U.S. ENVIRONMENTAL PROTECTION AGENCY
STUDY OF S02 AROUND THE
CRANE, RIVERSIDE AND WAGNER POWER
STATIONS IN METROPOLITAN BALTIMORE
FINAL REPORT
EPA-903/9-76-024
U.S. Environmental Protection Agency
Region Is I Information Resource
Center (3^52)
£-U niiv.iiut Street
Phiiadtipiiia, PA 19107 •'' ';"'
EPA Report Collection
Information Resource Center
US EPA Region 3
Philadelphia, FA 19107
MIDDLE ATLANTIC REGION-HI 6th and Walnut Streets, Philadelphia, Pennsylvania 19106
-------
I
I
I
I
I
I
I
I
I
I
I
I
IU ? E:,;;rrr>r.:.>y Prefedisn Agency
r -'. r; '' !; V.i'a!inn ReSGlifCQ
I
I
I
I
I
. _
b<- :. .1 (;-. ( ^f
I'l Gr.jrlnaJ St
fi)-.2!k!?J*ia, FA 19107
August 1976
Under Contract Number 68-0202331
William E. Bel anger, Project Officer
STUDY OF S02 AROUND THE
CRANE, RIVERSIDE AND WAGNER POWER
STATIONS IN METROPOLITAN BALTIMORE
FINAL REPORT
EPA-903/9-76-024
by
Robert C. Koch and Kenneth E. Pickering
GEOMET, Incorporated
15 Firstfield Road
Gaithersburg, Maryland 20760
prepared for
U.S. Environmental Protection Agency
Middle Atlantic Region - III
Philadelphia, Pennsylvania 19106
-------
I
This air pollution report is issued by Region III, Environmental
Protection Agency, to assist state and local air pollution control agencies _
in carrying out their program activities. Copies of this report may be I
obtained, for a nominal cost, from the National Technical Information Ser- ™
vice, 5285 Port Royal Road, Springfield, Virginia 22151.
I
I
I
I
I
I
I
I
I
I
I
This report was furnished to the Environmental Protection Agency |
by GEOMET, Incorporated, Gaithersburg, Maryland in fulfillment of EPA Con-
tract 68-02-2331. This report has been reviewed by Region III, EPA and
approved for publication. Approval does not signify that the contents •
necessarily reflect the views and policies of the Environmental Protection •
Agency, nor does mention of trade names or commercial products constitute
endorsement or recommendation for use. •
Region III Publication No. EPA-903/9-76-024 —
-------
I
I
• FOREWORD
_ Under Section 2 of the Energy Supply and Environmental
• Coordination Act, the Federal Energy Administration has issued orders
• to the Crane, Riverside, and Wagner Power Plants which prohibit the
burning of oil or natural gas in some of their boilers, thereby forcing
• a conversion to coal. EPA must determine the level of emission of
pollutants which can be tolerated from these units without jeopardizing
• national primary ambient air quality standards. In this study GEOMET,
m Incorporated, suggests possible upper limits on emission of sulfur diox-
ide from these power plants based on modeling.
I
| William E. Bel anger
Project Officer
U.S. Environmental Protection Agency
• Region III
I
I
I
I
I
I
I
-m-
-------
PREFACE
I
I
Three power plants in the vicinity of the Baltimore Metropolitan I
area, namely the Crane, Riverside, and Wagner plants, are to be prohibited
from burning gas or oil in some of their boilers. This study was under- |
taken to determine what limitations are required on the sulfur content of .
coal burned in these boilers and on the S02 emissions from these boilers. ™
The primary emphasis of the study is on determining the impact •
of S02 emissions from the three plants on the maximum 24-hour concentration
of S02 in the Metropolitan Baltimore area. The study was conducted in •
three phases; including a review of present and future (through 1979) S02
emissions for the affected area, development and validation of a model for •
representing 24-hour concentrations of S02 in the area, and use of the model •
to identify worst-case days and to determine the maximum SC^ emissions which
can be tolerated from each plant. •
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
ABSTRACT
A study of sulfur dioxide levels in the vicinity of the Crane,
Riverside, and Wagner power plants in Metropolitan Baltimore has resulted
• in an evaluation of the maximum allowable sulfur dioxide emissions from
these three plants, such that the primary National Ambient Air Quality
P Standards (NAAQS) will not be violated. The study considers the overlap
_ of plumes from the Metropolitan Baltimore area, from large industrial
™ plants, from other power plants, and from surrounding areas with each of
• the three power plants of interest. The persistence of meteorological
conditions associated with the worst overlap conditions (highest S02 con-
• centrations) for each plant was identified and used to define a worst-case
situation for each plant. The SCIM urban diffusion model was used to pre-
dict S02 concentrations for the worst-case situation associated with each
plant.
-v-
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
t
I
I
I
I
CONTENTS
Foreword i i i
Preface iv
Abstract v
Figures viii
Tables ix
Acknowledgement xi
1.0 Introduction 1
2.0 Conclusions 3
3.0 Data Acquisition 5
3.1 Emissions Inventory 5
3.2 Ambient S02 Data 13
3.3 Meteorological Data 17
4.0 Methodology 21
4.1 Air Quality Model 21
4.2 Model Modifications 22
4.3 Model Validation 29
5.0 Model Application and Analysis of Results 45
5.1 Determination of Worst Case Conditions 45
5.2 Determination of Sulfur Limits for Coal
Consumption 57
6.0 References 61
Appendices
A. A Multiple Source Gaussian Plume Model Using Sampled
Chronological Input (SCIM) 63
B. SCIM Model Calculations Compared to Measured 24-Hour
SO;? Concentrations 81
C. Emission Inventories for Baltimore AQCR 91
-v1f-
-------
FIGURES
I
I
I
Figure Page •
1 Identification Numbers for Monitoring Stations 33
2 Relationship Between Measured and Calculated 24-Hour •
S02 Concentrations 34
3 Ratio of Mean Measured to Mean Calculated (24-Hour •
Values) 36 "
Relationship Between Measured and Calculated 3-Hour •
S02 Concentrations 37
rea Sources with S02 Emissions Greater Than 1.5 pg/ •
2/sec 50 f
I
I
I
I
I
f
I
I
I
I
-vm-
-------
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLES
Table Page
1 Processing Emission Data 8
2 Building Dimensions of Major Point Sources 9
3 Rejected S02 Data 14
4 Air Quality Monitoring Stations in the Baltimore Area 16
5 S02 Stations with Exposure Limitations 18
6 SCIM Plume Heights Using Original and Revised Estimates 23
7 Hourly Space Heating Demand Threshold Temperatures
and Demand Factors 28
8 Meteorological Characteristics of 10 Days Selected
for Validation Analysis 31
9 Summary of 24-Hour Validation Comparisons for
Concentrations, yg/m3 32
10 Location, Date, and Wind Directions for 24-Hour Model
Estimates of S02 Which Vary From Measurements by More
Than 53 ug/m3 (0.02 ppm) 38
11 Measured and Calculated S02 Concentrations at the Sun
and Chesapeake Streets Monitoring Station on May 11,
1974 40
12 Measured and Calculated S0£ Concentrations at the
Robinson and Toone Streets Monitoring Station on
December 4, 1973 " 42
13 1-Hour S&2 Concentrations (Single Source) 46
14 Point Source Combinations Producing High 1-Hour Con-
centrations Under Plume Overlap Conditions 47
15 Number of Occurrences of Greater Than or Equal to the
Specified Number of Hours Having the Following Wind
Directions During a 24-Hour Period in January, February,
November and December (1971-1975) 48
(Continued)
-ix-
-------
I
I
TABLES (Concluded) •
Table • Page A
16 Critical Day - Wagner 52
17 Critical Day - Riverside 53 |
18 Critical Day - Crane 54 —
19 Maximum Concentrations Under Present Operating •
Conditions 56
20 Hypothetical Maximum Allowable Coal Sulfur 59 Jj
A-l Wind Speed Profile Exponents 70 g
A-2 Meteorological Stability Classifications for •
Characterizing the Diffusion Parameters (a and a ) 71
A-3 Net Radiation Index Values 71 •
A-4 Fitted Constants for Urban Diffusion Parameters Based g
on Turner Stability Classifications 73 •
I
I
I
I
I
I
I
I
-x-
-------
I
I
I
ACKNOWLEDGEMENTS
The cooperation of the Maryland Bureau of Air Quality and
Noise Control in furnishing S02 emission data and air monitoring data
• and in providing helpful information in using this data is gratefully
acknowledged.
• The guidance and constructive suggestions of Mr. William E.
IBelanger and Dr. Peter L. Finkelstein of the Region III Office of the
Environmental Protection Agency were most helpful to this study and
tt are strongly reflected in the general approach followed.
The efforts of Ms. Susan Steed of GEOMET, Incorporated, who
M performed most of the computer programming, is also greatly acknowledged,
I
1
i
i
i
i
i
i
-xi-
-------
I
I
I
I
I
I
I
t
I
I
I
I
I
I
I
I
I
I
Section 1.0
INTRODUCTION
-------
I
I
I
Section 1.0
INTRODUCTION
£ GEOMET, Incorporated has completed a study of sulfur dioxide
levels in the vicinity of the Crane, Riverside, and Wagner power plants
• in Metropolitan Baltimore. The study has resulted in an evaluation of
the maximum allowable sulfur dioxide emissions from these three plants,
m such that the primary National Ambient Air Quality Standards (NAAQS) will
m not be violated. The Federal Energy Administration has ordered certain
' boilers at the three aforementioned power stations operated by Baltimore
• Gas and Electric Company to cease burning oil and natural gas, thereby
facing a conversion to the burning of solely coal. GEOMET has determined
£ the maximum sulfur content of the coal that may be burned in each plant
^ without the level of sulfur dioxide in the ambient air in the vicinity of
W the plants exceeding the primary NAAQS. The study emphasized the impact
• of the SOp emissions from the three power plants on the 24-hour concen-
tration of S02 in the Metropolitan Baltimore area. Overlap of plumes
I from the Metropolitan Baltimore Area, from large industrial plants, from
other power plants, and from surrounding areas with each of the three
9 power plants of interest was considered. The persistence of meteorological
« conditions associated with the worst overlap conditions (highest SOg con-
* centrations) for each plant was identified and used to define a worst-case
M situation for each plant. The SCIM urban diffusion model was used to pre-
dict SOo concentrations for the worst-case situation associated with each
| plant.
I
I
-1-
-------
Section 2.0
CONCLUSIONS
-------
I
I
I
I
I
t
1
I
I
I
I
I
1
I
I
1
I
I
Section 2.0
CONCLUSIONS
Based on a determination of the worst-case meteorological con-
ditions associated with the Crane, Riverside and Wagner power plants in
the Metro!olitan Baltimore area and a modeling analysis of the contribution
of each plant to the maximum ground concentrations of S0£ under those
conditions, the following conclusions are drawn regarding each plant:
t Crane Power Plant - Although increased S02 emissions
from this plant will degrade air quality, the NAAQS
will not be exceeded if the present emissions are
tripled. The allowable sulfur content in coal from
this plant could be increased from 1 to 3 percent.
• Riverside Power Plant - S0£ emissions from this plant
could be doubled without exceeding ambient air quality
standards. The allowable sulfur content of coal
burned at this plant could be as high as 2 percent.
However, it is recommended that a more detailed study
of air quality resulting from this plant based on more
detailed emission and meteorological data than is
presently available be made before approving such an
increase.
• Wagner Power Plant - S02 emissions from this plant
should not be increased. The allowable sulfur content
of coal burned at this plant should not exceed the
current limit of 1 percent.
-3-
-------
I
I
1
I
i
t
Section 3.0
DATA ACQUISITION
-------
•
I
I
Section 3.0
DATA ACQUISITION
I
I
I
3.1 EMISSIONS INVENTORY
m
3.1.1 Metropolitan Baltimore Air Quality Control Region
y Sulfur dioxide emissions from the Metropolitan Baltimore Air
^ Quality Control Region (AQCR 115) were obtained from the Maryland Bureau
* of Air Quality and Noise Control (BAQNC). The inventory data which are
• presented in Appendix C were developed for use in the Air Quality Display
Model (AQDM) and include point and area sources for the base year 1973.
• Projected emission data were also available from BAQNC for the years 1980
and 1985. Appendix C shows the 1980 data which was used in this study.
V Sources emitting 25 tons/year or more of S0~ were included in the inventory.
IThe data for each point source included emission rate in tons/day, location
coordinates (Maryland State Grid), and stack parameters (height, gas exit
fl temperatures, diameter, and gas exit velocity). All point sources con-
taining more than one stack were combined into a single source with a
£ representative stack parameter. The exception was the Bethlehem Steel
plant at Sparrows Point where the emissions were distributed into 22
«• separate point sources with individual stack parameters. The point sources
•j account for 90 percent of the total sulfur dioxide emitted in the Baltimore
AQCR.
Area source emission rates were estimated for 338 grid square
of various sizes based on a grid system designed by Baltimore Gas and
Electric Company (BG&E) to record the number of residential and commercial
-5-
-------
I
I
meter connections. The grid squares ranged in size from 2 km to 16 km
on a side. Area source emissions were originally estimated by BAQNC on •
a county basis and were allocated to the particular grid squares on the
basis of the percent of county residential and commercial meters in each m
square. The emission height used was an average height based on standard «
stack heights of residential, commercial, and industrial sources.
Future emissions from point sources were estimated by BAQNC •
using the latest information available concerning phase-out, expansion
or new construction. If no information was available for a particular J
source, the emissions were kept constant. The sulfur content of residual _
fuel oil and coal was maintained at the present 1 percent and distillate "
oil was maintained at 0.3 percent sulfur. Power plant emissions were •
obtained from BG&E's 10-year projected fuel use.
Future emissions for area sources were calculated by BAQNC •
based on proportional changes in activities in each Regional Planning
District within the AQCR. The projected change in the number of house- •
holds was the determining growth factor for the residential portion of ^
the area source emission. The projected change in employment at *
commercial, industrial, and institutional establishments was the tt
growth factor for the remaining area source emissions. In general, coal
use was reduced from the 1973 levels for all four types of establishments •
for 1980 and 1985, and the new fuel mix was computed for each type of
source. •
In addition to the AQDM emission data, the National Emissions m
Data System (NEDS) file for Maryland was obtained. This provided infor-
mation concerning the number of days per week and hours per day that V
I
-------
I
I
the major point sources were in operation, so that the emissions could
m be distributed according to day of the year and hour of the day. The NEDS
data also provided figures of percent annual throughput by season for
• several sources. Generally, 25 percent was given for each season; there-
fore, the total emissions needed to be adjusted on a seasonal basis for
J[ only a few sources.
^ A second supplementary source of emission data was the Maryland
" BAQNC emission inventory. The data for the Metropolitan Baltimore AQCR
• was reviewed to insure that all major point sources had been included in
the AQDM emission listing. The State emission data is divided into three
• record types, the first being emissions due to industrial processes. The
second record type includes non-processing industrial emissions, commercial
$ and institutional emissions, while record type three is solely incinerator
M emissions. The processing emissions data provided the number of shifts
worked per day and days of operation per week. The amount of emission
1 due to space heating was determined by subtracting the process emissions
from the total. Data presented in Table 1 were used to estimate hourly
| point source emissions due to space heating and process operations.
Monthly fuel consumption data for the five major BG&E power
• stations were obtained from BAQNC for the years 1972 through 1975. Also,
• copies of Federal Power Commission Form 67 were available for 1973 and
1974. Therefore, monthly emissions could be calculated for the power
• plants.
Information concerning the dimensions of plant buildings was
| used to determine the effects of aerodynamic downwash for the major point
I
I
-7-
-------
TABLE 1. PROCESSING EMISSION DATA
AQDM
Source Number
(Appendix C)
2
13
20
24
34
42
55
57
58
59
60
61
62
66
68
72
73
75
76
78
82
87
88
92
98
106
111
113
114
118
119
124
126
127
132
Plant Name
Kennecott
U.S. Agri-Chemical
Many T. Campbell Corp. - White Marsh
Arundel Corp. - Maryland Slag
Lehigh Portland Cement
York Building Products
Exxon Company
GAE Corporation
Chevron Asphalt Co.
FMC Corp. - Organic Chemicals
Olin Matheson
Allied Chemical
Davison Chemical
Continental Oil
Glidden - Durkee (Hawkins Point)
Tomke A luminum
Proctor & Gamble
Maryland Glass Corp.
Southern Industries
National Gypsum Co.
American Oil Co.
American Sugar
M & T Chemical
Armco Steel
Shell Oil
Philadelphia Quartz
Bethlehem Steel - Penwood
Bethlehem Steel - 7th Street
Bethlehem Steel - Tin Mill
Bethlehem Steel
Bethlehem Steel
Bethlehem Steel - Coke Battery
Bethlehem Steel - Open Hearth
Bethlehem Steel - Plate Mill
Bethlehem Steel
SO2 Processing
Emissions
(tons/ day)
0.170
0,016
0.085
19.780
0.090
0.240
0.123
0.230
0.380
0.016
7.750
0.310
0.025
0.510
4.030
0.031
0.145
0.120
0.022
0.790
2.370
0.088
3.440
0.010
0.078
0.105
9.676
0.371
0.220
3.737
0.018
14. 123
3.737
2.827
3.737
Plant
Days Per
7
5
6
7
7
6
7
6
7
7
7
7
5
7
7
5
5
7
5
5
7
5
7
7
7
7
7
5
7
7
5
7
7
5
7
Operating Schedule
Week Shifts Per Day
3
3
1
3
3
1
3
2
3
3
3
3
3
3
3
3
3
3
2
3
3
3
3
3
3
3
3
2
3
3
2
3
3
3
3
-8-
I
I
I
I
I
I
I
I
I
I
I
I
I
1
I
I
i
i
-------
I
I
9
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
sources. This data was obtained by visits to the 16 major point sources.
The building height and length which were determined are presented in
Table 2. Based on this data it was found that pronounced downwash effects
occur near the American Sugar plant where the stacks are only 6 m higher
than the tallest portion of the building. For other plants the downwash
effects are smal1.
TABLE 2. BUIUDING DIMENSIONS OF MAJOR POINT SOURCES
AQDM
Source Number
(Appendix C)
1
14
16
17
51
52
55
58
59
60
68
73
87
88
96
124
Plant Name
Baltimore Gas & Electric (Wagner)
Spring Grove State Hospital
Baltimore Gas & Electric (Riverside)
Baltimore Gas & Electric (Crane)
Baltimore Gas & Electric (Westport)
Baltimore Gas & Electric (Gould St. )
Exxon Company
Chevron Asphalt Co.
FMC - Organic Chemicals
Olin Matheson
Glidden - Durkee (Hawkins Point)
Proctor & Gamble
American Sugar
M & T Chemical
American Oil
Bethlehem Steel - Coke Battery-
Building Height
(ft)
130
50
90
130
80
120
30
30
40
50
SO
40
130
25
20
100
Building Length
(ft)
250
70
300
150
450
300
200
175
250
250
350
100
200
250
300
1000
-9-
-------
I
I
3.1.2 Surrounding Air Quality Control Regions
An analysis was made of the extent of the area which the emission H
inventory should encompass by considering S(k concentrations as a function
of distance due to emissions by the three power plants under study and by ft
considering SC^ concentrations in the areas affected by these plants due m
to emissions from surrounding AQCR's.
The long-distance effects of the three power plants was investi- I
gated using the PTMAX and PTMPT programs from EPA's UNAMAP system. The
distance to maximum ground-level concentrations outside the Metropolitan I
Baltimore area from the emissions by the three power plants was determined
for a range of wind speeds. Highest concentrations occurred for the over- •
lap of plumes from two of the plants, at distances of 25 to 40 km from •
the upwind point. The only direction toward which the contribution from
the overlapped power plant plumes can be joined by sizable contributions •
from sources outside the Metropolitan Baltimore AQCR is toward Washington,
D.C. I
Highest ground-level concentrations at long distances from ^
sources are associated with stable conditions, but wind speeds must be *
sufficient to carry the plume the necessary distance to Baltimore during •
the time period that such conditions exist. The effects of the six AQCR's
bordering on the Metropolitan Baltimore AQCR were tested using a simple •
area source based on equations from Workbook of Atmospheric Dispersions
Estimates (Turner 1970). Total AQCR S02 emissions from the 1972 National •
I
I
I
-------
I
I
I
I
I
I
Emissions Report (EPA 1974) were used in making the estimates, treating
each AQCR as an area source. The calculations are based on:
u —j-y exp {-
I •"' l
where
H = source height
a (x) = vertical diffusion parameter
• x = distance from receptor to upwind edge of area source
x. = distance from receptor to downwind edge of area source
~ = relative concentration.
Based on the assumption of maximum persistence of E stability of 12 hours,
M the maximum exposure from each AQDM was estimated assuming a travel speed
for the plume of 9 km/hr. Maximum exposure times ranged from 1 hour for
I the Metropolitan Philadelphia AQCR to 7 hours for the Central Maryland
AQCR. Maximum 24-hour contributions from the surrounding AQCR's were
P significant for the National Capital, Metropolitan Philadelphia and South
_ Central Pennsylvania AQCR's, with values of 30, 33, and 105 ug/m3 respec-
™ tively. Based on this finding, a decision was made to include emissions
• from these three AQCR's in the analysis. Approximately three-fourths of
the emissions from the National Capital and South Central Pennsylvania
• regions are due to power plants. Hence, it was decided that these regions
can be adequately represented by the power plant emissions alone. There
-11-
-------
I
I
are no other large sources of S02 in the National Capital AQCR. However,
primary metal plants represent an additional 10 percent of emissions in m
the South Central Pennsylvania AQCR and were included in the emission m
inventory. Power plants comprised only about 40 percent of the emissions *
from the Metropolitan Philadelphia AQCR. Therefore all major industrial •
point sources were included along with the power plants for the Metropolitan
Philadelphia AQCR. |
Federal Power Commission Form 67 was the source of data for the
power plants in the three regions mentioned above. Emissions were calcu- ™
lated based on the monthly fuel consumption and emission factors from *
AP-42 (EPA 1973). Data for sources emitting greater than 100 tons/year
in the city of Philadelphia were obtained from the City Department of •
Public Health. Data for sources in the four Pennsylvania counties sur-
rounding Philadelphia and for five primary metal plants in South Central H
Pennsylvania were obtained from the Department of Environmental Resources _
in Harrisburg. The New Jersey Department of Environmental Protection ™
furnished copies of NEDS point source listing for the five New Jersey V
counties in the Metropolitan Philadelphia AQCR which were used to select
appropriate New Jersey source data. Seven major point sources in New I
Castle County, Delaware were selected from the emission inventory data
acquired from the Delaware Department of National Resources and Environ- •
mental Control. One additional large source which was judged to have a m
significant influence on the Baltimore area was also added to the emission
data for this study, namely the Morgantown power plant in southern •
Maryland. These data completed the emission inventory.
I
I
-------
I
I
I
3.2 AMBIENT SO,, DATA
3.2.1 Review of Data
I
I
I
^ Air quality data from monitoring sites in the Baltimore area
™ are necessary for the model validation phase of the study. The SAROAD
• file of sulfur dioxide monitoring data for the Baltimore area non-
continuous (sixth-day sample) stations was obtained for the years 1973
I and 1974. Continuous (hourly) S02 data were provided by the Maryland
BAQNC. Both sets of data were reviewed to verify the validity of high
S02 readings. Twenty-four hour average values in excess of 100 ug/m
3
m and hourly values in excess of 300 ug/m were checked to determine if they
were reasonable (i.e., downwind of major point or area sources for a
flj sufficient length of time, or other stations also recorded high values
on same day). Table 3 shows those values that were eliminated from con-
• sideration as being unreasonably high. The first two values were deter-
mined to be keypunching errors. The remainder of the rejected data
™ (half of which was from one station) was associated with wind directions
• which would not be expected to produce high concentrations.
• 3.2.2 Evaluation of Monitoring Sites
™ Sulfur dioxide data was available from a total of 36 stations
• in the Baltimore area for at least part of the period under consideration.
Nine of these stations recorded S02 levels continuously, while the remain-
• der recorded only 24-hour average values. Twenty-four hour ambient air
samples are collected by Research Appliance Company (RAC) portable five-
0 gas samplers. The wet chemical bubbler system samples fr.om 200 to 300
-13-
-------
TABLE 3. REJECTED SO2 DATA
SAROAD
Station Number
210080001
210080001
210120010
210120019
210080003
210120016
210120016
210120018
210120023
210120023
210120023
210120023
210120023
210120023
210120023
210120023
210120024
210680001
211360002
Location
Lintfaicum
linthicum
Robinson & Toone Sts.
Green & Lombard Sts.
Glen Burnie
Poly- Western High
Poly- Western High
Calvert & 22nd Sts.
Ft. McHenry
Ft. McHenry
Ft. McHenry
Ft. McHenry
Ft. McHenry
Ft. McHenry
Ft. McHenry
Ft. McHenry
S. Linwood & O'Donnell
Essex
Riviera Beach
Date
7/13/74
7/14/74
1/11/74
1/18/73
1/11/74
1/9/73
1/15/74
12/7/74
2/10/74
2/28/74
6/22/74
10/2/74
10/20/74
12/13/74
12/19/74
12/25/74
12/7/74
1/11/74
12/13/74
S02 (ptg/m3)
557.7
882.0
205.4
221.1
118.0
112.0
106.0
120.0
169.0
129.0
123.0
112.0
115.0
140.0
145.0
153.0
102.0
110.0
125.0
-14-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
g
1
1
liters per bubbler, depending on sampling rate and period. The RAC five-
gas instrument is one of the most widely used systems for sampling
gaseous pollutants in ambient air. Samples are analyzed for SO,, in the
ff laboratory according to a modification of the standard reference pararos-
aniline method, outlined in Appendix A of the April 30, 1971 Federal
Register. Concentrations of sulfur dioxide in the range of 25 to
3
1050 ug/m may be measured by this technique.
Most of the continuously monitoring stations in the Baltimore
4| area utilize Meloy Laboratories Model SA 185 total sulfur analyzers.
These instruments use a flame photometric detector to analyze sulfur gas
W pollutants. The range of the monitor is from the minimum detectable
sensitivity of 0.005 ppm to a maximum of 1 ppm, and the output is linearized.
£ The instruments are calibrated using permeation tubes (the only dynamic
I method for which National Bureau of Standards certification can be
"
obtained). Temperature control is maintained by a water-bath device
II specifically designed by the Maryland BAQNC which is not commercially
available. The remainder of the monitoring stations use Thermo Electron
£ Corporation pulsed S02 ambient air analyzers. The minimum detectable
concentration is 0.005 ppm. Using pulsating ultraviolet energy, S0?
W molecules are excited and they emit detectable radiation in direct
<• proportion to the S02 concentration in the gas sample for continuous
•
1
I
monitoring in real time.
All monitoring stations for which data were available are
listed in Table 4 along with their station numbers and Maryland State
Grid coordinators. Twenty-four stations in Baltimore City and Baltimore
-15-
-------
TABLE 4. AIR QUALITY MONITORING STATIONS IN THE BALTIMORE AREA
Reference
Number
25
29
4
s
6
13
u
30
31
23
26
IS
20
24
1
19
7
32
33
3
9
10
22
11
17
34
3
16
28
33
12
27
36
21
13
37
SAROAO
Station Number
210060002
210080001*
210080002
210080003
210080006
210080008
210120004*
210120006
210120007
210120010*
210120011*
210120012*
210120015
210120016
210120018*
210120019*
210120020
210120021
210120023
210120024
210120025
210140002
210140003*
210140004
210180001
210500001
210620001
210680001*
210920002
210960003
211040001
211060001
211300001
211300020
211360002
211720002
Location
Annapolis
xiiathicum
Odeaton
Glen Bumie
Harmoni
Harwood
200 Read St.
V . t. Police Station
M. W. Police Station
xRobison 0 Toone SB.
Sim & Chesapeake Sts.
X2000 Wllmuca Ave.
xMorgan State University
xPoiy-Western High School
Calven & 22nd Sa.
Green & Lombard Sts.
xjohns Hoptdns Hospital
Middle River
Ft McHenry
S. Linwood 5 O'Donneil
Patapsco STP
Ft. Howard
CanisoB Police Station
Citotuville
Bel Air
Cockeysville
Duadaik
Essex
Whiteiord
Simpsonville
Lansdowne
Laurel
Bowie
Belaville
Rivera Beach
Westminster
Maryland
Horizontal
943,400
398,500
390,000
904,000
386,300
911,000
909,330
917,900
388,100
921,400
917,450
396,150
917,000
900,600
909,000
906,400
913,300
967,000
918,000
920,000
924,000
957,000
366,600
373,650
983,600
397,600
939,300
948,600
985,200
333,700
397,500
340,100
354,230
323,000
938,400
302,600
State Coordinates
Vertical
418,600
499,000
449,000
488,300
482,300
360,730
534,300
549,400
331,400
326,700
511,400
524,100
551,000
351,300
339,500
329,300
534,250
348,000
522,000
327,500
510,400
497, 100
572,700
523,400
620,650
603,200
514,300
538,200
682,000
492,700
300,000
464,000
411,600
437,500
483,250
623,000
Continuous monitoring (hourly data).
Station no longer in operation.
See location in Figure 1.
-16-
I
I
I
I
I
I
I
I
I
I
I
1
I
I
I
t
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
County were visited to determine whether the sites have suitable exposure
and if they are susceptible to interference from nearby chimneys or major
sources of S02- The majority of the stations were located on the roofs
of schools, fire departments or police stations. These are locations
which generally give acceptable exposure as the free flow of air is not
interrupted. Six stations were judged to have possible interference
problems with certain wind directions. Table 5 indicates those stations
with interference problems, the cause of the problem and the wind direction
under which the problem would develop. Data from the Fort McHenry station
was omitted since there was a definite interference problem until the
station was moved on March 19, 1976 to its present site. The previous
site was on the roof of the nearby Naval Reserve Center but only 9 m below
the chimney. Of the five sites listed in Table 5, the station previously
operated at Robinson and Toone Streets probably had the worst interference
problem due to the Exxon plant, a major source of SO,, emissions located
approximately 700 m to the ESE of this station site. Information con-
cerning these exposures and interference problems was useful in determining
which data should be used in the model validation.
3.3 METEOROLOGICAL DATA
The SCIM diffusion model requires hourly surface weather obser-
vations of ceiling height, cloud amount, temperature, and wind speed and
direction. Such data was obtained on microfiche for Baltimore-Washington
International Airport for the years 1971-1975 from the National Climatic
Center (NCC) in Ashville, North Carolina. The airport is located in
-17-
-------
suburban Anne Arundel County approximately 13 km SSW of the center of
downtown Baltimore. The SCIM model also requires daily estimates of the
morning and afternoon mixing heights from atmospheric sounding data.
Radiosonde data for Dulles International Airport near Sterling, Virginia
(the closest National Weather Service upper air station to Baltimore)
was obtained on magnetic tape from NCC for the years 1971-1974. The data
included both standard and significant pressure levels.
TABLE 5. SOa STATIONS WITH EXPOSURE LIMITATIONS
SAROAD
Station Number
Location
Cause of Problem
Wind Direction
210120018
210120010
210620001
210120019
211040001
Calvert & 22nd Sts.
Robinson & Toone Sts.
Dundalk
Green & Lombard Sts.
Lansdowne
60 m from and 9m SW
below nearby small
power plant stack
700 m from 30 m ESE
industrial stack
School heating SE
plant chimney is
9 m above intake
Hospital heating NNE
plant stack 90 m
from trailer
School heating N
chimney 30 m
North and 5 m
above intake
Hourly surface meteorological data from Phillips Army Airfield
at Aberdeen Proving Grounds, Maryland, was obtained from NCC and was
examined for July and August for both 1973 and 1974. This data was used
to determine the frequency of a bay breeze from the Chesapeake Bay.
Phillips Field is located 6 km NW of the main section of the bay and 5 km
-18-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
SW of a portion of the bay that extends farther westward to the north of
• the proving grounds (see map in Figure 1). This location is much closer
to the bay than Baltimore-Washington International (BWI) and it was
• thought that the bay breeze might be apparent at Phillips Field when it
is not apparent at Baltimore Airport. The bay is approximately 20 km to
£ the east of Baltimore Airport. The bay breeze could be affecting a signif-
— icant part of the Baltimore Metropolitan Area and go undetected at the
™ Baltimore Airport.
• July and August were chosen to analyze the bay breeze influence
since they are the months usually characterized by strong solar radiation
V and relatively stagnant (weak pressure gradient) conditions - both necessary
ingredients for a well-developed bay breeze. The Phillips Field data were
£ available for 44 days during this period for each of the two years. Surface
^ weather maps were examined for these dates and a total of 31 days were
* chosen as prime possibilities for the development of the bay breeze. How-
• ever, only 1 day (2.2 percent) of the 44 1973 days was found to have a
well-developed bay breeze at Phillips Field which was not indicated at the
I Baltimore Airport. No days were found in the 44 days of 1974.
The frequency of a wind difference from the Baltimore Airport
• most likely would be greater closer to the shoreline of the bay than
• Phillips Field. However, the influence of the bay breeze would be limited
to travel distances of less than 6 km from the bay. Only a limited number
•
I
I
of sources and a limited geographical area would be affected. There is
no other source of meteorological data available that would characterize
the winds closer to the bay than Phillips Field at Aberdeen.
-19-
-------
I
I
However, based on the review of S02 data for the Baltimore
regions, the vast majority of high SO^ days occur in the winter, when |
the frequency of occurrence of a bay breeze should be even lower. Hourly »
meteorological data for 7 of the 10 model validation days were compared.
On these 7 days the Aberdeen data showed no clear evidence of a bay breeze •
occurring on these particular days.
I
I
I
I
I
I
I
I
I
I
I
I
-20- •
-------
Section 4.0
METHODOLOGY
-------
Section 4.0
METHODOLOGY
I
I
.
4.1 AIR QUALITY MODEL
* The methodology for relating S02 pollutant emissions to ambient
• air quality requires that both point and area sources be included in an
atmospheric simulation model capable of predicting both short- and long-
ff term SOg concentrations at any given receptor location. The methodology
used must be suitable for assessing the impact on ambient levels of S02
| due to changes in fuel used by power plants within the Metropolitan
_ Baltimore AQCR. The Sampled Chronological Input Model (SCIM) was selected
" to analyze and predict the transport and dispersion of pollutants in an
• urban and surrounding suburban region. SCIM in a steady-state Gaussian
plume model. The generality of the Gaussian plume equation and its appli-
• cation to urban diffusion analysis has been described by Calder (1970).
This basic equation has been shown to be valid over down-wind travel dis-
• tances of up to a few hundred kilometers (e.g., see Slade 1968 and Pasquill
• 1962). Since 1 to 50 kilometers is the distance scale of primary interest
in studying urban air pollution problems and since a great deal of experi-
• mental work has been devoted to defining parameters for the Gaussian plume
model, this concept has been the principal one used to analyze urban air
| pollution problems. A brief description of the Gaussian plume equation
_ and the modifications applied to it as it is used in SCIM are given in
^ Appendix A.
I
I
I
-------
I
I
4.2 MODEL MODIFICATIONS
Four modifications were introduced into the SCIM model for •
this study to represent: (1) an updated method of calculating plume
rise, (2) downwash, and (3) diurnal variations of industrial sources Q
and monthly variations of power plants. The need for a special modi- _
fication to represent the bay breeze and topography effects is dis- ™
cussed but no change in the model was introduced. •
4.2.1 Plume Rise
I
The SCIM program was modified to include an update of the 1969
Briggs equations. The revised equations (Briggs 1971) are listed in •
Appendix A. In general, the new equations result in higher plume heights.
Some typical comparisons for a light wind speed (2.6 m/sec) and neutral I
stability are shown in Table 6.
4.2.2 Downwash Effects
Rules for estimating the effects of downwash were taken from |
a paper by Briggs (1973). They are based on the widely accepted hypothe- _
ses that airflow is not affected by buildings above two and one-half "
times the building height and that below one and one-half times the build- •
ing height there is an aerodynamic cavity. The rules are also influenced
by wind tunnel experiments. The only additional information required to I
apply these rules are the heights and widths of buildings near the source
which are likely to create downwash influences. The rules, summarized |
below, have been incorporated in the SCIM model.
I
I
-22- •
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE 6. SCIM PLUME HEIGHTS USING ORIGINAL AND
REVISED ESTIMATES
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
Stack
Height
(m)
19
37
11
61
61
61
37
15
14
22
22
21
152
Stack
Diameter
(m)
2.0
1.6
1.5
2.7
2.7
2.7
1.4
1.8
1.2
1.0
1.1
0.9
12.8
Stack
Speed
(m/sec)
6
12
18
17
13
22
12
13
21
16
16
16
19
Stack
Temp.
(°k)
366
429
394
422
422
533
483
808
772
533
533
533
308
Plume
Original
(m)
48
86
52
187
170
246
83
92
75
58
61
53
513
Height
Revised
(m)
64
100
98
210
190
282
95
175
142
78
83
70
450
These results are for neutral stability, wind speed (at 6.4 meters) of
2. 6 m/sec and an ambient temperature of 65° F.
For each point source a building height and building width are
defined which may create downwash effects. If these are zero or blank,
the program will assume that no downwash effects are applicable. The
following data are used for each source to estimate downwash:
BH » building height, m
BW = building width, m
hs = stack height, m
d$ = stack diameter, m
v$ = stack gas exit speed, m/sec
-23-
-------
Using these parameters and the wind speed (u) at height hs» calculate
the following:
£B = min (BH. Bw)
hE - hs +2(^5.- 1.5)ds
The following rules are used to modify the ground concentration or the
effective stack height to account for downwash effects:
1.
2.
If hE >_ BH + 1.5 «,g or hs >_ BH + 0.25 £3.
no modification is required.
If the distance from source to receptor is less
than 3.5 a%, xk = M
u
where Q - source emission rate, ug/sec
0, hE > BH + 0.5 £B
K =
— , he > 0.35
BW2 E-
0.5 AB)
1.5
, hE ^0.35 (BH + 0.5
3. If
>. Bw, compute plume height (hp) as follows,
2 HE - BH - 1.5 BW, hE > BH
hE - 1.5 BW , hE <_ BH
use
hF = 0, if hp < 0.5 B
W
4.
If BH < BW, compute plume height as follows:
2 hE - 2.5 BH, hE > 1.5 BH
0 , hE <_ 1.5 BH
hF =
-24-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
•
4.2.3 Monthly and Diurnal Variations in SO? Emissions
Monthly fuel consumption for each of the five major Baltimore
Gas and Electric plants was obtained from the State of Maryland Bureau of
I Air Quality and Noise Control for the years 1972 through 1975. A modifi-
cation has been made to the SCIM model to allow the emissions for the
I power plants to vary by month. The S02 emission rate is computed using
g the standard EPA emission factors (EPA 1974) and monthly estimates of the
™ sulfur content and the amounts of fuel burned.
• Another modification has been incorporated into SCIM to allow
industrial emissions to vary according to the individual plant's working
I schedule. Data concerning the number of shifts worked per day and the
number of days worked per week were obtained from both the NEDS data file
• and the Maryland State Emissions Inventory. The State data tape also
• provided information concerning the amount of the total emissions from
industrial plants that could be attributed to processing. The remainder
V of the emissions for each plant could then be attributed to space heating
and adjusted according to ambient air temperature and time of day. The
| annual emission rate was adjusted as follows:
| Q - (QA - QP)FS + QF
• where:
* Q * hourly emission rate
• Q/\ = annual emission rate
Qp = annual emission rate excluding emissions related to
M space heating
* FS = hourly emission factor for space and water heating
I
-25-
I
-------
NS = number of shifts worked per day
NQ = number of days worked per week
p 1 if plant is in operation
0 if plant is not in operation
Whether or not a plant is in operation each hour is determined
by the following tables (where X indicates an operating time):
Beginning of Hour
00-07
08-15
16-23
Day of the Week
Monday
Tuesday
Wednesday
Thursday
Fri day
Saturday
Sunday
Number of Shifts Per Day
1 2
X X
X
Number of Days
5 6
X X
X X
X X
X X
X X
X
The emission factor for space heating is
FS = 1 + f (AT D S - 1)
1
where:
3
X
X
Worked Per Week
7
X
X
X
X
X
X
X
computed as follows:
f = fraction of space and water heating emissions associated
with space heating
Th - Ta if Ta < Th
0 if Ta > Tn
-26-
1
1
I
1
1
1
1
1
1
1
1
I
1
1
1
1
-------
1
1
1
^p
1
1
1
1
T, = hourly ambient air temperature
a
T. = hourly space hearing demand threshold temperature
D = hourly space heating demand factor
S = sensitivity factor
, 8760
s • 8760- £ ! °1
-------
TABLE 7. HOURLY SPACE HEATING DEMAND THRESHOLD
TEMPERATURES AND DEMAND FACTORS
Hour of the Day
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Demand Threshold
Temperature
( F)
56
55
55
55
56
58
59
61
63
64
65
65
65
65
65
65
65
65
65
65
65
64
62
60
Demand
Factor
0.3864
0.3288
0.3288
0.3432
0.4344
0.8400
1.4160
1.6848
1.4232
1. 1448
1.0464
0. 9864
0.9864
0.9984
0. 9792
1.0392
1.1520
1.2432
1.3128
1.3392
1.3056
1.2000
0.8160
0.4872
-28-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
• and the higher frequency of light winds and weak pressure gradients. The
• Aberdeen data was available for 44 days during the two-month period for
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
each of the two years. Only one case of a well-developed bay breeze was
found which was not reflected in the Baltimore Airport data.
4.3 MODEL VALIDATION
4.3.1 Choice of Ten Validation Days
After the air quality data had been reviewed and unreasonably
high values along with data from stations with interference problems were
rejected, the remaining data was judged to be reliable. Ten days were
chosen to validate the SCIM model for the Baltimore area, based on S02
monitoring data for 1973 and 1974 from the SAROAD data file for Maryland.
The procedure used to choose the 10 days was as follows:
1. The 10 highest S02 days at each monitoring station
over the two-year period were identified.
2. From step (1) the 20 days with the greatest number
of high station readings were determined and the
meteorological conditions were identified for each
of these 20 days.
3. General meteorological characteristics were identified
for each of these days. The following four categories
were found to be applicable:
(a) Days on which precipitation occurred
(b) Clear mornings followed by high overcast by
afternoon
(c) Clear mornings followed by low overcast
(d) Overcast all day - no precipitation.
4. Six days with precipitation were eliminated from
consideration.
-29-
-------
I
5. Eight of the remaining days were chosen as validation
days, attempting to represent as many wind directions —
as possible. An effort was also made to choose days I
having data available for the greatest number of •
stations. All eight days fell between November and
February. •
6. Two additional days with moderate $03 values were also
chosen using the criterion of having data for as many .
stations as possible. •
Hourly 862 data were also examined for high reading days and •
one day was selected to replace a day previously chosen that had data
for only a limited number of stations. Three additional days were chosen I
from those remaining in the original 20 to be preliminary test days to use
to test proposed changes in the SCIM program. This allows for model devel- I
opment to be independent of the validation data. The 10 validation dates •
are listed in Table 8 along with the general meteorological characteristics
for each. I
4.3.2 Validation Results •
A summary of 103 pairs of measured and calculated 24-hour
concentrations of S0£ is given in Table 9. The locations of the monitoring •
stations used in these comparisons are shown in Figure 1. Figure 2 is a
scatter diagram illustrating the correlation between the measured and |
calculated values. The center diagonal line represents a one-to-one —
relationship. The two parallel control lines represent a deviation from •
this line by +_ 53 yg/m3 (0.02 ppm). The correlation coefficient is 0.38. •
These results were for available comparisons at 25 monitoring stations on
10 selected days. Maps showing the comparisons for each day are presented I
in Appendix B. The ratio of mean measured to mean calculated concentration
-30-
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
S2
><
u
z
z
2
fAUDA
V
S SELECTED FOR
Q
O
1*-)
O
00
O
c
00
METEOROLOGICAL CHARACTERI
00
a
«
£
•o a
-a
si K
5 O § S
e! ^ ^<
•* a § a_
N ff U
•3 CM
J 00
2
V]
S
_o
*a
u
u
(4
S
T3
a
•l-l
1
3
•a
s
U
a
Q
(\J (jp
00 «
0 0
i 1
1 °s
O fsj
Overcast all day
Clear AM, high overcast PM
ro JJ>
& &
r*t **"•
*»
tx M
(M "-1
•8 i
tt, ^
2
*^
o
4!
a
0
S
CM
1
0
S
CM
in
CM
a
O
ro
CM
1
0
R
*"* o
< 7
« 0
3 2
*C Bill
> 2
Clear AM, low overcast PM
m
<*"
o
Q
^f
.H
O
O
0^
O
fj
Clear AM, low overcast PM
ro
CM
J
Q
CM
in
T-i
2
^oO
rt
0 0
O O
8 CM
r*
0 0
O O
in oo
Clear AM, low overcast PM
i
CM
O
Q
^
T-4
O
i
^ 2
o *"•
8 ^
a" 2
O " 0.
S 2«
o1 < S
Clear AM, low overcast PM
o\
^H
o"
i
b
C\
CM
g
•S
2"
00
s °0
jj S
3 w
'go
> S
2
2
•^
w_
M
•o
1
o
T)
i
2"|
•* g
2 -S
!i
u «
Tf
2
00
CM
^
U.
fs. 1C
in m
o
CM
1
0
o §
S J
2
o1 a-
0 -i
i-t
12 ^
0 rf 0
O „ 0
m 2 '•'
O •" ro
C
_2
4j O
S3 5
« JS
S .M
o •*
| 2
§ «
•S S
< 1
•^ u
3 ^
h O
6 3
S S
& **
~ ro
S N
>• 00
J3 3
2 <
to midnight)
<
M-t
rt
•3
^M
(M
O
O
ro
1
O
8
ro
o-f
O ^
CM **>
i |
2
2
**i
c-
Scattered clouds to 6 AM, cloudy i
3 PM), then scattered clouds
gj
r,*
>
o
-31-
-------
ff)
"SS
a.
VJ
X-H
o
p
1
U
§
U
CM
O
oo
Pi
o
u.
oo
2
i
2
rf
a.
f^
U
O
p
^J
§
>:
g
o
£
•*
O
cS
5
s
.^
00
TABLE 9.
Ps.
ro
CM
v^
oo
PS.
PS.
^
*"*
PS.
•s^
11
in
^4
P-.
00
CM
CM
Ps.
O
CM
ro
PS.
CM
CM
ro
ps.
CM
i^
«^
CM
11
ro
Ps.
I
CM
11
ro
PS.
rM
s^
JIJ
^
ro
PS.
CM
CM
Location
u ^
C^* fc»
fll
U J3
h e*
in
• — .
»-H
ro ffv
^4 00 00 Tj* O
t-l TH CM CM —I ^ »H
"S. ' -«. ^^ -s,. T* ~^,
ro 10 T(" CM ON oo oo
ro co ro ro CM CM ,. -s^
CM -t ro oo
O tv *-> ro
CM CM rv Tf
00 CM CJ •$>
in oo 10 to
^-i
"5
§
o
Q
| ,5 jSg^
^swSs^"^
-
O in -SO ij^ * ff S
ilCMcOr|.
~J^
•0
JS
a
o
2
a
ta
unibers indicate d
a
oo
a
•3
rt
01
X
*
T3
0)
J3
CO
a
a
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
/
C~ CARROLL / BALTIMORE
\ COUNTY / COUNTY
\
V HARFORD
\ COUNTY
V 17
FREDERICK
COUNTY
34
BALTIMORE CITY
22
HOWARD
COUNTY
11
31
24
16 32
MONTGOMERY
COUNTY
13
21
36
PRINCE
GEORGE
COUNTY
25
N
ANNE
RU^DEL
COUNTY
0 5 10
MILES
Figure 1. Identification numbers for monitoring stations.
-33-
28
£
-------
3-
-J-
Tf
N
T3
a
M
jj
W
O
I
I
I
I
I
I
I
I
I
e>
f
a
o
I
V
OS
I
I
I
I
I
I
I
I
I
-------
I
I
at each station is presented in Figure 3. In this figure it may be seen
• that in Baltimore City the ratio varies from 0.5 at the Linwood and O'Donnell
station to 2.5 at the Patapsco sewage disposal plant. Higher ratios occur
• well to the south of the city and to the northwest of the city. In general
the comparisons are best in the center of the metropolitan area. This is
I where the highest and most critical concentrations occur, for both measured
• and calculated values.
Hourly concentrations were measured at 8 of the 25 stations used
I in the validation analysis. The measurements and calculations at these
stations were averaged over 3-hour periods and compared. The comparison
| is shown in Figure 4. The correlation coefficient is 0.38, as it was for
24-hour averages.
• The scatter diagrams show that the model is reasonably accurate
• in that there is as much overprediction as underprediction. If anything
there is a tendency for the model to underestimate the lower measured
I concentrations. However, the precision of the model estimates is low in
that there are large discrepancies on any given day or at any given location.
| An examination of some specific differences will help illuminate the problem.
_ The locations, dates, and prevailing winds for 24-hour model
^1 ^
• calculations which vary from measured values by 53 pg/m (0.02 ppm) or
• more are listed in Table 10. It is useful to consider some of these
individual cases to try to discern possible causes for the errors. The
• first three cases are discussed below.
Patapsco, November 7
I The measured 24-hour concentration was 284 pg/m^ and the mean
24-hour calculated concentration was 43 ug/m^. Hourly measurements are
I
I
-35-
-------
FREDERICK
COUNTY
PRINCE
GEORGE
COUNTY
MONTGOMERY
COUNTY
MILES
Figure 3. Ratio of mean measured to mean calculated (24-hour values)
-36-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
-------
TABLE 10. LOCATION, DATE, AND WIND DIRECTIONS FOR 24-HOUR MODEL ESTIMATES OF SO2
WHICH VARY FROM MEASUREMENTS BY MORE THAN 53 /xg/m3 (0.02 ppm)
Location
Patapsco
Sun & Chesapeake
Robinson & Toone
Read St.
Unwood GO'Donnell
Robinson & Toone
Dundalk
Essex
Hopkins
O den ton
Dundalk
Read St.
Calvert
Garrison
Odenton
Essex
Date
Nov7
May 11
Dec 4
Feb 10
Feb 10
Feb 28
Feb 10
Nov 12
Feb 10
Feb 10
Feb 28
Dec 4
Dec 4
May 11
Feb 27
Dec 24
Model
Estimate
low
low
low
high
high
low
high
low
high
low
high
low
low
low
low
low
Prevalent Wind
Direction Octant
NW
NE then SE
SW variable, then SE
W thenS
W thenS
SW variable
W thenS
W variable
W thenS
W thenS
SW variable
SW variable, then SE
SW variable, then SE
NE then SE
NE
NE
-38-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
not available. However, the hourly calculations were highest for the
three hours when the reported wind direction was 300°. The reported wind
directions were in the range 280° to 320° for all but 3 hours on this day.
• However, 300° was only reported for 3 of 21 hours. Two large sources of
S02, the M and T Chemical Company and the Chevron Asphalt Refinery are
| about 1 km away at a bearing of about 300°. Clearly, an accurate wind
_ direction has a large bearing on the concentrations calculated for this
™ site when the wind direction is near an azimuth of 300°.
I Sun and Chesapeake, May 11
On this day the calculated 24-hour concentration was 35 ug/m3
I compared to a measured concentration of 157 pg/m3. Hourly calculated and
measured concentrations and wind directions are listed in Table 11 for this
I day. It is clear from these comparisons that the two sets of values are
• only in agreement on this day when the wind is from 070° or from 140°.
The first bearing is toward the M and T Chemical Company and the second is
I toward the Wagner Power Plant. It is difficult to decide what the cause
of the high concentrations is during the period of 10 AM to 5 PM. It was
I noted that the 24-hour concentration was 100 ug/m3 less at the Patapsco
station which is 2 km east of this site. Since the winds varied from
• northeasterly to southeasterly during the course of the day; it seems
• likely that a source between these two sites was the cause. There are
I
I
I
about six plants between these two sites any of which could have had high
emissions on that day which would account for the high concentrations.
Within a bearing of 110° to 160°, there are two primary sources including
the FMC Corporation plant and the Amoco Oil Refinery. Clearly, accurate
-39-
-------
TABLE 11. MEASURED AND CALCULATED SC>2 CONCENTRATIONS
AT THE SUN AND CHESAPEAKE STREETS MONITORING STATION
ON MAY 11, 1974
Hour of
Day
1
2
3
4
5
6
7
8
9
10
11
12
13
14
IS
16
17
18
19
20
21
22
23
24
Measured
(/Ig/m3)
80
0
53
53
53
80
80
80
80
133
453
559
240
399
293
106
213
266
80
80
53
53
80
53
Calculated
(MS/™3)
13
6
12
7
14
16
16
20
124
81
17
70
18
26
28
26
72
22
52
68
58
29
18
IS
Wind
Direction
(degrees)
50
40
50
40
40
20
50
40
70
70
40
110
160
120
130
120
110
130
140
140
140
160
160
150
-40-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
daily and even hourly emissions from such sources are necessary for
modeling validation studies.
Robinson and Toone, December 4
The hour to hour measured and calculated S02 concentrations at
the Robinson and Toone Streets Monitoring Station and the wind directions
for December 4, 1973 are presented in Table 12. There are two periods of
major discrepancies on this day. One occurs during hours 1 through 17.
The other occurs during hours 18 through 24. During the first period the
wind direction is primarily from the southwest quadrant. During the
second period the wind is primarily from the southeast quadrant. While
high hourly concentrations are calculated they are not quite as high and
not quite as frequent as the measured values. There are likely to be
other days on which the reverse is true.
Discussion of Validation Errors
Two possible causes of the validation errors found in this study
which result in lack of precision in the calculated concentrations are the
following:
• Imprecise meteorology, especially wind direction - A
mean wind direction which is representative of a city-
wide average for an hour is not available for the
Baltimore area.
t Impresice emissions - A detailed hour by hour accounting
of S02 emissions is not available. The crucial sources
are industrial plants. It would be necessary to study
each source in detail to get meaningful hourly emissions
and the distribution of emissions from each stack.
Although the validation analysis did not show a high degree of
one-to-one correspondence between calculated and measured values, the
-41-
-------
TABLE 12. MEASURED AND CALCULATED SO2 CONCENTRATIONS
AT THE ROBINSON AND TOONE STREETS MONITORING STATION
ON DECEMBER 4, 1973
Hour of
Day
1
2*
3
c
6
7
8*
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
Measured
(/Lig/m3)
0
106
106
133
106
160
320
666
320
213
0
213
0
266
373
133
240
426
213
506
532
Calculated
(Mg/m3)
80
325
240
97
113
80
54
115
29
91
13
22
161
60
413
36
36
201
428
76
43
Wind
Direction
( Degrees)
260
190
190
280
210
290
220
180
200
180
230
200
180
170
130
140
140
140
130
120
150
* Calm winds were reported for this hour and no com-
parison was made.
-42-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
comparisons do show that the model produces useful results from a plan-
I ning point of view. The distribution of concentrations at any given
• location and time cannot be accurately predicted with the level of detail
about meteorological conditions and source emissions which is available
I within the resources of this study. The calculated values depend on the
specific combinations of meteorology and emissions which are assumed at
| the time. However, the distribution of meteorological conditions which
_ are likely to occur over the course of a year can be systematically
™ examined and evaluated by a model. In a similar manner, the model can
• be used to evaluate air quality impacts at a much larger number of loca-
tions than can be economically analyzed by monitors. A monitor can only
• look at a single location at a single time. As long as both the monitors
and the model see similar distributions of concentrations over a period
I of time, the results of the model can be assumed to be valid. The results
• presented in the scatter diagrams attest to the adequacy of the agreement.
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
Section 5.0
I MODEL APPLICATION AND ANALYSIS OF RESULTS
I
I
I
I
I
I
I
I
I
-------
I
I
Section 5.0
• MODEL APPLICATION AND ANALYSIS OF RESULTS
I
I
5.1 DETERMINATION OF WORST CASE CONDITIONS
•
A method to determine the conditions associated with "worst
• case" air quality in the vicinity of the three power plants using model
calculations was developed. Two approaches were taken; the first assumed
• that worst case conditions can arise from the overlap of plumes from two
• or more major S02 sources. In this situation the location and conditions
of the highest S0£ concentration resulting from the overlap from point
I sources would identify the worst case conditions. The second approach
considered the overlap of plumes from area sources with each power plant
Jj plume. Calculations of concentration versus distance for various combin-
ations of wind speed, stability and mixing height for each of the wind
• directions causing significant overlap were used to identify locations
• and conditions of maximum impact.
_ 5.1.1 Plume Overlap of Major Point Sources
• The largest point sources of S02 in the Baltimore area were
• selected from the emission inventory. These included all five major
power plants operated by Baltimore Gas and Electric in the immediate
• Baltimore area. The 1-hour SOg concentrations expected from the nine
plants with greatest impact are listed in Table 13. Combinations of
• point sources lying along straight lines that would result in overlapped
• plumes were then identified for these sources. The EPA Multiple Source
Model (PTMPT in the UNAMAP System) was used to calculate SO,, concentrations
-45-
-------
resulting from each of 11 plant combinations that were chosen. Calcu-
lations were made separately for winds blowing in each direction along
each line for a total of 22 sets of computations. For each line 1-hour
ground-level concentrations were calculated for a series of receptors
located between the sources and downwind of the last source. The com-
binations that produced the highest S02 concentrations were identified
along with the corresponding wind directions, wind speeds and stability
classes. This information is presented in Table 14 for the five most
critical combinations.
TABLE 13. 1-HOUR SC>2 CONCENTRATIONS (SINGLE SOURCE)
Worst Case
Plant
W agner
Riverside
Crane
Arundel Corporation
Westport
Gould
Olin Matheson
Beth. Steel -
Penvrood
Beth. Steel - Coking
centration
/m3)
a oo
5 I
M
O
851
347
167
13,987
144
204
3,751
111
880
^,
s
«O
rt
4-1
C/J
A
A
A
C
A
A
A
A
A
•o
u
Qd ^
T3 ""J
e S
.-* )H
3.0
3.0
3.0
2.0
3.0
3.0
1.0
3.0
1.5
Downwind
Lin
-------
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
z
Q
Q
8
0-
s
w
o
w
I
a.
w
Q
§
i—
00
1
5
§
U
!
o
oo
5
S
1-1
X
O
X
o
§
D
Q
O
a.
Z
0
p
i
S3
8
w
u
8
H
g
2
a
PQ
rs
CA
31
••d
2
a
w
V)
TS
3)
J
•3
e
8
***
trt
S
00
I
_o
a
s
•j
o
s
o
u
(M
O
V)
0
!i!
1
s
a
S
x
M
2
'-}
Q
e
«M
3
1
§
5 s
—
^^
•3 "3 a
S j) j«
W ^* s
1/5 ~
t
c
'j 2
0— ••$
^
u |
cT "
V5
il
'J ^K
— g
S i
Q
S a -2.
> a s
•^
1
a
J |
••
CM *•*
O
VI
u
I!
.a -^
Q
_ "o"-
•2 « a
3 a ~~.
S wj* S
^^
,
c
u
S 3
"1
CM "
0
t/1
~3 -2
q u
5
i 1 i
i
Si«
Q JS i
U.
2
s
Z
•j
b*
O
oo
o
o
CM
^
ro
^
in"
•*
c\
0
m
t*.
^O
CO
O
rrt
00
in
^a"
o
§
CM
00
*» •
O eft
* or
O oo
S
J
CM
CM
*H
^
O
'-O
oo
M>
o
CM
rt
in
oo
•*"
PO
^;
o
I
'-O*
m.
0
CM
a
(^
O
3
oo
oT srl
° oo"
0 -•
3
s"
*>*
OJ
^"
1-*
"
«
5?
O
CM
CM
*H
^*
-
'-O
0
o
CM
00
^4
-------
In order to assess which plant combinations could contribute
the most to a worst case situation over a 24-hour period, wind persistence
frequencies were determined for each of the five directions based on hourly
surface weather observations from Baltimore-Washington International Air-
port for 1971-1975. Of the five directions of interest the most persistent
wind (defined in terms of maximum frequency of occurrence of the greatest
number of hours per day having a particular wind direction) was 280°, which
is along a line from Olin Chemical Corporation and through two major Bethle-
hem Steel point sources on Sparrows Point. The tabulated results are pre-
sented in Table 15.
TABIE 15. NUMBER OF OCCURRENCES OF GREATER THAN OR EQUAL TO THE SPECIFIED NUMBER
OF HOURS HAVING THE FOLD3WING WIND DIRECTIONS DURING A 24-HOUR PERIOD IN
JANUARY, FEBRUARY, NOVEMBER AND DECEMBER (1971-1975)
Number of Hours
14
13
12
11
10
9
8
7
6
5
4
3
2
1
280°
1
1
1
4
6
12
18
22
32
55
89
136
222
361
100°
_
-
-
-
-
-
-
1
2
2
8
25
56
137
080°
_
-
-
-
-
-
2
5
9
14
30
46
80
146
325°
_
-
-
-
-
-
-
4
8
12
21
46
87
189
145"
_
-
-
-
-
-
-
1
2
4
7
19
61
143
-48-
I
I
1
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
-------
I
I
I
I
I
5.1.2 Meteorological Conditions Selected for Overlapping Area and
Point Sources With Each Power Plant
As a result of reviewing concentrations calculated in the vali-
dation analysis and the overlaps of point sources, two types of situations
'•
were identified which produce critical days for $03 concentrations, (i.e.,
• critical in the sense of being most favorable to high S02 concentrations).
One type of situation consists of a cold day with low wind speeds and low
j§ mixing heights. The critical locations are downwind of the largest area
_ source emission rates. The area source emission rates in the Baltimore
1
™ area greater than 1.5 yg/m^/sec are shown in Figure 5.
• The second type of critical situation is downwind of one of the
two major industrial sites around Baltimore, i.e., Sparrows Point and
• Curtis Bay.
With these two situations in mind plus the fact that we are
• interested in locations which are affected by one or more of three power
§ pi ants, Crane, Riverside and Wagner, we have defined a worst case day for
1
each plant. Each day was defined in the following manner. A set of 24
A hourly stability classes and air temperatures was selected from a day with
relatively high calculated S02 concentrations. February 10, 1973 was used
| as a model. This was one of the 10 days selected for the validation anal-
_ ysis. The wind speeds observed on this day were changed slightly to con-
•
• form to a more normal pattern, i.e., higher wind speeds during the day and
flower wind speeds during the evening. A very similar set of wind speeds,
stabilities and temperatures was used for all three worst case days, i.e.,
• one corresponding to each of the three power plants. A uniform mixing
-49-
-------
Figure 5. Area sources with SC>2 emissions greater than 1.5^g/m2/sec (all are located in Baltimore City)
-50-
I
I
I
I
I
I
I
I
I
s
1
I
I
I
I
1
I
I
-------
I
' height was defined for each day. It was set at a value just above the
• maximum effective plume rise expected from each plant.
Wind direction and persistence were selected specifically with
f each power plant in mind. A prevailing wind direction was defined such
that the plume from the power plant would overlap locations which would
• be in one of the two critical situations. For the Wagner plant it was
ft determined that the most critical situation occurs when the plume from
that plant overlaps the plume from the 01 in Matheson Chemical plant.
• This would require a wind coming from a bearing of 130°. The wind was
allowed to vary through the 24 hours such that it would have a persistence
|| of about 10 hours in the direction of the prevailing wind. In order to
^ make the wind directions more realistic, a uniform random number over the
™ range of -5 to 5 was drawn and added to each selected wind direction. The
• meteor!ogical parameters selected for the critical day for the Wagner plant
are listed in Table 16.
• For the Riverside plant it was determined that the most critical
situation exists when the plumes from the Sparrows Point area overlap the
plumes from the power plant and both extend over the downtown Baltimore
4| area. This occurs with a wind direction of 140°. The meteorological con-
ditions for a critical day involving the Riverside plume are presented in
J Table 17.
For the Crane plant the most critical condition was found to
| exist when the power plant plume extends over the central Baltimore area.
I The highest concentrations occur along a bearing of 260° from the plant
which corresponds to a wind direction from 80°. Accordingly, a pattern
t
I
-51-
-------
TABLE 16. CRITICAL DAY - WAGNER
Hour
00
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Mixing Height
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
700m
Wind Direction
85
97
113
95
108
123
133
131
130
137
107
124
134
128
131
131
131
131
131
131
130
142
129
116
Wind Speed
2 m/sec
2 m/sec
2.5 m/sec
3 m/sec
3 m/sec
3 m/sec
2.5 m/sec
2 m/sec
3 m/sec
3.5 m/sec
4.0 m/sec
4.5 m/sec
4.5 m/sec
5.5 m/sec
4.5 m/sec
2.5 m/sec
2.5 m/sec
2.5 m/sec
2.5 m/sec
2.5 m/sec
2.5 m/sec
2 m/sec
2.5 m/sec
2.5 m/sec
Stability
E
E
E
E
E
E
E
D
D
C
C
C
B
D
C
D
D
D
D
D
D
E
D
D
Temperature
18°F
17°F
15°F
14°F
14°F
13°F
12°F
12°F
9°F
14°F
20°F
27°F
30°F
31°F
32°F
29 °F
29 °F
29°F
29 °F
29°F
29 °F
27°F
29 °F
30°F
-52-
I
I
I
I
I
I
I
I
I
I
I
I
I
f
I
I
I
I
-------
I
I
I
I
f
i
8
I
t
I
t
f
1
I
t
I
I
I
TABLE 17. CRITICAL DAY - RIVERSIDE
Hour
00
01
02
03
04
05
06
07
08
09
10
11
12
13
14
IS
16
17
18
19
20
21
22
23
Mixing Height
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
400m
Wind Direction
195
187
183
165
158
153
143
141
140
127
157
154
144
138
141
130
162
162
139
136
143
134
137
146
Wind Speed
4 m/sec
2 m/sec
2.5 m/sec
3 m/sec
3 m/sec
3 m/sec
2.5 m/sec
2 m/sec
3 m/sec
3. 5 m/sec
4 m/sec
4.5 m/sec
4.5 m/sec
5.5 m/sec
4.5 m/sec
3 m/sec
4 m/sec
4 m/sec
4 m/sec
4 m/sec
2.5 m/sec
2 m/sec
2.5 m/sec
2.5 m/sec
Stability
E
E
E
E
E
E
E
D
D
C
C
C
B
D
C
C
D
E
E
D
D
E
D
D
Temperature
18°F
17°F
15°F
14°F
14°F
13°F
12° F
12°F
9°F
14° F
20° F
27° F
30° F
31°F
32°F
32°F
32°F
29°F
29° F
300 F
29° F
27° F
29° F
30°F
-53-
-------
with 10 hours of wind persistence having an average direction from 80
was used for the Crane critical day. The meteorological conditions for
the Crane worst case day are presented in Table 18.
TABIE 18. CRITICAL DAY - CRANE
Hour
00
01
02
03
04
05
06
07
08
09
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Mixing Height
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
800m
Wind Direction
125
117
113
95
88
73
83
81
80
57
87
74
84
78
81
60
92
69
76
73
34
87
76
60
Wind Speed
2 m/sec
2 m/sec
2.5 m/sec
3 m/sec
3 m/sec
3 m/sec
2.5 m/sec
2 m/sec
3 m/sec
3.5 m/sec
4 m/sec
4.5 m/sec
4.5 m/sec
5.5 m/sec
4.5 m/sec
3 m/sec
4 m/sec
3 m/sec
3 m/sec
3 m/sec
2.5 m/sec
2 m/sec
2.5 m/sec
2.5 m/sec
Stability
£
E
E
E
E
E
E
D
D
C
C
C
B
D
C
C
D
D
E
D
D
E
D
D
Temperature
18 °F
17°F
15 °F
14 °F
14 °F
13°F
12 °F
12 °F
9°F
14 °F
20 °F
27 °F
30°F
31°F
32 °F
32 °F
32°F
31 °F
29 °F
30°F
29°F
27°F
29°F
30 °F
-54-
I
I
I
1
t
i
i
i
t
i
t
i
i
t
t
i
i
i
-------
I
t
5.1.3 Air Quality on Worst Case Days
ft Using the emission inventory for present conditions and the
worst case meteorology for each day and a set of receptor locations, the
• model was run for each day. The receptor locations were initially selected
to represent maximum impact areas within which the power plant would have
m a contributing influence. The receptor locations were systematically varied
A to identify the maximum concentrations. Variations in the emission rates
^ of area source emissions and large point sources were used to guide the
I receptor selection process.
ft
ft
•
V
I
I
Based on repeated runs, the values listed in Table 19 were
obtained. These results show that neither the 3-hour National Ambient Air
Quality Standards (NAAQS) (1300 yg/m3) or the 24-hour NAAQS (365 ug/m3) is
exceeded by areas affected by the Crane and Riverside plants. In both of
these cases the contribution of the plant to the 24-hour maximum concentra-
tion is small. For both plants the contributions are not out of line with
the maximum concentrations which might be expected. The Riverside contri-
bution is about 25 percent of the maximum 1-hour S02 concentration from
the plant under neutral stability conditions (i.e., 109 ug/m3, see Table 13).
However, the Crane contribution is smaller than probably occurs elsewhere.
It is only 12 percent of the maximum 1-hour S02 concentration from the
plant under neutral stability conditions. At other locations, closer to
the plant, where higher contributions from the Crane plant are likely, con-
tributions from other sources will be much smaller. Very large changes in
emissions from the Crane plant will make very small increases in $02 concen-
trations in the Baltimore area.
-55-
-------
TABLE 19. MAXIMUM CONCENTRATIONS UNDER PRESENT OPERATING CONDITIONS
24-Hour Maximum
Plant
Wagner
Riverside
Crane
Concentration (/ig/m )
Power Plant
Total Contributions
370 105
320 26
ISO 4
Location
Bearing
(deg)
310
320
260
from Plant
Distance
(km)
7.3
5.5
22.5
3-Hour Maximum
Concen-
tration
(jug/mS) Location
Bearing
Total (deg)
1360 310
750 320
580 260
from Plant
Distance
(km)
7.3
5.5
22.5
With respect to the Riverside plant, increases in emissions
from this plant must be considered much more critically. If the S02
emissions were doubled, the maximum 24-hour S02 concentration at what
was found to be the most critical location will increase to 346 yg/m3,
still below the NAAQS of 365 pg/m3, but getting close to it.
Our analysis of emissions from the Wagner plant shows that
under critical conditions which can be reasonable expected to occur,
the 24-hour and 3-hour NAAQS for S02 is exceeded at locations in the
Curtis Bay area. Furthermore, the Wagner plant makes very significant
contributions to the 24-hour S02 concentrations under these conditions.
This result clearly indicates the need for very careful study of the air
quality in this region, perhaps by obtaining more detailed emissions and
meteorological data than are routinely available. Further increases in SOp
emissions from the Wagner plant or other sources in the area are highly
undesirable.
A separate modeling analysis to determine the impact of the three
power plants under projected 1980 emissions was found to be unnecessary.
The changes in emissions, based on linear extrapolation between the 1973
-56-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
E
£ emission inventory and the 1980 inventory would be very small. These
™ changes would not impact on the findings based on the 1973 emission
II intentory.
The two inventories are listed in Appendix C. A careful analysis
• of these two inventories will confirm the validity of the above hypothesis.
• 5-2 DETERMINATION OF SULFUR LIMITS FOR COAL CONSUMPTION
Based on the results presented in the preceding discussion, it
• is evident that S02 emission increases from the Wagner plant are not accept-
able, and that increases from the Riverside and Crane plants may be pre-
" mitted with little danger of exceeding the NAAQS for S02- We shall proceed
f to develop recommendations for sulful limits for coal based on the follow-
ing assumptions. In converting to coal, the loss of thermal efficiency due
• to combustion of coal in place of oil is small and may be neglected. There-
fore, if the same S02 emission limits are maintained as a mass percent of
J fuel burned, the fuel conversion does not require any change in S02 emissions.
tlf the sulfur content of the fuel consumed by the power plants is increased
the S02 emissions will increase in direct proportion to the increase in the
IB sulfur content. Similarly, the contributions of each plant to air quality
at a point will increase in direct proportion to the increase in the sulfur
M content of fuel.
Using the air quality models results and the preceding assumptions,
• it is a very simple matter to determine sulfur limits for coal at each plant.
* The contributions to S02 concentrations in the modeling results were all
developed assuming each plant burns fuel with a sulfur content of 1 percent
• or something very close to 1 percent. By determining the maximum concentration
-57-
-------
1
I
contribution of each plant to any location in the Baltimore area, a con-
servative rule can be used to determine a maximum allowable sulfur content •
for the plant's fuel. From concentrations calculated for a wide range of
meteorological conditions, 20 twenty-four periods associated with the 0
highest SC>2 measurements during 1973 and 1974, we determined the maximum g
concentration contribution of each plant to each of 25 locations in the *
Baltimore area. The maximum concentration contribution of each plant was jf
as follows:
|
Maximum 24-hour $02 Concentration
Plant Contribution (yg/rPJ^
Wagner 110 9
Riverside 34
Crane 6
The highest legitimate 24-hour S02 concentration measured in the Baltimore
area during 1973 and 1974 was 284 yg/m3. It may be reasonably assumed that
I
I
concentrations as much as 15 percent higher may have occurred. The maximum f
o o 4W
ambient S02 concentration is estimated to be 326 yg/nr, which is 39 yg/m13
below the 24-hour NAAQS. Theoretically, plant emissions could be increased •
to just offset that 39 yg/m3. If this were done, the sulfur limits shown
in Table 20 would develop. However, other model calculations involving con- Jjf
tributions from the Wagner plant have shown that under some conditions the g,
calculated S02 concentration in the Curtis Bay will exceed 365 yg/m3 with- *
out any increases from the Wagner plant. It is recommended that no increases •
in fuel sulfur be incurred at the Wagner plant. The same type of findings
did not occur for the Riverside and Crane plants. It seems reasonable to I
-58-
I
I
-------
I
i
i
i
t
I
I
t
i
I
i
i
i
i
i
t
i
i
expect that a fuel sulfur content increase to 2 percent may be incurred
at the Riverside plant without exceeding the NAAQS. More detailed model-
ing of emissions in the vicinity of this plant may be required to justify
this proposal. Much larger increases in the sulfur content of fuel from
the Crane plant are permissible. However, we are reluctant to recommend
an increase of more than three-fold, primarily because the available coal
supplies make it unnecessary to consider the use of coal in excess of
3 percent.
TABLE 20. HYPOTHETICAL MAXIMUM ALLOWABLE COAL SULFUR
Plant
W agner
Riverside
Crane
Maximum 24-Houi Concentration
Contribution (/ig/m^)
110
34
6
Ratio of Allowable Increase
to Maximum Contribution
0.35
1.15
6.50
Maximum Allowable
Coal Sulfur (Percent)
1.35
2.15
7.50
-59-
-------
Section 6.0
REFERENCES
I
I
I
I
I
I
I
-------
I
I
I
1
1
I
i
I
i
l
i
i
i
i
I
I
i
i
REFERENCES
Briggs, G.A. 1971. "Some Recent Analyses of Plume Rise Observations,"
in Proceedings of the Second International Clean Air Congress.
Academic Press, New York.
. 1973. Diffusion Estimation for Small Emissions (draft). ATDL
Contribution File No. 79. NOAA, Oak Ridge, Tennessee.
Calder, K.I. 1970. "Some Miscellaneous Aspects of Current Urban Diffu-
sion Models," in Proceedings of Symposium for MultipleSource Urban
Diffusion Models.Environmental Protection Agency, Research Triangle
Park, North Carolina.
Davidson, B. 1967. "A Summary of the New York Urban Air Pollution
Dynamics Research Program." Journal of Air Pollution Control Associ-
ation, 17, pp. 154-158.
DeMarrais, G.A. 1959. "Wind Speed Profiles at Brookhaven National
Laboratory." Journal of Meteorology 16(4):181-190.
Engineering-Science, Inc. and Howard, Needles, Tammen, and Bergendoff.
1974. Development of a Trial Air Quality Maintenance Plan Using the
Baltimore Air Quality Control Region.EPA-450/3-74-050.
Environmental Protection Agency. 1973. Compilation of Air Pollutant
Emission Factors (Second Edition). EPA Document AP-42.Research
Triangle Park, North Carolina.
. 1974. 1972 National Emissions Report. EPA-450/2-74-012. Research
Triangle Park, North Carolina.
Koch, R.C. and G.E. Fisher. 1973. Evaluation of the Multiple-Source
Gaussian Plume Diffusion Model. EF-186, GEOMET, Incorporated, Gaithersburg,
Mary!and.
Maryland Bureau of Air Quality and Noise Control. 1974. Technical Support
Document for Sulfur Content Regulation Change. Baltimore, Mary!and.
McElroy, J.L. and F. Pooler, Jr. 1968. St. Louis Disperion Study,
Vol. II - Analysis. AP-53, Environmental Protection Agency, Research
Triangle Park, North Carolina.
Pasquill, F. 1962. Atmospheric Diffusion. Van Nostrand, London.
-61-
-------
I
Slade, D.H. (Editor). 1968. Meteorology and Atomic Energy. Atomic Energy
Commission, Oak Ridge, Tennessee.
Turner, D.B. 1964. "A Diffusion Model for an Urban Area." Journal of I
Applied Meteorology,3(1):83.
. 1970. Workbook for Atmospheric Dispersion Estimates. AP-26. £
Environmental Protection Agency, Research Triangle Park, North Carolina.
Walden Research Division of Abcor, Inc. 1974. Modeling Analysis of •
Power Plants for Evaluation of Impact of Ambient SO? Concentrations -
Metropolitan Baltimore AQCR 115. Cambridge, Massachusetts.
I
i
t
i
I
i
i
i
i
i
i
i
-62-
-------
I
I
I
I
I
I
I
I
Appendix A
• A MULTIPLE SOURCE GAUSSIAN PLUME MODEL USING SAMPLED
CHRONOLOGICAL INPUT (SCIM)
I
I
I
I
1
I
t
I
I
-------
I
•• Appendix A
A MULTIPLE SOURCE GAUSSIAN PLUME MODEL USING SAMPLED
_ CHRONOLOGICAL INPUT (SCIM)
A.I THE GAUSSIAN PLUME EQUATION
1 Experimental data describing the distribution of concentration
« in plumes from point sources show that, although wide variations occur,
these plumes exhibit a strong tendency toward a Gaussian or normal dis-
• tribution as a statistical average (Pasquill 1962). To describe this
distribution, consider a period of time with a constant point source
• emission rate of Q (units of mass per unit time) and a constant (in the
horizontal plane) mean wind velocity (with magnitude u). In a standard
H three dimensional coordinate system with the origin beneath the point
• source which is at a height h above the ground, the x axis oriented in
the direction of the mean wind and the y axis orineted normal to the
• mean wind direction, the ground-level concentration x is:
I x = _a_exp(if£
• TlUO CJ_ I C I O.
I
• /'
where
. co ^
y2 x
-------
f
I
u x dy dz = Q. w
\> —CO •
The parameters ay and az define the horizontal and vertical
dimensions, respectively, of the pollutant cloud in a vertical plane •
perpendicular to the mean wind velocity. Estimates of av and az as
I
functions of downwind distance from the source give a complete specifi- 0
cation of the concentration distribution. _
Certain modifications have been introduced in the above *
equation to account for more realistic boundary conditions. In parti c- •
ular, one such modification is concerned with the marked reduction in
vertical diffusion which is caused by a stable layer aloft, in addition I
to an imprevious ground surface. A simple approximation which avoids
the use of an infinite series as a solution was developed by Pasquill •
(1962). At some distance downwind, continued "reflection" of pollutants m
between the effective mixing ceiling and the ground surface will lead
to a uniform vertical distribution. Further reductions in pollutant con- V
centrations will be due only to crosswind diffusion. The uniform vertical
distribution will be approximately achieved at a downwind distance from |
the source at which az is equal to the height of the effective mixing
ceiling. Out to a distance at which az is equal to half this ceiling no •
effect due to "reflection" needs to be considered, as long as the plume •
height is significantly lower than the mixing ceiling. If az is repre-
sented as a simple power law of x, the following equations follow from •
these rules:
a (x) = bxq, x <_ x-, (5)
-64-
I
-------
I
I
I
1
1
I
I
I
I
I
I
I
I
I
I
I
I
I
oz(x) = L, x >_
oz(x) -
1 +1
< X <
xl
A2
L
oz(x)
X
b, q
.
2b
~ distance from source at v/hich a (x)
distance from source at which a (x)
vertical mixing ceiling,
vertical diffusion parameter,
dov/nwind distance from source, and
empirical constants.
L/2
L5
It may be noted that this approach provides a simple method
of using the same diffusion model for all travel distances from a source,
and avoids the necessity of changing the diffusion model formulation at
critical distances.
The SCIM model contains an option to modify the continuity
condition assumed above by an exponential decay factor to account for
various po-llutant removal processes in the atmosphere. With a pollutant
half-life of t5Q and a decay time of x/u, equation (1) with the decay
factor included becomes:
f
V2 0.693x
(6)
(7)
(8)
x -•
exp
(9)
-65-
-------
I
I
I
where:
x = concentration (mass/unit volume)
Q = source emission rate (mass/unit time)
u = mean wind speed (distance/unit time) I
ay = standard deviation of the horizontal displacement
of pollutants in the crosswind direction (distance) ff
az a standard deviation of the vertical displacement of
pollutants (distance) •
x = horizontal coordinate in mean wind direction
(distance)
y = horizontal coordinate in crosswind direction *
(distance)
tso = pollutant half-life (time) j|
h = effective height of the source (distance) _
0.693 * natural logarithm of 2 *
In this study it was assumed that there is no significant decay £
of sulfur dioxide (i.e., tsn is very large).
«
A.2 PLUME RISE
It is characteristic that emissions are released from a stack |
in order to reduce their polluting effect in the immediate vicinity and M
in the nearby downwind area. In general, these emissions must also be *
hot and fast-moving for the stack to discharge its effluents adequately. •
As a result, they leave the stack with a considerable amount of upward
momentum and thermal buoyancy. The effluent gases are accelerated up- •
ward; however, the upward momentum is continually diluted due to tur-
•
bulent mixing with the ambient air. The resulting effect is a general •
leveling off of the effluent plume at some distance downwind. •
-66-
I
-------
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
Several formulas have been developed to predict the plume rise
effect. A comprehensive review of these formulas and available plume rise
observations by Briggs (1971) resulted in a set of recommended formulas
which are simplifications and combinations of previous findings. The
following recommended formulas are obtained from his conclusions:
where:
h = Hs + AH (10)
h = effective stack height (m)
Hs = physical stack height (m)
AH = plume rise (m)
For neutral and unstable conditions (Pasquill stability classes A through
D).
AH «
where: „ /U-T
(11)
(12)
u = wind speed at stack height (m/sec)
k
K
F <_ 55 m4/sec3
0.75
21.4
F > 55 mVsec3
0.6
38.7
F * buoyancy flux parameter (nrVsec3)
Ta = ambient air temperature (°K)
u = wind speed at stack height (m/sec)
-67-
-------
Ts = stack gas temperature (°K)
Vs = stack gas exit speed (m/sec)
Ds = stack diameter (m).
g = acceleration due to gravity (m/sec^)
For stable conditions (Pasquill stability classes E and F)
1/3
AH = min
2 6 -
Z'b us
5.0
K Fk
V u
(13)
-3/8
where:
•T?
£0
az
2§. a vertical gradient of potential temperature (°C/m)
Certain limitations in the use of these formulas were noted by
Briggs. In flat and uniform terrain, he concludes that observed plume
rise may deviate from Equation (11) by ^10 percent; in the vicinity
of substantial terrain steps or near a large body of water, the deviations
may be +_40 percent. These findings relate to neutral stability conditions,
In unstable conditions, the deviations may be larger and occur somewhat
irregularly, however, the data presented by Briggs are not adequate to
quantify this. With regard to stable conditions, the data presented by
Briggs suggest that deviations in the nondimensional rise (i.e., Ah/
(F/us) 0-33) as large as +_0.5 may occur. When compared with a mean
-68-
I
I
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
-------
1
1
•
1
1
•
1
nondimensional plume rise of 2.6, this deviation shows that deviations
of +_ 20 percent
data.
A. 3 WIND SPEED
from Equation (13) have been observed in the developmental
AND DIRECTION
The model hypothesizes the existence of a mean wind speed and
direction. In some urban areas, wind monitoring equipment installed with
1
I
I
1
1
1
1
air quality moni
can be vector! a!
toring equipment from several representative points which
ly averaged to obtain a mean wind direction and speed.
However, it is often necessary to use a single airport observation to
define the wind
Since
is desirable to
associated with
wind profile may
atmosphere by a
direction and speed.
wind speeds near ground level increase with height, it
account for the increased dilution due to transport speed
emissions from sources with large effective heights. The
be represented empirically in the lowest layers of the
power law. An extensive summary of such power laws observed
on a 125-meter tower at Brookhaven National Laboratory were reported by
DeMarrais (1959)
where:
u(h) -
UT »
a =
h =
nl =
. Let
u(h) * ui(-7~)a (14)
nl
wind speed at height h,
reported wind speed,
empirical wind profile parameter
effective stack height (m)
observing height (m)
-69-
-------
Values of "a" are assigned in SCIM on the basis of the computed
stability class using three values, corresponding to unstable (classes A,
B, and C), neutral (class D), and stable (classes E and F) conditions.
The values shown in Table A-l are judged to be representative for a suburban
area such as is representative of most of the area in this study. They are
a compromise between values observed for a downtown (New York City) and a
more rural area (Brookhaven). Hourly wind speeds were obtained from surface
observations at the meteorological observatory at Baltimore-Washington
International Airport.
TABLE A-l. WIND SPEED PROFILE EXPONENTS
Atmospheric
Stability
Class
Unstable (A,B,C)
Neutral (D)
Stable (E,F)
Wind Speed
Profile Exponent
0.1
0.2
0.3
A.4 DIFFUSION PARAMETERS
A large number of empirical functions have been proposed by
investigators to represent the diffusion parameters ay and CTZ. In each
set of ay and az parameters, the values vary with the stability of the
atmospheric boundary layer. Two sets of diffusion parameter vlues,
one from Pasquill (1962) for rural conditions and one from McElroy and
Pooler (1968) for urban conditions, and the associated meteorological
parameters used to characterize stability are used in the SCIM program.
In this study the urban diffusion parameter values provided
by McElroy and Pooler (1968) were used. The system of stability
-70-
I
I
I
I
I
I
I
I
I
1
I
I
I
I
I
I
I
I
-------
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
classification is the same one used with the Pasquill diffusion parameter
values. The system, suggested by Turner (1964), for determining these
classes is described in Tables A-2 and A-3 and has been incorporated into
the SCIM program.
TABLE A-2. METEOROLOGICAL STABILITY CLASSIFICATIONS FOR CHARACTERIZING THE
DIFFUSION PARAMETERS (oy and !2
Stability Class for Indicated Net
•1
A
A
A
B
B
B
C
C
C
3
A
B
B
B
B
C
C
C
D
2
B
B
C
C
C
C
D
D
D
1
C
r
D
D
D
D
D
D
D
Radiation Index*
0
D
D
D
D .
D
D
D
D
D
-1
E
E
E
E
D
D
D
D
D
-2
E
J%
E
E
E
£
E
D
D
* Net Radiation Index Values are given in Table 6.
TABLE A-3. NET RADIATION INDEX VALUES
Time
of
Day
Night
Night
Night
Night
Day
Day
Day
Day
Day
Day
Total
Cloud
Amount (t)
t<0.4
O.-KiO.O
t = 1.0
t= 1.0
• t<1.0
0.57,000
c>16,CCO
c<7,003
le.or^cO.coo
c<7,030
!6,0:0>c>7,000
c>16,Cm
*<;-°
-2
-1
0
-1
1
1
1
0
1
1
Ket Radi.iuon
Indicated Solar
15°60°
--
—
— .
_™
4
2
3
0
2
3
-71-
-------
I
I
The solar altitude required in this system may be computed by •
the following formula (expressed in arctan rather than arcsin form):
I
I sin sin 6 + cos $ cos 6 cosHp,(h-12)p
« = arctan <— {]Z ^—) (15)
I 1/1 -{sin sin 6 + cos $ cos 6 cos(T5-(h-12)H'J"! I
\ I 11<- j / / •
L r i) *
f _ OO F/ " \ ^ 4 „ / ~ \-> A / m 1\ j. .I 00 V
o c c J. J\ -i on ' \~\ RH p5U \ul— I j i Q-OUI.
1J;
where:
a * solar altitude V
6 * solar declination
m ~ month of year I
d « day of month
h s hour of day H
<)) = latitude. •
I
The diffusion parameters presented by McElroy and Pooler
(1968) have been fitted to power laws for use in SCIM. The curves have •
been extrapolated from 600 to 10 meters. The fitted equations have the
forms,
av » axP •
y (17)
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE A-4. FITTED CONSTANTS FOR URBAN DIFFUSION PARAMETERS
BASED ON TURNER STABILITY CLASSIFICATIONS
Stability
Index
B
C
D
E
Crosr.riud
Constants '
a
IK
1.42
1.26
1.13
Q.992
1)
P
0. 745
0.730
0,710
0.650
Constants for
x600
b
0.0720
0.169
1.07
1.01
(2)
q
1.22
1.01
0.682
0. 554
(l) a = ax , where x is downv/ind distance from SOIDTCC; g and x are in meed's.
D
(2) " z - br.1; a and ^ are in moters.
(3) Not: available from McElroy and Pooler data; use Class B values.
The Turner stability categories can be determined from routine
airport weather observations. Thus, although the Turner stability classes
may result in greater uncertainty regarding the true diffusive capability
of the atmosphere at any given time, they are preferable to other systems
of stability classification due to their greater availability.
A.5 MIXING HEIGHT
The vertical mixing ceiling is that height above ground level
at which there is a marked reduction in vertical diffusion. Such barriers
are observed as a sharp drop in the concentration observed in a vertical
sounding (e.g., Davidson, 1967). It may be observed as a delineation
between the smoke-fille layer and cleaner air aloft over many cities in
the early morning. Much higher ceilings typical of afternoon hours are
clearly visible to air travelers in climbing to or descending from cruising
altitudes. The ceiling may vary from 100 meters at night to over 1500
-73-
-------
1. Use the morning minimum from midnight to 6 a.m.
2. Linearly interpolate between the minimum and the maximum
between 6 a.m. and 2 p.m.
3. Use the afternoon maximum between 2 p.m. and midnight.
I
I
meters during the day. Hourly estimates of the ceiling are required for
use in the model. ft
The ceiling may be inferred from temperature soundings which
are routinely observed twice daily at certain airports by the National I
Oceanic and Atmospheric Administration (NOAA). The closest one to Baltimore
is located at Dulles International Airport. •
The procedure used to define the mixing ceiling is the following: m
Determine the general rural vertical temperature profile from the Dulles
radiosonde. Use the Baltimore airport minimum morning and maximum after- I
noon air temperatures as representative of the urban ground temperatures.
Determine the height at which an adiabatic temperature profile from the J
urban ground temperatures intersect the Dulles temperature profile. The _
heights of these intersections are assumed to be the daily minimum and "
maximum mixing ceilings. The method of interpolating between these values •
to give hourly estimates is:
I
^
Although this pattern introduces a sharp discontinuity at midnight, this
is not a serious limitation. The actual process of decay of the afternoon •
mixing height and establishment of the new mixing height observed in the m
morning radiosonde is more complex. It begins after sunset with the ^
formation of an inversion layer near ground level in the rural areas V
surrounding an urban region. The height of the rural inversion layer
I
-74-
-------
I
I increases through the right and gradually extends over the urban area.
« Within the urban area mixing continues between the ground and the base of
the new inversion layer which gradually forms over the city. The intensity
ff of mixing below the mixing height gradually decreases after sunset. This
is reflected by increasing atmospheric stability. As stability increases
I the height of the mixing layer becomes less of a limit to the dispersion
process, since stable plumes are much less likely to extend vertically to
• the top of the mixing layer than unstable daytime plumes. It is assumed
• in the SCIM program that the new mixing height becomes established over
the urban area at 1 a.m.
I
I
I
I
I
I
I
A.6 THE GAUSSIAN PLUME EQUATION FOR MULTIPLE SOURCES
The basic Gaussian plume model describes the concentration
field from a single point source during a short-term (i.e., steady-state)
period. However, in an urban environment it is often necessary to con-
sider emissions from many small sources (e.g., residential heating units),
in numbers too large to treat each source individually. As a practical
matter, individual large sources whose emission rates are significantly
* greater than other sources in the general vicinity are treated as point
• sources. The remaining emissions are treated as an area source whose
emission rate and effective height are specified as a function of hori-
| zontal position using appropriate emission inventory techniques. As
described by Calder (1970), it is a simple matter to adapt the point
source model to an area source. First, consider a reorientation of the
-75-
-------
I
I
coordinate system so that the origin is at a receptor point of interest
and the x axis is pointed upwind into the mean wind direction. In this I
coordinate system, equation (9) gives the concentration at the receptor
from a point source with horizontal coordinates (x,y). |
Let q dx dy be the total amount of pollutant emitted per unit
time in a horizontal element of area dx dy. Assuming that the total •
concentration at a receptor is the sume of concentration contributions j|
from all individual area source elements, the concentration, x/\> at the
receptor location due to the area source is: I
/•Xl fy2 q(x, y) / y2 hA2 I
= ./« /,. *u(hJo.,uK(xr expi - rlry" ^T^T
' " i
•A/uyv-/uz
I
dy dx
(18) -
where: ^
hA = effective area source emission height, ™
q(x,y) « area source emission rate per unit area, •
x-| = distance to most upwind portion of emission area, and
y-j ,y2 = distances to furthermost crosswind portions of I
emission area.
The effect of the two types of sources (area and point) are analyzed
separately and added together to give a resultant concentration.
-76-
I
I
I
I
I
-------
I
•
•
XP ^-) exp - 21 - try - Trr 09)
_
9
•
•
Q
I
I
I
I
The concentration xp at a receptor point due to N point sources
is given by:
N [ Qj ( y,2 0.693 x., h,2 }]
where the script i denotes values for the ith point source.
Equations 18 and 19 give the concentrations from area and
point sources in an urban area during a single quasi steady-state period.
The total concentration is given as the sum of that from the area and
the point sources.
The basic model described thus far defines the urban concen-
tration field for a single quasi steady-state period. However, mean
concentrations over long-term periods during which both emission and
meteorological conditions are significantly variable are also of interest.
_
™ The short-term model is used to estimate long-term concentrations by
• assuming that the long-term period may be resolved into a succession of
short-term quasi steady-state periods. This assumption is basic in the
• long-term models which have been developed by various investigators.
Although relationships between diurnal variations in emis-
• sion rates and meteorological characteristics are difficult to estab-
• lish with currently available data, there are reasons to believe that
significant correlations exist. For instance, the marked diurnal vari-
ation in diffusive capability of the atmosphere is correlated with the
variation in emissions associated with the diurnal cycle of business
activity. Similarly, cold and windy weather, associated with strong
-77-
-------
I
I
emissions from fuel consumption for space heating, tends to be correlated
with northerly wind directions. A study by Koch and Fisher (1973) resulted •
in a method used in SCIM to estimate hourly variations in emission rates
using hourly temperature observations and the hour of the day. •
SCIM reduces the calculations required to estimate long-term
mean concentrations by statistically sampling the quasi steady-state •
periods which make up the long-term period. Concentrations calculated •
for the selected periods are averaged to obtain the long-term mean con-
centration. These calculations are used to approximate the frequency I
distribution of short-term concentrations over the long-term period, by
ordering the concentrations from lowest to highest value. This approach J
has been applied with considerable success (Koch and Fisher 1973). Sta- _
tistical sampling can reduce the calculations required for seasonal con- •
centrations by 90 percent or more. •
The SCIM computer program provides the user with a tool for
estimating short-term maxima of pollutant concentrations for a sample •
of short-term periods selected from a specified long-term period. The
sample is then used to estimate the long-term mean concentration, the m
geometric standard deviation and the statistical frequency distribution •
of short-term concentrations for all specified locations. The expected
annual maximum concentration may be determined from the frequency distri- I
butions or by means of statistical distributions based on the log normal
distribution. The calculations are made for specified receptor locations.
-78-
I
I
I
-------
I
Since the geometric standard deviation is determined for specific locations
• using hourly concentrations generated by the model, the statistical
extrapolation is not influenced by existing characteristics of source
J emission rates and spatial distributions of sources. Any changes in the
• sources which are input to the model are reflected in the computed geo-
metric standard deviations.
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
• SCIM MODEL CALCULATIONS COMPARED TO MEASURED 24-HOUR
I
I
I
I
I
I
I
I
I
Appendix B
S02 CONCENTRATIONS
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
24-Hour SO2
Feb. 27, 1973
FREDERICK
COUNTY
PRINCE
GEORGE
COUNTY
MONTGOMERY
COUNTY
\
V
Calculated
10
MILES
-81-
-------
24-Hour
Nov. 12, 1973
t^% CARROLL / BALTIMORE
\ COUNTY / COUNTY
1
\. HARFORD
\ COUNTY
FREDERICK
COUNTY
BALTIMORE CITY
0 51/10
HOWARD
COUNTY
MONTGOMERY
COUNTY
122/103
96/99 * 88/
39/28000JW
o
033/10.
166/84
PREVAILING
WINDS
I
N
PRINCE
GEORGE
COUNTY
ANNE
rARUNDEL
10
MILES
-82-
o32/21
Measured
Calculated
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
24-Hour SO2
Dec. 4, 1973
FREDERICK
COUNTY
PRINCE
GEORGE
COUNTY
MONTGOMERY
COUNTY
PREVAILING
WIND
MILES
-83-
-------
24-Hour SO2
Dec. 12, 1973
FREDERICK
COUNTY
MONTGOMERY
COUNTY
PREVAILING
WINDS
PRINCE
GEORGE
COUNTY
Calculated
MILES
-84-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
24-Hour SO2
Dec. 24, 1973
FREDERICK
COUNTY
MONTGOMERY
COUNTY
I
10
MII£S
PREVAIUNC
WINDS
FRINGE
GEORGE
COUNTY
-85-
-------
24-Hour SO2
Feb. 10, 1974
\CARROLL / BALTIMORE
COUNTY / COUNTY
\
VHARFORD
COUNTY
_ 65/4S
FREDERICK
COUNTY
BALTIMORE CITY
FRINGE
GEORGE
COUNTY
MONTGOMERY
COUNTY
/\
PREVAILING
WINDS
N
10
MILES
-86-
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
24-Hour SO2
Feb. 28, 1974
W CARROLL / BALTIMORE
Y COUNTY / COUNTY
!
V HARFORD
\ COUNTY
^k / X^ 0 6S/49
ICK I J \^ o
COUNTY
PRINCE
GEORGE
COUNTY
MONTGOMERY
COUNTY
/ PREVAILING
/ WIND
10
MILES
-87-
-------
24-Hour SO2
May 11, 1974
FREDERICK
COUNTY
BALTIMORE CITY
O 751 IS
PRINCE
GEORGE
COUNTY
MONTGOMERY
COUNTY
\
f
5 10
MILES
PREVAIUN
WINDS
-88-
I
I
I
I
1
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
24-Hour SO2
Aug. 23, 1974
»*•% CARROLL / BALTIMORE
\ COUNTY / COUNTY
\
V HARFORD
\ COUNTY
PREVAILING
WINDS
FREDERICK
COUNTY
BALTIMORE CITY
HOWARD
COUNTY
,17/5
,56/20 oo
LO/22 o
>38
MONTGOMERY
COUNTY
PRINCE
GEORGE
COUNTY
ANNE
fARUNDEL
O
Measured
Calculated
-89-
-------
24-Hour SO2
Nov. 7, 1974
FREDERICK
COUNTY
PREVAILING
WIND
PRINCE
GEORGE
COUNTY
MONTGOMERY
COUNTY
\
f
10
MILES
-90-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
Appendix C
• EMISSION INVENTORIES FOR BALTIMORE AQCR
I
I
I
I
I
I
I
I
I
-------
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
-
—
f
~
i r
~ t±
*
" t^
•^ r
_h
" e
u.
i
I
f
i
1
i
i
~
j.'
X
-L
X
T
-
'r
-
-L
LL
1
i<
i. ^
-L _
>-
^ >V
r j.
X — •
J7 J
1 X
-1- -L
-
± -
7^
- 2
_
"
^
C C
f —
^ J-
- -
•X 3-
-r
L _
ki —
—
- .-
-
•^ -H"
-'
<
"
^rv
f IT
* J-
r c
X fx
f *
" -
--
~* X
£?
„
T >-
k --
n c
t —
-. rx
4
- -
f1 -T
* ~
•*" *
- -
- -
t ^
^
£ k
.- i
£!
i r
v "~
C „
r ^
- „
r ;
"* T
rr -r
- -
_ -
^ r
x ^
^ ^
S5
x r\
- —
-
- ~
-. -
- :
; :
4- J
r :
j f
7 -
P S'
f\ ^
(f —
: c
"^ -T
- -
- ^
. ^
t 4
- -
- T
rv ^
C C
£ *
> (\
f rx
- _
i fx
x
— I
'
t -
— -4-
X <
f [f
^
~-- j:
* *
- IT
--
- _
f, rr
X ?r
f -£
r a
; T
- 5
: s
r- :
T 7fc
^
: -
- c
r rx
- •
C- -3
- -
E''
' -r
; -
- -
. -
r ^
f -
r f
r-
-
£•-
~ c
x c\
- —
J
- -
r i
1 IX
- -
- -
b- ;
f ~
f f
:
£ ?
- -
?£
N ,-c
- ^
-5 4
- -
-3" ^
•* J"
L ^
X X
- J
: 3
kt f\
-
;
- -
.-x <
r r
J- -J-
X rr
- -
r "v,
-
C '^
P 3-
— t
- -
r :
f ''f
Vr ^
f?
N rr,
* ^
- »
r r
s. -
• -
F *_
-,
\ s
r- x
r j-
V L-T
- —
f f
j ;
f- ^
^ -f
C
v r\
—
f^ x
f -r
— r
• •
c c
r. -:
• -
- -
•• ,-r
' -
t ;
r x
t —
<• x
^ "^
f -
"" ~"
r*- >;
- ^
r --
-
- x
f :
- -
-
r %
K .
r-
r-
t " ir -
- : r -
> - r x
i
- - - -
r I r" "
I
i
• •» L" -
K -r
-
- „- i'v
• - :
_ cv --
k , K ,
-91-
-------
^
~ *
r -c
r ~
~ '
If -a
- "-
"V
c c
f- .T,
- _
r r
.
1-
—
r
r -
ri
-
t;
x -
^ <
: ~
: =
v -?
r ,x!
t c
c X
"
- r
S. "f
I ~
- -
tf *
- -
t*
;
J -I
le -
— _
X X
: -
L- .-
; -
(V
S «•
r j-
V !X
.
;. —
— _
1
V -r
f X
•--'
r -
( ' C
K -"
£ -c
r (x
.,
-'
If 7
"
; ,^
r j-
'
r ^
- .'
- f.
= t
x^
X, ,J
..
14 -
KJ- ,X
.- f
• -
-
"- "
f- -n
r ^
•r*
, fC.
..
r -
£ ;
- -x
~ i
i ^
- lC
r >
•— :*.
rx ^
"
i -r,
- .-
^
m
f -
v i
- =
»- -
^
r ?
- !\
:: -c
n in
r <
•^
"*" r*-
~ f
'- IX
"
•^ _^-
f if
- :.
f ^
S. -f.
- t.
: jv
v _ ,
•r, 3,
* *
r -
"
v «
<" IT
V ~
V F~
f -C
- c
: c
". tv
^ rr
if
r. !Nl
_
*
. .-
r •_-
; ^
c ^
n c
C IT
- ^ .
.IX
n t\J
r. c
- -4-
:"
r !V
r if
v «-
- -
C C
- C '
n c
-
r, ^.
: .r
p r- i
r ^
y ^«
<• X.
r T
"
- c
-> c ;
Z -n
' "*
• ,
" w
~~
1 «•
•'
" "~
T -r
-•n •
- _
X CV
f- ^T
r -"
fc rv
~
.4 __,
r.?
:• ^
~ c
r c !
: c
u
-
r, -*
-- f
r ^
~ *"
f -f
r,r.
~ -t
— ^
- -
i- C
- ^
^ IX
- T.
* r
t r
. (X
• s
r IT
- X
— X
r -s, ••
- c
— IX
t ,-r
r .r
- .x
"' -
-, -r
f CX
r s
-'
~ "
*> in
: tx
v _ ,
"^
; s-
"
~ ^
N. ^^
: -
- •*
• ^
c .
^ -
v f •
f 'X
- -'
r ^i
"*
* ,-x
\ •»
: ":
- z \
•m, J- .
^- <~
1
- -
x. X
* -T
. :
: ,
" .r .
- _
^
~
f i
r r
\ .
"'
i i
" C -T -"
I
' F =
z t -
"\ t- r
_ L _
T J -fc •*"
i
L
r /•
v ^ t ^
i
- - .- -
'• - F ^
. '. u .
i
f -
- ; f .-
, V t ^'
<~ -^ If
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
- -*
r j-.
- c
r ^
r j-
f -
: v
£ -
r f
' !\
f- r
r c'
~ .r
— >»•
C _
r c
r. c
^ -i
t -j
<• —
\ ^
- —
T :
H~
r f
r-
7
—
- -
p. if.
t •*
£ —
\ ~
* >J
•^ N
r f\
r *r
"- "N
r- -
h- —
: c
" .n -
t >fr ;
- —
- c
r -r
s. —
7 ,-
r -
L ^*
f .
V "
J **•
-, -
tf ^*
\. &
r j-
— f
- ^~
"-
- r~
• c
J- J-
t „
•i rv
^ rf
«• .-V
, "T
- ""f
7 »•»
- r^
- r
V. •%!
" «T
[ -
; C
- X
L 1 ^~
" C
- J^
T _
ff
— •*
U T
K -r
r- '.
r .-
[.-
- ~
-
--
t~ t\
•? f^
^ re
C j-
- ^
r
C T
- <
L* f
:-.
rv ,T
"
- ,-
» ^.
n rr
"* >•
c z
r ro
* >3-
-£
r x
3 ^
3 \f
"
- ^
- -c
t. •*
- _-
•c *.
£ -
-, ^~
T T
- 3
if .r
^ rv.
f X
>f r
£ •>•
- xi
"• ^
J- —
— —
- j-
* (X
IT ^
- X
(X
r -J-
r .%
<- ^
-r x
? ir
4 <
; r
: &
I c\
1 \n
f* X
* C
f X
•4
- ^*
- rv,
V (N.
: r^
^ -c
f ir
- j"
S- rr
^ -
r >r
r. r
J- -I
<• <
- c.
- C
ft IV
£ rv.
f f
— (\j
|M(
r c
N. in
-i r
v j~
r x
^". -^
- j>
r. r-
r ^>
•^ —
r (»•
<^. *-
r -t.
r J
i if
3- r
- =
""
: c
t c
f i»-
• <
: c
^ r*
"• i/>
-
*• tv
r .*
- IT,
:
-------
? i
r- 4
- -
r -~
r >-
r
k 4
- .r
k '
h
m >•
r- =
X (N.
- c
t. —
^ -
- <
* IT
^ rr
~ •
-c
:v
I -p
p ?
- cr
^ i1^-
r :
r <
94-
', C\.
—
t: =
!>y
^ 3"
r r
-•
C^
4- •*
^ -
F =
I ^
r -r
? -3
r r
; x
: ;
p. c
X?
u. "-
t .--
•4
- -4-
v cc
*»(
C r\.
i r^
t 4
_ ^
s. r'"
- —
P -"•
r I
r -f
C >£
r -f
r- x.
N rr.
- -
; ;
V "3"
\ c
^ 4"
r- ;
c -c
- 3
i -
r. -*~>
r •*
X ."**!
^- -
f -r
': ?
r ir
»
;x
-
r
^
r «"
\ —
: -
^ i
_ -
\ LT
f.
^ -<"
tC\ f
|
j- —
iC.^
- ~
f -
" '
-C -C
-
1:
t- .
i
1
1
1
1
-•I- 1
;-•>&? •
- ,- u <-
1
: z \z : \
,
-------
1
1
1
1
1
1.!
|
1
1
1
1
1
' ?1
— ^
^ f
-r f
< -
X
- -
* 7
N '.T
^-- •£
_
T <
< <•
r c
r c
_
: ^
T
. -
- x
J S
C '
- -f
r «•'
r z
: ^
-
:>
v" _,.'
— ;
"
<
£ C
f sT
; j
-
. '-
|r -r
r "'
•t i
X T
- -
••- —
- -
—
- .'
r *
t -f
•: f
^ -r
!T£
- X
X "V.
— T
- ^.
r ^~
• r\i
~ -r
T
>?
'
\ "\
1
1
' -' i
1
f. -f:
* •*•
~ -r
* -
. *
V £
\ r
-
"• x
- -
r s
- J,
- <•
- C
r ^
- <>•
k I
"_ .X
— I—
r -"
; &
r ^a-
- y
-r-
f
L '"^
- j
-
j-
3 r-
~i
I. ••?•
- -t
r C
c J"
i*. f
~ -
^ -
V ••"
15
c ^
f -»•
- ^
. ^
f- r\j
: ~
1- -
^1
* *r
i <
r =
i •-
•*
• r-
u -
?5
\ i\.
r 2
-
: c
n c
-< rf.
: f-
- x
i J-
£ c
_ ^
i 5
•/ f
•— 4 !\
I .-
95-
: J
t •*
n x.
t «••
n c
X (V
L -
x tv
M r^
•*- rv
; r-
:
1 C
f "">
r^
. „
r iT
' r*
^
T >
F. J\
*•' •*
\j rr.
r "•
V; iXJ
r ,-x
; rr
, ^
-, If
f ~*
X -
f »-
r c
C -3-
: -t
-4
r -*. i
: 't
^ _
J- -A
r IT
s. *••
^ .^
: ^
: -t
:'.-
' *
• f-
. _
:»
V. "\
-
- J ^
^ -r -
<• *•
\- -
_
V X
r -r i
f *
~ ~~
\ -r
r c
r -c
r --
- j-
«•
r f
<-' f<-'
V. •••
1 r'
•\ >?
- .r
-. -
•?
' !x
H ^.
- r
'-
= ;.
v
-------
A *
>r i— *
=• --f
i •<•>
«• -"^
- -
- ^
t: •
f -T
\ a
•V i—
,C "-
n x\
4- •*•
C i
: ^
- 1
~ -
- ,-r
T *
r f
i^--
r^" -v
r. (^
4- ^
r.-r.
; £
~ •!
^.
; -r
- •<:
- -
* f
e- f
- ^
-
~- t~
r. .-<•.
*\ ^
K. -
3 —
T rr
£ -
r ;
•r
x r
.- -T
< <
; -t.
x -J
c. -f
, -
; ~*
— t r«<
-
U N
r j-
^ <
c A
(\i
\. t\;
^ -£
; ?
a- "-
•- !~
I- IT
I £
4 -S-
X -
1 C
; C
n (v
«^
r sv
r. -c
- OJ
i-M
* —
M T.
~i CSt
J «-,
^- r*
s. rr
. «•
s. x>
r in
r (^
* J
- -
N fx
~ T>
C *
\ C
\: -f
•* i
r> (r.
- i^
•4 Z-
r <*•
r, —
- -
f. -C
^ IT-
1 >f
^ "•?
V rr
- ?'
? r-
C x
= C
^ -T>
^ (S*
**
•f O
>£ r
f S*
; •*
_ —
A T
f\ ir\
J
•v, if
f i-V
-1 X.
r '--
\; if
- s-
r x\
* f-
i -4-
i >
'
n c
-> c
fl X
- C
* >3
\i -fr
j— i
r ^
f- C
r-
i -*•
-» a-
r —
7 «
" 2C
4- <
T r^
T I*1
.- -
1
-96-
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-903/9-76-024
3. RECIPIENT'S ACCESSION-NO.
4. TITLE AND SUBTITLE
Study of SC>2 Around the Crane, Riverside and Wagner
Power Stations in Metropolitan Baltimore
5. REPORT DATE
January, 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Robert C. Koch and Kenneth Pickering
8. PERFORMING ORGANIZATION REPORT NO.
EF-558
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Geomet, Inc
15 Firstfield Rd
Gaithersburg, MD 20760
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-2331
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Region III
6th & Walnut Streets
Philadelphia, PA 19106
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA - Region III
15. SUPPLEMENTARY NOTES
Contract performed pursuant to the Energy Supply and Environmental Coordination Act
of 1974
16. ABSTRACT
Based on a determination of the worst-case meteorological conditions associated with
the Crane, Riverside and Wagner power plants in the Metropolitan Baltimore area and a
modeling analysis of the contribution of each plant to the maximum ground concen-
trations of SC>2 under those conditions, the following conclusions are drawn regarding
each plant: Crane Power Plant - Although, increased SC^ emissions from this plant
will degrade air quality, the NAAQS will not be exceeded if the present emissions are
tripled. The allowable sulfur content in coal from this plant could be increased
from 1 to 3 percent. Riverside Power Plant - 862 emissions from this plant could be
doubled without exceeding ambient air quality standards. The allowable sulfur con-
tent of coal burned at this plant could be as high, as 2 percent. However, it is
recommended that a more detailed study of air quality resulting from this plant based
on more detailed emission and meteorological data than is presently available be made
before approving such an increase. Wagner Power Plant - 862 emissions from this plant
should not be increased. The allowable sulfur content of ocal burned at this plant
should not exceed the current limit of 1 percent.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
sulfur dioxide
Energy Supply and Environmental
Coordination Act
air pollution modeling
power plants
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |