[PA 904/9-76-005C
NORTH CAROLINA
AIR QUALITY MAINTENANCE
AREA ANALYSIS
APRIL 1976
FINAL REPORT
VOLUME III
TSP DISPERSION MODELING AND ANALYSIS
FOR CHARLOTTE, WINSTON-SALEM, AND GREENSBORO
AQMA'S FOR 1973, 1975, 1980, AND 1985
U.S. ENVIRONMENTAL PROTECTION AGENCY
REGION IV
AIR AND HAZARDOUS MATERIALS DIVISION
ATLANTA, GEORGIA 30309
-------
, (tk£>
STATES ENVIRONMENTAL PROTECTION
REGION IV n i
I In-r I Q
J-121 PEACHTF?EEST., N, E. ' ' vJ 13
ATLANTA, GEOFJGIA 30309
September 10, 1976
i^orth Carolina Air Quality Maintenance Area Analysis is now
Available from- National Technical Information Service
Springfield, Virginia 22l6l
using the following ordering information:
Vol. I: TSP Area .Source Emission Inventories for Charlotte. Winstcn-
Salem, and Greensboro AQMA's for 1973, 1975, 1980 & 1985! 100 m>
EPA 90V9-76-005a. NTIS order # - PB 255 218/AS, $5.00
Vol. II:TSP Point Source Emission Inventories, TSP Monitoring Data and
Meteorological Data for Charlotte, Winston-Salem, and G^e^sboro
AQMA s for 1973, 1975, 1980 and 1985. 6h pp.
EPA 90U/9-76-005b. I\TTIS order # PB 255 219/AS, $^.50.
Vol. Ill; TSP Dispersion Modeling and Analysis for Charlotte. Y?inston-
Salem, and Greensboro AQMA. s for 1973, 1975, 1980, and 1955. 85 T>X>
EPA 90V9-76-005c. WTIS order # PB 256 227/AS, $5.00
VOL. IV: Project Suir^ary and Recommendations for the Charlotte, WJnsfot'-
Salen, and Greensboro AQJ-li's for 1973, 1975, 19&0, and 1935." 5^ pp. *
EPA 90U/l-?6-005d. KTIS order # PB 255 22-0/AS, $4.50. '
All volumes are available in microfiche for $3.00 each.
A limited number of copies is still available at the EPA Region IV office.
-------
EPA 904/9-76-005C
NORTH CAROLINA AIR QUALITY MAINTENANCE AREA ANALYSIS
VOLUME III: TSP DISPERSION MODELING AND ANALYSIS
FOR CHARLOTTE, WINSTON-SALEM, AND GREENSBORO AQMA'S
FOR 1973, 1975, 1980, 1985
FINAL REPORT
APRIL 1976
by
Richard C. Haws
Harry L. Hamilton, Jr.
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY - REGION IV
ATLANTA, GEORGIA 30309
Under
Contract Number 68-02-1386
Task 15
RESEARCH TRIANGLE INSTITUTE
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27709
-------
This air pollution report is issued by Region IV,
Environmental Protection Agency, to assist state and local
air pollution control agencies in carrying out their program
activities. Copies of this report may be obtained, for a
nomimal cost, from the National Technical Information Service,
5285 Port Royal Road, Springfield, Virginia 22151.
This report was furnished to the Environmental Pro-
tection Agency by Research Triangle Institute, Research
Triangle Park, North Carolina 27709, in fulfillment of
EPA contract no. 68-02-1386, Task 15. This report has been
reviewed by the Air and Hazardous Materials Division, Region
IV, 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 IV Publication No. EPA 904/9-76-005C
ii
-------
TABLE OF CONTENTS
Page Number
LIST OF TABLES iv
LIST OF FIGURES vi
1.0 Introduction 1
2.0 Projected Air Quality Through 1985 5
2.1 Charlotte (Mecklenburg County) AQMA 6
2.2 Greensboro (Gullford County) AQMA . 20
2.3 Winston-Salem (Forsyth County) AQMA 30
3.0 Calibration Procedures 43
3.1 Mecklenburg County, Final Calibration . 70
3.2 Guilford County and Forsyth County, Final Calibration 72
REFERENCES 76
Hi
-------
LIST OF TABLES
Table Number Page Number
1 Mecklenburg County Receptor Concentration Data for
TSP by Year 10
2 1973 Mecklenburg County Source Contributions to Five
(5) Receptors with the Highest Calculated Concen-
trations (From AQDM Source Contribution Tables) ... 15
3 Mecklenburg County Source Contributions to Receptor
with Highest Calculated Concentrations (From AQDM
Source Contribution Tables) 19
4 Guilford County Receptor Concentration Data for TSP
by Year 26
5 1973 Guilford/Forsyth Source Contributions to Five
(5) Receptors with the Highest Calculated Concen-
trations (From AQDM Source Contribution Tables) ... 29
6 Guilford County Source Contributions to Receptor with
Highest Calculated Concentrations (From both Guilford
and Forsyth Sources) 34
7 Forsyth County Receptor Concentration Data for TSP by
Year 37
8 Forsyth County: Source Contributions to the Two
Forsyth County Receptors with the Highest 1985
Calculated Concentrations (From both Guilford and
Forsyth County Emissions) 39
9 Comparison of the Ranges of (Unadjusted) Calculated
TSP Emissions for Mecklenburg County by Source,
Category (1973 Annual Arithmetic Average, yg/m ) . . 51
10 Comparison of Observed and (Unadjusted) Calculated
Concentrations of TSP (by Source Category) for
Mecklenburg County (1973 Annual Arithmetic Average)
(yg/m3) . 52
iv
-------
LIST OF TABLES (Continued)
Table Number Page Number
11 Comparison of Effect of Effective Plume Height
Changes on Calculated Concentrations of TSP for
Mecklenburg County Using 1973 Unpaved Road
Emissions Only (yg/m3, Regression Slope 1.0,
Intercept 0.0) 56
12 Comparison of (Unadjusted) Calculated Concentrations
and Observed Concentrations of TSP at Sampling sites
in Mecklenburg County Using AQDM with the Briggs
Plume Rise Equation and the Holland Plume Rise
Equation (1973JVnnual Arithmetic Average Concen-
trations, yg/m ) 58
13 Comparison of Mecklenburg County Receptor Concen-
tration Data for 1973 Using AQDM with Briggs Plume
Rise Equation and Holland Plume Rise Equation .... 59
14 Comparison of (Unadjusted) Calculated and Observed
Concentrations of TSP from Forsyth County and
Guilford County Emission Sources when Modeled
Separately with the AQDM (1973 Annual Arithmetic
Averages, yg/nr) 66
15 Comparison of Contribution of Source Categories
to Total Emissions (1973 Tons/Year and percen-
tages) 67
16 Comparison of (Unadjusted) Calculated and Observed
Concentrations at Calibration Monitoring Stations in
Guilford and Forsyth Counties and Calculated
Regression Parameters with: 1) None Removed, 2) One-
half Removed, and 3) Total Removal of Fugitive Dust
from Unpaved Roads from the Emissions Inventory
(1973 Annual Arithmetic Averages, yg/m3) 69
17 Final AQDM Calibration Run Results for Mecklenburg
County (1973 Annual Arithmetic Averages, yg/m3) ... 71
18 Final AQDM Calibration Run Results for Gull ford and
Forsyth Counties (1973 Annual Arithmetic Averages,
yg/m3) 74
-------
LIST OF FIGURES
Figure Number Page Number
1 Mecklenburg County Area Source Grid Network 7
2 Mecklenburg County N.C. AQDM Receptor Grid 8
3 Mecklenburg County N.C. 1973 (Calibrated) AQDM
Calculated Concentrations of Total Suspended
Particles (Micrograms/Cu. Meter) 9
4 Mecklenburg County N.C. Projected 1975 Concentrations
of Total Suspended Particles (Micrograms/Cu. Meter) . 16
5 Mecklenburg County N.C. Projected 1980 Concentrations
of Total Suspended Particles (Micrograms/Cu. Meter) . 17
6 Mecklenburg County N.C. Projected 1985 Concentrations
of Total Suspended Particles (Micrograms/Cu. Meter) . 18
7 Gui1 ford County Area Source Grid Network 21
8 Forsyth County Area Source Grid Network 22
9 Gui1 ford County N.C. AQDM Receptor Grid 23
10 Forsyth County N.C. AQDM Receptor Grid 24
11 Guilford County N.C. 1973 (Calibrated) AQDM
Calculated Concentrations of Total Suspended
Particles (Micrograms/Cu. Meter) 25
12 Guilford County N.C. Projected 1975 Concentrations
of Total Suspended Particles (Micrograms/Cu. Meter) . 31
13 Guilford County N.C. Projected 1980 Concentrations
of Total Suspended Particles (Micrograms/Cu. Meter) . 32
vi
-------
LIST OF FIGURES (Continued)
Figure Number Page Number
14 GUI1ford County N.C. Projected 1985 Concentrations
of Total Suspended Particles (Micrograras/Cu. Meter) . 33
15 Forsyth County N.C. 1973 (Calibrated) AQDM
Calculated Concentrations of Total Suspended
Particles (Micrograms/Cu. Meter) 36
16 Forsyth County N.C. Projected 1975 Concentrations
of Total Suspended Particles (Micrograms/Cu. Meter) . 40
17 Forsyth County N.C. Projected 1980 Concentrations
of Total Suspended Particles (Micrograms/Cu. Meter) . 41
18 Forsyth County N.C. Projected 1985 Concentrations
of Total Suspended Particles (Micrograms/Cu. Meter) . 42
19 Locations of Sampling Sites in Mecklenburg County . . 44
20 Locations of Sampling Sites in GUI1 ford County ... 46
21 Locations of Sampling Sites in Forsyth County .... 47
22 Mecklenburg Final Calibration, Total Suspended
Particles 1973 Annual. Arithmetic Average 73
23 Guilford/Forsyth Final Calibration, Total Suspended
Particles 1973 Annual Arithmetic Average 75
vii
-------
l.Q INTRODUCTION
The Research. Triangle Institute (RTI) is under contract* to the
Environmental Protection Agency (EPA), Region IV, to assist the Air Quality
Section, Division of Environmental Management, North Carolina Department of
Natural and Economic Resources (NCAQS), in the Air Quality Maintenance Area
(AQMA) analysis. Three (3) AQMA's in North Carolina were designated for
suspended particulate matter only: the Charlotte AQMA (Mecklenburg County),
the Greensboro AQMA (Guilford County), and the Winston-Sal em AQMA (Forsyth
County). The analysis steps have included the updating of the emissions
Inventories and their projections, the "calibration" of the dispersion
model for the specific AQMA's for the baseline year (1973), and the sub-
sequent modeling of projected air quality for the years 1975, 1980, and
1985. This report summarizes the analytical work which has been done by RTI.
RTI's primary responsibility has been to perform dispersion modeling
of air quality for total suspended particulate matter (TSP). In doing so,
RTI has assisted in the validation of air quality monitoring station data,
meteorological data, and particulate emissions data from both point and
area sources. All data have been furnished, directly or indirectly, to
RTI by the NCAQS.
Engineering Sciences, Inc. (ES), under contract to EPA, Region IV,
has had the responsibility for preparing the area source emissions inven-
tories, their projections, and the allocation of county-wide emission
totals to sub-county grid squares for modeling. Details of this effort
are contained in the ES final report (Ref. 1).
The NCAQS, with the assistance of other state and local agencies, was
responsible for preparing updated point source emissions inventories and
*Contract Number 68-02-1386, Task 15
-------
their projections, jneteorolocjtcal data and analyses, and monitoring station
data and its validation. Details of this effort are contained in the NCAQS
technical report (Ref. 2).
RTI was instructed to use the Air Quality Display Model (AQDM) (Ref. 3),
as modified by EPA Region IV for the dispersion modeling. During cali-
bration, the AQDM was effectively used to detect data errors, and/or
anomalies, and to systematically test hypotheses concerning the emissions
and their impact on air quality. During the course of the modeling, results
with suggested interpretations were jointly reviewed and discussed by RTI,
NCAQS, ES, and EPA.
Mecklenburg County was modeled as an entity and Guilford and Forsyth
Counties were modeled together. Therefore, calibration procedures des-
cribed in Section 3 discuss Guilford and Forsyth Counties jointly. Section
2 presents the air quality analyses separately for each AQMA. Section 2
presents the analyses in both isopleth and tabular form for the baseline
year 1973 and the projection years 1975, 1980, and 1985. Sections 3.1 and
3.2 present the final dispersion model calibration results with graphical
(Figures 22 and 23) and tabular (Tables 17 and 18) comparisons of calculated*
and observed TSP concentrations.
Although not a specific objective of this effort, several comparisons
of computed concentrations are contained in Section 3 which may be of
general interest to others involved in dispersion modeling analysis. In-
cluded are: AQDM results with the Briggs Plume Rise Equation (Ref. 4)
versus the Holland Plume Rise Equation (Ref. 3); the results of varying
throughout this report calculated emissions are referred to as "adjusted",
where the best fit linear regression coefficients have been applied, and
"unadjusted" where they have not been applied, i.e. effectively a slope of
1.0 and an intercept of 0.0.
-------
area source grid sizes.; the results of varying the plume heights for area
sources; and the contributions to receptor concentrations of several sub-
categories of area sources.
North Carolina is one of the first states, (mid-1975), to include
non-exhaust particulate emissions from roadway vehicles in large area dis-
persion modeling. Because these emissions represented such a significant
portion of the total emissions inventories, a great amount of time and
effort was expended in analyzing their effects.
-------
-------
2.0 Projected Air Quality Through 1985
The year 1973 was used as a base for the calibration of the AQDM
dispersion model as described in detail in Section 3 below. Projected
emissions inventories for Mecklenburg, Guilford, and Forsyth Counties for
both point and area source categories for the years 1975, 1980, and 1985
were provided by the NCAQS. The calibrations were made using mean annual
stability wind roses for 1973, while for all projected years, long term
(5 year) mean stability wind roses were used. The 1973 meteorological
data used with the AQDM for calibration for both the Charlotte AQMA
(Mecklenburg County) and the Greensboro-Winston-Sal em AQMA's (Guilford
and Forsyth Counties) were modified sets that included data only for those
days on which high-volume samples had been collected. These data sets are
designated "Lund Winds" in this discussion.* Their use was agreed upon to
make the wind data represent the actual monitoring days.
The reader is cautioned in interpreting the analyses for 1975, 1980,
and 1985 that projected data and long term average meteorological parameters
were used. For example, 1975 projected air quality can be expected to agree
with observed values only to the extent that 1975 meteorological conditions
were similar to the averages over the five-year period used and actual 1975
emissions agree with the emissions projected on the basis of expected
economic activity, emissions control and compliance schedules and other
factors. Thus, if the economic activity projected was too optimistic,
then projected emissions will be too high and projected air quality will
be worse than actually observed.
*These modified stability wind roses were prepared by Steven Lund of the Air
Quality Section, Division of Environmental Management, Department of Natural
and Economic Resources, State of North Carolina.
-------
2.1 Charlotte (Mecklenburg County! AQMA
The regression coefficients from the calibration* of the AQDM dispersion
model (see Figure 22) were used with the 1973 emissions inventory for
Mecklenburg County and 1973 meteorological data for Douglas Airport. The
emissions inventory included 97 point sources and area source emissions
allocated to 123 area source grid squares. Figure 1 presents the Mecklen-
burg County area source grid network (Ref. 1). A rectangular grid with 180
receptors with 4 km by 4 km spacing was used and 12 extra (non-grid) recep-
tors were added. AQDM calculated concentrations at each receptor were
plotted and analyzed. Figure 2 presents the AQDM receptor grid for
Mecklenburg County. Figure 3 presents an isopleth analysis for 1973.
Table 1 includes 1973 calculated concentrations for all receptors in
Mecklenburg County. Table 2 presents an analysis of the source contribu-
tion tables from AQDM for the five receptors with the highest calculated
concentrations. This table compares the contribution from point sources
and from area sources to the total concentration.
For 1975, 1980, and 1985 projected air quality, the same regression
coefficients were used as for 1973. The mean stability wind roses based
on the five (5) year period January 1, 1969 through December 31, 1973 at
the Douglas Airport were used. Emissions from both point and area sources
for each of the three projected years were input to the dispersion model.
Figures 4, 5, and 6 present the isopleth analyses of projected air quality
for 1975, 1980, and 1985, respectively. Table 1 includes the calculated
concentrations for all receptors in Mecklenburg County for each of the
projected years. Table 3 presents an analysis of the source contribution
*During the calibration of AQDM, the standard procedure is to calculate
concentrations at sampling site receptors only (Ref. 3).
-------
T,<5m -. —
•XQOO -
O9OU
,
I
V
l
r
*
t
V
V"
490
r
(.
1
\
1
/
xx'\
>
^
500
1
4
V
?
i
\
^
r
j
&
<
i
iji
\
(
>
j-^
J
t
V
j
S
r
•>,
'(j1
J
\
j
4
v-
5JC
\
>
\
N
•
\
\
\
i
w
/
-------
39?9 -
3921
3913
3905
D
o
3897
3889
3881
3873
494
502
510
5P6
518
UTM (EASTING)
Figure 2. Mecklenburg County N.C. AQDM Receptor Grid
544
-------
3873
494
502
510 518
UTM (EASTING)
526
Figure 3. Mecklenburg County N.C. 1973 (Calibrated) AODM Calculated Concentrations
of Total Suspended Particles (Micrograms/Cu. Meter)
-------
Table 1 . Mecklenburg County Receptor Concentration Data for TSP by Year
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
18
19
20
33
34
35
36
37
38
39
40
49
50
51
52
53
54
55
56
57
58
59
60
63
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
498.0 3881.0
498.0 3885.0
498.0 3889.0
502.0 3881.0
502.0 3885.0
502.0 3889.0
502.0 3893.0
502.0 3897.0
502.0 3901.0
502.0 3905.0
502.0 3909.0
506.0 3885.0
506.0 3889.0
506.0 3893.0
506.0 3897.0
506.0 3901.0
506.0 3905.0
506.0 3909.0
506.0 3913.0
506.0 3917.0
506.0 3921.0
506.0 3925.0
506.0 3929.0
510.0 3881.0
EXPECTED ANNUAL ARITHMETIC MEAN (MICROGRAMS/CU. METER)
1973
38.8
40.2
41.8
40.5
42.6
44.3
47.2
51.9
53.7
52.4
50.4
51.6
47.9
52.3
55.8
58.8
54.8
52.4
50.0
48.7
46.5
43.9
42.3
43.1
1975
40.8
42.4
43.9
43.0
45.3
47.0
49.7
52.0
51.1
48.3
45.8
54.7
51.3
55.6
58.1
57.1
51.4
47.8
44.9
42.8
41.3
40.1
39.2
47.1
1980
42.2
44.0
45.6
44.6
47.4
49.2
52.1
55.1
54.0
50.6
47.6
58.9
54.2
59.0
62.2
60.6
54.1
49.9
' 46.5
4^.3
42.5
41.2
40.1
49.5
1985
43.4
45.5
47.2 "
46.2
49.3
51.2
54.4
57.7
56.5
52.7
49.2
63.6
57.0
62.0
65.8
63.5
56.6
51.7
48.0
45.4
43.5
42.1
41.0
51.9
-------
Table 1 . (Continued)
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
64
65
66
67
68
69
70
71
72
73
74
75
77
78
79
80
81
82
83
84
85
86
87
88
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
510.0 3885.0
510.0 3889.0
510.0 3893.0
510.0 3897.0
510.0 3901.0
510.0 3905.0
510.0 3909.0
510.0 3913.0
510.0 3917.0
510.0 3921.0
510.0 3925.0
510.0 3929.0
514.0 3877.0
514.0 3881.0
514.0 3885.0
514.0 3889.0
514.0 3893.0
514.0 3897.0
514.0 3901.0
514.0 3905.0
514.0 3909.0
514.0 3913.0
514.0 3917.0
514.0 3921.0
EXPECTED ANNUAL ARITHMETIC MEAN (MICR06RAMS/CU. METER)
1973
47.2
52.0
62.1
65.2
63.9
58.0
52.1
50.4
47.5
. 45.6
44.5
42.7
41.3
43.7
47.5
54.1
62.9
73.0
69.1
60.4
54.2
50.7
47.1
45.5
1975
52.7
57.8
68.3
69.2
64.5
55.9
50.1
46.6
44.3
42.3
41.3
39.9
44.9
47.9
52.6
60.6
71.2
82.5
71.2
61.0
52.8
48.4
45.4
43.5
1980
56.1
61.9
75.1
75.2
69.8
59.8
52.6
48.5
45.9
43.7
42.5
40.9
46.9
50.3
55.7
64.9
77.1
92.1
77.3
65.9
55.8
50.6
47.1
45.0
1985
59.4
65.6
80.6
80.4
74.2
63.2
54.8
50.2
47.3
44.9
43.5
41.8
48.8
52.7
58.7
68.9
82.5
99.6
82.6
70.0
58.4
52.7
48.7
46.3
-------
Table 1 . (Continued)
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
89
90
92
93
94
95
96
97
98
99
100
101
102
103
104
105
107
108
109
110
111
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
514.0 3925.0
514.0 3929.0
518.0 3877.0
518.0 3881.0
518.0 3885.0
518.0 3889.0
518.0 3893.0
518.0 3897.0
518.0 3901.0
518.0 3905.0
518.0 3909.0
518.0 3913.0
518.0 3917.0
518.0 3921.0
518.0 3925.0
518.0 3929.0
522.0 3877.0
522.0 3881.0
522.0 3885.0
522.0 3889.0
522.0 3893.0
EXPECTED ANNUAL ARITHMETIC MEAN (MICR06RAMS/CU. METER)
1973
44.2
42.8
41.9
44.1
47.1
51.6
58.0
63.5
62.4
56.0
52.7
49.1
46.3
44.4
43.2
41.9
42.0
43.9
46.2
48.9
52.9
1975
42.1
40.7
44.6
47.3
51.1
56.7
64.5
71.4
68.3
58.4
53.2
48.4
45.4
43.2
41.9
40.6
43.7
46.0
49.0
52.6
57.7
1980
43.4
41.8
46.5
49.7
54.1
60.4
69.2
77.7
72.8
62.5
56.1
50.6
47.1
44.7
43.1
41.7
45.5
48.2
52.6
55.8
61.8
1985
44.5
42.8
48.4
52.0
57.1
64.1
73.8
83.3
77.9
66.1
58.6
52.6
48.7
46.1
44.3
42.8
47.2
50.3
54.2
58.9
65.7
-------
Table 1. (Continued)
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
112
113
114
115
116
117
118
123
124
125
126
127
128
129
130
131
139
140
141
142
143
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
522.0 3897.0
522.0 3901.0
522.0 3905.0
522.0 3909.0
522.0 3913.0
522.0 3917.0
522.0 3921.0
526.0 3881.0
526.0 3885.0
526.0 3889.0
526.0 3893.0
526.0 3897.0
526.0 3901.0
526.0 3905.0
526.0 3909.0
526.0 3913.0
530.0 3885.0
530.0 3839. 0
530.0 3893.0
530.0 3897.0
530.0 3901.0
EXPECTED ANNUAL ARITHMETIC MEAN (MICROGRAMS/CU. METER)
1973
53.6
51.9
49.8
48.0
45.9
44.2
42.6
42.8
44.2
45.6
46.5
46.7
45.8
44.7
43.9
. 43.3
42.0
42.2
42.6
42.6
41.9
1975
59.2
57.0
54.1
50.4
46.6
44.2
42.3
44.1
46.0
48.2
50.1
51.4
50.5
48.9
47.0
45.1
43.3
44.3
45.7
46.5
46.0
1980
63.2
60.9
57.7
53.3 .
48.8
45.9
43.7
46.0
48.1
50.8
52.8
54.2
53.2
51.2
49.3
47.1
45.0
46.2
47.7
48.6
48.1
1985
67.0
64.5
60.9
55.9
50.7
47.5
45.0
47.8
50.3
53.3
55.4
57.0
55.9
53.9
51.4
48.8
46.7
48.0
49.7
50.7
50.2
-------
Table 1 . (Continued)
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
144
155
156
157
158
171
172
181
182
183
184
185
186
187
188
189
190
191
192
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
530.0 3905.0
534.0 3889.0
534.0 3893.0
534.0 3897.0
534.0 3901.0
538.0 3893.0
538.0 3897.0
510.6 3897.8
516.1 3900-7
518.4 3892.3
513.3 3897.8
514.4 3928.1
509.2 3928.7
514.7 3895.0
513.1 3900.8
531.9 3892.5
522.5 3906.5
511.4 3880.3
501.9 3887.3
EXPECTED ANNUAL ARITHMETIC MEAN (MICROGRAMS/CU. METER)
1973
41.5
39.9
40.1
40.0
39.5
38.4
38.3
69.6
66.3
55.8
75.0
43.1
42.8
65.8
68.3
41.2
48.8
42.8
43.7
1975
45.3
41.6
42.7
43.3
43.1
40.8
41.3
73.9
69.9
61.9
82.3
41.0
39.7
74.3
70.7
43.9
52.5
46.9
46.1
1980
47.3
43.0
44.2
45.0
44.8
42.0
42.6
81.3
76.1
66.3
91.7
42.2
40.8
80.8
76.9
45.6
55.9
49.3
48.1
1985
49.2
44.5
45.7
46.5
46.4
43.2
43.9
87.4
81.4
70.7
99.3
43.2
41.7
86.6
82.1
47.3
58.9
51.5
50.1
-------
Table 2. 1973 Mecklenburg County Source Contributions to Five (5) Receptors
with the Highest Calculated Concentrations (From AQDM Source Contribution Tables)
Receptor Receptor
Number Name
184
82
181
83
188
Charlotte Community
Hospital*
82
Fire Station #10*
83
Beatties Ford Water
Plant*
Receptor Location
(Kilometers)
HORIZ VERT
513.3
514.0
510.6
514.0
513.1
3897.8
3897.0
3897.8
3901.0
3900.8
Point Source
Contribution
(Percent)
7.96
7.00
8.73
11.38
10.19
Expected
Area Source Total Concentration
Contribution at Receptor
(Percent) (Micrograms/Cu. Meter)
92.04
93.00
91.27
88.62
89.81
75.03
73.03
69.67
69.12
68.40
*See Figure 19 for relative location.
-------
3929 -
3881
3873
494
510
518
UTM (EASTING)
526
534
Figure 4. Mecklenburg County N.C. Projected 1975 Concentrations of Total
Suspended Particles (Micrograms/Cu. Meter)
16
-------
3929 -
3921 -
3913 -
5 3905
i—
at
f
3897 -
3889 -
3881
3873
494
510
518
UTM (EASTING)
526
534
Figure 5. Mecklenburg County N.C. Projected 1980 Concentrations of Total
Suspended Particles (Micrograms/Cu. Meter)
17
-------
3929 -
3921
3913
g 3905
' :'
D
• .
3897
3889
45
3881
3873
I I I I
494
502
510
518
UTM (EASTING)
526
534
Figure 6. Mecklenburg County N.C. Projected 1985 Concentrations of Total
Suspended Particles (Micrograms/Cu. Meter)
18
-------
Table 3. Mecklenburg County Source Contributions to Receptor with Highest Calculated Concentrations
(From AQDM Source Contribution Tables)
Year
1973
1975
1980
1985
Receptor
184*
82
82
82
Receptor Location
(Kilometers)
HORIZ VERT
513.3
514.0
514.0
514.0
3897.8
3897.0
3897.0
3897.0
Point Source
Contribution
(Percent)
7.96
3.36
3.34
3.40
Area Source
Contribution
(Percent)
92.04
96.64
96.66
96.60
Total Concentration
at Receptor
(Micrograms/Cu. Meter)
75.03
82.54
92.14
99.69
vo
*Charlotte Community Hospital, see Figure 19 for relative location.
-------
tables from AQDM for the receptor with the highest calculated concen-
trations for each of the projected years, specifically, the percentage
of the total receptor concentrations from point sources and from area
sources.
2.2 Greensboro (Guilford County) AQMA
The regression coefficients from the calibration*of the AQDM dispersion
model (see Figure 23) were used with the 1973 emissions inventory and the
1973 meteorological data for the Greensboro-High Point-Winston-Salem Airport.
Guilford and Forsyth Counties were combined for dispersion modeling purposes;t
after modeling, most of the analyses are presented for each county separately.
The inventory for the two-county area included 255 point sources and area
source emissions allocated to 253 area source grid squares. Figures 7 and 8
present the Guilford and Forsyth County area source grid network, respectively
(Ref. 1). A rectangular grid with 180 receptors with 5 km by 5 km spacing was
used and 12 extra (non-grid) receptors were added. Figures 9 and 10 present
the AQDM receptor grids for Guilford and Forsyth Counties, respectively.
AQDM-calculated concentrations at each receptor were plotted and analyzed.
Figure 11 presents the isopleth analysis for 1973. Table 4 includes 1973
calculated concentrations for all receptors located within Guilford County.
Table 5 includes the 1973 source contributions for all point sources and for
all area sources (for both Guilford and Forsyth sources) to the five receptors
with the highest calculated concentrations.
For 1975, 1980, and 1985 projected air quality, the same regression
coefficients were used as for 1973. The mean stability wind roses
based on the five (5) year period from January 1, 1968 through December 31,
*During the calibration of AQDM, the standard procedure is to calculate
concentrations at sampling site receptors only (Ref. 3).
This procedure is discussed in Section 3.
20
-------
4020
4010
4000
ce.
o
3990
398O
3970
590
600
620
630
UTM (EASTING)
Figure 7. Guilford County Area Source Grid Network
-------
4010
| 4000
r
\
3990
3980
'x
550
a
570
•v
I
I
I
f
I
]_„
L
58O
UTM (EASTING)
Figure 8. Forsyth County Area Source Grid Network
590
22
-------
90
4011
4001
88
8_Z_
86
3991
TOO
98
97
96
95
110
109
108
107
1QIF
105
120
119
J_LB_
117
116
115
130
129
128
127
126
T40
139
138
137
150
149
147
146
145
160
58
57
55
T70
169
168
167
166
165
179
178
177
176.
175
84
94
104
114
L24_
134
144
54
164
74
3981
83
Q3
113
23
133
143
53
163
173
3971
81
MIGHT))]
101
11
21
II
141
51
161
72
71
584
594
604 614
UTM (EASTING)
624
634
Figure Q. Guilford County N.C. AQDM Receptor Grid
-------
4021
4011
4001
I
c
3991
3981
3971
539
10
20
I
549
30
29
28
27
2fi
25
24
22
ML
39
38
37
36
50
49
48
47
35
34
33
32
31
M
43
42
60
59
58
SL
54
53
70
69
68
67
66
ffi_
^_
62.
80
79
78
77
Z£.
75
73
559 569
UTM (EASTING)
579
90
89
^.
SL
86
85
83
82
81
100
98
95
96
93
_|91
589
Figure 10. Forsyth County N.C. AQDM Receptor Grid
-------
4021
4011
4001
'-
z
. •
i
3991
3081
3971
S84
70
Greensooro-
HipffPoint-
ston-Sal
JL
_L
L
_L
594
624
604 614
UTM (EASTING)
Figure 11. Guilford County N.C. 1973 (Calibrated) AQDM Calculated
Concentrations of Total Susoended Particles (Micrograms/Cu. Meter)
634
25
-------
Table 4. GUIIford County Receptor Concentration Data for TSP by Year
rv>
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
92
93
94
95
96
97
98
99
102
103
104
105
106
107
108
109
112
113
114
115
116
117
118
119
122
123
124
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
589.0 3976.0
589.0 3981.0
589.0 3986.0
589.0 3991.0
589.0 3996.0
589.0 4001.0
589.0 4006.0
589.0 4011.0
594.0 3976.0
594.0 3981.0
594.0 3986.0
594.0 3991.0
594.0 3996.0
594.0 4001.0
594.0 4006.0
594.0 4011.0
599.0 3976.0
599.0 3981.0
599.0 3986.0
599.0 3991.0
599.0 3996.0
599.0 4001.0
599.0 4006.0
599.0 4011.0
604.0 3976.0
604.0 3981.0
604.0 3986.0
EXPECTED ANNUAL ARITHMETIC MEAN (MICROGRAMS/CU. METER)
1973
86.6
89.8
68.2
65.8
64.0
62.2
62.1
61.1
59.5
78.1
76.4
71.5
67.3
62.4
62.1
60.6
55.0
65.4
70.5
79.0
71.8
67.7
64.3
63.4
52.2
61.9
82.8
1975
103.5
94.2
70.6
65.6
61.6
57.9
55.9
53.2
74.6
89.1
76.5
70.0
64.1
58.0
56.5
53.6
66.0
74.0
74.9
79.1
67.7
62.5
58.4
55.7
62.1
71.3
93.4
1980
88.8
84.2
66.6
62.3
58.4
54.8
52.8
50.5
71.5
80.7
73.1
67.2
61.1
55.4
53.9
51.4
64.6
74.1
74.6
79.4
66.2
60.4
56.2
53.8
60.9
71.0
96.1
1985
88.8
82.6
65.8
61.5
57.3
53.4
51.3
49.3
72.0
81.0
73.5
67.5
60.7
54.8
53.0
50.8
65.3
76.5
76.4
82.2
66.9
60.2
55.9
53.7
61.4
72.4
100.6
-------
Table 4. (Continued)
ro
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
125
126
127
128
129
132
133
134
135
136
137
138
139
142
143
144
145
146
147
148
149
152
153
154
155
156
157
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
604.0 3991.0
604.0 3996.0
604.0 4001.0
604.0 4006.0
604.0 4011.0
609.0 3976.0
609.0 3981.0
609.0 3986.0
609.0 3991.0
609.0 3996.0
609.0 4001.0
609.0 4006.0
609.0 4011.0
614.0 3976.0
614.0 3981.0
614.0 3986.0
614.0 3991.0
614.0 3996.0
614.0 4001.0
614.0 4006.0
614.0 4011.0
619.0 3976.0
619.0 3981.0
619.0 3986.0
619.0 3991.0
619.0 3996.0
619.0 4001.0
EXPECTED ANNUAL ARITHMETIC MEAN (MICROGRAMS/CU. METER)
1973
93.3
87.5
75.2
69.7
65.4
49.1
57.4
70.2
108.2
99.6
85.6
76.3
69.6
45.6
52.3
62.7
77.5
89.7
83.1
76.6
70.3
43.5
49.3
55.2
64.3
68.9
71.3
1975
96.1
81.8
67.4
61.6
56.7
58.6
67.1
81.6
108.9
91.2
73.7
65.4
59.2
54.4
60.0
68.5
77.7
79.0
69.5
64.3
58.9
50.6
54.8
58.5
62.9
62.1
60.3
1980
98.4
83.4
66.9
60.4
55.3
57.2
66.2
82.9
114.5
93.5
74.0
64.8
58.2
52.9
58.6
67.8
77.8
78.9
69.5
63.7
58.2
49.0
53.1
57.1
61.8
60.7
58.9
1985
102.4
86.7
68.2
60.8
55.4
57.2
66.8
85.6
121.2
97.6
75.8
65.7
58.7
52.7
58.6
68.3
79.4
80.4
70.8
64.7
59.0
48.5
52.6
56.9
62.0
60.6
58.8
-------
Table 4. (Continued)
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
158
159
162
163
164
165
166
167
168
169
172
173
174
175
176
177
178
179
188
189
190
191
192
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
619.0 4006.0
619.0 4011.0
624.0 3976.0
624.0 3981.0
624.0 3986.0
624.0 3991.0
624.0 3996.0
624.0 4001.0
624.0 4006.0
624.0 4011.0
629.0 3976.0
629.0 3981.0
629.0 3986.0
629.0 3991.0
629.0 3996.0
629.0 4001.0
629.0 4006.0
629.0 4011.0
609.7 3989.1
598.1 3992.6
587.6 3977.4
589.6 3979.1
609.2 3992.8
EXPECTED ANNUAL ARITHMETIC MEAN (MICR06RAMS/CU. METER)
1973
71.3
68.3
41.1
45.9
51.8
58.2
61.8
62.5
64.7
63.7
39.2
43.8
48.6
53.8
56.1
56.7
58.5
58.9
92.9
75.2
75.1
113.9
106.0
1975
59.0
56.5
46.9
50.1
53.4
56.1
56.3
53.9
53.7
52.0
43.9
46.6
49.1
51.1
51.3
49.9
49.5
48.4
101.5
72.6
90.9
116.8
103.0
1980
57.8
55.5
45.4
48.4
51.8
54.8
54.5
52.2
52.1
50.7
42.5
45.1
47.6
49.7
49.8
48.2
47.8
46.8
106.7
71.4
78.3
100.3
106.9
1985
57.9
55.9
44.8
47.9
51.4
54.7
54.0
51.7
51.5
50.4
42.0
44.6
47.1
49.4
49.3
47.6
47.2
46.3
115.4
72.8
77.2
101.0
112.3
ro
00
-------
Table 5. 1973 Guilford/Forsyth Source Contributions to Five (5) Receptors
with the Highest Calculated Concentrations (From AQDM Source Contribution Tables)*
ro
to
Receptor
Number
191**
135
192*
136
125
Receptor Location
(Kilometers)
HORIZ VERT
589.6
609.0
609.2
609.0
604.0
3979.1
3991.0
3992.8
3996.0
3991.0
Point Source
Contribution
(Percent)
32.53
11.77
8.05
7.12
9.32
Area Source
Contribution
(Percent)
67.47
88.23
91.95
92.88
90.68
Total Concentration
at Receptor
(Micrograms/Cu. Meter)
113.95
108.26
106.04
99.72
93.37
* All Five of the Receptors are Located in Guilford County.
** S. Main, W. Green (see Figure 20 for relative location)
1-1 Davie Mebane (see Figure 20 for relative location)
-------
1972 at the Greensboro-High Point-Winston-Sal em Airport were used. Emission
inventories from both point and area sources for both Guilford and Forsyth
Counties for each of the three projected years were input to the dispersion
model. Figures 12, 13, and 14 present the isopleth analyses of projected air
quality for 1975, 1980, and 1985, respectively. Table 4 includes the cal-
culated concentrations for all receptors located within Guilford County for
each of the projected years. Table 6 presents an analysis of the source
contribution tables from AQDM for the receptor with the highest calculated
concentration for each of the projected years, specifically, the percentage
of the total receptor concentration from point sources and from area sources.
2.3 Winston-Sal em (Forsyth County) AQMA
As discussed in Section 2.2, Guilford and Forsyth Counties were
modeled together. Part of the detail of that section is repeated here for
completeness.
The regression coefficients from the calibration of the AQDM dispersion
model were used with the 1973 emissions inventory and the 1973 meteorological
data for the Greensboro-High Point-Winston-Sal em Airport. Guilford and
Forsyth Counties were combined for dispersion modeling purposes;* after
modeling, most of the analyses are presented for each county separately.
The emissions inventory for the two-county area included 255 point sources
and area sources allocated to 253 area source grid squares. Figures 7 and 8
present the Guilford and Forsyth area source grid squares, respectively (Ref. 1).
A rectangular grid with 180 receptors with 5 km by 5 km spacing was used and
12 extra (non-grid) receptors were added. Figures 9 and 10 present the AQDM
receptor grids for Guilford and Forsyth Counties, respectively. AQDM-calculated
*This procedure is discussed in Section 3.
30
-------
4021 r
4011
4001
|
3991
3981
Greensboro^,
High Poiirt-
3971
JL
L
584
594
Figure 12.
604 614
UTM (EASTING)
624
Guilford County N.C. Projected 1975 Concentrations of Total
Suspended Particles (Micrograms/Cu. Meter)
634
31
-------
ex
IN3
4011 -
4001
z: 3991
3981
3971
584
Greenshdro-
High/oint-
Wins^on-Salem
_L
_L
^_
J L
594
Fiaure 13.
604 614
UTM (EASTING)
624
634
Guilford County N.C. Projected 1980 Concentrations of Total
Susoended Particles (Micrograms/Cu. Meter)
-------
4021
4011
4001
3991
3981
45
3971
L
J L
J L
584
594
Figure 14.
604 614
UTM (EASTING)
624
Gui1 ford County N.C. Projected 1985 Concentrations of Total
Suspended Particles (Micrograms/Cu. Meter)
634
33
-------
Table 6 . Guilford County Source Contributions to Receptor with Highest Calculated Concentrations
(From both Guilford and Forsyth Sources)
Year
1973
1975
1980
1985
Receptor
Number
191*
191*
135
135
Receptor Location
(Kilometers)
HORIZ VERT
589.6
589.6
609.0
609.0
3979.1
3979.1
3991.0
3991.0
Point Source
Contribution
(Percent)
32.53
25.66
11.74
11.97
Area Source
Contribution
(Percent)
67.47
74.34
88.26
88.03
Total Concentration
at Receptor
(Micrograms/Cu. Meter)
113.95
116.95
114.60
121.28
CO
* S. Main, W. Green (see Figure 20 for relative location)
-------
concentrations at each receptor were plotted and analyzed. Figure 15
presents the isopleth analysis for 1973. Table 7 includes 1973 calculated
concentrations for all receptors located within Forsyth County. In 1973, 1975,
and 1980 none of the five receptors with the highest calculated concentrations
for the two-county area were located in Forsyth County. In 1985 receptor
184 (R. J. Reynolds) and receptor 181 (Government Center) were the receptors
with the fourth and fifth highest calculated concentrations, respectively.
Table 8 presents a comparison of the percentage contribution from total
point sources and total area sources to each of these Forsyth County
receptors.
For 1975, 1980, and 1985 projected air quality, the same regression
coefficients were used as those for 1973. The mean stability wind
roses based on the five (5) year period from January 1, 1968 through December
31, 1972 at the Greensboro-High Point-Winston-Salem Airport were used.
Emissions inventories from both point and area sources for both Guilford
and Forsyth Counties for each of the three projected years were input to
the dispersion model. Figures 16, 17, and 18 present the isopleth analyses
of projected air quality for 1975, 1980, and 1985, respectively. Table 7
includes the calculated concentrations for all receptors located in Forsyth
County for each of the projected years.
35
-------
4021 r
4011
4001
U
01
3991
Greensboro-
High Point-
Winston-Sal em
Airport
3981
3971
_L
._
_L
539
Figure 15.
519
559 569
UTM (EASTING)
579
589
Forsyth County N.C. 1973 (Calibrated) AQDM Calculated Concentrations
of Total Suspended Particles (Micrograms/Cu. Meter)
-------
Table 7 - Forsyth County Receptor Concentration Data for TSP by Year
CO
-j
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
15
16
24
25
26
27
28
29
34
35
36
37
38
39
44
45
46
47
48
49
54
55
56
57
58
59
65
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
549.0 3991.0
549.0 3996.0
554.0 3986.0
554.0 3991.0
554.0 3996.0
554.0 4001.0
554.0 4006.0
554.0 4011.0
559.0 3986.0
559.0 3991.0
559.0 3996.0
559.0 4001.0
559.0 4006.0
559.0 4011.0
564.0 3986.0
564.0 3991.0
564.0 3996.0
564.0 4001.0
564.0 4006.0
564.0 4011.0
569.0 3986.0
569.0 3991.0
569.0 3996.0
569.0 4001.0
569.0 4006.0
569.0 4011.0
574.0 3991.0
EXPECTED ANNUAL ARITHMETIC MEAN (MICROGRAMS/CU. METER)
1973
37.6
36.6
39.5
39.5
39.8
37.6
36.9
35.2
43.9
46.3
45.9
44.6
43.6
42.0
48.4
54.9
57.7
54.8
51.6
48.6
48.8
68.2
83.2
69.0
61.1
56.2
57.0
1975
41.3
39.8
45.6
44.8
43.8
40.8
39.4
36.9
51.7
53.7
50.6
47.7
45.1
42.2
57.0
64.4
62.7
56.6
51.6
46.9
58.2
72.9
82.5
64.6
57.1
51.9
63.7
1980
40.9
39.3
45.6
44.6
43.4
40.1
38.6
36.4
52.1
54.6
50.5
47.1
44.3
41.5
57.5
66.9
64.5
57.2
51.6
46.7
57.8
76.4
87.6
66.5
57.5
51.7
62.2
1985
41.1
39.4
46.1
45.1
43.5
40.0
38.4
36.3
53.2
56.2
51.2
47.2
44.2
41.4
58.7
70.0
66.9
58.4
52.3
47.2
58.3
80.7
93.3
69.0
53.5
52.2
61.8
-------
Table 7 . (Continued)
CO
CO
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
66
67
68
69
75
76
77
78
79
84
85
86
87
88
89
181
182
183
184
185
186
187
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
574.0 3996.0
574.0 4001.0
574.0 4006.0
574.0 4011.0
579.0 3991.0
579.0 3996.0
579.0 4001.0
579.0 4006.0
579.0 4011.0
584.0 3986.0
584.0 3991.0
584.0 3996.0
584.0 4001.0
584.0 4006.0
584.0 4011.0
568.0 3994.8
567.5 3995.1
568.0 3999.2
568.3 3995.1
563.3 4010.8
560.7 3993.1
576.7 4007.7
EXPECTED ANNUAL ARITHMETIC MEAN (MICROGRAMS/CU. METER)
1973
64.4
65.1
63.0
59.2
53.3
58.6
60.5
62.0
61.0
49.4
56.2
61.1
61.2
62.3
61.1
77.5
71.2
73.4
82.7
47.8
51.8
62.4
1975
65.9
61.2
57.5
53.2
59.2
60.3
58.0
56.4
53.8
58.1
60.7
60.7
57.8
56.1
53.2
86.0
77.6
71.0
88.9
46.4
58.9
56.4
1980
66.7
60.6
56.9
52.6
57.0
58.3
55.7
54.2
52.2
55.5
57.7
57.5
54.7
53.1
50.7
94.3
82.9
74.1
96.9
46.1
61.0
55.0
1985
68.5
61.0
57.2
52.9
56.0
57.6
54.7
53.3
51.5
54.6
56.5
56.1
53.1
51.6
49.5
102.9
88.7
77.8
105.3
46.4
63.8
54.5
-------
Table 8. Forsyth County: Source Contributions to the Two Forsyth County Receptors with the
Highest 1985 Calculated Concentrations
(From both Guilford and Forsyth County Emissions)
Receptor Location
/i/,-irtm04.QV,c\ Point Source Area Source
Receptor imometersj Contribution Contribution
Number HORIZ VERT (Percent) (Percent)
184* 568.3 3995.1 8.62 91.38
181** 568.0 3994.8 5.78 94.22
Total Concentration
at Receptor
(Micrograms/Cu. Meter)
105.3
102.9
* R. J. Reynolds (see Figure 21 for relative location)
** Government Center (see Figure 21 for relative location)
-------
4021
4011
4001
.pi
C
|
3991
3981
60
Greensboro-
High Point-
Winston-Salem
Airport
3971
539
549
579
Figure 16.
559 569
UTM (EASTING)
Forsyth County M.C. Projected 1975 Concentrations of Total Suspended
Particles (Micrograms/Cu. Meter)
589
-------
402 J
4011
4001
3991
3981
50
Greensboro-
High Point-
Winston-Salem
•
Airport
3971
I
539
549
559 569
UTM (EASTING)
579
Figure 17. Forsyth County N.C. Projected 1980 Concentrations of Total Suspended
Particles (Micrograms/Cu. Meter)
589
-------
4021 r
4011
4001
4*
PO
3991
3981
Greensboro-
High Point-
Winston-Sal em
Airport
3971
539
549
559 569
UTM (EASTING)
579
Figure 18. Forsyth County N.C. Projected 1985 Concentrations of Total Susoended
Particles (Micrograms/Cu. Meter)
589
-------
3.0 Calibration Procedures
The AQDM as modified to use the Briggs Plume Rise Equation instead of
the Holland Plume Rise Equation was used for the dispersion modeling of
emitted particulate matter and air quality for the North Carolina AQMAs.
The first step in the modeling process was to prepare base year (1973)
emissions data (from both point sources and area sources), meteorological
data, and TSP concentrations data at monitoring (sampling) sites for input
to the AQDM to calibrate the model for each area in the study.
The Charlotte (Mecklenburg County) AQMA was modeled as a unit, but the
Greensboro (Guilford County) AQMA and the Winston-Sal em (Forsyth County)
AQMA were modeled together. The common boundary of these latter AQMAs and
the availability of only one appropriate set of meteorological data were
determining factors in the decision.
Mecklenburg County calibrated on the first attempt using stability
wind rose data (Lund Winds) for Douglas Airport. Point source (93* sources)
and area source (123 grid squares) emissions data (1973) were input to the
dispersion model. Annual arithmetic average concentrations for each of the
12 Mecklenburg County monitoring sites for 1973 were used for calibration.
Figure 19 presents the locations of the sites. As will be discussed in
detail later, several emissions inventory corrections/modifications and
further analyses of the representativeness of certain monitoring stations
required additional calibration runs for Mecklenburg County.
Guilford and Forsyth Counties did not calibrate on the first attempt.
Annual stability wind rose data (Lund Winds) for Greensboro-High Point-
Winston-Salem Airport, emissions data for 255 point sources and 250 area
*During later calibration runs, an additional source (stack) was added at one
location and three (3) Rowan County sources were added: the final number of
point sources used for calibration was 97.
43
-------
Davidson Pumping ation
Davidson Filer Plant
North 29
Patrol Station
Beatties Ford
Fire Station
Charlotte Community Hospital
County Health Department
Fire Station
#10
Fire Station
Carpenter Airport
McAlpine Creek
Sewage Plant
Sampling sites used in final calibration
Jfc Sampling sites rejected for final calibration (Ref. 2)
Figure 19. Locations of Sampling Sites in Mecklenburg County
44
-------
source grid squares, and 1973 annual arithmetic average concentrations
at thirteen 03)' Guilford and Forsyth County monitoring sites were used for
calibration. Figures 20 and 21 present the locations of sites in Guilford
and Forsyth Counties, respectively. In a technical review meeting at RTI,
we discussed the emissions inventory and AQDM source-receptor contribution
tables with NCAQS project personnel and noted the overwhelming contributions
(*9Q% of the total emissions inventory) of the area sources and particularly
the dominance of fugitive dust emissions from vehicles. During the analysis,
a systematic error was discovered in the apportioning of unpaved road emissions.
The area source emissions inventory was corrected and another AQDM calibration
run was made. This run resulted in a "successful calibration". However,
as will be discussed in detail later, corrections to the emissions inventory,
further analyses of representativeness of monitoring stations, and an analysis
of the AQDM source-receptor contribution tables caused additional AQDM
calibration runs to be made for the Guilford and Forsyth County areas.
The additional calibration runs were made for the AQMAs to include several
changes in the emissions inventory, and to investigate new approaches and
techniques (applied to one of the areas but with general, or overall,
applications). The following narrative is a description of the procedures
and approaches used; the ramifications of whether the techniques applied
in one AQMA solved a problem peculiar to that AQMA or had more general
applicability were at all times considered.
The need for additional calibration runs evolved as we attempted to
resolve what were, in the judgement of the participants in the N.C. AQMA
analysis, unacceptable calibration results or because changes were made in
the emissions inventory or other data. In the case of the Guilford/
45
-------
Rockingham Co.
: )
P
Swing Road
Davie Mebane
City Garage
(B
< j
1 ':
S. Main & West Green
ational Guard Armory
Randolph Co.
Sampling Sites used in Final calibration
Sampling Sites rejected for final calibration (Ref. 2)
Figure 20. Locations of Sampling Sites in Guilford County
46
-------
Stokes Co.
Rural Hall
County Courthouse
Agricultural Building
South Forks
Clemmons
Walkertown
Hanes Hosiery Park
R. J. Reynolds
Government Center
Davidson Co.
Sampling sites used in final calibration
Samnling sites rejected for final calibration (Ref. 2)
Figure 21. Locations of Sampling Sites in Forsyth County
o
!
i i
i '
47
-------
Forsyth-areas, we were having problems in obtaining statistically significant
regression parameters and in the case of Mecklenburg County we were looking
for "better" regression parameters and studying the contributions of the
various source categories. Of the two areas, Mecklenburg County, being
smaller geographically and with a smaller number of total sources to model,
was selected to systematically study the contributions of the various source
categories. Several comparisons of the effect of changing input values to
AQDM on calculated concentrations are presented. Where appropriate, a table
of AQDM calculated concentrations at monitoring sites is presented. In some
cases the comparison is not one-for-one because a correction to the emissions
inventory may have also been made — these are duly noted in the discussion.
The (single copy) computer printouts are archived with the NCAQS. It
should also be noted that as the emissions inventories were modified, the
total emissions, and hence the percentage of the total emissions for a
particular area source category or a subset of point sources changed; the
relative effect, however, is still useful when studying the interrelation-
ships between the source contributions, the concentrations as measured at
the monitoring sites, and the meteorological parameters. In more than
one case, when the concensus of opinion of the group of participants in
the N.C. AQMA study was that certain changes in the input data would
produce certain effects, when the changes were incorporated and the AQDM
was re-run, the modeling results simply did not follow the predictions.
The objective of reporting these comparisons is twofold; first, that
the results presented in Section 2 above are the result of these detailed
analyses of the AQMAs, and secondly, that others may benefit from these
48
-------
dispersion modeling efforts.
No attempt is made, nor is it considered pertinent, to present AQDM
modeling results in a chronological order. However, the very nature of the
calibration process and the one-by-one testing of various hypotheses suggest
that a step-by-step (hence chronological) presentation is significant.
Subsequent sections discuss the effects of systematically varying these
AQDM input values: area source plume heights, size of the area source grid
squares, and the magnitude of area source non-exhaust TSP emission factors
for the vehicle miles traveled (VMT) on paved and unpaved roads. In
addition, the effect of using the Holland Plume Rise Equation rather than
the Briggs Plume Rise Equation was tested as was also the use of National
Weather Service (NWS) stability wind roses, based on all 365 days in 1973
in contrast to wind data for only those days when sampling was done, i.e.
the Lund Wind roses.
In an attempt to gain insight into the contribution of the several
source categories (data which is not directly available from the AQDM outputs)
RTI modeled Mecklenburg County using a 12 x 12 receptor grid with 1 km
spacing over the urban area (Charlotte). The southwest corner of the receptor
grid was located at the UTM coordinates of 510.0 km easting and 3893.0 km
northing. Twelve (12) additional receptors were specified; some of these
receptors, however, were near the geographic limits of the county and were
not representative, especially when wind directions were from neighboring
counties. (The Mecklenburg emissions inventory does not include area sources
<
outside the county.) It should be noted that the 12 x 12 receptor grid
(1 km spacing) was located near the center of the county, and therefore,
the (unadjusted) calculated concentrations from the several source category
49
-------
combinations are, in our opinion, juore representative at these receptors
than at the geographically outlying ones.
The 1973 emissions inventory of point sources was separated into two
subsets of data, (13) large point sources (those equal to or greater than
50 tons per year), and (81) other point sources (those less than 50 tons per
year). The area source emissions, apportioned to 123 grid squares, were
input to AQDM in five combinations: 1) all categories, 2) non-exhaust
emissions from unpaved roads only, 3) non-exhaust emissions from paved roads
only, 4) road vehicle exhaust emissions only, and 5) other area source
categories, i.e., l-(2+3+4) above. The 1973 Lund Winds, a slope of 1.0 and
an intercept of 0.0 were input to AQDM. Table 9 presents the (unadjusted)
calculated minimum and maximum concentrations at the grid receptors for the
various combinations of source categories. Table 10 contains the observed
and (unadjusted) calculated concentrations at the twelve (12) extra receptors
for the combinations of sources as described above. As noted above, these
AQDM runs were made before the final (corrected) area source emissions
inventory had been compiled.
Attempts were made to quantitatively estimate the magnitude of
contributions from emissions not included in the inventory available for
the areas being modeled. Study of the stability wind roses make it obvious
that treating all emissions from neighboring counties as contributing to
a constant, county-wide background concentration is not realistic and that
actual contributions probably are responsive to wind direction frequencies
and the particular sources upwind of each monitoring site. However, 1973
base year emissions data for neighboring counties were not available and
the concensus of opinion was that time and manpower required to obtain
50
-------
Table 9. Comparison of the Ranges of (Unadjusted)
Calculated TSP Emissions for Mecklenburg County
by Source Category (1973 Annual Arithmetic
Average,
SOURCE CATEGORY
LOWEST
CONCENTRATION ON
RECEPTOR GRID
HIGHEST
CONCENTRATION ON
RECEPTOR GRID
1. Point Sources 1 50 tons/yr.
2. Point Sources < 50 tons/yr.
3. Total Area Sources
4. Fugitive Dust from Unpaved
Roads Only
5. Fugitive Dust from Paved
Roads Only
6. Road Vehicle Exhaust Only
7. Other Area Sources
[3-(4+5+6)]
1.34
0.44
13.07
5.88
3.69
1.79
1.42
4.95
6.21
25.97
8.50
10.33
5.00
4.74
51
-------
Table 10. Comparison of Observed and (Unadjusted) Calculated Concentrations of TSP (by Source Category)
for Mecklenburg County (1973 Annual Arithmetic Average) m-3
CJl
Receptor
Fire Station 110
Fire Station #11
Fire Station #14
Charlotte Community Hospital
Davidson Filter Plant
Davidson Pump Station
Mecklenburg County Health Dept.
Beatties Ford Water Plant
Mint Hill
North 29 Patrol Station
McAlpine Creek Sewage Plant
Carpenter Airport
UTM Coordinates
Horizontal Vertical
510.6
516.1
518.4
513.3
514.4
509.2
514.7
513.1
531.9
522.5
511.4
501.9
3897.8
3900.7
3892.3
3897.8
3928.1
3928.7
3895.0
3900.8
3892.5
3906.5
3880.3
3887.3
Point Point
Sources Sources
>50t/yr <50t/yr
Only Onlv
2.12
3.22
1.42
2.20
2.30
2.58
1.75
2.55
0.79
1.34
1.42
2.18
1.50
1.87
0.60
1.61
0.30
0.25
0.88
1.56
0.17
0.40
0.38
0.48
Total Unpaved
Area Roads
Source Only
21.93
19.82
14.97
24.97
7.13
6.66
20.01
21.12
7.09
11.51
7.22
7.58
6.49
6.63
5.93
6.39
4.24
4.01
6.20
6.59
4.37
5.73
3.82
4.30
Paved Road Veh-
Roads icle Ex-
On1y haust Only
8.17
6.86
4.90
9.50
1.42
1.34
7.47
7.66
1.41
3.00
1.77
1.67
3.98
3.33
2.38
4.62
0.69
0.65
3.62
3.72
0.69
1.46
0.86
0.82
Other All
Area Categories
Source (Point & Area)
3.28
3.02
1.76
4.48
0.79
0.66
2.72
3.15
0.63
1.32
0.78
0.78
25.55
24.91
16.99
28.78
9.73
9.49
22.64
25.23
8.05
13.25
9.02
10.24
1973
Observed
71.
66.
51.
79.
49.
35.
59.
108.
42.
58.
39.
32.
-------
such data was beyond the scope of the N.C.AQMA study. Crude approaches to
apportioning total area source and point source emissions in neighboring
counties were abandoned because the geographic positioning of the sources
of emissions was so sensitive to the wind direction frequencies and the
orientation and proximity of receptors to these outlying sources. The
projection of any values used for 1973 into future years would also have
been a formidable task, again far beyond the manpower and time available.
In reviewing the source contribution tables and investigating the very
low concentrations calculated by the AQDM at monitoring sites within
Mecklenburg County, but near the county border, it became apparent that any
regression analysis using those sampling stations would be biased because
emissions from a large wind direction sector are not included in the inventory;
in one case (Carpenter Airport, in the southwest corner of Mecklenburg County)
the station was downwind to emissions included in the inventory for only
approximately 70 degrees of the compass. By a concensus of opinion, three
(3) sampling sites -Carpenter Airport, Davidson Filter Plant, and Davidson
Pump Station — were removed from further consideration and were not used in
any subsequent calibration attempts.
At this time revised area source emissions data for Guilford and Forsyth
Counties were received by RTI, and another calibration run was made for those
counties. In reviewing the AQDM run outputs, it appeared that the Walkertown
monitoring site (rural) was an outlyer in the data array relating observed
and calculated air quality. This monitoring station was located interior
to a 100 km grid square (#19) containing over 1450 tons/year of (area
source) partlculate emissions; over 95% of the 1450 tons/year was attributed
to fugitive dust from vehicles traveling on unpaved roads. To reduce the
53
-------
effect of treating area sources as virtual point sources by AQDM and to
provide more resolution for emissions from unpaved roads, the large (100
2 2
km ) grid square was subdivided into four (4) smaller (25 km ) grid squares
(numbered 19a, 19b, 19c, and 19d)* and the 1450 tons/year of particulate
emissions were then reallocated to these smaller grid squares to better
represent the infulence on the Walkertown monitoring site. Based on VMT
analysis, RTI was instructed to allocate the unpaved road emissions into the
four subdivided grid squares (as compared to the large grid square) in
the following percentages: NE, 38.6%; SE, 23.2%; SW, 22.0%; and NW, 16.2%.
The area source emissions from other source categories were evenly allocated
to the four smaller grid squares. The AQDM was again run and the (unadjusted)
3
calculated concentration at Walkertown changed less than 0.1 yg/m , i.e.
3 2
it rounded to 27 yg/m for both the large grid square (100 km ) and the
2
four (4) smaller grid squares (25 km ) cases. This was the only change
made in the total emissions inventory between the two runs.
It had also been suggested during a technical review with EPA and
NCAQS personnel that our dispersion modeling problems with fugitive dust
from paved and unpaved roads might result from the fact that emission
factors in the literature are not representative of N.C. Questions were
also raised regarding the filtering effect of vegetation and the
appropriateness of a 10-ft. effective plume height for area sources in N.C.
(because vehicular activity dominated the area source emissions totals). By
modeling only fugitive dust emissions from unpaved roads at more than one
effective plume height, and then comparing the results, we could thus
evaluate the effect of plume height.
*Subsequently, from this point in time, all runs for Guilford and Forsyth
Counties were made with 253 area source grid squares.
54
-------
Because we already had on hand the area source emissions, by category,
for Mecklenburg County, the "height of plume" runs were made with Mecklenburg
emissions from unpaved roads. The AQDM input deck for unpaved road emissions
was duplicated three (3) times using 3 meters, 10 meters, and 20 meters
effective plume heights. Three (3) AQDM runs were then made using a slope
of 1.0, and intercept of 0.0, and an 8 x 8 km receptor grid spanning the
entire county. Table 11 presents the resulting concentrations calculated
at the locations of each of the monitoring sites; also given in the table
are the minimum and maximum values calculated anywhere on the rectangular
receptor grid. The changes in effective plume height obviously had very
little impact on the calculated concentrations —generally less than 0.1
yg/m3.
Before evaluating the effect of possible vegetation filtering, two
other questions posed during technical review discussions were resolved:
1) Assuming a correct emissions inventory, might we have erred in selecting
the AQDM, as modified to use the Briggs Plume Rise Equation, and therefore
be using a model that was under-predicting? (AQDM using the Holland Plume
Rise Equation reportedly over-predicts [Ref. 4].) 2) Were the Lund Winds
causing a latent problem, that is, how would the use of National Weather
Service winds based on all 365 days in 1973 compare with the use of Lund
Winds?
The AQDM with the Holland Plume Rise Equation was run with the same
emissions inventory and stability wind rose tables as for a previous Briggs
Plume Rise Equation run for Mecklenburg County. The calculated concentrations
at monitoring stations bear out higher (unadjusted) calculated concentrations
using the Holland Plume Rise Equation. The calibration (regression) equation
55
-------
Table 11. Comparison of Effect of Effective Plume Height Changes
on Calculated Concentrations of TSP for Mecklenburg County Using 1973
Unpaved Road Emissions Only (pg/m3, Regression Slope 1.0, Intercept 0.0)
UTM COORDINATES
EFFECTIVE PLUME HEIGHT (METERS)
en
RECEPTOR
Fire Station #10
Fire Station #11
Fire Station #14
Charlotte Community Hospital
Davidson Filter Plant
Davidson Pump Station
Mecklenburg County Health Dept.
Beatties Ford Water Plant
Mint Hill
North 29 Patrol Station
McAlpine Creek Sewage Plant
Carpenter Airport
Minimum (on Receptor Grid)
Maximum (on Receptor Grid)
Horizontal
510.6
516.1
518.4
513.3
514.4
509.2
514.7
513.1
531.9
522.5
511.4
501.9
Vertical
3897.8
3900.7
3892.3
3897.8
3928.1
3928.7
3895.0
3900.8
3892.5
3906.5
3880.3
3887.3
3.0
6.5
6.6
5.9
6.4
4.2
4.0
6.2
6.6
4.4
5.7
3.8
4.3
1.8
7.5
10.0
6.5
6.6
5.9
6.4
4.2
4.0
6.2
6.6
4.4
5.7
3.8
4.3
1.8
7.4
20.0
6.5
6.6
5.9
6.4
4.2
4.0
6.2
6.6
4.3
5.7
3.8
4.3
1.8
7.2
-------
using the Holland Plume Rise Equation had a lower slope, a slightly higher
intercept, but a slightly smaller correlation coefficient. At the monitoring
stations, the comparison between (unadjusted) calculated concentrations, for
3
the two plume rise equations ranged from "no change" to 3 pg/m higher for
the Holland equation. Table 12 is a comparison of the (unadjusted) calculated
concentrations at the monitoring stations and the observed concentrations.
Table 13 presents a comparison of the (adjusted) calculated concentrations
for each receptor on the grid and for twelve (12) extra receptors using, in
turn, the Briggs Plume Rise Equation and the Holland Plume Rise Equation
versions of the AQDM. The regression equations used (from calibration) were:
Briggs, £ = 1.6126X + 27.5 (computed regression coefficient = 0.936, 5%
confidence level = 0.707); Holland, £ = 1.4395X + 28.1 (computed regression
coefficient = 0.933, 5% confidence level = 0.707). It should be noted that
emissions inventory data were modified subsequent to these comparison runs,
therefore, 1973 (adjusted) calculated emissions differ from those in Table 1
in Section 2; not withstanding, the comparison of the results of using the
two plume rise equations is considered of enough interest to report.
During early calibration attempts with the Guilford and Forsyth areas,
parallel AQDM runs were made where the only difference in the runs was the
stability wind rose tables used; one run was made using the 1973 Lund Winds data,
and a second run using the 1973 Greensboro-High Point-Winston-Sal em Airport
stability wind rose data from the National Weather Service (NWS) where all
365 days of 1973 were included. The results of the two runs did not indicate
significant differences. This suggests that for 1973 the 61-days on which
hi-vol samples were obtained were representative days, from a meteorological
standpoint, of the entire year. This may be true for any year. However,
57
-------
Table 12. Comparison of (Unadjusted) Calculated Concentrations and Observed
Concentrations of JSP at Sampling Sites in Mecklenburg County
Using AQDM with the Briggs Plume Rise Equation and the Holland Plume
Rise Equation (1973 Annual Arithmetic Average Concentration, yg/nr)
UTM COORDINATES
Ol
00
SAMPLING SITE
Fire Station #10
Mecklenburg County Health Dept.
Fire Station #11
Fire Station #14
Mint Hill
Charlotte Community Hospital
North 29 Patrol Station
McAlpine Creek Sewage Plant
HORIZONTAL
510.6
514.7
516.1
518.4
531.9
513.3
522.5
511.4
VERTICAL
3897.8
3895.0
3900.7
3892.3
3892.5
3897.8
3906.5
3880.3
BRIGGS
26
23
25
17
8
29
14
9
HOLLAND
29
25
28
19
9
32
14
10
OBSERVED
71
59
66
51
42
79
58
39
-------
Table 13. Comparison of Mecklenburg County Receptor Concentration Data for
1973 Using AQDM with Briggs Plume Rise Equation and Holland Plume
Rise Equation
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
18
19
20
33
34
35
36
37
38
39
40
49
50
51
52
53
54
55
56
57
58
59
60
61
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
498.0 3881.0
498.0 3885.0
498.0 3889.0
502.0 3881.0
502.0 3885.0
502.0 3889.0
502.0 3893.0
502.0 3897.0
502.0 3901.0
502.0 3905.0
502.0 3909.0
506.0 3885.0
506.0 3889.0
506.0 3893.0
506.0 3897.0
506.0 3901.0
506.0 3905.0
506.0 3909.0
506.0 3913.0
506.0 3917.0
506.0 3921.0
506.0 3925.0
506.0 3929.0
510.0 3881.0
(ADJUSTED) CALCULATED ANNUAL
ARITHMETIC MEAN
(MICR06RAMS/CU. METER)
BRIGGS
37.4
39.8
42.1
39.6
42.9
45.2
48.5
54.2
56.2
55.5
54.4
59.0
49.5
54.3
56.9
61.9
57.9
56.7
54.6
53.5
50.2
46.0
43.4
37.6
HOLLAND
38.6
41.5
45.1
40.5
43.9
47.4
51.3
59.2
60.7
60.4
65.6
58.5
50.7
55.9
58.6
64.4
62.3
64.1
63.8
63.0
55.9
49.3
46.0
38.2
59
-------
Table 13. (Continued)
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
64
65
66
67
68
69
70
71
72
73
74
75
77
78
79
80
81
82
83
84
85
86
87
88
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
510.0 3885.0
510.0 3889.0
510.0 3893.0
510.0 3897.0
510.0 3901.0
510.0 3905.0
510.0 3909.0
510.0 3913.0
510.0 3917.0
510.0 3921.0
510.0 3925.0
510.0 3929.0
514.0 3877.0
514.0 3881.0
514.0 3885.0
514.0 3889.0
514.0 3903.0
514.0 3897.0
514.0 3901.0
514.0 3905.0
514.0 3909.0
514.0 3913.0
514.0 3917.0
514.0 3921.0
(ADJUSTED) CALCULATED ANNUAL
ARITHMETIC MEAN
(MICROGRAMS/CU. METER)
BRIGGS
48.6
53.0
61.8
65.7
65.1
61.0
54.4
54.2
50.3
48.0
46.7
44.1
40.4
43.2
47.2
53.6
61.6
71.3
70.2
63.2
57.7
54.2
49.3
47.6
HOLLAND
49.4
53.1
62.0
65.6
66.5
62.9
56.1
56.8
52.9
50.9
59.2
46.4
40.9
43.6
47.5
53.5
61.3
70.5
70.1
63.1
57.7
54.7
50.0
49.0
60
-------
Table 13. (Continued)
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
89
90
92
93
94
95
96
97
98
99
100
101
102
103
104
105
107
108
109
110
111
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
514.0 3925.0
514.0 3929.0
518.0 3877.0
518.0 3881.0
518.0 3885.0
518.0 3889.0
518.0 3893.0
518.0 3897.0
518.0 3901.0
518.0 3905.0
518.0 3909.0
518.0 3913.0
518.0 3917.0
518.0 3921.0
518.0 3925.0
518.0 3929.0
522.0 3877.0
522.0 3881.0
522.0 3885.0
522.0 3889.0
522.0 3893.0
(ADJUSTED) CALCULATED ANNUAL
ARITHMETIC MEAN
(MICROGRAMS/CU. METER)
BRIGGS
45.9
44.0
40.8
43.3
46.4
51.4
57.4
62.8
63.6
57.8
55.7
51.6
47.9
45.5
44.2
42.4
40.7
43.1
45.7
48.9
53.0
HOLLAND
47.4
45.6
41.2
43.6
46.6
51.3
57.1
61.9
62.4
57.0
55.1
51.5
48.1
46.0
45.0
43.3
41.0
43.2
45.7
48.7
52.6
61
-------
Table 13. (Continued)
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
112
113
114
115
116
117
118
123
124
125
126
127
128
129
130
131
139
140
141
142
143
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
522.0 3897.0
522.0 3901.0
522.0 3905.0
522.0 3909.0
522.0 3913.0
522.0 3917.0
522.0 3921.0
526.0 3881.0
526.0 3885.0
526.0 3889.0
526.0 3893.0
526.0 3897.0
526.0 3901.0
526.0 3905.0
526.0 3909.0
526.0 3913.0
530.0 3885.0
530.0 3889.0
530.0 3893.0
530.0 3897.0
530.0 3901.0
(ADJUSTED) CALCULATED ANNUAL
ARITHMETIC MEAN
(MICR06RAMS/CU. METER)
BRIGGS
54.0
52.4
50.1
48.5
46.7
44.7
42.7
41.8
43.5
45.4
47.0
47.6
46.3
44.8
44.0
43.5
41.1
41.6
42.6
42.6
41.5
HOLLAND
53.4
51.8
49.5
48.2
46.6
44.8
43.0
42.0
43.8
45.4
46.8
47.3
46.0
44.5
43.8
43.4
41.3
41.8
42.7
42.6
41.5
62
-------
Table 13. (Continued)
RECEPTOR CONCENTRATION DATA
RECEPTOR
NUMBER
144
155
156
157
158
171
172
181
182
183
184
185
186
187
188
189
190
191
192
RECEPTOR LOCATION
(KILOMETERS)
HORIZ VERT
530.0 3905.0
534.0 3889.0
534.0 3893.0
534.0 3897.0
534.0 3901.0
538.0 3893.0
538.0 3897.0
510.6 3897.8
516.1 3900.7
518.4 3892.3
513.3 3897.8
514.4 3928.1
509.2 3928.7
514.7 3895.0
513.1 3900.8
531.9 3892.5
522.5 3906.5
511.4 3880.3
501.9 3887.3
(ADJUSTED) CALCULATED ANNUAL
ARITHMETIC MEAN
(MICR06RAMS/CU. METER)
BRIGGS
41.1
38.9
39.4
39.2
38.4
37.1
36.9
69.4
68.3
55.4
74.6
44.3
44.2
64.6
69.0
40.8
49.3
42.5
44.6
HOLLAND
41.0
39.1
39.6
39.4
38.6
37.4
37.2
69.5
68.5
55.1
74.7
45.8
46.6
64.4
68.8
41.0
48.9
43.0
46.1
63
-------
because of the reasonableness of the Lund concept of using wind data only
for those days on which hi-vol samples were obtained, the NC AQMA study
group decided to proceed with the use of the Lund winds for the AQMA analysis.
Another question which arose during a joint technical meeting was
whether modeling of the Forsyth and Guilford County areas might be biased
against the "down-wind" county considering that emissions were not also
included for other neighboring counties. After much discussion, it was
decided to separate the emissions inventory into two subsets, one
containing only Forsyth County emissions (both point source and area source)
and the other containing only those emissions in Guilford County. Likewise,
the Forsyth County monitoring stations were paired with the Forsyth County
emissions and the Guilford County monitoring stations were paired with the
Guilford County emissions. The Greensboro-High Point-Winston-Sal em Airport
meteorological data (Lund Winds) were used for both counties. Separating the set
of sampling sites obviously would cause the required correlation (5%
confidence level) to be higher because of the smaller number of sampling
sites for each AQDM calibration run. All sampling sites for each county
were used — nine (9) in Forsyth County and five (5) in Guilford
County. The entire two-county receptor grid was used for both AQDM runs
and twelve (12) extra receptors were included; these were located at
sampling sites in both counties for each of the AQDM runs. It was reasoned
that even though some sampling sites had been tentatively considered as not
representative from earlier analyses, we should have AQDM calculate an
unadjusted concentration for each, and by using a desk-top programmable
calculator, one could evaluate regression equations using any subset of
sampling sites.
64
-------
Neither AQDM run, however, produced an acceptable calibration. The
contribution of Guilford County sources in both counties, and likewise, the
contribution of Forsyth County sources to receptors in both counties, are
presented in Table 14. The concentrations in parentheses are
contributions to the other county's receptors (at sampling sites). The
calculated values are unadjusted; further, the observed 1973 annual
arithmetic average concentrations are presented for comparison.
The dominance of fugitive dust emissions from vehicle activity on
unpaved roads in the emissions inventories, and the debate over whether the
emission factor used was valid for N.C. roads led to an investigation of
reducing these emissions by a constant percentage and using the AQDM to
evaluate the impact on calculated concentrations. With the emissions
inventories in hand at that time, the unpaved roads emissions represented
a much higher contribution to total emissions than most sources and in the
cases of Guilford and Forsyth Counties were the primary source category.
Table 15 presents the 1973 particulate emissions totals for paved roads,
unpaved roads, total area sources, point sources, total for all sources and
the percentage of the total emissions contributed by each of these source
categories.
It was decided to prepare the AQDM input data for two (2) runs in which
point sources emissions would be included but in one run one-half of the
emissions from unpaved roads would be removed from each area source grid
square and in the second run all of the emissions from unpaved roads would
be removed from each area source grid square. To limit computer time, a
3x3 receptor grid was used with twelve (12) extra receptors (all located
at sampling sites), the AQDN calibrate option was used with the eight (8)
65
-------
Table 14. Comparison of (Unadjusted) Calculated and Observed
Concentrations of TSP from Forsyth County and
Gui1 ford County Emission Sources when Modeled
Separately with the AQDM (1973 Annual
Arithmetic Averages, pg/nr*)
UTM COORDINATES
RECEPTOR
County Courthouse
Government Center
R. J. Reynolds
Agricultural Bldg.
*
Hanes Hosiery Park
Cl emmons
Rural Hall*
South Forks
*
Mai kertown
City Garage
Davie Mebane
Swing Road
South Main-
West Green
National Guard
Armory*
HORIZ
568.0
568.0
568.3
567.5
568.0
556.0
563.3
560.7
576.7
609.7
609.2
598.1
589.6
587.6
FORSYTH
VERT SOURCES ONLY
3994.9
3994.8
3995.1
3995.1
3999.2
3986.3
4010.8
3993.1
4007.7
3989.1
3992.8
3992.6
3979.1
3977.4
29
29
34
26
31
11
17
16
24
(7)f
( 8)
(10)
( 7)
( 7)
GUILFORD
SOURCES ONLY
-
(5)*
( 5)
( 5)
( 4)
-
( 3)
( 5)
( 5)
39
43
30
55
31
OBSERVED
95
84
91
86
87
74
49
63
39
95
100
53
95
65
Eight (8) sampling sites used in most calibration attempts.
Parentheses indicate contribution from other county.
66
-------
TabTe 15. Comparison of Contribution of
Source Categories to Total Emissions
(1973 Tons/Year and Percentages)
EMISSIONS
SOURCE CATEGORIES
Total Emissions (Tons/Yr.)
Point Sources (Tons/Yr.)
Percentage of Total Emissions
Area Sources {Tons/Yr.)
Percentage of Total Emissions
Unpaved Roads (Tons/Yr.)
Percentage of Area Source
Emissions
Percentage of Total Emissions
Paved Roads (Tons/Yr.)
Percentage of Area Source
Emissions
Percentage of Total Emissions
MECKLENBURG
47670.2
35286.0
74.0
12384.2
26.0
7116.5
57.5
14.9
2689.5
21.7
5.6
FORSYTH
20144.3
1746.3
8.7
18398.0
91.3
15665.0
85.2
77.8
1439.7
7.8
7.2
GUILFORD
32076.8
3088.1
9.6
28988.7
90.4
24580.8
84.8
76.6
2182.9
7.5
6.8
67
-------
monitoring stations in Forsyth and Guilford Counties which were being used
for calibration attempts.
The Guilford/Forsyth area was selected because of the percentages that
unpaved road emissions represented of the total emissions (see Table 15).
Table 16 is a comparison of the (unadjusted) calculated and observed concen-
trations at calibration monitoring stations and the calculated regression
parameters for the three (3) runs (the original one plus the two as described
above). The range of contributions to the (unadjusted) calculated emissions
at the eight (8) monitoring stations, of emissions from unpaved roads is
from 32.4% to 72.2%. As would be expected, with the removal of such a
significant percentage of the total emissions in the inventory, the regression
slope and intercept increased, but so did the correlation, approaching but
not reaching the 5% confidence level (see Table 16). For any given monitoring
station the effect of removal of 0%, 50%, and 100% of the emissions from
unpaved roads is linear although the percentage reduction across the several
monitoring stations varied. It was therefore possible to use a programmable
desk-top calculator to test the impact of removing any percentage of emissions from
unpaved roads and to calculate the resulting slope, intercept, and correlation.
Several percentage reductions were tested including 25% and 75% with no
success. It was then obvious that pursuing the appropriate unpaved roads
emission factor, considering vegetation filtering and N.C. soil types would
not, in itself, solve the Guilford and Forsyth calibration problem; hence,
that approach was abandoned.
RTI was then requested to test the effect of removing emissions from
both paved and unpaved roads for a more "conventional" AQDM run, i.e., is
the fugitive dust from roadways amenable to dispersion modeling,
specifically with the AQDM. The requested run was made and the resulting
68
-------
Table 16. Comparison of (Unadjusted) Calculated and Observed Concentrations at Calibration Monitoring Stations
in Guilford and Forsyth Counties and Calculated Regression Parameters with: 1) None Removed,
2) One-half Removed, and 3) Total Removal of Fugitive Dust from Unpaved Roads
from the Emissions Inventory {1973 Annual Arithmetic Averages,
{AQDM Run 1)
UJH Unpaved
Coordinates Emissions
Receptor HORIZ. VERT Unchanged
Agricultural Bldg. 567.5
Hanes Hosiery Park 568.0
Rural Hall 563.3
South Forks 560.7
Ualkertown 576.7
City Garage 609.7
Swing Road 598.1
National Guard 587.6
Armory
Regression Parameters:
AQDM Run 1
AQDM Run 2
AQDM Run 3
3995. 1 29
3999.2 33
4010.8 18
3993.1 19
4007.7 27
3989.1 43
3992.6 36
3977.4 34
Slope
1.2690
1.5966
1.9453 *
(AQDM Run 2) (AQOH Run 3) Contribution to Total Contribution to Total
Removal of 1/2 Removal of All Unadjusted Calculated Unadjusted Calculated
Unpaved Roads Unpaved Roads Concentrations of Concentrations of
Emissions from Emissions from Unpaved Road Emissions Unpaved Road Emissions
Each Area Source Each Area Source , Observed
Grid Square Grid Square (wg/m ) (X) Concentrations
22
27
12
13
18
33
27
29
Intercept
29.0
31.1
38.6
15
20
5
7
9
23
16
23
Correlation
0.534
0.609
0.684
14 48.3
13 39.4
13 72.2
12 63.2
18 66.7
20 46.5
20 55.6
11 32.4
5% Confidence Level
.707
.707
.707
96
87
49
63
39
95
53
65
-------
intercept was near 50, the slope near 1.7, and the correlation coefficient
was less than the correlation coefficient obtained after the "unpaved roads"
only had been eliminated.
The N.C. AQMA study group also considered forcing an intercept for the
Guilford/Forsyth area and least squares fitting a slope to the calculated-
observed concentration data, however, with a poor correlation, that approach
did not appear to have merit. The concensus of the group was that we should
not abandon attempts to calibrate the AQDM for the Guilford/Forsyth area
as long as a valid alternative remained. In reviewing, during a joint
technical meeting, all calibration attempts to date, the discussion again
centered on the dispersion modeling efforts producing apparently adequate,
or even high, calculated concentrations in the non-urban areas and
significantly low concentrations in the urban areas. The source of, and
confidence in, the emissions factor for paved road emissions was discussed
and especially its validity for N.C. The group whose task it was to update
and project the area source emissions (Ref. 1) and the NCAQS personnel
studied further the emission factor used. The paved road emission factor
originally used was determined to be too low and RTI was provided modified
area source emissions reflecting the changes for Guilford, Forsyth, and
Mecklenburg Counties. Using these modified area source emissions, the AQDM
calibrated successfully for both the Mecklenburg and Guilford/Forsyth
modeling areas.
3.1 Charlotte (Mecklenburg County) AQMA, Final Calibration
The final Mecklenburg AQDM calibration run calculated a regression
equation with a slope of 0.8194, an intercept of 31.6, a computed
correlation of 0.928 and a 5% confidence correlation of 0.707. Table 17
70
-------
Table 17. Final AQDM Calibration Run Results for Mecklenburg County
(1973 Annual Arithmetic Averages,
TSP
Sampling Site
Fire Station 10
Mecklenburg County
Health Department
Fire Station 11
Fire Station 14
Mint Hill
Charlotte Community
Hospital
North 29 Patrol
Station
McAlpine Sewage
Plant
Location
(KILOMETERS)
SAROAD* NO. HORIZ VERT
34-0700-010 510.6 3897.8
34-0700-011 514.7 3895.0
34-0700-003 516.1 3900.7
34-0700-004 518.4 3892.3
34-2580-001 531.9 3892.5
34-0700-002 513.3 3897.8
34-0700-008 522.5 3906.5
34-0700-014 511.4 3880.3
TSP Concentration
(MICROGRAMS/CU. METER)
OBSERVED CALCULATED CALCULATED
(Unadjusted) (Adjusted)
71
$9
66
51
42
79
58
39
46
42
42
30
12
53
21
14
70
66
66
56
41
75
49
43
*Storage and Retrieval of Aerometric Data (SAROAD)
-------
is a presentation of the TSP sampling site receptors, their locations and
SAROAD* number, and the observed, (unadjusted) calculated, and (adjusted)
calculated concentrations. Figure 22 is a graph of the AQDM (unadjusted)
calculated concentrations (X axis), observed concentrations (Y axis), and
the least squares best fit regression line (equation).
3.2 Greensboro and Winston-Salem (Guilford County and Forsyth County) AQMAs,
Final Calibration
The final Guilford/Forsyth AQDM calibration run calculated a regression
equation with a slope of 0.8204, an intercept of 23.1, a computed correlation
of 0.705 and a 5% confidence correlation of 0.632. Table 18 is a
presentation of the TSP sampling site receptors, their locations and SAROAD*
number, and the observed, (unadjusted) calculated, and (adjusted) calculated
concentrations. Figure 23 is a graph of the AQDM (unadjusted) calculated
concentrations (X axis), observed concentrations (Y axis), and the least
squares best fit regression line (equation).
*Storage and Retrieval of Aerometric Data (SAROAD)
72
-------
100
90
80
Charlotte Comnunlty
Hospital
F1re Station
No. 10
70
F1re Station
No. 11
60
North 29
Patrol Station
O
Mecklenburg County
Health Department
i 50
o>
u
i
8
s
40
O
Fire Station
No. 14
Hint Hill
O'
HcAlplne Sewage
Plant
30
20
10
X - 0.8194X + 31.6
r - 0.928
(5% confidence level • 0.707)
10
20
Figure 22.
30
40
50
60
70
80
90
Calculated Concentrations (wg/m )
Mecklenburg Final Calibration, Total Suspended Particles
1973 Annual Arithmetic Average
73
-------
Table 18. Final AQDM Calibration Run Results for Guilford and Forsyth Counties
(1973 Annual Arithmetic Averages,
Location
TSP
Sampling Site
Government Center
Agricultural Building
Hanes Hosiery Park
R. J. Reynolds
Rural Hall
South Forks
Wai kertown
City Garage
Swing Road
National Guard Armory
SAROAD* NO.
34-4460-008
34-4460-002
34-4460-009
34-4460-003
34-1480-002
34-4460-005
34-1480-001
34-1740-002
34-1780-010
34-2000-003
(KILOMETERS)
HORIZ VERT
568.0
567.5
568.0
568.3
563.3
560.7
576.7
609.7
598.1
587.6
3994.8
3995.1
3999.2
3995.1
4010.8
3993.1
4007.7
3989.1
3992.6
3977.4
TSP Concentration
(MICROGRAMS/CU. METER)
OBSERVED CALCULATED CALCULATED
(Unadjusted) (Adjusted)
84
86
87
91
49
63
39
95
53
65
66
59
61
73
30
35
48
87
64
63
77
71
73
83
48
52
62
95
75
75
*Storage and Retrieval of Aerometric Data (SAROAD)
-------
100
90
80
70
60
50
01
I
40
30
20
10
R. J.
Hanes Hosiery
Park
Agricultural
Building
Reynolds
O
0
City
Garage
s
Government
Center
South Forks
O
Rural Hall
National Guard
Armory
O
Swing Road
O
Ualkertown
x • 0.8204X + 23.1
r « 0.705
(5% confidence level = 0.632)
I
I
10
20
Figure 23.
30
50
Calculated Concentrations (ug/m )
60
70
80
90
Rullford/Forsyth Final Calibration, Total Suspended Particles
1973 Annual Arithmetic Average
75
-------
REFERENCES
1. North Carolina Air Quality Maintenance Area Analysis, Volume I: TSP Area
Source Emissions Inventories for Charlotte, Winston-Salem, and Greensboro
AQMAs for 1973, 1975, 1980, and 1985, (April 1976).
2. North Carolina Air Quality Maintenance Area Analysis, Volume II: TSP Point
Source Emissions Inventories, TSP Monitoring Data, and Meteorological Data
for Charlotte, Winston-Sal em, and Greensboro AQMAs for 1973, 1975, 1980, and
1985, (April 1976).
3. TRW Systems Group, 1969: Air Quality Display Model. Contract No. PH 22-68-
60 USDHEW, Public Health Service, National Air Pollution Control Administration.
Washington, D.C.
4. Busse, A. D., and J. R. Zimmerman: User's Guide for the Climatological
Dispersion Model, EPA-R4-73-024, December, 1973.
76
-------
BIBLIOGRAPHIC DATA
SHEET
1. Report No. •
EPA 90V9-76-005c
4. Title and Subtitle North Carolina Air Quality Maintenance Area Anal-
ysis, Volume III: TSP Dispersion Modeling and Analysis for
Charlotte, Winston-Salem, and Greensboro AQMA's for 1973, 1975»
3. Recipient's Accession No.
S. Report Date
April 1976
6.
..1.Q8C
7. Author(s)
Richard C,
Havs and Harry L. Hamilton, Jr.
8. Performing Organization Kept.
NO.
9. Performing Organization Name and Address
Research Triangle Institute
Research Triangle Park, N.C. 27709
10. Projcct/Task/Work Unit No.
15
11. Contract/Grant No.
68-02-1386
12. Sponsoring Organization Name and Address Air & Hazardous Materials DlV. ,
KPA, Region IV, 1»+21 Peachtree St., N.E. Atlanta, GA 30309 AND
Air Quality Section, Division of Environmental Management, N.C.
Department of Natural and Economic Resources, P. 0. Box 27687,
Raleigh. N.C. 27611 •
13. Type of Report &
Covered
14.
Supplementary Notes
•- ..u.,r»CI, This report gives the results of a study designed to advance air quality main-
tenance planning for North Carolina on the basis of area and point source «J»^JJ-
ventories supplied respectively by Engineering-Science, Inc. and theN. C.Air Quality
Section. The study involved calibration of the Air Quality Display Model to detect
data errors and to test hypotheses on emissions and their air quality impact. Final
dispersion model calibration results and both graphical and tabular J0^1^8 ?t ?£'
culated and observed TSP concentrations are presented. Mecklenburg County vas modeled
separately; Guilford and Forsyth Counties vere modeled together A unique feature of
this study stems from the fact that North Carolina was one of the first to include non
exhaust particulate emissions from roadway vehicles in total large area -n *
modeling; because these emissions represented such a significant part c
a great amount of time and effort vas expended to find the best vay to deal with
7- Key Words and Document Analysis.
Air Pollution
Statistical Inference
Suspended Particulates
17o. Descriptors
Identifiets/Open-Ended Terms
7«. COSAT1 Field/Group
• Availability Statement
Unlinited release
19.. Security Class (This
Report)
u ^\t_jif *lr *-"!..--••**** ••-•
20. Security Class (This
Pa1TNC.l.ASSlFlF.D
21. No. of Pages
85
22. Price
NTis-39 i«ev. io-7si ENDORSED BY ANSI AND UNESCO.
USCOMM-OC 82SS-P74
------- |