURCES; therefore, the lower-end of the
appropriate regulations apply. Table C-2 lists the standards for area
sources.
In order to scale down the emissions in Table C-l, one must first
calculate the emissions of the sources, on a unit basis, under present
conditions (or, under the conditions when the data in Table C-l was valid).
This can turn out to be a difficult task, because control of those emis-
sion sources which are included in an "area source" has not been very well
defined. For the most part, analysts have tended to compute emissions
from area sources by assuming that they are uncontrolled. This is probably
the most reasonable approach, and will be followed here. Emission factors
were obtained from Reference 3.
-------
TABLE C-l. COMPOSITION OF CINCINNATI AREA SOURCES
Component
Fuel Combustion Sources
Bituminous Coal
Resid
Distillate
Gas
Fuel Combustion Mobile Sources
Refuse Disposal Sources
Open Burning
Other
Percent of Total
Particulates
42.6
37.1
.5
1.3
3.7
22.8
34.6
23.4
11.2
100%
Emissions
so2
83.5
74.7
3.8
5.0
0
14.3
2.3
2.2
.1
100%
-------
TABLE C-2 305(a) STANDARDS FOR AREA SOURCES
PARTICIPATE ALLOWABLE EMISSIONS
Fuel Combustion - .6 pounds/10 BTU
Industrial Process - 11 pounds/ton input
Solid Waste Disposal - 6 pounds/ton refuse charged
S02 ALLOWABLE EMISSIONS
Fuel Combustion - 1.46 pounds/10 BTU
or 1 percent sulfur coal
or 1.38 percent sulfur oil
Industrial Process 500 parts/million
-------
In order to scale down the emissions in Table 3-3, one must first
calculate the emissions of the sources, on a unit basis, under present
conditions (or, under the conditions when the data in Table 3-3 was
valid). This can turn out to be a difficult task, because control of
those emission sources which are included in an "area source" has not
been very well defined. For the most part, analysts have tended to com-
pute emissions from area sources by assuming that they are uncontrolled.
This is probably the most reasonable approach, and will be followed here.
Emission factors were obtained from Reference 3.
Particulates
Bituminous Coal
Uncontrolled emission, pounds per ton £ 5A
where A = ash content by weight
S 8 percent in Cincinnati
Uncontrolled emissions = 40 pounds per ton
= .6 pounds/106 BTU
=15 pounds per ton
Allowable emissions = .6 pounds/10 BTU
(Heat content of coal is about 25 x 10 BTU per ton)
SCALE FACTOR = 15/40 = .375
Residual Fuel Oil, Distillate and Natural Gas
All three fuels produce very low particulate emissions
and satisfy.the fuel combustion standard.
SCALE FACTOR =1.00
Fuel Combustion, Mobile
This source is unaffected by the 305(a) standards.
SCALE FACTOR =1.00
Open Burning
The 305(a) report calls for elimination of open burning,
substitution of land fill.
SCALE FACTOR = 0
-------
Other Refuse Disposal
Uncontrolled emissions =.17 pounds per ton charged
Allowable emissions = 6 pounds per ton charged
SCALE FACTOR = 6/17 = .35
Bituminous Coal
Average sulfur level is 1 percent in Cincinnati, satisfies
standard.
SCALE FACTOR =1.00
Residual and Distillate
Average sulfur level in Cincinnati is, respectively, 1.36
percent and less than 1 percent, both of which satisfy
the standard.
SCALE FACTOR - 1.00
Gas
Natural gas emits essentially zero S02<
SCALE FACTOR =1.00
Fuel Combustion. Mobile
Unaffected by 305 (a) standards
SCALE FACTOR =1.00
Open Burning
Elimination of open burning.
SCALE FACTOR = 0
Other Refuse Disposal
Unaffected by 305 (a) standards.
SCALE FACTOR =1.00
The overall scale factor for each pollutant is simply:
SF = (Fraction of total emission, component i) x
i (Appropriate scale factor for i)
-------
SCALE FACTOR, PARTICULATES = .463
SCALE FACTOR, SO,
= .978
Table C-3 lists the scaling factors for all the detailed inventory
strategies investigated in this report.
-------
TABLE C-3 AREA SOURCE SCALING FACTORS
AQCR
Cincinnati
St. Louis*
Washington, D.C.
so2
Strategy
305(a)
305(a)
305(a)
1/2 305(a)**
Scale Factor
.978
.477
1.00
.7795
PARTICULATES
Strategy
305(a)
305(a)
1/4 305(a)
305(a)
1/2 305(a)
Scale Factor
.463
.463
.338
.73
.6323
* Data on breakdown of area sources was not available; Cincinnati
breakdown was used instead.
** 1/2 305(a) = twice as stringent as 305(a)
-------
APPENDIX D
EMISSION STANDARDS
The emission standards used as a base in this report are those
utilized in the 1970 "Cost of Clean Air" and the 1971 "Economics of
Clean Air" reports submitted to Congress in compliance with Section
305(a) of the Clean Air Act, as amended. There are five regulations
for the control of particulates and sulfur oxides:
Particulates
1. Fuel combustion sourcesMaryland Particulate
Emission Standard for Fuel Burning Installations,
allowable emissions versus equipment capacity
rating (Figure D-l).
2. Industrial process sourcesSan Francisco Bay
Area Pollution Control District Process Weight
Regulation, allowable emissions versus process
weight (Table D-l).
3. Solid Waste DisposalNew York State Particulate
.. Emission Standard for Refuse Burning Equipment,
allowable emissions versus refuse charged
(Figure D-2); banning of open burning.
S02
4. Fuel Combustion Sourcesallowable emission rate
of 1.46 pounds of sulfur dioxide per million BTU
input (equivalent to one percent restriction on
sulfur content, by weight, in coal, and a 1.38
percent restriction in oil).
5. Industrial Process Sourcesallowable emission
of 500 parts per million of sulfur dioxide.
-------
TABLE D-l. Allowable Rate of Particulate Emission
Based on Process Weight Rate
Process weight rale
lb:./hr Tons.hr
100 ..
2IK)
40(1
I IX)
t(XI
I nun
I.JWI .
2.0UII.. .
2.M!I>. .
3 11, d
3 Mill
4 000
bOOU
6000
7.floo...
8000
9000
lo.oco. .
12,000. .
005
0 10
0?0
030
0.40
ObO
07b
1 00
1 ?5
1 SO
1.7b
?.oo
?,50
3.00
.. 3.50
4.00. . . .
4.JO
5.00
6.00
Rale ol
(mission,
Ibs.lir
0.5M
.877
1.40
1.83
2.22
2. b8
3.38
4.10
4.76
5.38
b.96
6.52
7.58
8.56
9.49
10.4
11.2
12.0
13.6
Process weight rate
Lbs/hr Tons/hi
IfcOOO ..
1C 000
20(100
30000
40.000
bOOW)
CO.OdO.
71) OCO
80 OPO
911.1'UO
luoono
1?OCOU
140000
IGO.OOO .
200,000..
1 000 000
2,000 COO.
c.ooo.ooo
8
9
10
IS .
. 20
25
30 . ..
35
<0
... 45
bO
CO
70
80
. 100
500
1 000 .
3000
Rale ol
emission,
Ibs/hr
16.5
17.9
19.2
25.2
30. b
35.',
40.0
41.3
42.5
43.6
44.6
46.3
47.8
49.0
51.2
69.0
77.6
92.7
i Data in (Ills table can be interpolated for process weichl 'ales up to 60,POOIb1ii by using equation: E-4.10 P""; and
can pe interpolated and f iliapol.iM lur process weipht lates in excess ol 60,000 Ib. hr by using equation: £ = 55.0 P» " -40
(E' rale of emission in lb.hr; P^picctss weight fate in tcnsyhr).
-------
Figure D-l. Maryland Particulate Emission Standards
For Fuel Burning Installations
sqvinaifr urtciTt uiixc, 10* ITOA.r
Figure D-2. New York State Particulate Emission Regulation
For Refuse Burning Equipment
50,000 1C".V1
------- |