-------
Table 3.2 Temporal Allocation Factors for the Filling of Auto-
mobile Gasoline Tanks.
Hour
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Weekdays
6.45 x 10"5
5.82 x 1Q-5
1.62 x 10"5
3.23 x 10"6
3.23 x lO'5
1.62 x 10-5
9.70 x TO"5
2.20 x ID'4
1.94 x TO'4
1.29 x ID'4
1.29 x TO'4
1.46 x 10'4
1.46 x ID'4
1.46 x TO'4
1.52 x TO'4
2.43 x TO'4
2.72 x ID'4
2.85 x TO'4
2.10 x TO'4
2.07 x 10'4
1.62 x 10"4
1.23 x TO'4
9.70 x 10-5
Saturday
9.70 x 10~5
9.70 x 10~f
7.12 x 10"5
2.59 x ID'5
2.59 x 10-5
3.23 x 10-5
4.53 x 1Q-5
6.79 x lO"5
9.70 x ID'5
1.07 x lO'4
1.23 x TO'4
1.29 x 10"4
1.29 x TO'4
1.16 x ID"4
1.13 x 10'4
1.13 x 10'4
1.20 x 10-4
1.10 x 10-J
9.06 x 10"5
8.73 x 10"5
9.70 x 10"5
1.10 x 10'4
1.10 x 10'4
Sunday
9.70 x IQ-jj
8.09 x TO"5
5.82 x TO'5
1.62 x ID'5
1.29 x ID'5
1.62 x 10-5
1.94 x 10'5
2.59 x TO'5
3.23 x 10-5
3.88 x 10-5
5.18 x 10"b
6.47 x 10"5
6.47 x 10"5
7.76 x 10'5
7.76 x TO"5
7.44 x 10'5
7.12 x TO'5
7.12 x ID'5
2.47 x 10-5
6.14 x 10"b
5.82 x 10"5.
5.82 x 1Q-5
5.18 x 10"b
34
-------
variation in seasonal driving habits is necessary. Figure 3.2 shows the
monthly gasoline sales for the state of Missouri in 1973 and 1974. Similar
data is available for Illinois. As shown, there are marked differences
between summer and winter gasoline sales. The monthly sales factor shown
in the figure is based upon the average of the 1973 and 1974 values and
the relative fraction of sales for a composite average month.
The temporal allocation is then:
(from 8:00 a.m. to 5:00 p.m.) (EQN 3.1)
Evaporative HC_ -, *
Emissions = JTAQ [annual dry cleaning emissions +
annual surface coating emissions +
(0.5 x annual gasoline marketing
emissions x monthly sales factor)]+
(automobile allocation factor x
0.5 x annual gasoline marketing
emissions x monthly sales factor)
(from 5:00 p.m. to 8:00 a.m.) (EQN 3.2)
Evaporative HC_ automobile allocation factor x 0.5 annual
Emissions = gasoline marketing emissions x monthly
sales factor
* This number is based on a workday from 8:00 a.m. to 5:00 p.m., five
days a week and 52 weeks per year.
35
-------
260--
250--
240--
2-Year Average
(106 Gal.)
Monthly Sales
Factor
3 AN
FEB
MAR
APR MAY
JUNE JULY
AUG
SEP
OCT
NOV
DEC
Figure 3.2 State of Missouri Gasoline Sales.
"36"
-------
SOLID WASTE DISPOSAL
Emissions from solid waste disposal through open burning and incinera-
tion were considered in this study. Conversations with the St. Louis
6
County Air Pollution Control Board and with the Missouri Air Conservation
Commission indicated that open burning was banned from the populous areas
of AQCR 70. It is allowed by law only where no public or commercial
refuse collection service is available and in places where population
density is less than 100 dwelling units or less per square mile. The
emissions from open burning are considered insignificant in this study.
Information sources on solid waste incineration included a nationwide
study and the emission inventory prepared by St. Louis County. The latter
source of information appeared to be the most reliable as it included
actual emissions data other than nationwide statistics. This study indi-
cated that residential incineration was negligible. According to the
Pollution Control Regulations, for the St. Louis Metropolitan Area, only
multiple chamber incinerators may be used and must not exceed 0.3 grains
of particulate per dry standard cubic foot of exhaust gas. Since the
emission standards are stringent and the cost of multiple chambered units
high, residential incineration emissions were considered insignificant in
this study. The only significant source of incineration pertinent to
this study was the commercial-institutional area source category.
Treating the St. Louis County figures as being representative of the
entire AQCR, Table 4.1 summarizes the emissions estimates from solid
waste disposal by county.
The spatial and temporal allocation procedures are similar to those
for emissions from the dry cleaning process; that is, in proportion to
the commercial-institutional land use in each grid square, and an 8:00 a.m.
to 5:00 p.m. workday.
37
-------
Table 4.1 Solid Waste Commercial-Institutional Incineration
Emissions (tons/year) in AQCR 70 for 1973.
County
Missouri
Franklin
Jefferson
St. Charles
St. Louis
St. Louis City
Illinois
Bond
Clinton
Madison
Monroe
Randolph
St. Clair
Washington
Part.
3.21
6.05
5.47
50.00
29.86
0.74
1.52
13.30
0.95
1.63
15.09
0.74
SOX
0.38
0.73
0.66
6.00
3.58
0.09
0.18
1.60
0.11
0.20
1.81
0.09
NOX
0.58
1.09
0.98
9.00
5.38
0.13
0.27
2.40
0.17
0.29
2.72
0.13
HC
2.57
4.84
4.37
40.00
23.89
0.59
1.22
10.68
0.76
1.30
12.07
0.59
CO
5.13
9.67
8.75
80.00
47.78
1.18
2.44
21.37
1.51
2.61
24.14
1.18
38
-------
STRUCTURAL FIRES
Data was not available for wildfires and forest fires. The available
data on structural fires is such that a high degree of reliability cannot
be placed on the emissions estimates. The available data included:
1) the number of fires and dollar damage on a monthly basis for Illinois^
2) statewide statistics on the total number of fires (10,000) doing
9
$45,500,000 worth of damage in 1974; and 3) nationwide statistics indi-
cating that 40 percent of the typical structure is consumed in a fire and
that the average structure contains approximately 17 tons of combustible
10
material.
For Missouri the statewide figures were disaggregated to the county
level on the basis of population. Using the emission factors in AP-42 for
open burning of municipal solid waste, the emission estimates were derived.
These values are summarized in Table 5.1
The spatial apportionment is most appropriately conducted on the
basis of the number of housing units per tract:
Grid = no. of housing units in grid x County
Emissions no. of housing units in county Emissions
The temporal distribution of these fires is random. The apportion-
ment figures can be utilized on an annual basis as a reasonable approxima-
tion. However, for the purpose of the calibration of dispersion models,
the only appropriate approach is for the modeler to be aware of whether or
not a structural fire has occurred that may affect short-term model
results. The fire can be treated as a ground level point source. An
estimate of emissions can be included in the grid area based upon 40 per-
cent of the 17 tons of combustible material being consumed in a four hour
period using the emission factors for incinerator without controls from AP-42.
39
-------
Table 5.1 Estimated Annual 1973 Emissions (tons/year) from
Structural Fires in AQCR 70.
County
Illinois
Bond
Clinton
Madison
Monroe
Randolph
St. Clair
Washington
Missouri
Franklin
Jefferson
St. Charles
St. Louis
St. Louis City
Part.
0.7
2.1
49.7
2.2
3.8
56.2
1.1
7.0
13.2
12.0
109.3
65.3
sox
<0.1
0.1
3.1
0.1
0.2
3.5
0.1
0.4
0.8
0.8
6.8
4.1
CO
3.8
11.0
264.2
11.6
19.9
298.5
6.1
37.3
70.2
63.6
580.9
347.1
HC
1.3
3.9
93.2
4.1
7.0
105.4
2.1
13.2
24.8
22.4
205.0
122.5
NOX
0.3
0.8
18.7
0.8
1.4
21.1
0.4
2.6
5.0
4.5
41.0
24.5
40
-------
SUMMARY
General
The Regional Air Pollution Study has unique requirements of its
component emission inventories. Generally it requires degrees of spatial
and temporal resolution that have not been previously achieved. This
document brings together data on fuel usage, land use, sales of gasoline
and paints, the use of dry cleaning fluids, solid waste disposal and
uncontrolled fires for the St. Louis region and develops methodologies
for estimating the pollutant emissions from stationary residential and
commercial-institutional area sources on an hour-by-hour basis for com-
ponents of a spatial grid system developed for RAPS.
The methodologies presented in this document are a series of sub-
elements integrated into a single system for deriving the required
emission data. This system seeks to provide the best emission estimates
possible from the available data. In order to provide the RAPS with as
much flexibility as possible to meet the multiple and varied demands upon
it, the inventory is presented as direct statements of weight of pollutant
emitted by this class of source as a function of location for every hour.
Specific Results
Space Heating - The emissions from space heating were based upon the
emission factors utilized by the National Emissions Data System and from
AP-42 for each fuel. The distribution within the RAPS grid system was
determined by allocating total county fuel use for residential, and, in
a separate calculation, commercial-institutional fuel usage to each grid
area in proportion to the number of units using that fuel in the grid
area. The temporal fuel use variation and variation with meteorological
41
-------
parameters was established by statistically analyzing the detailed data
available on gas use. This analysis identified a base component of
usage which is largely independent of meteorological influence and has
a distinct relationship with the time of day. The remaining component was
shown to be strongly affected by the ambient temperature and wind velocity.
Evaporative Emissions
The major components of residential and commercial-institutional area
sources of evaporated hydrocarbons were identified as surface coating
(primarily painting), gasoline handling, and dry cleaning. Characteristic
emissions on a per capita basis were established and the emissions deter-
mined for each grid area by allocating the total projected county emissions
in proportion to commercial land use in the grid area for gasoline handling
and dry cleaning, and in proportion to population for surface coating.
Structural Fires and Solid Waste Disposal
Emissions from structural fires were difficult to quantify; however,
average data was used and total estimated county emissions were allocated
to grid areas based on the number of homes in the grid. However, it was
realized that significant structural fires would be best handled as a point
source.
Emissions from the disposal of solid wastes is largely restricted to
commercial-institutional enterprises and large incinerators because of the
restrictive air pollution regulations. Thus the NEDS county emissions were
allocated to grid areas in proportion to the commercial-institutional land
use in the grids.
42
-------
REFERENCES
1. Personal communication between Mr. Don McQueen of Environmental
Science and Engineering, Inc. and representatives of the LaClede
Gas Company, May, 1975.
2. Ryan, Richard R., Gas Sendout Forecasting with Weather Sensitive
Loads, LaClede Gas Company, May 1974.
3. Hydrocarbon Emission Sources in the Metropolitan Boston Intrastate
Air Quality Control Region, Vol. 1, prepared for the U.S. Environ-
mental Protection Agency by GCA/Technology Division, May, 1974.
4. Personal communication between Mr. Robert Holden of Environmental
Science and Engineering, Inc. and Mr. Ray Conner of the National
Paint Coating Association, May, 1975.
5. Personal communication between Mr. Don McQueen of Environmental
Science and Engineering, Inc. and the Mobil Oil Distributors,
St. Louis, Missouri, May, 1975.
6. Personal communication between Mr. Robert Holden of Environmental
Science and Engineering, Inc. and Mr. Michael L. Farley, Engineer,
Air Pollution Control Branch, Division of Environmental Health
Care Services, March, 1975.
7. Personal communication between Mr. Robert Holden of Environmental
Science and Engineering, Inc. and Mr. Hoffman of the Missouri Air
Conservation Commission, May, 1975.
8. Illinois State Fire Marshal's Office, May, 1975.
9. Personal communication between Mr. Pete Burnette of Environmental
Science and Engineering, Inc. and the Missouri State Fire Marshal,
May, 1975.
10. Derived from data provided by the National Fire Protection Asso-
ciation of Boston, Massachusetts.
-------
APPENDIX A
Description of the Annual Emission Inventory Calculation
-------
Emissions from residential space heating were determined by allocating
total county residential fuel usage to each grid base on the number of
homes in the grid using a given fuel type:
SHR
1;j,k
,
NHCf.j
Equations and Input Parameters
Where: SHR-j^-^ = residential space heating emissions
of pollutant k_
from fuel type j_
in grid number i_
NHGFU-j s j = number of homes in grid i using
fuel type j
NHCjjf = number of homes using fuel type j
in the county f in which grid is located
TCFUR-; f = total residential usage of fuel type j
sJ 3 '
in the county f in which grid is located
EFj}k = emission factor (AP-42) for fuel
type j and pollutant k.
Emissions from commercial fuel usage was determined by allocating
total commercial fuel usage in a given county to each grid based on
commercial land use within a grid and the ratio of the number of homes
using a given fuel type in a grid to the total number of homes within
that grid.
x NHC, f
J > '
A-l
-------
Where: COM-jj^ = emissions from commercial fuel usage
of pollutant k_
from fuel type j_
in grid number j_
CLUG-j = commercial land use (km2) in grid i
CLUCf = commercial land use (km2) in the county
in which grid is located
NHGj = total number of homes in grid i
TCFUCj,f = total commerical usage of fuel type j
in the county
= is defined above
^ f
J >T
NHCT = total number of homes in county f
Evaporative hydrocarbon emissions for the three processes indicated
below were determined for each grid by allocating total county emissions
from the three processes on the basis shown.
Process Allocation Basis
1) Surface Coating Grid Population
2) Gasoline Handling Commercial Land Use
3) Dry Cleaning Commercial Land Use
GHCEij = CHCE1>f x
GHCE-}}2 = CHCE2,f x CLUGi
GHCE,.,3 - CHCE3,f
A-2
-------
Where: GHCE..- -• = evaporative hydrocarbon emissions
1 »J
in grid i from process j
CHCEj f = total county f emissions from process j
GPOP-j = population of grid i
CPOPf = population of county in which grid is located
Emissions from structural fires were determined by allocating total
county emissions to each grid based on the number of homes in the grid.
Where: EMSF-j ^ = emissions from structural fires
of pollutant j< in grid i_
ESFC^ f = total county emissions from structural
fires of pollutant k from county in
which grid is located
NHCTf = total number of homes in county
Emissions from the disposal of solid wastes was determined by
allocating total county emissions to each grid based on commercial land use
within that grid.
ESWi i< = ESWCk f x CLUGi
1,K K,T
Where: ESW-j^ = emissions from solid waste disposal
of pollutant J£ in grid j_
A-3
-------
total county emissions from solid waste
disposal of pollutant k for county in
which grid is located.
A-4
-------
APPENDIX B
Sample Calculation
-------
I. EMISSION FACTORS
Fuel Type
Bituminous Coal (tons/ton)
Domestic
Commercial
Residual Oil (tons/103 gal)
Commercial
Distillate Oil (tons/103 gal)
Domestic
Commercial
Natural Gas (tons/106cu. ft.)
Domestic
Commercial
Bottled (LP) Gas (tons/103gal)
Domestic
Part
SO,
Pollutant
CO
HC
NO.
0.0100 0.00475 0.0450 0.0100 0.0015
0.0140 0.00475 0.0036 0.0010 0.0046
0.0115 0.1590 0.0020 0.0015 0.0300
0.0050 0.0144 0.0025 0.0015 0.0060
0.0075 0.0144 0.0020 0.0015 0.0300
0.0050 0.0003 0.0100 0.0040 0.0400
0.0050 0.0003 0.0100 0.0040 0.0600
0.00095 7.2xlO"6 0.0010 0.0004 0.0400
B-l
-------
II.
COUNTY AND GRID ANNUAL DATA
A. Attributes of County No. 2:*
Population
Commercial Land Use
Number of Homes Using Fuel Type 1 Oil
2 NG
3 LPG
4 Coal
Total Residential Usage of
Fuel Type
(Table 2.1) (TCFUR)
Total Commercial Usage of
Fuel Type
(Table 2.1) (TCFUC)
Total Evaporative Hydrocarbon
Emissions
(Table 3.1) (CHCE)
Total County Emissions from
Structural Fires Pollutant
(Table 5.1) (ESFC)
Total County Emissions
Solid Waste Disposal
Pollutant
(Table 4.1) (ESWC)
from
Total Number of Occupied Homes
in the County
(NHCT)
1
2
3
4
Dist-35,000
Residents
2
3
4
1
2
3
1
2
3
4
5
1
2
3
4
5
= 215479
Part
S09
or
HC
NO.
CPOP 568100
CLUC 11.965 km2
NHC 16949
186093
2523
3375
13560
26400
16619
18560
0
11100
0
112000
872
3078
151
65 Tons
4
347
123
25
30 Tons
4
48
24
*A11 units are standard NEDS units (emissions as tons/year, fuel oil as
103 gal/year, natural gas as 10° cu ft/year, LPG as 103 gal/year, and
coal as tons/year.
B-2
-------
B. Attributes for Grid No. 895, located in the City of St. Louis
which is treated as County No. 2:
Population GPOP 5096
Area 1.0 km
Commercial Land Use CLUG 0.15 km2
Number of Homes NHG 1547
Homes Using Fuel Type 1 NHGFU 139
2 1302
3 26
4 41
B-3
-------
III. GRID EMISSION CALCULATIONS FOR POLLUTANT NO. 4 (HYDROCARBONS)
A. Residential Fuel Consumption
1. Fuel Type 1, Fuel Oil
x TCFUR(l) x EF = _139 x 13560(103gal) x
Ib949
0.0015(tons/103gal) = 0.1668 (tons/year)
2. Fuel Type 2, Natural Gas
NHGFU(2) x TCFUR(2) x EF = 1302 x 26,400(106cu.ft.) x
NHC(l) 186093
.0004 (tons/106cu.ft.) = 0.7388 (tons/year)
3. All other Fuel Types similar
B. Commercial and Institutional Fuel Usage
1. Fuel Type 1, Fuel Oil
a. Distillate Oil
CLUG x NHGFU(l) x NHCT x EF x CFT(dist.) =
CLUC NHG NHC(l)
0.15 x 139 x 215479 x 0.0015 / tons \ x
11.965 T547 16949 M03gal'
35000(103gal) = 0.7518 (tons/year)
b. Residual Oil
CLUG x EF x TCFUC (res.) = 0.15 x .0015 (tons/103ga1) x
CLUC 11.965
0(103gal) = 0.0 (tons/year)
c. Total Commercial Fuel Oil HC Emissions
Results of "a" + "b" = 0.7518 x 0.0 = 0.7518 (tons/year)
B-4
-------
2. Fuel Type 2, Natural Gas
CLUG x NHGFU(2) x NHCT x EF x TCFUC(2) =
CLUC NHG NHC(2)
0.15 x 1302 x 215479 x .004(tons/106cu.ft.) x
11.965 1547 186093
11100 (106cu.ft.) = 0.5424(tons/year)
3. All other Fuel Types similar to la and 2 above.
C. Emissions from Structural Fires
ESFC(4) x NHG = 123(tons) x 1547 =
NHCT 215479
0.8831(tons/year)
D. Emissions from Solid Waste Disposal
ESWC(4) x CLUG = 24(tons) 0.15 = 0.3009(tons/year)
CLUC 11.965
E. Evaporative Hydrocarbons
1. Surface Coating
CHCE(l) x GPQP = 872(tons) x 5096 = 7.822(tons/year)
CROP 568100
2. Gasoline Handling
CHCE(2) x CLUG = 3078(tons) x 0.15 = 38.5875(tons/year)
CLUC 11.965
3. Dry Cleaning
CHCE(3) x CLUG = 151(tons) x 0.15 = 1.8930(tons/year)
CLUC 11.965
B-5
-------
IV. GRID HOURLY EMISSIONS FOR HC
A. Hourly Characterization
1. Time: 8:00 a.m. = 9:00 a.m.
2. Wind Speed: 15 mph
3. Temperature: 40°F
4. Day: Tuesday
5. Month: February
B. Fuel Oil
[Residential + Commercial & Institutional Fuel Oil Yearly
Emissions] x [4.8499 x 10'4 - 7.0986 x 1Q-6T + 1.4614 x
10'6WS] = [Results III A.I plus III B.I] x Eqn =
[0.1668 + 0.7518] (tons/year) x [4.8499 x 10~4 - 7.0986 x
TO"6 x (40) + 1.4614 x 10-6(15)](y_ear) = (0.9186) x (7.9086 x 10-4")=
hour
7.2648 x 10"4 tons/hour
C. Natural Gas
[Residential + Commercial & Institutional Natural Gas Yearly
Emissions] x 4.8499 x 10~4 - 7.0986 x 1Q-6T + 1.4614 x 1Q-6WS +
0.4832 x PF] = [Results III A. 2 and III B.2] x [Eqn with PF
8760
from Table 2.4] = [0.7388 + 0.5424] (Ms, x [4.8499 x 10'4 -
7.0986 x 10'6 x (40) + 1.4614 x 10~6 x (15) + 0.4832 x 1.1 5j
8760
) = (1.2812) x (8.5429 x 10~4) (~r)= 1.0945 x 10~3 tons/hour
^hours' nour
D. Other Fuels
1 . LP Gas - same as C above
2. Coal - same as B above
B-6
-------
E. Structural Fires
Hourly emissions based on a random distribution may be used.
However, for use in dispersion models these values would be
meaningless. For this source, the modeller must investigate
the nature of any fires. The random distribution is obtained
by dividing the yearly figure by 8760.
F. Solid Waste
[Results from III D] x 4.27 x ICT4 = 1.28 x 1CT4 |2QS.
J hour
G. Evaporative Hydrocarbons
4.27 x 10~4 x [annual dry cleaning emissions + surface coating
emissions + (0.5 x gasoline marketing emissions x monthly sales
factor*)] + (automobile allocation factor** x 0.5 x gasoline
marketing emissions x monthly sales factor) = 4.27 x 10 x
[1.8930 + 7.822 + (0.5 x 38.5875 x 0.88)] + (1.94 x 10~4 x
0.5 x 38.5875 x 0.88) = 1.14 x 10-2 + 0.33 x 10'2 = 1.47 x
10~2 tons/hour
B-7
-------
TECHNICAL REPORT DATA
(Please read Inunctions on the reverse before completing}
1. REPORT NO. 2.
EPA-450/3-75-078
4. TITLE ANDSU8TITLE
Residential and Commercial Area Source
Emission Inventory Methodology for the Regional Air
Pollution Study
7. AUTHOR(S)
R. E. Holden and W. E. Zegel
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Science and Engineering, Inc.
P.O. Box 13454
Gainesville, Florida 32604
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, N.C. 27711
3. RECIPIENT'S ACCESSION«NO.
5. REPORT DATE
September 1975
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-1003
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
One of the major objectives of the Regional Air Pollution Study (RAPS) is to
provide data on the emissions of air pollutants, meteorological conditions and
ambient air quality with unprecedented density and resolution as to allow the
testing and development of a spectrum of mathematical models to simulate relation-
ships between emissions of pollutants and air quality. As part of this effort a
methodology for estimating the pollutant emissions from stationary residential
and commercial-institutional area sources on an hour-by-hour basis, and
apportioning them to the RAPS grid system is presented.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Regional Air Pollution Study
Emissions
Emission Models
13. DISTRIBUTION STATEMENT
Release Unlimited
b.lDENTIFIERS/OPEN ENDED TERMS
19. SECURITY CLAS.S (This Report)
Unclassified
20. SECURITY CLASS (This page)
Unclassified
c. COSATI field/Group
- •
21, NOy-Q-P PAGES
22. PRICE
EPA Form 2220-1 (9-73)
-------
JULY 1976 AMC7010.T0108D-FCR
REGIONAL AIR POLLUTION STUDY (RAPS)
100% COMPLETION REPORT
FOR
TASK ORDER 108-D
STATIONARY INDUSTRIAL AREA SOURCE
EMISSION INVENTORY
Prepared for
Environmental Protection Agency
Office of Air & Water Management
Office Of Air Quality Planning Standards
Research Triangle Park, N.C. 27711
by
Fred F. Littman
Kevin Isam
-------
TABLE Of CONTENTS
PAG!:
1.0 INTRODUCTION I
2.0 METHODOLOGY AND APPROACH 6
-------
TABLES
PAGL
TABLE 1 SOURCES AND EMISSIONS IN THE ST. LOUIS AQCR 3
TABLE 2 EMISSIONS FROM INDUSTRIAL AREA SOURCES 8
TABLE 3 EMISSION LIMITS FOR INDUSTRIAL AREA SOURCES 13
TABLE 4 EMISSION FACTOR SUBSTITUTIONS 13
-n-
-------
1.0 INTRODUCTION
The Regional Air Pollution Study (RAPS) emission inventory was designed
to provide emission data for the St. Louis Air Quality Control Region (AQCR-70)
to complement the meteorological data and ambient concentrations of pollutants
gathered in the framework of RAPS. The emission inventories encompass several
segments:
a. Major Stationary Point Sources
b. Minor Stationary Point Sources
c. Airports, Railroads and Vessel Emissions
d. Off-Highway Mobile Sources
e. Stationary Residential/Commercial Area Sources
f. Stationary Industrial Area Sources
g. Mobile Sources
h. Hydrocarbon Inventory
i. Heat Emission Inventory
j. Non-Criteria Pollutant Inventory
k. Particulate Size Distribution Inventory
1. Fugitive Dust Inventory
All inventories are stored in the RAPS data handling system using a Univac
1110 computer at the Environmental Protection Agency's headquarters at Research
Triangle Park, N.C. The major stationary point source inventory is the most de-
tailed one. It contains hourly measured values for fuel consumption or process
rates which can be transformed into emission rates using emission factors. Other
inventories are less detailed, but all will yield hourly values.
Major Stationary Point Sources includes all sources which individually con
tribute more than approximately 0.1% of the total emissions of a given pollutant
in the AQCR. Minor ones, for which less detailed data are available, includes
sources emitting more than 0.01% of a given pollutant.
All remaining emissions are designated "Area Source Emissions." These in-
clude industrial area sources as well as residential and commercial area sources.
The latter are discussed in EPA report no. 450/3-75-078 entitled "Residential
-1-
-------
and Commercial Area Source Emission Inventory Methodology For RAPS." The for-
mer are the subject of this report. Table 1 shows the amounts involved.
As indicated in Table 1, point sources account for the bulk of the emis-
sions of S02, particulates and oxides of nitrogen, but only for a minor part
of hydrocarbons and in particular, carbon monoxide emissions. Of all the point
sources, major sources account for 85 to 95% of the emissions. Industrial Area
Sources, that is, sources emitting individually less than 0.01% of a given pol-
lutant, account for only 0.2 to 0.6% of the total amount of pollutants. Nor
do they contribute significantly to the area source emissions, since they ac-
count for only 0.2 to 13% of other area emissions.
All area emission sources are assigned to a system of grid squares de-
veloped for RAPS^ which divides the AQCR into squares from 1 to lOOKm .
Emissions from residential and commercial sources are assigned to the grid
squares on a population density basis using various correction factors to ac-
count for the effect of ambient temperature, etc. Unfortunately, the distribu-
tion of industrial area sources on the micro scale of grid squares does not
lend itself to any estimations based on industrial employment, which have been
(?\
used successfully for larger scale estimates'1 ;. Thus, the only available tech-
niques consisted in an actual count of the sources, coupled with determinations
of their exact locations (in UTM coordinates), estimates of their annual fuel
consumption or process rates and work patterns.
(1) The Regional Air Pollution Study (RAPS) Grid Study
R.C. Haws and R.E. Paddock, Research Triangle Inst., Dec. 1975
(2) Residential and Commercial Area Source Emission Inventory
Environmental Science & Engineering, Inc., Gainesville, Fla.
EPA 450/3-75-078
-2-
-------
TABLE 1
SOURCES AND EMISSIONS IN THE ST. LOUIS AQCR
A. PARTICULATES
/
Tons/Yr. # Of # Of Companies I Of
Sources Total
Total Emissions In AQCR 354,665 - 100
Point Source Emissions 317,140 512 94 89.4
Major Sources (> 250T) 304,936 72 29 86.0
Minor Sources (> 25T) 10,100 138 24 2.8
Industrial Area Sources
( > IT) 2,104 302 41 0.6
Other Area Sources 37,525 - - 10.6
B. SULFUR DIOXIDE
Total Emissions In AQCR 1,234,395 - - 100
All Point Sources 1,187,294 349 62 96.2
Major Point Sources
( > 1000T) 1,142,906 62 20 92.5
Minor Point Sources
( > 100T) 38,003 120 13 3.1
Industrial Area Sources
( > IT) 6,385 167 29 0.5
Other Area Sources 47,101 - _ 3.8
-3-
-------
TABLE 1 (CON'T)
C. OXIDES OF NITROGEN
Total Emissions In AQCR
All Point Sources
Major Point Sources
( > 300T)
Minor Point Sources
( >30T)
Industrial Air Sources
Other Area Sources
Total Emissions In AQCR
All Point Sources
Major Point Sources
( > 100T)
Minor Point Sources
( > TOT)
Industrial Area Sources
(> T)
Other Area Sources
Tons/Yr.
433,547
310,992
291 ,438
17,010
2,544
122,554
# Of
Sources
382
55
194
133
-
# Of Companies
83
15
30
38
-
% Of
Tota
100
71,7
67.?
3.9
0.6
28.3
D. HYDROCARBONS
295,123
78,497
71,051
5,893
1,533
216,646
-
455
103
163
189
-
-
69
30
24
15
-
100
26.5
24.1
2.0
0.5
73.4
-4-
-------
TABLE 1 (CON'T)
E. CARBON MONOXIDE
All Point Sources
Major Point Sources
( > 100T)
Minor Point Sources
( > 10T)
Industrial Area Sources
( > IT)
Other Area Sources
ons/Yr.
1,079,522
63,045
51,570
9,030
2,446
1,016,477
# Of
Sources
-
202
31
69
102
-
# Of
Companies
-
59
18
12
29
•/, Of
Total
100
5.8
4.8
0.8
0.2
94.?
-5-
-------
2.0 METHODOLOGY AND APPROACH
The data shown in Table 1 (column 3) indicates a total number of be-
tween 15 to 41 companies in the Industrial Area category. Others were
added from listings of these manufacturers. Over 60 potential industrial
area sources were contacted; satisfactory data were obtained from 55.
There is a good deal of overlap, since combustion sources, which constitute
the bulk of the emission sources, emit all five "criteria" pollutants (S0?,
NO,., particulates, hydrocarbons and carbon monoxides.)
On the other hand, gas-fired sources emit only N0y and hydrocarbons,
most important hydrocarbon sources originate from evaporation rather then
combustion; and sources of particulates frequently do not emit other pol-
lutants.
The first step in inventorying these sources consisted of the preparation
of a cross-referenced list of the companies involved. This was followed up
by actual contact with the appropriate officials to obtain current consump-
tion or process rates as well as work patterns.
For the purposes of the industrial area emission inventory a source is
an individual company. By the term "company" is meant a plant location of
industrial character which is treated as a separate entity, though it may be
one of several subsidiary plants of a larger or parent firm. Annual 1975 in-
formation on fuel usage, incineration, paint and solvent usage, and produc-
tion was obtained from the companies. Emissions of each of the five criteria
pollutants were calculated using the 1975 data along with AP-42 emission fac-
(3)
tors v . Also, each of the companies was assigned to a grid square after de-
termination and verification of its UTM coordinates.
In those cases where data was unobtainable due to lack of cooperation ot
the companies involved, either National Emission Data System (NEDS), Illinois
EPA Emission Inventory, or Missouri Emission Inventory (MEI) data from 1973
was used to calculate emission. Those few cases are identified by an aster-
isk in Table 2, which lists all companies included in the inventory, its grid
(3) Compilation of Air Pollutant Emission Factors (2nd. Ed.)
Environmental Protection Agency, Research Triangle Park, N.C. No. AP-42
-6-
-------
square location and total pounds of emissions of each of the five {ril.eriri
pollutants. As a starting point for the inventory, NEDS and MEI were con-
sulted, using the numerical criteria shown in Table 3 for placing a source in
the industrial area category. Other NEDS or MEI companies listed were added
to the list of candidates if there was either no data given for them, or the
1973 data appeared questionable.
Knowing the SCC code for each process, the units processed, and the emis-
sion factor(s) corresponding to the SCC code, it is straightforward to calcu-
late pounds of pollutant emitted:
EMISSIONS (LBL) = EMISSION FACTOR X UNITS PROCESSED X (1-CONTROL EFFICIENCY)
(Ibs./unit)
At no time was there a problem in finding an appropriate SCC code to
classify the data received from the companies. The emission factors for in-
process fuel SCC codes are zero because the emissions should be included in
the associated SCC process code. In several cases, however, emission data is
not available to include inprocess fuel emissions with the process emissions.
To circumvent this obstacle, emission factors of inprocess fuel SCC source
codes were assigned industrial boiler codes. This was done only when inprocess
fuel emissions were not accounted for in the process emission factors. All such
substitutions appear in Table 4.
Some of the companies examined were not included in Table 2 because the
magnitude of their emissions placed them in the Minor Source category. On the
other hand, a few companies whose emissions of each of the five pollutants fall
below one ton have been included in Table 2 anyway. Data listed in this table
were entered on coding sheets, keypunched on cards and entered into the RAPS
Emission Inventory System at Research Triangle Park, North Carolina.
-7-
-------
TABLE 2
EMISSIONS FROM INDUSTRIAL AREA SOURCES
COMPANY NAME
A.B. Chance
Transformer Works
Washinqton, Mo.
Beall Tool Mfq.
Div. of Varien
Corporation
East Alton, 111.
Coates Steel Prod.
Granville, 111.
Continental Can
Plant*
2419 Lemp
St. Louis, Mo.
Corn Sweetners
Granite City, 111.
Drug Package,
Inc.
0' Fall on, Mo.
East St. Louis
Castings Co.
East St. Louis,
Illinois
Eaton Corp.
Washington, Mo.
Excelsior Foundry
Belleville, 111.
Gilster-Mary Lee
Chester, 111.
G & S Foundry
Freeburg, 111.
Havin Material
Service
St. Clair, Mo.
GRID
SQUARE
#39
#1240
#1795
#958
#1128
#2034
#1252
#2011
#1561
#1579
#1584
#2022
EMISSIONS
PART SOX NOX
111 110 742
1,356 4,192 12,421
11,730 42,182 51,720
000
417 25 9,591
540 32 28,170
2,023 0 0
1,910 0 0
410 1,562 1,650
48 3 576
509 454 480
2,702 0 0
(LBS)/YEAR
HC
107,4^
5C
2,4C
6,OC
12
16
8
1
2
CO
98
2,099
3,988
0
125 709
162 918
0
0
83
14
24
1,725
0
no
82
136
-8-
-------
TABLE 2 (CON'T)
COMPANY NAME
Havin Material
Service
Sullivan, Mo.
International
Shoe Company*
St. Clair, Mo.
J.C. George
St. Clair, Mo.
Jennison-Wright
Corporation
Granite City,
111.
Kellwood Co.
Finishing Div.
New Haven, Mo.
Kohen Concrete
Products
Germantown, 111.
Kurtz Concrete
St. Charles, Mo.
Maclay Concrete
Festus, Mo.
Mascoutah Grain
and Feed
Mascoutah, 111.
Masters Bros.
Sand Co.
Pevely, Mo.
Micro Al loy
Corporation
O'Fallon, Mo.
Missouri
Meerschaum
Washington, Mo.
GRID
SQUARE
#2002
#2022
#56
#1128
#3
#1768
#2126
#467
#1637
#292
#2034
#47
EMISSIONS (LBS)/YEAR
"PART " "SOV' NOV"
A A
2,366 0 0
465 1,761 1,920
000
7,576 3,956 8,844 6,7
180 108 2,160
1,506 0 0
11,000 0 0
7,200 0 0
13,702 0 0
4,000 0 0
167 10 2,004
262 2,685 34 23,8
HC
93
54
0
0
0
50
CO
124
9,102
306
284
45
-9-
-------
TABLE 2 (CON'T)
COMPANY NAME
Missouri Portland
Loading Terminal
St. Louis, Mo.
Mon Clair Grain
St. Clair County
Belleville, 111.
Mon Clair Grain
Waterloo, 111.
National Mine
Service
Nashville, 111.
New Baden Grain
Company
New Baden, 111.
Peavy Flour Mills
Alton, 111.
Permaneer Corp.
Union, Mo.
Pre-Coat Metals
St. Louis, Mo.
Koesch Enamel
Mfg.
Belleville, 111.
Ruprecht Quarry
Lemay, Mo.
St. Louis Grain
Corporation
Duncan Street
St. Louis, Mo.
Washington Metal
Products
Washington, Mo.
GRID
SQUARE
#972
#1484
#1146
#1829
#1714
#1020
#2015
#852
#1511
#744
#858
#59
EMISSIONS (LBS)/YEAR
PART SOY NOY HC
A A
1,800 000
16,977 000
15,400 000
0 0 1,500 750
6,616 000
12,750 000
600 593 4,011 15,100
3,400 204 40,800 16,700
1,780 107 40,940 534
48,000 000
179 0 0 0
1,080 4,090 4,320 216
CO
n
0
0
2,500
0
0
531
5,780
3,026
0
0
288
-10-
-------
TABLE 2 (CON'T)
COMPANY NAME
St. Louis Grain
Corporation
Foot of E. Grand
St. Louis, Mo.
St. Louis Grain
Corporation
Cahokia, 111.
St. Louis Steel
Casting
St. Louis, Mo.
Spartan Alum-
inum Products
Sparta, 111.
Sterl ing Steel
Casting Co.
Sauget, 111.
Sunoco Petroleum
255 E. Monroe
Kirkwood, Mo.
Sunoco Petroleum
1252 E. Road
Manchester, Mo.
Thompson Asphalt
Alton, 111.
Trautman Quarry
Pevely, Mo.
Troy Grain Co.
Troy, 111.
Vitro Products*
St. Louis, Mo.
Weber, Inc., Fred
O'Fallon Plant
O'Fallon, Mo.
GRID
SQUARE
#1008
#997
#849
#2448
#1095
#404
#789
#2281
#355
#1624
#1072
#118
EMISSIONS (LBS)/YEAR
PART SOX NOX HC
1,607 000
535 0 0 0
13,520 0 2,080 0
6,840 000
1,414 988 5,460 162
000 4,105
00 0 728
7,025 4 870 22
13,336 000
4,300 000
000 22,400
3,250 5,680 6,000 300
CO
r\
V
0
0
0
697
0
0
123
o
0
0
A;::
-11-
-------
TABLE 2 (CON'T)
COMPANY NAME
Weber, Inc., Fred
Festus Asphalt
Plant
Festus, Mo.
Weber, Inc. , Fred
South Asphalt Plant
Lemay, Mo.
Weber, Inc. , Fred
North Asphalt Plant
Creve Coeur, Mo.
Western Litho
Plate
St. Louis, Mo.
Wirco Castings,
Inc.
New Athens, 111 .
GRID
SQUARE
#467
#2245
#2147
#281
#1683
EMISSIONS (LBS)/YEAR
PART SOX NOX HC
3,590 6,305 6,660 333
13,230 41,350 43,680 2,184 2
16,565 93 18,600 465 2
240 14 2,880 72
1,050 000
CO
444
96
* ASTERISK INDICATES DATA DERIVED FROM 1973 INVENTORIES
-12-
-------
TABLE 3
EMISSION LIMITS FOR INDUSTRIAL AREA SOURCES
PARTICULATES
1 < X < 25
sox
1 < X < 100
NOX
1 < X < 30
HC
1 < X < 1 0
CO
1 -, X -
X - Tons/Year OF POLLUTANT EMITTED
Upper limits in Table 3 are based on emissions of less than 0.01% of a
«
given pollutant.
TABLE 4
EMISSION FACTOR SUBSTITUTIONS
STANDARD
SCC CODE
3-90-004-XX
3-90-005-XX
3-90-006-XX
3-90-008-XX
3-90-010-99
SUBSTITUTE
SCC CODE
1-02-004-OX
1-02-005-OX
1-02-006-OX
1-02-007-OX
1-02-010-OX
-13-
-------
METHODOLOGY FOR THE DETERMINATION OP EMISSION LINE SOURCES
Environmental Protection Agency
Contract No. 68-02-141?
February 28, 1975
Washington University
School of Engineering and Applied Science
Department of Civil Engineering
St. Louis, Missouri 63130
Dr. Lonnie E. Haefner
Principal Investigator
-------
TABLE OF CONTENTS
Page
CHAPTER I - INTRODUCTION - STUDY DESIGN 1
A.) Introduction - Objectives of Research - -- -- 1
B.) Limited Literature Search -_---_---___--- 2
1.) Relevance of Air Standards ----------- 2
2.) Relationships of Emissions to Key Traffic
Engineering Stimuli -__ 3
C.) Formal Research Plan' _______________ 12
1.) Emissions Related Traffic Flow Research Aspects - 12
2.) Specific Work Plan of Research 15
CHAPTER II - DATA DEVELOPMENT 22
CHAPTER III - MODEL DEVELOPMENT 27
A.) Overview of Model Logic ---------------- 27
B.) Software System Development - ___________ 33
CHAPTER IV - DOCUMENTATION OF EMISSION LINE SOURCES 40
A.) Introduction - Issues in Defining Line Sources - - _ - 40
B.) Basic Definitions - Delineation of Gross Line Sources - 40
C.) Refined Level - Ultimate Definition of Line Sources - - 49
D.) Synthesis of Line Source Information- -_--_--__ 63
E.) Related Processing Costs ----------- 75
F.) Limited Sensitivity Analysis ___ ________ 73
CHAPTER V - CONCLUSION 87
A.) Use and Applicability of Present Research and
Modelling Results 87
B.) Recommendations for Further Research --------- 88
C.) Closing Comments on Status of Line Source-
Traffic Attribute Modelling 91
i (Continued)
-------
TA.BLE OF CONTENTS (Continued)
Page
BIBLIOGRAPHY 93
Selected Research Bibliography --------------- 94
St. Louis Area Traffic Data Sources ____ 95
Professional Interviews ------------------- 96
APPENDIX A - COMPUTER PROGRAM DOCUMENTATION 97
A-l NETSEN Program 98
A-2 KRCVRT Program 123
A-3 Data Management Program - ----- ---135
A-4 Interface Program --- _______________ 137
APPENDIX B - EXAMPLES OF DATA COLLECTION AND CODING FORMAT 143
ii
-------
LIST OF TABLES
Page
TABLE 1 - ADT RANGES USED FOR GROSS LEVEL LINE
SOURCE DEFINITION 42
TABLE 2 - GROSS LEVEL LINE SOURCE MISSIONS
(FUNCTIONAL CLASS-FREEWAYS) 44
TABLE 3 - GROSS LEVEL LINE SOURCE EMISSIONS
(FUNCTIONAL CLASS-PRINCIPAL ARTERIALS) 46
TABLE 4 - GROSS LEVEL LINE SOURCE MISSIONS
(FUNCTIONAL CLASS-MINOR ARTERIALS) 48
TABLE 5 - ADT RANGES USED FOR ULTIMATE LINE SOURCE DEFINITION - - 51
TABLE 6 - ULTIMATE LINE SOURCE MISSIONS
(FUNCTIONAL CLASS-FREEWAYS) 53
TABLE 7 - ULTIMATE LINE SOURCE MISSIONS
(FUNCTIONAL CLASS-FREEWAYS) 55
TABLE 8 - ULTIMATE LINE SOURCE MISSIONS
(FUNCTIONAL CLASS-FREEWAYS) 57
TABLE 9 - ULTIMATE LINE SOURCE MISSIONS
(FUNCTIONAL CLASS-FREE//AYS) 59
TABLE 10 - ULTIMATE LINE SOURCE MISSIONS
(FUNCTIONAL CLASS-PRINCIPAL ARTERIALS) 61
TABLE 11 - ULTIMATE LINE SOURCE EMISSIONS
(FUNCTIONAL CLASS-PRINCIPAL ARTERIALS) 64
TABLE 12 - ULTIMATE LUTE SOURCE EMISSIONS
(FUNCTIONAL CLASS-PRINCIPAL ARTERIALS) 66
TABLE 13 - ULTIMATE LINE SOURCE MISSIONS
(FUNCTIONAL CLASS-PRINCIPAL ARTERIALS) 6s
TABLE 14 - ULTIMATE LINE SOURCE MISSIONS
(FUNCTIONAL CLASS-MINOR ARTERIALS) 70
TABLE 15 - ULTIMATE LINE SOURCE MISSIONS
(FUNCTIONAL CLASS-MINOR ARTERIALS) 72
TABLE 16 - ULTIMATE LINE SOURCE EMISSIONS
(FUNCTIONAL CLASS-MINOR ARTERIALS) 74
(Continued)
-------
LIST OF TABLES (Continued)
Page
TABLE 1? - SUMMARY OP ULTIMATE LINE SOURCE EMISSIONS INFORMATION - ?6
TABLE 18 - PERCENTAGE OF TOTAL EMISSIONS CONTRIBUTED BY
FUNCTIONAL CLASS OF LIKE SOURCE 77
TABLE 19 - COST SUMMARY FOR TYPICAL SOFTWARE SYSTEM RUN 79
TABLE 20 - EMISSIONS SUMMARY FOR FREEWAY CORRIDOR DIVERSION 82
TABLE 21 - EMISSIONS SUMMARY FOR PARALLEL ARTERIAL DIVERSION 85
-------
LIST OF FIGURES
?age
FIGURE 1. Speed Correction Factors for 1968 Model Year
Vehicles in Low Altitudes ----------- 4
FIGURE 2. Speed Correction Factors for 1968 Model Year
Vehicles in Denver- -------------- 5
FIGURE 3. Speed Correction Factors for 1971 Model Year
Vehicles in Denver- -------------- 6
FIGURE 4. Pollution Levels Along Transverse Street Cross Section
of Centered Expressway with Joint
Development Structures --- ----- 8
FIGURE 5» Pollution Levels Along Transverse Street Cross Section
of Centered Expressway without Joint
Development Structures ----- ___-_ 9
FIGURE 6. Pollution Levels Along Transverse Street Cross Section
of Centered Expressway-Boulevard - ___- 10
FIGURE 7» Comparison of City Street and Freeway Conditions - 11
FIGURE 8. Flow Parameters Related to Emissions and/or
Air Quality Phenomena - ------ -_ 16
FIGURE 9. Research Work Plan 17
FIGURE 10. Master Logic of 1TETSM Model 28
FIGURE 11. Traffic Emissions Software System 34
FIGURE 12. Gross Level Line Sources (Functional Class-Freeways) - 43
FIGURE 13. Gross Level Line Sources (Functional Class-
Principal Arterials) 45
FIGURE 14. Gross Level Line Sources (Functional Class-
Minor Arterials) --------------- 47
FIGURE 15. Ultimate Line Sources (Functional Class-Freeways) - - 52
FIGURE 16. Ultimate Line Sources (Functional Class-Freeways) 54
FIGURE 17. Ultimate Line Sources (Functional Class-Freeways) - - 56
FIGURE 18. Ultimate Line Sources (Functional Class-Freeways) 58
v (Continued)
-------
LIST OF FIGURES (Continued)
Page
FIGURE 19. Ultimate Line Sources (Functional Class-
Principal Arterials) 60
FIGURE 20. Ultimate Line Sources (Functional Class-
Principal Arterials) 62
FIGURE 21. Ultimate Line Sources (Functional Class-
Principal Arterials) 65
FIGURE 22. Ultimate Line Sources (Functional Class-
Principal Arterials) _____ -_ 67
FIGURE 23. Ultimate Line Sources (Functional Class-
Minor Arterials) 69
FIGURE 24. Ultimate Line Sources (Functional Class-
Minor Arterials) ------ - ____ 71
FIGURE 25. Ultimate Line Sources (Functional Class-
Minor Arterials) 73
FIGURE 26. Emissions Summary for Freeway Corridor Diversion - - - 81
FIGURE 27. Emissions Summary for Parallel Arterial Diversion - - 84
FIGURE A-l Program GIN Flowchart 136
FIGURE A-2 Program RUMMY Flowchart 138
vi
-------
CHAPTER I
INTRODUCTION - STUDY DESIGN
-------
CHAPTER I
INTRODUCTION - STUDY S3SIGN
A. INTRODUCTION - OBJECTIVES OF RESEARCH
The study of pollution concentrations in a metropolitan area requires
accurate characterization of pollution emitted due to the presence and
operation of transportation corridors and grids, termed line sources.
Accurate reporting of emissions depends on efficient monitoring of
traffic flow and system design and location parameters critical to the
emission process, in addition to adequate characterization of vehicular
emissions under a range of operating conditions.
The objective of this research is to develop a methodology which
documents the bases and criteria for determining which major freeway and
arterial links should be considered emission line sources in a metropolitan
area, and their geographic and temporal sensitivity to frequency and dura-
tion of monitoring. In addition to developing the methodology, it is to
be verified through testing in the St. Louis Air Quality Control Region,
determining specifically what links in this region shall be considered
as line sources. In accomplishing the above, several specific performance
objectives will be attained, which are:
a.) Obtainment and documentation of the most recent traffic
data relevant to emissions phenomena for the St. Louis Region.
b.) Development of the methodology, and use of the above data
to estimate emissions levels, through the use of the
Department of Transportation model SAPOLLUT, which computes
aggregate emissions and concentrations of CO, NO and HC
Jv
for a traffic network. Detailed discussion of the model
operation as used in this research is found in Chapters
III, IV and Appendix A.
- 1 -
-------
- 2 -
c.) Additional conceptualization of the methodology to
formulate a sensitivity analysis which analyzes emis-
sions information provided by different combinations of
line source components, and allows comparison of information
output vs. different specifications of traffic network inputs.
d.) Verification of the sensitivity analysis on the St. Louis
Area, ultimately yielding appropriate specifications of
emission line sources in the area.
B. LIMITED LITERATURE SEARCH
The objective of a limited literature search into traffic behavior
and air pollution emissions in a project such as this one is to reinforce
"basic knowledge of relationships of emissions to key traffic behavioral
variables. In so doing, the search allows the research team to see
the rationale for their model building, its particular relationship to
emissions stimuli, and to develop an overview and fluency with the
traffic engineering literature dealing with the problem.
1.) Relevance of Air Standards
A basic research program with such a broad and intensive scope as
the RAPS program is interested in accurately uncovering the phenomena
of behavior of line sources and their resultant emissions, and the part
these emissions play in the region-wide emissions problem.. In so doing,
the capability exists to add to basic knowledge which may ultimately
lead to improved strategies for meeting ambient air standards.
-------
The highway system acting as a line source may act as a primary
cause of CO. It may be a contributor to NO^ and HC emissions. Cer-
tain aspects of the highway transportation system will now be investigated
in relation to these.
2.) Relationships of Emissions to Key Traffic Engineering Stimuli
The pollutants of CO, HC and NO have documented associations with
X
2
speed. Results of recent research on such associations is graphically
shown in Figures 1, 2, and 3« These figures are the result of developed
equations of:
—2
LN HC = A + BS + CS in grams/mile
LN CO
= A1 + B'S + C'S2 in grams/mile
NO = A" + B"S in grams/mile
.X.
In the above, S is the average speed of the driving sequence. To de-
termine the speed correction factor at any particular speed in the
range of 15 to 45 miles per hour, a ratio of the above equations is
used. The emissions are determined at the desired speed and ratioed
with the emissions at a speed of 19.6 miles per hour, the average
speed over the federal driving schedule. Figure 1 represents vehicle
model year 1968 in low altitudes, Figure 2 represents the 1968 vehicle
model year in Denver, and Figure 5 represents the 1971 model year in
Denver. It should be noted the relationship is quadratic with respect
to HC and CO, and linear with respect to HO , yielding decreasing
emissions with increased average route speed for HC and CO, and in-
creasing emissions of NO with increased average route speeds.
-------
10
Average Route Speed, KPH
30 50
70
1.5
o
4J
0
-------
10
Average Route Speed, KPH
30 50
70.
I
1.5
o
+j
o
c
o
•r—
•«J
O
OJ
5
Q.
CO
1.0
0.5
I
f
I
15 30
Average Route Speed, MPH
45
FIGURE 2. Speed Correction Factors for
1968 Model Year Vehicles in Denver
-------
10
Average Route Speed, KPH
30 50
70
1.5
o
«O
c
o
O
O
O
•o
o>
(U
CL
to
1.0
CO
0.5
1
I
15 30
Average Route Speed, MPH
45
FIGURE 3. Speed Correction Factors for
1971 Model Year Vehicles in Denver
- 6 -
-------
- 7 -
A further finding is that complex highway design configurations,
unique localized meterology, presence of topography and rough terrain
and downtown or high-rise street canyons play readily identifiable,
but relatively less understood roles in air quality. In the simplest
sense, average speed and vehicle miles of travel on a link are relevant
inputs as indicators of CO, EC and NO emissions. The realistic design,
Jt
environmental and neighborhood attributes modify the impact of these
two basic stimuli on air quality, and the mechanics of these modifica-
tions are not developed in depth in the basic research to date. Figures
4, 5 and 6 are offered as visual examples of research output of the
effect of geometric configuration on air quality. A further behavioral
input, closely related to average speed, is the smoothness of traffic
flow and capability of avoiding traffic congestion effects. Figure 7
shows the concentrations for smooth, uninterrupted flow of 50 nrph of
typical freeway movement versus higher emissions induced by poor signal
timing, pedestrian and parking interference which increases delay on a
typical interrupted flow arterial street operating poorly. As such,
from a traffic engineering point of view, V/C ratios and acceleration
noise parameters of the traffic stream are relevant to emissions levels.
In conclusion, the literature search revealed:
1.) The basic stimuli of emissions to be well documented
against inputs of average speed and VKT for CO and HC,
with more questionable data and relationships with respect to NO .
2.) Further complexities in air quality overlaid on the above when
cut, fill, and complex cross section and geometric design
configurations are included, as illustrated in Figures 4»
5 and 6,
-------
-5
I I
c t
0.0$
..0.01.
0.03
SO
30
20
60
do
JO
0.02 . 10 -
Level 3 70
Level 2 25
Levef I 0
/ W
25 Feet
I
FIGURE 4. Pollution Levels Along Transverse Street Cross Section of
Centered Expressway with Joint Development Structures
- 8 -
-------
0.05
0.04
0.03
0.02- 10-
Level 3 601 1
Level 2 25^ \-
Level 1 oL
I
FIGURE 5. Pollution Levels Along Transverse Street Cross Section of
Centered Expressway Without Joint Development Structures
- 9 -
-------
K F.«t
FIGURE 6. Pollution Levels Along Transverse Street Cross Section of
Centered Expressway-Boulevard
- 10 -
-------
8
17H n.p.h.,
7H m.p.
200 400 600 800 1000 1200 1400
Vehicle Flow Per Hour
FIGURE 7. Comparison of City Street and Freeway Conditions
- 11 -
-------
- 12 -
3.) Further complexity in emissions when system delay from
congestion is introduced, as illustrated in Figure 7.
4.) Further complexity in air quality when overlaid by local,
uniquely complex topography and terrain, and localized
unique meterologic conditions.
5.) The apparent need for further basic research on line
sources where such complexities in 2-4 above are introduced.
That is, further research into locating them, stratifying
their attributes, classifying them in an orderly data
system, and relating their attributes to their resulting
link emissions and the air quality of their locale,
C. FORMAL RESEARCH WORK PLAN
1.) Emissions and Air Quality-Related Traffic Flow Research Aspects
In light of the previous literature search, it has been categorically
found that volume, or vehicle miles of travel, and average speed are
critical inputs which relate to emissions of nitrogen oxides, carbon
monoxides and hydrocarbons. However, exhaustive use of typical traffic
flow data related to the above two inputs should allow a refined and
meaningful statement of flow phenomena and traffic systems design which
pollution output is sensitive to. Thus, a short discussion of such
flow-related parameters is warranted, to demonstrate their categorization
as inputs:
Volumes - Average Daily Traffic and peak hour volumes, historically
as indicators of use of the facility.
-------
— 13 —
Average Travel Speed - As an indicator of efficiency of the
facility and adequacy of design.
General Interzonal Origin-Destination Patterns - As a regional
mapping of incidence of travel, and proximity of
travel paths to other regional activities.
Functional Classification of Highways - Classed as whether freeway,
arterial, collector or local, as an indication of importance
and frequency of use and level of design standards employed.
Delay Information - which modifies or refines information on
average speed, above, through studies of volume to
capacity ratios, travel time profiles, travel time
contours, or waiting time or delay contours. Locations
are detected in the system where speeds are radically
altered due to delay and congestion.
Locations of^Design Related Phenomena - such as extremely
complicated route or interchange configurations, and
areas of cut or fill, or frontage roads with struc-
tures, v/hich induce localized alterations in air quality
and emissions, the latter when average speed is modified
due to the design phenomena.
Unique Areas of Progression - in addition to areas which can be
reviewed from speed and delay information as stated
above, these are unique in the network, in that some
engineering or planning alteration exists to eliminate
congestion by specific means with highly predictable
results, such as one-way street flows, or progressive
signaliaation, thus allowing atypical consistency in
volumes or average speeds, with stable emissions output.
-------
- 14 -
Areas of Critical Land TJse Adjacent to or Within the Network -
The first type is an area adjacent to the traffic system
which is a highly sensitive land use to emissions output
and local air quality, or a land use type such as industrial,
which supplements and confounds the emissions level and
air quality in the vicinity of the corridor. The second
type is the downtown or core area, or like areas of
high-rise, high-density buildings. The building heights
or "street canyons" affect air quality in the vicinity of
the grid and corridor sources.
Vehicle Mix - the composition of traffic, in terms of percentage
of autos, intermediate size trucks and large size trucks
is relevant, due to differing emissions from vehicle type,
and the impact of the traffic composition on average
speed and traffic flow throughout a link. The composition
of vehicles by age also determines the level of emissions
from the traffic stream.
Frequency of Monitoring - all links under study will encompass some
or many of the above flow related phenomena which have
an impact on emissions and/or air quality. The frequency
of observing such network components with respect to
adequate characterization of CO, NO , and HC information
Jv
is critical. Typical choices of duration of traffic
volume counting periods include 1, 8, 12, 24 hours,
weekly, and peak-off peak combinations.
-------
Thus, the above considerations are relevant to comprehensively
using traffic flow information to develop the "optimal" network inputs
for specifying emissions line sources. Use of this information may be
characterized by the following three dimensional array in Figure 8.
The research proceeds by essentially categorizing each apparently
relevant link component with respect to the above array, then testing
and trading off combinations of these with respect to emissions levels
to ultimately produce a set of line sources. The process pro-
ceeds interactively, making full use of local traffic engineering
knowledge about network components, yet developing the taxonomy in a
rigorous and consistent logic.
2.) Specific WorkPlan of Research
To accomplish the stated objectives, and develop the output dis-
cussed above, a six-phase work program has been pursued, as shown in
Figure 9«
Phase 1 has four tasks, which were carried on simultaneously.
Task 1.1 consisted of review of the flow aspects discussed above with
respect to their impact on emissions. Tasks 1.2 and 1.5 included the
procurement of St. Louis Air Quality Control Regional Traffic data, and
review of important links and their traffic operating attributes (Figure 8) with
local professionals. Through the principal investigator's local knowledge and
contacts, use was made of East-West Gateway Regional Coordinating Council profes-
sionals, and engineers with Missouri State Highway Department, Illinois
Department of Transportation, and St. Louis, St. Clair, Madison and
other appropriate county engineering personnel. The results of 1.2 and
1.5 yielded realistic insights into the operation of corridors and
arterial highway grid components. In TasV 1.4, the traffic input and
emissions output aspects of the model SAPOLLUT was reviewed, and the
model was procured and put on the Washington University Software Library.
-------
%
3
o
•p
•H >-.
C 0
0 C
SO
CO D
(H 0)
3£
w
CO
X
-p
C
o
3
rH
0)
0)
^
•^
•=*
CM
CM
I
fl>
0)
O
-p
O
CD
O
O
O
3
o
- 16 -
-------
Review of
Traffic Flow
Parameters
(1.1)
Review
St. Louis Area
Traffic Data
(1.2)
V
Categorization of
Appropriate Corridor,
Link and Grid
Components
(2.1)
t
^L
Develop Combinations of
Network Components, Estimate
Emissions with SAPOLLUT
(3.D
A/
Review
Link Attributes
with Local
Professionals
(1.3)
Procure and
Study SAPOLLTJT
Model
(1.4)
Review Mechanics
of Sensitivity
Analysis
(2.2)
1
1
1
A
1
1
L-? -
x__
\
/
Documentation of
Emissions Estimates vs.
Network Combinations vs.
Monitoring Frequency
(4.0)
Sensitivity Analysis
of Different Network
Combinations
(3.2)
\
\
X
/
Final
Specification
of St. Louis
Air Quality
Control Region
Line Sources
(5.0)
Write
Final
' (6?0)
FIGURE 9. Research Work Plan
- 17 -
-------
- 18 -
Phase 2 included two overlapping tasks. The initial task, 2.1
categorized appropriate corridors, links and street grid components
with respect to the flow phenomena illustrated in the previous three-
dimensional array. The coding was developed so that a complete set of
descriptors exists for each corridor, link or arterial street grid level con-
ceptualized and developed as input. Overlapping with this, task 2.2 familiarized
the study team with the sensitivity analysis, and the mechanics of
its performance in order to assure continuity between Phase 3 and
later phases.
Phase 3 had two tasks. Task 3.1 developed levels and combinations
of network hierarchial components as inputs to studying emissions.
The following sequence of components were developed for use with
SAPOLLUT and sensitivity testing:
Round 1; All freeway corridors, link components broken up by
average speed differentials and volume differentials.
Round 2; All freeway corridors, additional breakdown by sites
of specifically complex configuration.
Round 3: Addition of the arterial street grid to above, broken
down by speed, delay and volume differentials.
Round 4: All the above, additional breakdown by arterial sites
of specifically complex configuration.
Round 5t Addition of refined locations adjacent to sensitive
land use areas, and areas of exceptional progression.
Round 6: All of the above, cross classified by vehicle mix.
-------
- 19 -
Thus, the approach is to ultimately yield a descriptive network of
line sources N, which is composed of
5~ 9
N ~ <~~~ A, ijkmnopq t
i,3,k,m,n,o,p,q, t
where JL is a specific link or network component and i, j, k, m, n, o, p, q .....
are its flow parameters shown in the three-dimensional array, and t is
the monitoring frequency. At each of the above rounds, the network
could be input to SAPOLLUT, and emissions information estimated. The
network is synthesized in a sequential manner through the several rounds,
to develop clarity about its composition and emissions output, allowing
for a more orderly sensitivity analysis.
Task J>.2 articulated changes in the above synthesized network
and site monitoring frequency. Prom knowledge of the local area,
specific link and network components developed through all of the above
rounds were removed of added, noting through the use of SAPOLLUT in a
sensitivity analysis, the change in resulting emissions information.
The sensitivity analysis of the coded link and network components was
developed in a batch software format, to allow a certain programmed
sequence of network alterations to occur. This process was hooked
with SAPOLLUT, to allow network input operations and emissions estima-
tion to occur in one continuous software run.
Phase 4 documented the results of the use of SAPOLLUT and the
sensitivity analysis carried on in J>.2. Reference to Figure 9 shows
feedback and interaction across 3.1, 3.2 and 4 to adequately formulate
test, converge and document the above processes.
-------
- 20 -
Phase 5 utilizes the above array of sensitivity and convergence
information and specifically states network component locations, their
frequency of emissions monitoring, and the required traffic flow-related
data. This specification defines the emission line sources for the
St. Louis Air Quality Control Region.
Phase 6 combines previous conceptual analysis, the computer software,
associated network synthesis, use of SAPOLLUT and the sensitivity analysis
and its results, with specifications for refinement and further research
into a final written report contained herein. The following text will
elaborate on data collection, software development, use of SAPOLLUT,
sensitivity analysis and definition line sources, and provide appropriate
appendices on software documentation and data formats.
-------
Footnotes Chapter I
Federal Register, Volume 36, No. 84, Friday, April 30, 1971.
2
Scott Research Laboratories, Inc., Development of Representative Driving
Patterns at Various Averse Poute Speeds, EPA rlo. 68-02-1301 (6-1 J>).
February 11, 1974, San Sernadino, California.
., pp. 3-1 through 4-1S.
^Sturman, G. M. , The Effect of Highways on the Environment, May, 1970.
Steering Group and Working Group Appointed by the Minister of Transport,
Great Britain, Cars for Cities, 1967.
- 21 -
-------
CHAPTER II
DATA DEVELOPMENT
-------
CHAPTER II
DATA DEVELOPMENT
An assessment of highway line sources with respect to emissions requires that
the road network under study be classified according to a set of para-
meters that allows appropriate hierarchial analysis, since some links
will be much more critical from an emissions standpoint than others.
The data collection effort focused on those areas designated
as urban and urbanizing. The East-West Gateway Coordinating Council
defines these areas to include the City of St. Louis, St. Louis County,
and parts of St. Charles and Jefferson Counties in Missouri. In
Illinois part of Monroe, and Madison and St. Glair Counties are included.
Roadway data was collected within this area in light of emissions and
their sensitivity to highway functional class, volume and composition of
vehicles present, and the operational characteristics of vehicles related
to both traffic volume and average speed alterations due to roadway align-
ment and profile. Other data collected, such as the intensity and type of
the adjacent land use is relevant, along with certain highway design
characteristics, to the localized air quality.
A number of agencies were contacted to compile the inventory.
They included the East-Vest Gateway Coordinating Council, Missouri
State Highway Department Jefferson City Office and St. Louis Office,
the Region VII Office of the Illinois Department of Transportation,
the Office of the Deputy Commissioner of the City of St. Louis Street
Department, St. Louis County Division of Highways and Traffic, and
the City of St. Charles.
A working set of highway links was established through the use
of East-West Gateway's 7001 link-node map for the entire region under
- 22 -
-------
study. Actual street names for all links were recorded from maps
provided by the Auto Club of Missouri.
The initial major data collection effort was for the freeway
functional class. Volume and vehicle attributes sought for each link
were average daily traffic, peak hour traffic, hourly distribution of
traffic, percent of heavy duty vehicles, and the directional distribu-
tion of traffic. All the above were obtained for the Missouri counties
except percent heavy duty vehicles. However, in Illinois, only average
daily traffic and percent heavy duty vehicle data was available.
Link attributes sought to describe vehicle operating characteristics
were the volume to capacity ratio (V/C), peak hour speed by direction,
off peak speed by direction, frequency of complex interchanges, lane
drops and existing freeway bottleneck sections. The capacity informa-
tion obtained from East-West Gateway was a representative daily capacity
factored down to obtain a peak hour V/C ratio at level of service E. An
hourly V/C ratio table was given in the SAPOLLUT users manual based on
an hourly speed distribution. Current peak hour speed data by direc-
tion was available for 75?° of the area freeway links from East-West
Gateway. Through the cooperation of the Illinois Department of
Transportation and local agencies in St. Louis and St. Louis County,
a complete set of links with complex interchanges, lane drops, and
major freeway bottlenecks was compiled. The study team drove the
network to collect information on roadway topography as a freeway
link descriptor. Four terrain types were categorized: high fill,
deep cut, rolling terrain, and level terrain.
-------
- 24-
As a final set of link descriptors for the freeway functional
class, land use by type and intensity was sought. East-West Gateway
provided an area map delineating high density regionally oriented
land uses. Types included commercial, educational, medical, recrea-
tional, and airport.
The next major data collection effort was for principal arterial
roadways. These roadways are characterized "by a minimum number of
intersections at grade. Such intersections are typically designated
to provide channelization and signal synchronization to enhance the
traffic movement along the arterial. The link attributes sought to
describe volume and vehicle composition were as before, average daily
traffic, peak hour traffic, hourly distribution of traffic, percent
of heavy duty vehicles, and the directional distribution of travel.
The same data gaps that existed for freeways exist for principal
arterials in this attribute set. In addition to volume to capacity
ratios, peak hour directional speed, off peak directional speed, lane
drops, and general roadway bottlenecks were collected as link attributes
for vehicle operating characteristics. Farther, information concerning
the degree of progressive movement was sought. Progressive movement
is typified by a continuous flow of a platoon of vehicles over long
stretches of highway. Such movement can be induced by the- type of
signal systems employed at intersections and the distribution of
one-way streets. It is desirable to separate out links that have
progressive movement since the degree of vehicle delay is much less
than on arterials without it. Attributes for progression included
links with pre-timed signal systems, physically interconnected signal
-------
- 25 -
systems, and one-way street flows. The City of St. Louis, St. Louis
County, Missouri State Highway Department, and the Illinois Department
of Transportation provided us with complete information on this attri-
bute set.
As in the freeway case, sensitive land uses were included as link
descriptors. In addition to those already listed, the central business
district was included as an area type in the inventory. This was done
to aid in the delineation of an additional topography attribute, the
categorization of street canyons.
The final data collection report was for minor arterial roadways.
Such roadways provide for both traffic movement and land access. The
data sought was the same for principal arterials and had the same data
gaps.
In total, 28 link descriptors across three functional classes of
highway were assembled. This data was compiled on an individual link
basis, coded for keypunching and readied for input into the software
system to be described in Chapters III and IV. The collection and
coding format is illustrated in Appendix B.
-------
Footnotes Chapter II
Level of Service E represents operation of the system with volume at or
near capacity. Operating speed is relatively low, flow is unstable
and momentary stoppage occurs; the system is on the verge of complete
jam and saturation with attendant congestion effects.
- 26 -
-------
CHAPTER III
MODEL DEVELOPMENT
-------
CHAPTER III
MODEL DEVELOPMENT
A. OVERVIEW OF MODEL LOGIC
This chapter describes the logic construct of the network sensitivity
model NETSEN, and its interface with the emissions estimation model
SAPOLLTJT. The flow chart of the master logic for the model NETSEN is
shown in Figure 10. In general, the model works by defining a series
of sequential tests of presence of network related attributes shown in
steps 2.0-13.0. These are presence of the link within the Central Business
District (i.e., the Downtown Core Commercial Area), functional class of the
link, its ADT, presence of special topography, capacity alterations, pre-
sence of sensitive land uses, presence of progressive movement, speed
difference, truck volumes, and V/C ratio. After reading the coded link data
records in, with step 2.0, each link is tested to sort and classify it ac-
cording to combinations of attributes present in it. Those groups of links
thus containing certain combinations specified in the control card in step
1.0 are then output to SAPOLLUT for use in estimating emissions. A complete
description of the software documentation is given in Appendix A. Two
specific points are important in overviewing the logic at this point:
1.) The network and any link subset component of it can be
tested at any level of data attributes relevant, from
very gross descriptions containing only ADT information,
to very refined descriptions of the network, classifying
and locating all of the attributes shown in steps 2.0-13.0
on the network.
- 2? -
-------
- 28 -
,' Read in Control Card (l.O) /
^7 Read in Link REG (2.0)
Modify by CBD?
(3.1)
Modify by
Functional Class
(4.1)
Test for
Functional Class
(4.2)
Yes
'<-
la
-¥-
V'
(Continued)
FIGURE 10. Master Logic of NETSEN Model
-------
Modify by ADT?
(5.1)
Test for ADT
(5.2)
Modify by
Special Topography
(6.1)
Test for
Topography
Modify by
Capacity Alterations
(7.1)
Test for
apacity Alteration
(Continued)
-------
Test for
Sensitive Land Use
(8.2)
No
Modify by
Progressive Movement
„ (9.1)
Test for
Progressive Movement
(9.2)
Modify by
Speed Difference
(10.1)
(Continued)
-------
3a
. No
No
No
Difference
Modify by
Truck Volumes
(11.1)
Test for
Truck Volumes
(11.2)
Modify by
V over C
12.1)
over
(12.2)
(Continued)
-------
4
OUTPUT;
LINKS MEETING
ASSIGNED CHARACTERISTICS
(13.1)
ARE ALL LI1
IS FILE TESTED
GO TO 2
Yes
TO EMISSIONS MODEL
WITH TAGGED
LINK DESCRIPTORS
(14.0)
-------
- 33-
2. The level of attribute refinement chosen to be tested for
may be varied with the refinement of detail of data present
on the network the user has access to, or the level of re-
finement deemed necessary for the user to study emissions.
Thus, complete flexibility exists in describing the traffic
related behavioral aspects of the network as related to
emissions estimation.
B. SOFTWARE SYSTEM DEVELOPMENT
A discussion of the formal software system and interface with
SAPOLLUT should be prefaced with brief mention of the traditional
Urban Transportation Planning (UTP) background of SAPOLLUT. The
implication is that the model was designed to link to the FHWA pro-
gramming battery which performs the UTP process of trip generation,
trip distribution and traffic assignment, with the attendant problem
of using a loaded traffic assignment network versus actual ground
counts. The design of the UTP process makes it necessary to modify
some software usage in order to use realistic ground counts in
SAPOLLUT.
The completed software system correctly employing these modifica-
tions is shown in Figure 11. The system begins with the sequential
file of the network link records of data attributes (l.l). The
complete documentation of the format of this file is in Appendix A.
The system branch containing program modules 1.2, 1.3 and 1.4 is an
initialization routine executed only one time to set up the system's
-------
-34-
LINK RECORD FILE
(1.1)
FORMAT REVISION PROGRAM
(1.2)
NETWORK SENSITIVITY PROGRAM
(2.1)
NODE PAIR OUTPUT
(2.2)
INTERFACE PROGRAM
(3.0)
DATA MANAGEMENT PROGRAM
(1.3)
MODIFIED LINK FILE;
(1.4)
BUILDHR (FHWA)
(4.0)
\t
(5.0)
(FHWA)
HRMOD (FHWA)
(6.0)
SAPOLLUT
(7.1)
EMISSIONS OUTPUT SUMMARY/
(7.2)
FIGURE 11. Traffic Emissions Software System
-------
operation. The format revision program HRCVHT (1.2), takes the link
record file as input and transforms it into output which will eventually
be in a usable format for input into the FHWA program BTJILDHR in step 4.0.
As such, module 1.3» referred to as the Data Management Program,
accepts as input the link records from the Format Revision Program.
The program creates what is called an indexed sequential data set.
Essentially, this implies attaching a key to each link record so that
it may later be retrieved with a single command, obviating a search
procedure. The output of the program, termed a modified link (1.4)
file, is essentially the same as its input except for the above keyed
reorganization to speed access, it is important to emphasize that this
program does not make any functional changes in the link records and
its function could be performed elsewhere in the system. As noted
earlier, this program is only executed once for all runs of NETSEN.
The next program' in the software is the Interface Program (3«0)»
and as its name implies, it is the heart of the interfacing procedure.
It accepts input from two places. First, it accepts a node pair output
(2.2) from the NETSEN Program (2.1) which has passed all logic tests in
that program. It then uses this node pair as a key to retrieve with a
single statement the link record from 1.4 which is identified by the
node pair. The program then processes the link by producing a dummy
link to connect to its A-node if the previous link already processed
does not have a B-node which is the same as the current A-node. The
program then checks to see if the A-node is numerically less than the
B-node, and if so the two are reversed. Further action takes the sum
-------
of the A-B volume count and the B-A volume count and places this sum
minus 1000 in an A-B count location. The program then makes a series
of edit checks to assure that valid links or dummy links do not violate
any of the traffic assignment coding conventions for leg numbers and
to assure that the proper count volumes are passed to SAPOLLUT. The
output is a network compatible to the historical record building pro-
gram, BUILDHR, in 4.0.
The next program in the software system is the FHWA BUILDHR program.
The program accepts link records as input, and functionally, the program
performs edit checking on them for consistency in coding, ultimately
outputting a binary historical record for each link and also one for
each node. The types of edit checking done by BUILDHR include checking
for unusually long links, excessively large volume-to-capacity ratios
as well as duplicate node and leg numbers.
Another FHWA program, PRIKTHR (5«0), follows in the software system.
The program accepts the binary historical records as input from BUILDHR
and prepares a printed summary of information in the records which is
useful in checking for proper operation of preceding programs as well
as for interpreting the output of SAPOLLUT on the basis of a specific
set of links. Although this program is not functionally necessary
for operation of the system, it provides useful information at a small
cost.
The final interface program preceding SAPOLLUT is the FHWA program,
HRMOD (6.0). This program is necessary because of SAPOLLUT's orienta-
tion toward traffic assignment loaded networks. Although the basic
historical record has ground count information included in it, SAPOLLUT
-------
- 37 -
can not use it in its storage location. It is necessary to relocate
actual ground count data from their storage locations in the historical
record to those storage locations where traffic assignment 'oads would
normally be. The program KRMOD is used to shift the ground counts to
the locations where the loads are normally situated. It then outputs
this modified historical record to SAPOLLTJT.
The last program in the software system is the emissions model
SAPOLLUT (?.l). It receives the modified historical record from HBMOD
and several control cards as input. It then proceeds to compute three
types of emissions (HCt CO, 110 ) for three different area types (CBD,
jC
r)
Central City, Suburb) and two different functional classes, freeways and
arterials, using vehicle miles of travel and average speed input, the latter
developed by one of three alternative methods available to the program.
The emissions output is currently available only in the aggregate, broken
down by area type, hour of the day, and functional class across each emission
type. By further dividing gross kilograms of emissions by vehicle-miles
traveled it provides emissions in grams per vehicle-mile and grams per
p'assenger-mile, given an average auto occupancy level.
In its current state the software system is fully automated in
batch mode. Thus, when the network inventory is loaded onto a tape
or disk data set and the initialization programs (Format Revision
Program (1.2) and Data Management Program (1.3)) are run once, the
system can execute several runs of WETSEN, examining the network at
several levels of refinement, with one submittal to the computer. By
providing a series of control cards to NETSSN, separate member data
-------
- 38 -
sets for each run are created, saving the individual sets of node pairs
to "be processed all the way through SAPOLLUT on successive runs through
the system. This enables the user to rapidly analyze the network with
a set of pre-determined runs. It should be emphasized that no manual
interface is necessary during this process. The following chapter will
demonstrate the flexibility and solution properties of the system opera-
tion in documenting line sources for the St. Louis Air Quality Control
Region.
-------
Footnote? Chapter III
Program Documentation, Urban Transportation Planning, March 1972,
Federal Highway Administration. A loaded network is defined
as a transportation planning network in which the traffic volumes
on various links are the result of theoretical computation. A
ground count is defined as the actual recorded count of vehicles
on a roadway link, generally made by mechanical counting devices.
2
Central Business District is the downtown commercial core; the Central
City is the non-commercial downtown core area, and the suburban
area is the outlying area of lesser density.
-------
CHAPTER IV
DOCUMENTATION OF EMISSION LINE SOURCES
-------
CHAPTER IV
DOCUMBKTATION OP MISSION LUTE SOURCES
A. Introduction-Issues in Defining Line Sources
As previously discussed, useful definition of line sources hinges
on the capability to analyze the highway network and its traffic and
design attributes at varying levels of detail, depending on the data
availability and the level of spatial refinement sought in emissions
information from SAPOLLUT. As such, this chapter demonstrates the deve-
lopment of a very unrefined definition of line sources, termed gross
line sources, and the sequential refinement of such to a set of descriptors
termed ultimate line sources. Each of these extremes of definition are
consistent with the basic definition of a line source, given immediately
below in Section B.
B. Basic Definitions-Delineation of Gross Line Sources
The following basic definition of a line source was employed in
documenting emissions for the St. Louis Regional network:
"the smallest sentient of inventoried roadway depictable with a
given specific set of attributes for the roadway."
At the grossest level, the line sources were broken down in the categories
depicted by Table 1. The X's indicate which ADT ranges were used for
functional classes of roadway. In addition, separate baseline runs were
made to select all freeways, all principal arterials, and all minor
arterials within these ADT classes. The emissions produced by this
gross ADT breakdown of line sources are summarized in the following set
of figures and tables for 24-hour periods:
- 40 -
-------
- 41 -
1. Figure 12 represents those freeway links in each of the ADT
ranges from Table 1. Table 2 presents vehicle-miles travelled (VMT),
total emissions, and emission rates for freeways in each of these ADT
ranges.
2. Figure 13 represents those principal arterial links in each
of the ADT ranges from Table 1. Table $ presents WIT, total
emissions, and rates for principal arterials in each of these ADT ranges.
3. Figure 14 represents those minor arterial links in each of
the ADT ranges from Table 1. Table 4 presents VMT, total emissions
and rates for minor arterials in each of these ADT ranges.
A composite analysis of these tables leads to conclusions generally
consistent with the literature, subject to subtle interpretation which
must be employed when aggregating emissions information over all links
under study in a particular functional class. Total kg of emissions
over all categories (CO, NO , and EC) rank lowest for minor arterials,
• J\.
reflecting lowest VMT exposure. Both the freeways and principal
arterials have very similar VMT totals. However, the freeways exhibit
lower totals for CO and HC, and higher NO totals than the principal
-A.
arterials, reflecting consistently lower average speeds on the arterials.
The CO rates (grams/vehicle mile) show the most noticeable change across ADT
ranges for all functional classes, with the most sharp changes associated
with the last three ADT ranges in each class. This is apparently due to
the distinct reduction in average speed associated with links at these
particular ADT ranges nearing or exceeding saturation for their functional
class. HC and NO , however, show generally stable emissions totals across
.X.
all functional classes. The HC rate rises slightly, reflecting the aggre-
gate impact across all links in the network of reduced average speed
-------
ADT
Ranges
(Thousands)
Freeways
Principal Arterials
Minor Arterials
1-30
30-40
40-50
X
50-60
60-70
70-200
5-10
10-15
15-20
20-25
25-30
X
30-35
35-40
40-100
20-40
TABLE 1
ADT RANGES USED FOR GROSS
LEVEL LIKE SOURCE DEFINITION
-------
- 45-
GROSS LEVEL LINE SOURCES
(JTCCTIOIIAL CLASS-FREEWAYS)
SYSTEM 7001
IDT Ranges (Thousands)
1-50 IU4UI
30-40 ^=*
40-50 ca»
50-60 • « * •
60-70
70-200
-------
-44-
1-30
ADT Ranges (Thousands)
30-40 40-50 50-60
60-70 70-200 Total
VMT
(24 hours)
Emissions (kilc
grains /24-hr, pe
CO
NO
X
HC
Emissions
Rates
(grams per
vehicle-mile)
CO
N0x
^ HC
624,407
riod)
10,906
5,809
2,257
17
9
4
563,007
10,221
5,133
2,073
18
9
4
889,163
16,876
7,929
3,340
19
9
4
1,015,38^
20,849
8,701
3,956
21
Q
4
1,728,011
35,901
14,747
6,773
21
Q
4
2,774,162
60,37^
23,16^
11,126
22
8
4
7,594,134
155,128
65,482
29,525
TABLE 2
GROSS LEVEL LIKE SOURCE EMISSIONS
(FOECTIOML CLASS-FREEWAYS)
-------
-45-
T*
O*AP*»C SCAtf IN Miff
VfHsa.
SYSTEM 7001
„..„„„ LINE SOTIRCSS
(;"TOcrio:uL CIASS-PHIIICIPAL ARTERIALS) 10-15
15-20
20-25
25-50
30-35
35-40
40-100
tiff Ranges (Thousands)
?"~
-------
ro
ei
6
GH
O
C
^
ft!
4-^
O
E-i
O
O
rH
I
O
*3
o
1
10,
LO
to
/O
O
to
1
If"1-
CM
10
CM
1
O
CM
O
CM
rH
to
*~l
rH
O
rH
LO
ON
CO
C—
00
KJ~
•»
o
VO
to
Vl
LO
o
CM
VO
ON
•v?
to
rH
rH
CO
CM
CM
CM
VO
VO
O
KN
cT
ON
rH
#«
rH
LO,
^3-
vk
^-
vo
rH
rH
00
C—
CM
CO
rH
ON
O
f*^^
[^
LO
rH
CO
CM
LO
•s
CM
C-
vo
/*— s
m
l\
o
,r!
r ^ "^'
r ' i CM
[;. v_x
, — s
i
o
•H
!H
1 0
rH
•H fa
^W £3
^^ O
r^H
W |
r* ^J"
O CM
"ft ^***N»
03 CO
w E3
•H n?
f-^ ?H
w ^
ITN
•^j-
CO
^p
VO
CM
to
H
VO
ft
CO
CM
^
LO
rH
O
CM
VO
CM ,
C--
CM
LO
*^-
0
ho,
KN
«.
K>
l/^
00
^,
H
LO
CM
O
CO
CM
LO
LO
*,
H
CM
O
0
rH
r—
ON
rH
•^
\
CM
!O
H
ft
rH
CO
*^+-
£•—
O
^
rX;
1
VO
VO
CO
co"
0
00
ro
«^
CO
rH
ON
06"
CO
CO
•^
N^
W
o
r i
|
^j~ \
CM
00*"
K"\
J
LO
LO
H
•t
rH
O
VO
C~-
00
ON
KN
CO
vo
£
ON
pt
r—
o
rH
VO
f —
CM
CO
ON
r~
LO
CM
to
v>>
to
r.i
t
I
x«s
Ol
rH
W ;-\
n P
f3 W 1
u' '•:! .it,
':" 1 ^-^ t,!j
'•' ' ^
U ">
. , -J > J-l
• . (-1 l'l
l-i '!< Ol
VO
^d"
CM
^"
CM
CO
to
LO
•sf
IO
CM
to
O
C j
VO
vo
VO
VO
VO
.
VO
VO
VO
X
p
? 1
VO
VO
VO
LO
LO
LO
LO
LO
ff]
i
CO
o
PH
[
00
o
o
-46-
-------
- 47-
Banges (ThousandB)
SYSTEM 7001
GROSS LEVEL LIKE SOURCSS '
(RJNCTIONAL CLASS-MINOR ARTERIALS) 15_2Q
20-40
(
-------
- 48 -
ADT Ranges (Thousands)
5-10
-] 5
15-20
20-40 Total
VMT
(24 hours)
Emissions (kill
grams/24-hour ;
CO
NO
X
HC
Emissions
Rat GO (c/voh,-
mile)
CO
NOX
HC
612,870
period)
20,330
3,496
3,076
35
6
5
594,058
19,713
3,369
2,994
33
6
5
319,406
12,611
1,779
1,744
39
6
5
191,680
9,695
1,012
1,193
51
5
6
1,718,014
62,349
9,656
9,007
TABLE 4
GROSS LEVEL LIME SOURCE EMISSIONS
(.FUNCTIONAL GLASS-MIHOR ARTKRIALS)
-------
- 49 -
due to increased flow. The NO rate shows some reductions in the latter
JL
ADT ranges of each class. This is apparently due to reductions in the
average speed on these higher volume sets of links which operate at or
near saturated levels of congestion.
C. Refined Level - Ultimate Definition of Line Sources
The most refined level of line source definition used involved
classifying the links "by narrow ranges of ADT and combinations of
special characteristics defining their attributes. These attributes
have been noted in depth in Chapters II and III. The following para-
graphs show some examples of typical 24-hour graphic and tabular informa-
tion which form the ultimate definition of line sources for the St. Louis
Air Quality Control Region. Detailed commentary will be reserved until
synthesis of information occurs in Table 17.
The initial presentation in Table 5 exhibits the basic .ADT
ranges used in conjunction with the various combinations of attributes.
Figure 15 represents those freeway links in the various ADT ranges
for which none of the attributes of special topography, capacity
alterations, or sensitive land use were present. Table 6 is an
emissions summary of these links from Figure 15. Further, Figure 16
depicts those freeway links in various ADT ranges for which capacity
alterations were present but not special topography or sensitive land
use attributes. Their emissions are summarized in Table 7. Likewise,
Figure 17 depicts those freeway links in various ADT ranges for which
both special topography and capacity alterations were present, but
sensitive land use attributes were not. Table 8 summarizes emissions
for these particular links. Figure 18 r:presents those freeway links,
-------
in various ADT ranges, for which the combination of attributes of
special topography, capacity alterations and sensitive land use were
present. Their emissions are summarized in Table 9 .
The next component series of figures and tables illustrates the
most refined level of line source definition for principal arterials,
using ADT ranges as noted in Table 5 . Figure 19 exhibits those
principal arterial links in various ADT ranges which do not have any
of the attributes of capacity alteration, progressive movement, or
sensitive land use present. Table 10 summarizes emissions for
these links. Figure 20 depicts those principal arterial links in
various ADT ranges which had attributes of capacity alteration, but no
attributes of progressive movement, or sensitive land use. Their emis-
sions are summarized in Table 11. Further, Figure 21 illustrates
those principal arterial links in various ADT ranges which have attributes
of capacity alteration and progressive movement,but not sensitive land
use. The emissions summary for these links is shown in Table 12.
Figure 22 exhibits those principal arterial links in various ADT
ranges which have all the attributes of capacity alteration, progres-
sive movement, and sensitive land use. Their emissions are summarized
in Table 1?.
The final example of line source definitions is composed of
minor arterials in the ADT ranges previously noted in Table 5. Figure
2J illustrates those minor arterial links in various ADT ranges which
did not have any attributes of capacity alterations, progressive move-
ment, or sensitive land use. Their emissions are summarized in
Table 14« Figure 24 shows those minor arterial links in various
-------
-51-
ADT
Ranges
(Thousands)
Freeways
Principal Arterials
Minor Arterials
5-10
10-15
15-20
20-25
X
20-40
25-30
30-35
X
35-40
40-45
45-50
50-55
55-60
60-65
65-70
X
70-200
X
40-100
TABLE 5
APT RANGES USED FOR ULTIMATE
LINE SOURCE DEFINITION
-------
-52-
1OT Ranges (Thousands)
30-55
55-40
Attributes! Special Topography i not present 40-45 *
Capacity Alterations: not present 45-50
Sensitive Land Use: not present 50-55
65-70 .'.V
70-200 w/m
-------
•XJ
CO
0)
to
i-H
-P
O
EH
O
0
CM
ci
f-
o
*!"
in
vo
VO
vo
o
VO
in
in
in
1
o
in
o
in
£
in
Vl
^
O
1
10
I
Cl)
N^
i>-
vo
*
rH
VO
CO
a\
in*
CM
in.
^^*
C7\
«^
CM
rH
O
m
CM
KN
KN
IfN
rH
O
0
O
O
CM
VO
•*^*
•»
O
ON
^^*N
n
jL|
r-j
b
TJ
O
•H
I q>
O P
H
'H ^
v_x 0
CO |
S ^~
O CM
•H ~\
0 CO
H
CM
*
CM
CM
CM
rH
00
ON
in
VO
o
^
CO
ON
CM
0
o
o
0
t—
ON
VO
•k
CM
O
ON
rH
*vT
y*
o
r—
^
rH
ON
O
00
O
O
in
en
rH
O
O
O
O
0
ON
VO
M
p
f •*
00
CM
«.
rH
VO
r-T
ON
O
00
O
CM
00
m
o
0
0
o
CM
^*
O
1
•
'p
«""C
f,\ ^0
O ' — '
•1-1
CO ^} Ci
"-J -P rH
m
KN
m
o
ON
rH
O
O
O
o
o
KN
O
O
VO
VO
O
ON
o
o
o
o
00
w
0
-
VO
o
^
o
o
o
o
, in
p
c
0)
CO
• o
M
CO
VO
H
W
s
•8
0) -P
M C
_ ft £
, . ft rt ct
EH CO O K
O
ca
0)
-p
- 53 -
-------
-54-
..,*'
iOT Ranges (Thousands)
ULTIMATE UHE SOTjnCSS JO-J5 W8
(fUTCTIOIIAL CLASS-?SSEtfAYS) 35-40 »=«=•
Attributes: Special Topographyi not present 40-45 •«•«
Capacity Alterations: all categories 45-50 ^5$s
Sensitive Land Use: not present 50-55 t-i~j
55-60 -v.v
60-65 35-CQr
65-70 ..V.'
70-200 '/////M
-------
n
t5
§
M
§
£
CO
p
fca
rH
CVJ
-P
O
EH
0
0
CM
ci
C-
O
1
IT
O
VD
O
1
in
in
in
in
0
if-
0
in
i
in
3
o
in
in
o
if*
CO
CO
in
vo
CO
rH
O
0
O
O
CO
CO
••
in
0
0
CM
KN
KN
rH
O
0
O
d
(J
£J
O
p"* ^^"
L-4 CM
r" v~-'
s~~.
0
•H
fH
\
rH
•H !4
N~ ' O
W |
£j -3-
O CM
t: Vl
C'J S
•/-( CD
rj H
M ci3
in
CO
CM
O
0
O
O
in
in
O
CM
in
. CM
o
O
o
o
0
^
i-H
in
rH
0
o
0
0
CM
CM
ON
CO
rH
0
O
0
W
p
V£>
^
O
O
0
0
VO
o
o
in
o
0
0
0
u:
it
Q)
Wv.
f? fcO
o > —
•rl
'Jl 0) O
•H -<•> rH
ti d -H
H a; u
o
o
o
o
3
CJN
rH
0
0
O
O
U
O
O
o
o
c—
ON
o
o
o
M
*- ,
o
0
o
o
VO
^.
o
o
o
a
•H
I-
s
CO
O
M
CO
CO
O
O
CO
M
PH
-p HO-P
C CD C!
W -P Q>
w d M
o o
ftHft
rH
-p d -P
o o
c .. c
CO
£> O Q>
,C -H W
ft-P !=>
d d
H fn ti
Cl> S
ftrH »3
O «J
EH co
rH -P -H
Ci iH +"
•H O --H
O d CO
0) ft C
ft d o
co o co
62
0>
-P
- 55 -
-------
- 56-
/
IDT Ranges (Thousands)
ULTIMATE LIKE SOURCES JO-J5 «W
(r,~T:io!:AL CLASS-FHSJ-WAYS) 55-40 *=s
Attributes I Special Topography: all categories 40-45 •»••
Capacity Alterations: all categories 45-50
Sensitive Land Use: not present 50-55
55-60
60-65
65-70
70-200
-------
§
W
g
VI
C.I
to
EH
rH
ni
•P
o
EH
0
o
CM
£
o
f-
lA
vo
LT>
VO
|
O
vo
o
vo
1
l/N
LfN
0
LO
O
tr-
lO
,
VO
rH
0
C—
• v^
/^^N
rrj
O
•rH
rH
1 G)
O Pi
H
•rH ?H
0
43
« 1
O CM
"rH ^^^
11 W
I'J £
•rH Cu
H F
|JH ^3
rH
00
f-
KN
KN
rH
ON
o
VO*
rH
IfN
ON
CM
cT
rH
O
H
^-
00
•<=)-
vo
KN
O
O
00
LfN
t-
CM
CM
CM
LfN
O
O
ON
ON
rH
C—
00
O
•v
CM
VO
f—
00
rH
^
CM
H
CM
VO
IfN
rH
o .
O
CM
rH
CM
H
H
i-T
ON
VO
CM
W
0
JL^
o
CO
"*
VO
rH
0
CM
0
O
•«*•
rH
O
t—
^
ON
vo
o
o
vo
ON
KN
LfN
CO
0
rH
CJ
|
^
"^1
O "^
•rH
n ti ^-— ,
w o O
•r-l -I' rH
i ' ri •'!
!u (-: f '
o
IfN
o
IfN
CM
o
CM
o
o
LfN
CM
00
rH
CO
i-H
O
0
vo
f-
00
ON
o
o
CO
ON
ON
.,
(y
K-i
vo
IfN
^
o
0
-
»
f 5
fl'l
00
EH
CO
O
M
CO
CO
CO
W
CO 'H rt
SS-F
0
to
CD
>3 O
X! -H
PH -P
c3 co
fH rH 'O
tD O C
<; O -P rt
3 p
l
t
3
a
H
•rH -P
O -H
ri M
CQ
•H
H
- 57 -
-------
- 58-
JtDT Ranges (Thousands)
E LINE SOURCES 30- J5
(/;:.CI'IC::AL CLASS- FRIEV/AYS) 35-40 =====
Attribute3i Special Topography: all categories 40-45 ••»»
Capacity Alterations: all categories 45-50 ^t^Z
Sensitive Land Use: all categories 50-55 "=» , - *..»v\
55-60 «v.v 60-65 3.-W33
65-70 .-.V
70-200
-------
K
-tf
(-*
a
w
2
o
g
CO
o
to
i
0,«
-p
o
EH
70-200
o
r-
1
Uv
VO
tf>
vo
|
o
VO
o
VO
IT-
1
0
UN
O
IT,
LT.
vi-
0
I
rX
IT-,
O
NN
CM
O
-=t-
CO
rH
VO
CM
M
CO*
MA
rH
CM
rH
CM
r—
VO
CM
CO
in
rH
•>
rH
rH
O
O
0
ON
00
in
r*-
rH
O
W
S
0
rj
£-•* *Nj"
|; CM
0
•H
rl
1 «>
O ft
rH
•rl M
*— ' O
M I
O CM
.r| ^v^
C3 W
li B
vo
CM
CM
C-
ON
CM
t~-
IfN
^J-
KN
00
CO
ITN
rH
IfN
0
0
O
VO
VO
VO
0
0
0
t—
VO
r-
fOl
ON
VD
CM
CO
O
O
•k
rH
MA
ON
CM
CM
H
C—
O
O
O
VO
CM
rH
O
C3*
y -
rH
CO
O
CM*
VO
rH
!
•rl -P rH
•M ? "^
CM
KN
KN
CM
CM
ON
O
O
O
CO
o
o
CJ
r—
ON
vo
o
o
o
r-
o
X
P
in
^
VO
o
o
o
m
o
o
M
CO
H
CO
CO
w
I
CO
I
EH
0)
ra -H to
•H P -rl
rl §0 M
0) rt a>
-P O -P
cd n3
O rH O
Co'rH nJ rH
rH rH
rt •• Cd
CQ
r-3 O 0)
..•ass
I <\) cd
CO f-| >H TJ
oo to
-------
- 60 -
SYSTEM 7001
AM1 Ganges (Thousands)
ULTIMATE LINE SOURCES 5-10 n—a
(i'Tnc'i'io.;AL CLAss-riirrciFAL ARTERIALS) 10-15 mmm
C, Attributes: Capacity Alterations: not present 1^-20 ~s=s:
Progresoive Kovcnent: not present 20-25 ••«»»
Sensitive Land Usej not present 25-30 "fz&S
JO-35 •*•••%•
J5-40
40-100
-------
ra
•ti
§
w
o
to
rH
ti
O
EH
O
O
rH
d,
•v!
O
If
in
in
f*"v-
1
0
O
I
in
OJ
in
CM
C
(M
O
OJ
.A
rH
IfN
r-i
d>
rH
O
i-!
lA
^-
H
VO
t—
vo
ON
ON
,_!
rH
ON
r—
i-l
KN
CM
rH
XH^
^,
VO
CM
t—
vo
*t
00
vo
vo
rH
CM
rH
CM
O
N"N
£
t—
ON
r-T
CM
ON
CO
VO
VO
9t
O
""^
~
^
0
r£Tj
\—\ *cj-
£>'•£!.
TlJ
O
•H
1 0)
O PH
•H ^
^W r-J
v— ' O
CO 1
g CM"
•ri "^"^
CJ "
CO §
•r) P
(3 ^
ON
VO
ON
ON
CM
rH
ON
-st
• 00
CM
t-
0
rH
CM
VO
CM
*,
^-
H
t—
^
rT
O
00
^
H
CM
O
O
ON
o
H
O
O
00
C—
o
CM
l>-
o
VO
KN
i-H
rH
N^N
^
rH
C—
0
•v
CM
t—
KN
CM
in
g
KN
KN
CM
in
CM
CM
in
*
CM
W
O
t*-— t
CM
rH
0
ON
rH
in
vo
rH
rH
O
O
in
r-T
in
ON
ON
•s
rH
0
CO
^
^
C—
vo
CM
KN,
VO
"?
CO
f—
H
•»
CM
O
£M
0)
H
11 -rl
q H
ti d t
fl ft •
O 1(0 .1'!
1" J>
U C!)
•'H "i J J- 1
} ,'J 'T
111 f'n Ml
5
rH
ON
ON
KN
KN
£
K^
CM
KN
O
O
in
vo
in
vo
vo
vo
vo
vo
X
P
vo
in
vo
in
in
in
in
in
o
W
o
iH
CO
H
CO
CO
O
I
CO
H
PH -H
H -f
O nj
a
-P -p
c c
0) 0) -P
w w C
0) 0) 0)
rl fH W
ft ft CO
rl
•P -P ft
O O
C«S
CO
CO
to -p
c c ••
O <1> 0)
?:
O TJ
-HO)
>
-H
•H Q) H-"
O
td _
ft o ..
rt fn CD
O PH CO
H -
oO
O CO
0)
H
-p
-61-
-------
- 62-
(VSTEM 7001
IDT Ranges (Thousands)
ULTIMATE LIKE SOURCES 5-10 «="»
(FutiCTioriAL CLAS-vpFiinciFAL AOTERIALS) 10-15 flwtw-
Attributes: Capacity Alterations: all categories 15-20
Progresnive Movement: not present 20-25
Sensitive Land Use: not present 25-JO
50-55
55-40
40-100
-------
- 63 -
ADT ranges which have attributes of capacity alteration, but no pro-
gressive movement or sensitive land use. Emissions from the link sub-
set are summarized in Table 15. Figure 25 depicts those minor
arterial links in various ADT ranges which have attributes of capacity
alteration and progressive movement, but no sensitive land use attributes.
Table 16 summarizes these particular emissions. Finally, there are no
minor arterial links in various ADT ranges which have all the attri-
butes of capacity alteration, progressive movement, and sensitive land use,
The synthesis of all possible component attribute groups over all func-
tional classes represents the most refined and accurate level of line
source definition for the St. Louis Air Quality Control Region on the
basis of current network data. Discussion of the implications of this
synthesis will occur in the immediately following section.
!>• Synthesis of Line Source Information
The synthesis of 24-hour detailed definition and description of line
source information provided by NBTSEN, and their resulting emissions
computations from SAPOLLUT is shown in Table 17« The aggregate
network of line sources encompasses approximately 1,370 miles of roadway,
with a total of nearly 17 million vehicle miles of travel daily.
Approximately 45$ of this VMT exposure occurs on freeways, 44$ on
principal arterials, and the remaining 11$ on minor arterials. In
terms of mileage, 195 miles is composed of freeway line sources,
representing the freeway corridors of the region, and the principal
and minor arterials comprise the remaining 1,175 miles of line sources.
The aggregate 24-hour emissions are 482,322 kg. of 00,117,109 kg. of NO
.A-
and 76,777 kg. of HC. A component analysis in Table 18 shows that
-------
• 1
-p
o
E-t
O
o
rH
O
to
to
1
O
KN
C
10,
t
u
CM
10
CM
0
CM
O
CM
H
to.
rH
A
rH
O
rH
1
L»
vo
rH
KN
•,
CO
to,
vo
O
O
ON
0
rH
to,
CO
*^~
•s
CO
CO
o
o
to,
CM
CO
to,
ON
vo
•s
C—
ITN
rH
CM
r—
CM
i-H
rH
CO
ITN
ON
o
VO
rH
O
0
CM
•,
LP>
CM
^
W
J-i
r|
O
.^•i
n ~j
^
o
•H
rl
1 fl)
O ft
1 — 1
v-* O
CO |
^ ^J*
O CM
•rf'**1^
w vi
w §
*ri Cu
t ] t-(
(*:i ^2
rH
10,
CM
O
CM
10,
CO
O
»,
K^
vo
vo
rH
to,
ON
• ON
rH
ON
rH
"*
vo
ON
CM
CO
O
o
rH
VO
to,
O
CM
VO
to,
ON
^j-
0
vo
o
t—
CO
t—
CM
VO
VO
O
ON
to,
rH
X
C_)
* *
c~-
to,
to,
O
IfN
CO
^~
to,
^J"
•^~
O
CO
ON
ON
CM
CM
00
CM
rH
O
1 ' -4
O
EJ 'fl
t) i] i
V 1 f-1 *
'"•* ' ' ,^"i
"r-f '*_rf' *"^
1 * ^"
01 O
•i -j -5-J ? j
1 ."1 (!)
O
to,
to,
ON
10,
CO
10,
CO
to.
t—
to,
hfN
to,
to,
O
c >
o
vo
vo
vo
VO
vo
VO
VO
X
*o
K-1
o
«
LTN
IP,
ITN
10,
10,
ir\
r>
1-1
to
0)
•H
fj JJ
p C
^-N ft) 0> -P
CO 0) CO C
)J -p QJ Q)
*S rf fH W
I-H O ft 0)
^
, , rH -P ft
EH rH O
W d C-P
co +>
. . f! C ••
H O CD O
O -H B CO
'~ 4»
M O Ti
s»l
rH CU t-H
^ri ^
-H > co >
O HJ W -H
•H CD HJ
O H -H
rt to w
ft o C
ns j-i 0)
O PH CO
o
CO
-------
-65-
1OT Banges (Thousands)
ULTIMATE HNS SOURCES 5-10 t=ss
CTIOI.'AL CLAS3-rHi::ciPAL ARTERIALS) 10-15 ttntntn
Attributes! Capacity Alterations! all categories 15-20 ^==w
Progressive Hovenent: all categ-ories 20-25
Sensitive Land Use: not present 25-30
30-35 .V.V
35-40 ie>:-y-:e
40-100 '/////j&.
-------
rH
ci
o
£:'
.
O
0
^
q
0
NN
O
T
If •
CM
LO
CJ
CM
15-20
IT,
1
rH
O
rH
t--
f-
CM
KN
rH
O
0
o
t-
"*
o
C—
VO
CO
+,
vo
o
o
CO
t—
rH
0
rS
•»
00
o
s~*\
V)
£-j
P
o
n v^
f-4
oi o
;1 ^ fi
M Cl C)
rH w P,
o
o
o
t_
o
t--
KN
o
o
o
o
o
vo
VO
o
-
vo
o
cf
o
0
IfN
IfN
o
VO
UN
o
P
CM
0} CO
0) 0)
•H -H
M
IALS
ate
ate
sent
s
g
CO
l-H
T
O O C)
rH rH P<
AR
a
a
t
C C
•H O
-P OT -H
•H (U -P
O fi-t CO
o
•H
•P
- 66 -
-------
- 67-
AOT fiancee (Thousands)
ULTIMATE LIKE SOURCES 5-10
U:;AL CUSS-PRIMCIFAL ARTSKIALS) 10-15
Attributes: Capacity Alterations: all categories 15-20
Progressive Movement: all cate
-------
03
1
03
8
rH
crt
-P
O
EH
40-100
o
10
K>
o
CM
10
CM
d,
CM
O
CM
to
rH
H
rH
O
rH
0
CM
*,
00
0
O
o
o
o
o
CM
CO
O
W
fj
p
o
£H ^sj~
*£j CV
p '*~s
o
•H
1 o
0 Pi
rH
•H fj
<^ 0
-•?
C "3-
O CM
•H \
03 W
03 S
•H Co
_E3 H
W G
c—
CM
O
O
O
o
o
o
r—
c-
o
o
o
vo
*^t"
o
o
o
o
o
0
$
o
ox
CM
*^~
0
o
o
o
o
o
CM
O
g
03 -H
03 CO 1
C K •
03 Q)
•tH -P !^f
B cd O
[2 w ft
O
o
o
o
0
o
*
o
o
o
o
0
o
o
0
o
v£>
o
g*
o
o
o
0
0
o
in
o
g
10!
rH
EH
m 03
0) QJ
iH **H
*i ^|
CO
S
SSI
0
o
CO
i
EH
ERI
0> CO
-P -P
td rt
o o •
o
a
H
OjrH
'rf
W -P
C C "
O O 0>
•H B M
rH 0) KH
^ -H O
>5 W >
+> W iH
•H
-------
-69-
M1MK SCAU Ifl »»!«
SYSTEM 7001
l/LTIMATK Ll.'fS EOUHCBS
-< (FUHCTI01IAL CLASS-MIiiOa Affl'ERIALS)
Attributes! Capacity Altcrationo: not nreoent
progreooivo Kovonient: not present
Sensitive Land Uaei not preaont
ADT Ranges (Thousands)
5-10 tm
10-15 tattmt
15.20 »«•«•
20-40 i-J5
-------
- 70-
ADT Kanr;o3 (Thousands)
5-10
10-35
35-20
20-40
Total
Vffl!
(24 hours)
Euissions (kilc
grams/24-hour j
CO
NO
X
HO
Emissions
Raton ({./voh.-
irdle)
CO
NOX
KG
539,917
eriod)
18,020
3,078
2,717
33
6
5
488,099
16,144
2,772
2,455
33
6
5
218,731
9,043
1,216
1,218
41
6
6
90,525
4,357
488
547
48
5
6
1,377,272
47,564
7,554
6,937
TABLE 14
ULTIMATE LUTE SOURCE HUSSIONS
(FUNCTIONAL CLASS-MINOR ARTERIALS)
Attributes: Capacity Alterations: not present
Progressive Movement: not present
Sensitive Land Use: not present
-------
- 71 -
ADT Ranges (Thousands)
STSTEM 7001
ULTIMATE LIKE SOURCES 5-10
(KUNCTICWAL cuss-ranon AOTEPJALS) 10-15
Attrituteni Capacity Aiterationo: all catecories 15-20
Progressivo Movement: not present 20-40
Sensitive Land Une: not present
-------
- 72 -
ADT Ranges (Thousands)
10-V
15-20
20-40
Total
VMT
(24 hours)
Emissions (kilo-
grams/24-hour pi
CO
NO
X
HC
Emissions
Ratec (g/veh.-
mile)
CO
1IOX
HC
18,032
uriod)
575
103
89
32
6
5
24,161
839
135
125
35
6
5
81,595
2,944
455
430
36
6
5
33,458
1,985
175
22?
59
5
7
157,246
6,343
868
8?1
TABLE 15
ULTIMATE LIIIE SOURCE EMISSIONS
(FUNCTIONAL CLASS-Hi:roR ARTEHIALS)
Attributes: Capacity Alterations: all categories
Progressive Movement: not present
Sensitive Land Use: not present
-------
- 75-
ALT Ranges (Thousands)
ULTIMATE LINE SOURCES ^-10 «=*
(IWCTIOKAL CLASS-KIIIOR AKTEIUALS) 10-15 tntrmr
Attributes: Capacity Alterationn: all cateeorlos 15-20
Progressive Movements all catef^ries 20-40
Sensitive Land Daei not present
-------
- 74 -
ADT Ranges (Thousands)
5-10
10-15
15-20
20-40
Total
7MT
(24 hours)
Emissions (kilc
grams/24-hour j
CO
NO
HC
Emissions
Rates (g/veh.-
mile)
CO
NO
HC
0
eriod)
0
0
0
0
0
0
16,947
622
93
90
37
6
5
19,080
624
108
96
33
6
5
61,305
2,935
315
373
48
5
6
97,332
4,181
516
559
TABLE 16
ULTIMATE LINE SOURCE EMISSIONS
(FUNCTIONAL CLASS-MII-TOB ARTEHIALS)
Attributes: Capacity Alterations: all categories
Progressive Movement: all categories
Sensitive Land TJse: not present
-------
- 75 -
freeway line sources consistently contribute J>0-60fo to totals in all
emission types, with typically ^Gfo of total contribution emanating
from arterial sources, and typically near 12$ being- contributed from
minor arterial operations. These percentages correlate closely with
the logic of discussion of emission stimuli of average speed and VMT
referred to in Chapter I, and component analytic discussions earlier
in this immediate chapter. The average rates are aggregates over all
ADT ranges considered in each of the specific functional classes
under study. They show a disturbing stability across all functional
classes, essentially due to the aggregation and averaging of specific V/C,
average speed and VMT ranges such as those displayed in Tables 6 through
16. The research team feels that more meaningful and accurate rates exist
at the dissaggregate levels such as those shown in Tables 6 through 16
where rates specific to ADT ranges for particular components of the line
sources containing specific attributes are exhibited.
E. Related Processing Costs
The traffic emissions software system developed herein is a series
of eight basic programs coupled to a variable number of utility programs.
Two of the eight basic programs are run only once. As such, this dis-
cussion will deal only with the remaining six involved in software processing
with SAPOLLUT. It is theoretically possible to make 95 separate runs
of the system under one batch job, however, only four runs were made
in one job for the duration of the project, in order to facilitate
turn around time. A threshhold fixed cost figure for one run through
the system which produced emissions on only line link would be approxi-
mately $3«00. This cost represents the //stem overhead and individual
-------
-76-
Functional Class
Freeway
Principal
Arterials
Minor
Arterials
Totals
Total VMT*
(24 hours)
Total Emissions
(kilograms /24-hc
CO
N0x
HC
Average Rate
of Emissions
(g. /vehicle-
mile)
CO
NO
X
HC
7,594,154
ur period)
155,128
65,482
29,525
7,448,789
264,3^3
41,971
38,245
i
20
9
4
36
6
5
1,718,014
62,349
9,656
9,007
36
6
5
16,760,937
482,322
117,109
76,777
*For all links with volumes reported.
TABLE 17
SUMMARY OF ULTIMATE LIES
SOURCE EMISSIONS INFORMATION
-------
Percent
Contribution
- 77 -
Principal Minor
Freeway Arterials Arterials Total
CO
NO
X
HC
32$
56$
38/o
55$
36$
50$
13$
8$
12$
100$
100$
100$
TABLE 18
PERCENTAGE OF TOTAL EMISSIONS
CONTRIBUTED BY FUNCTIONAL
CLASS OF LI1IE SOURCE
-------
- 78 -
program overhead costs. Due to the rather large number of combinations
of link attributes which may be tested for, results of using the system
on the St. Louis Air Quality Control Region yield a typical itiaTn'Ti"™ cost
for one pass of $8.00, with an average cost of $5 to $6. Table 19
illustrates a typical run of the system showing time requirement,
percentage of total time, and the amount of main core storage required
for each program. It is interesting to note that SAPOLLUT required
only 34»5$ of all Central Processing Unit (CPU) time. The cost of the
run is computed on the rate shown, and charges for lines printed
(approximately $1 per run) should be added to the CPU time cost.
If four or more passes through the system are made, which is typical
of the operating rationale used in the performance of the research on
the St. Louis Air Quality Control Region, the cost drops to $6.30 due
to the spreading of overhead costs over several runs. It is felt
this cost figure is quite tolerable, given the level of detail possible
for examining the network, and the nominal number of runs required to
synthesize information over all functional highway classes. All figures
are for an IBM S/360 Model 65 machine.
F. Limited Sensitivity Analysis
Limited amount of sensitivity analyses were performed subsequent
to the definition of line sources, to examine gross changes in the
supply of highway facilities and resultant alterations of traffic flow
and emissions. Two separate analyses relating to 24-hour periods were performed,
-------
- 79 -
Program
CPU* Time % Core Required**
NETSEN
RUMMY
BUILDER
PRINTER
ERMOD
SAPOLLUT
MISC. UTILITIES
TOTALS
8.59
1.8?
4.25
7.60
1.82
14.35
3.06
41.54
20.7
4.5
10.2
18.3
4.4
34.5
7.4
100.0
60
46
82
40
42
60
34
*Central Processing Unit, in seconds
**in K's (1024 bytes)
***A11 Freeways, 145 links
TABLE 19
COST SUMMARY FOR TYPICAL
SOFTWARE SYSTEM RUN***
-------
- 80 -
typical of near-term alteration of emissions estimates which might occur
as typical modification of corridor facilities occurs in one case, re-
quiring the placement of total corridor loads onto other corridors in
the region during construction alterations. In the second case, a
badly needed distributor facility is completed, as highway planning is
brought progressively to the implementation and completion phase, yield-
ing resulting diversion of presently congested crosstown flows onto the
new high-type design facility.
The first analysis was the examination of emissions performance re-
sulting from the deletion of availability of a freeway corridor and the resultant
loading of the displaced traffic onto two adjacent freeways. This was
performed through deletion of 1-44 from 1-55 to 1-244. An average ADT
was taken for 1-44 and half of this was loaded onto each link of 1-55
from 1-44 to 1-244 and the remaining half onto each link of U.S. 40
from 1-55 to 1-244. Figure 26 shows the location of these three
freeways. Table 20' summarizes original VMT's and emissions on
each freeway and their aggregate. It should be noted that because
of limitations in Mfl/A Battery program BUILDER, only a maximum ADT
of 99,999 can be used. Therefore, since some links on U.S. 40 are
either currently near or over 100,000 ADT, or would exceed 100,000
-------
FIGOBE 26
HMSSIOHS SDKHARY FOB FBEEWAT COHRIBOR DITERSI05
1-55, IT.S. 40 >;>£0ffi
1-44 %V.V
-------
- 82 -
Freeway
1-55
Original
1-44
Original
U.S. 40
Original
Total
Original
Facilities
1-55, U.S. 40
with Diversion
VMT
(24 hours)
Emissions
(kilograms/24-hot;
CO
NO
X
HC
Emissions
Bates
(g./veh.-mile)
CO
N°x
HC
789,093
r period)
17,763
6,452
3,220
23
8
4
718,290
'1,092,597
,
15,131
6,070
2,834
21
8
4
24,473
8,987
4,446
22
8
4
2,599,980
57,367
21,509
10, 500
2,502,875
59,534
19,982
10,513
24
8
4
TABLE 20
MISSIONS SUMMARY FOR FREEWAY CORRIDOR DIVERSION
-------
- 83 -
ADT with diversion, the VMT's for U.S. 40 are underestimates in
this sensitivity analysis.
The results reflect the result of eliminating a corridor from
use with current traffic loads, or the present impact if 1-44 had
not been completed. Again, the results are consistent with the
literature with respect to impact of increased VMT and lowered average
speed as saturation is approached on remaining 1-55 and U.S. 40 after
diversion. Total CO emissions are higher, total HC emissions are
slightly higher, reflecting the above speed and VMT issues, and total
N0xis lower, reflecting impact of lowered operating speeds. Again,
the total YMT's shown diverted, and resulting increase in emissions
output is understated, due to the FHWA input limitations on ADT dis-
cussed above. The aggregate rates have raised slightly for CO, reflect-
ing increased VMT loads on the remaining 1-55 and U.S. 40 corridors,
however, the lowered average speed component resulting from these
flow increases on each corridor have yielded a stable aggregate NO
Jx
rate. The aggregate combinations of altered speeds and changed VMT's
over all links of both remaining corridors appeared to have a balancing
effect on HC rates, with no apparent aggregate rate change.
The second sensitivity analysis examined the Innerbelt (Mo. 725)
North-South corridor. One component section of this route is complete,
and the analysis centered around hypothetically extending it North and
South to become an effective North-South Freeway Corridor, and divert-
ing traffic onto it from parallel high volume arterials. Figure 27
illustrates the hypothetical Innerbelt Freeway and the parallel
arterials under study. The analysis consisted of diverting 50$ of the
traffic off each arterial and placing it on the Innerbelt. Baseline
-------
- 64 -
SYSTEM 7001
PIOTEE 27
DQSSIOKS STOMART TOE PARALLEL ARTERIAL DIVBIiSIOH
Faxallel Arterials HllllU
Hypothetical Innerbolt • • •
-------
- 85 -
Boutes
(1)
Parallel Arterials
before Diversion
(2)
Parallel Arterials
after Diversion
(3)
Hypothetical
Innerbelt
(2) + (3)
VMT
(24 hours)
Emissions
(kilograms/24-ho-u
CO
NOX
HC
Emissions
Bates
(g./veh.-mile)
CO
NOX
HC
451,042
r period)
15,513
2,565
2,299
34
. 6
5
220,896
6,811
1,279
1,069
31
6
5
261,992
5,079
2,312
993
19
9
4
482,888
11,890
3,591
2,062
25
7
4
TABLE 21
EMISSIONS SUMMARY FOR
PARALLEL ARTERIAL DIVERSION
-------
- 86 -
runs were made of arterial street emissions before and after diversion,
and the hypothetical presently loaded Innerbelt. Table 21 summarizes
the results of this sensitivity analysis. Present provision of the
new freeway corridor facility, in light of current parallel arterial
operations, yields a drop in total emissions of CO and HC, when compared
to operation of arterials alone in column 1. Again, this is due to
diversion of part of the aggregate traffic load to a facility with
higher average speeds. Likewise, this increase in speed yields
higher total N^ emissions. The rates on the Innerbelt, given the
constancy of total WIT and provision of higher operating speeds, have
dropped as expected for CO and HC, and increased slightly for NO ,
JC
when compared to arterial rates before diversion.
-------
CHAPTER V
CONCLUSION
-------
CHAPTER V
CONCLUSION
In concluding the reported research, it is relevant to point out
the capabilities and information which have been provided, comment on
further needed research to advance the state of the art, and discuss
the general status of line source emissions modelling with respect to
current knowledge about related traffic engineering phenomena.
A. Use and Applicability of Present Research and Modelling Results
The data analysis, model development, and integration with SAPOLLUT
has yielded several tangible outputs. They are:
1.) A complete quantitative and relevant qualitative data base
for freeways, principal arterials and minor arterials in the St. Louis
Air Quality Control Region.
2.) A model format - NETSEN - which is capable of sorting and
describing any subset of components of the above traffic network, at
continually varying levels of detail, from gross geographic description
of volumes only, to highly refined geographic locations possessing
multiple attributes of traffic, geometric design, topographic, control
and land use conditions which are significant in the link's operation
and its relationship to emissions and/or air quality.
3.) The interface of the above format with the present operating
rationale of SAPOLLUT, thus yielding the capability to model and
describe the total emissions of CO, N0xand HC emanating from a network
described at a desired level of attributes, which have been used as
input to NETSEN.
4«) Based on the most refined use of the above in the St. Louis
Area, an extremely accurate statement of the line sources. This statement
- 87 -
-------
- 88 -
encompasses the sources, descriptions, attributes and total emissions
resulting from 1,370 miles of roadway, composed of 195 miles of freeways,
and 1,175 miles of principal and minor arterials. The complete descrip-
tion and discussion of these line sources is contained in Chapter IV.
5.) A capability to perform link by link sensitivity analysis
on the types of attributes existing on one individual link, a corridor
of several links, or an area of several individual street links. Thus,
design characteristics and/or traffic loads may be altered or eliminated,
for purposes of using SAPOLLUT to yield the resulting changes in
aggregate emissions.
B. Recommendations for Further Research
As is expected from intensive research on any problem, the activities
of data collection, logic development and model construction of NETSEN,
and construction of its interface with SAPOLLUT, have yielded some in-
sights into current gaps in the state of the art of merging information
on the traffic and network phenomena with present emissions modelling.
As such, several specific items represent areas of new or improved
research focus which the research team feels should be pursued in future
activity on this topic. They are:
-------
- 89 -
1.) Future use of a model such as SAPOLLTJT for basic research on
current emissions behavior should be pursued in a highly different
software format than SAPOLLTPT currently employs. Specifically:
a.) Use of any format requiring a loaded assignment
network, including assignment on present networks should
not be considered due to:
1.) The complexities of intermediate software.
2.) The theoretical issues surrounding the validity
of assignment model forecast volumes, or
assignment model loaded volumes on the
present network vs. current ground count
data. This is particularly relevant in the
St. Louis Area at this time, due to vague
relationships of assignment output vs. actual
future foreseeable auto and transit networks
in the region, as the comprehensive transporta-
tion planning begins a period of revision.
2.) The study team further encourages the development of a
capability to use a highly detailed network descriptor model such
as NETSEN, with an emissions model which can be interfaced directly
with one of the output parameters from NETSEN without requiring an
intermediate software battery to change input form. Preferably, the
variable input from NSTSEN to the emissions model would be a traffic
flow theoretic variable having significance to both the tagged links
in NETSEN and the emissions computation process.
-------
3.) To this immediate end, it is recommended that NETSEN be run
in conjunction with the Modal Emissions Model, which employs speed
modes as input to emissions computations. The speed mode concept,
as part of the traffic flow theoretic envelope of speed and delay
studies, and acceleration noise, it is a relevant and meaningful traffic
flow parameter, and can be output as another link descriptor in NETSEN.
4.) Further, appropriate development, collection and use of speed
mode or speed profile data should be undertaken by interested profes-
sional groups. Such activity is currently underway in the St. Louis
Area, through contracts to East-West Gateway from the Federal Highway
Administration and the Department of Transportation Systems Center.
The activity focuses on driving patterns throughout the metropolitan
area and the inherent speed profiles, volumes and spot speeds in such
driving patterns. Effort should be directed to matching speed mode and
profiles to links with specific groups of attributes, thus facilitating
the capability of using speed mode as the critical transfer parameter
from NETSEN through emission computations in a Modal Emissions Format.
Comprehensive results relating research to speed modes should be
possible through use of relevant field collection information to date,
synthesized with appropriate use of the literature and flow-theoretic
computations,
5.) Development and use of the capability to output emissions
information on a link by link mapping is necessary. Current SAPOLLUT
output is aggregate emissions by area type and functional class, rendering
investigation of emissions intensity and sensitivity analysis somewhat
cumbersome at the individual link and corridor level. It appears
-------
- 91 -
that employment of the Modal Emissions Model in a manner discussed in
the previous paragraphs has the capability of allowing mapping of
emissions output at an individual link level over the entire network
entered for study.
C. Closing Comments on Status of Line Source-Traffic Attribute Modelling
In final conclusion, effort should be directed toward detailed
filtering out of locations of attributes, and measurement of resultant
emissions at these link locations, thus cataloging the simultaneous
impact of these attributes on emissions. Further, the categorization
of traffic operation on facilities should proceed by capturing theoretically
sound aspects of flow activity categorized by situation type which are of
relevance to emissions, such as queuing and delay descriptions at inter-
sections, and shock wave phenomena on uninterrupted flow links and freeway
bottlenecks. Thus, a mapping of network description, refined traffic
flow parameters, and emissions will ultimately result, yielding a
comprehensive format from which to investigate and calculate emissions.
-------
- 92 -
Footnotes Chapter V
Automotive Exhaust Emissions Modal Analysis Model. SPA No. 460/3-74-005i
United States Environmental Protection Agency, Office of Air and
Water Control Program, Office of Mobile Source Air Pollution Control,
Certification and Surveillance, Ann Arbor, Michigan, January, 1974.
-------
BIBLIOGRAPHY
-------
Selected Research Bibliography
A Study of Traffic Flow on a Restricted Facility, Irterim Report
PhaseOne, Department of Civil Engineering, University of Maryland,
College Park, Maryland, June 1973*
"Air Pollution Controls for Urban Transportation," Highway Research
Record 465, Highway Research Board, National Research Council, 1973«
"Air Quality and Environmental Factors," Transportation Research Record
492. Transportation Research Board, National Research Council, 1974.
"An Introduction to Traffic Flow Theory," Highway Research Board
Special Report 79. Highway Research Board, National Research Council,
1964.
Automotive Exhaust Emission Modal Analysis Model. SPA Ho. 460/3-74-005.
Ur.ited Gtatos Environmental Protection Agency, Office of Air and
Water Control Programs, Office of Mibile Source Air Pollution Control,
Certification and Surveillance Division, Ann Arbor, Michigan, January, 1974.
Design of An Urban Speed Characteristics Study, Research Triangle
Institute, Center for Development and Resource Planning, May 1974.
Drew, Donald R., Traffic Flow Theory and Control. McGraw-Hill, 1968.
"Highway Capacity Manual," Highway Research Board Special Report 87.
National Research Council, 1965.
"Highways and Air Quality," Highway Research Board Special Report 141,
Highway Research Board, National Research Council, 1973*
Hillier, Fredrick S. and Liebermann, Gerald J., Introduction to Operations
Research, Holden-Day, Inc., 1970.
Littman, Fred E., Semrau, Konrad T., Rubin, Sylvan, Dabberdt, Walter P.,
A Regional Air Pollution Study (RAPS) Preliminary Emissions Inventory,
Stanford Research Institute, Menlo Park, California, January 1974.
Rossano, A. J. Jr., Ed., Air Pollution Control Guidebook for Management.
Environmental Science Services Division, E.R.A. Inc., Stamford, Conn.,
1969.
Scott Research Laboratories, Incorporated; Malcom Smith, Dovelopment of
Representative Driving Patterns at Various Average Route Speeds;
Prepared for the Environmental Protection Ardency, Office of
Adminis+ra+ion, Research Triangle Par!;, North Carolina, SPA Contract
Number 66-02-1501 (6-73), February 11, 1974, San Bernadino, California.
- 93 -
-------
- 94 -
Selected Research Bibliography (Continued)
"Social, Economic, and Environmental Factors in Transportation," Highway
Research Record 336, Highway Research Board, National Research
Council, 1971.
Special Area Analysis Final Manual, Federal Highway Administration,
Urban Mass Transit Association, Federal Aviation Administration,
Office of the Assistant Secretary for Policy, Plans, and Interna-
tional Affairs, August 1973.
"Traffic Engineering A Tool to Reduce Air Pollution," Traffic
Engineering Magazine, Vol, 44» No. 9» June 1974.
Venezia, Ronald A., "The Impact of Transportation Alternatives on
Ambient Air Quality," Unpublished Ph.D. Dissertation, Washington
University, January 1972.
-------
- 95 -
St. Louis Area Traffic Data Sources
Peak ilour Travel Analys5-3, St. Louis Area. East-West Gateway Coordinating
Council, F'jbru^ry 1974.
Headway Functional Classification Study for the Saint Louis Area.
East-Vest Gateway Coordinating Council, July 1973.
3tat\is of Missouri State Hi^iways by Routes & Systems. Missouri State
Highway Department, December Jl, 1973*
St. Louis Area Transportation Study. Streets. Highways and Transit.
East-West Gateway Coordinating Council.
St. Louis City Traffic Volume Studies. 1973* St. Louis City Department of
Streets, Ilighwcy Division.
St. Louis County Traffic Volume Studies. 1973* St. Louis County Department
of Highways and Traffic, April 1974.
St. Louis Metropolitan Area Traffic Volume Summary. March 1973-April 1974.
Missouri State Highway Department, May 1974.
Traffic Characteristics on Illinois Highways. State of Illinois Department
of transportation, 1972.
-------
•"vofessional Interviews
Bob Weldin^r
Lave Sohmi.lt
Donald C. Pri: ster, Jr.
?aul Huang
Dick S, T. rlsu
J-iCk iLretser
Bob Vatson
Tom Dalton
James P. Eaumann
Frank Kriz
Tom Dollous
Harold Ruffner
Fred Bartlesmeyer
Ted
Richard Wilcox
Brown
Missouri State Highway Department, Division
of Planning, Tral'fis Section, Jefferson City, Mo,
City of St. Charles, Ko.
Geoinet, Rockville, Maryland
East-West Gateway Coordinating Council,
St. Louis, Mo.
Chief Traffic Engineer, St. Louis County
Division of Highways & Traffic
Deputy Traffic Commissioner, City of St. Louis
Missouri State Highw?.y Department, Division of
Traffic, District VI Office, Kirkwood, Mo.
Illinois Department of Transportation,
Traffic Division, East St. Louis District Office
Illinois Dapartment of Transportation,
Planning Division, East St. Louis District Office
-------