United States Office of Mobile Source Air Pollution Control EPA 460/3-85-004
Environmental Protection Emission Control Technology Division March 1985
Agency 2565 Plymouth Road
Ann Arbor, Michigan 48105
Air
Recommended Revisions to
Gaseous Emission Factors From
Several Classes of Off-Highway
Mobile Sources
-------
EPA 460/3-85-004
Recommended Revisions to Gaseous
Emission Factors From Several Classes of Off-
Highway Mobile Sources
by
Melvin N. Ingalls
Southwest Research Institute
6220 Culebra Road
San Antonio, Texas 78284
Contract No. 68-03-3162
Work Assignment 8
EPA Project Officer: Craig A. Harvey
EPA Branch Technical Representative: Robert I. Bruetsch
Prepared for
ENVIRONMENTAL PROTECTION AGENCY
Office of Mobile Source Air Pollution Control
Emission Control Technology Division
2565 Plymouth Road
Ann Arbor, Michigan 48105
March 1985
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are available
free of charge to Federal employees, current contractors and grantees, and
nonprofit organizations - in limited quantities - from the Library Service
Office, Environmental Protection Agency, 2565 Plymouth Road, Ann Arbor,
Michigan 48105.
This report was furnished to the Environmental Protection Agency by Southwest
Research Institute, 6220 Culebra Road, San Antonio, Texas, in fulfillment of
Work Assignment No. 8 of Contract No. 68-03-3162. The contents of this
report are reproduced herein as received from Southwest Research Institute.
The opinions, findings, and conclusions expressed are those of the author and
not necessarily those of the Environmental Protection Agency. Mention of
company or product names is not to be considered as an endorsement by the
Environmental Protection Agency.
Publication No. EPA 460/3-85-004
ii
-------
FOREWORD
This project was conducted for the U.S. Environmental Protection Agency
by the Department of Emissions Research of Southwest Research Institute. The
project was begun in August 1983 and completed in September 1984. The
project was conducted under Work Assignment 8 of Contract 68-03-3162, and
was identified within Southwest Research Institute as Project 03-7338-008.
Mr, Robert J. Gar be of the Emission Control Technology Division, Office
of Mobile Source Air Pollution Control, Environmental Protection Agency, Ann
Arbor, Michigan, served as EPA Project Officer until August 1984. Mr. Craig
A. Harvey of the same office served as Project Officer after August 1984. Mr.
Robert I. Bruetsch, of the same EPA Division was the Branch Technical
Representative. Mr. Charles T. Hare, Manager, Advanced Technology,
Department of Emissions Research, Southwest Research Institute, served as the
Project Manager. The project was under the supervision of Melvin N. Ingalls,
Senior Research Engineer, who served as Project Leader and principal
investigator.
iii
-------
TABLE OF CONTENTS
Page
FOREWORD iii
LIST OF FIGURES vii
LIST OF TABLES viii
SUMMARY xi
I. INTRODUCTION 1
II. LOCOMOTIVES 5
III. CONSTRUCTION EQUIPMENT 33
IV. MARINE VESSELS 39
V. DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS 65
REFERENCES 69
APPENDICES
A. EXAMPLE OF PLOTS FROM THE EKMA COMPUTER
PROGRAM SHOWING THE RELATIONSHIP BETWEEN HC,
NOX AND O3
B. SUPPORTING MATERIAL FOR LOCOMOTIVE EMISSION
FACTOR DEVELOPMENT AND IMPACT DETERMINATION
C. SUPPORTING MATERIAL FOR MARINE EMISSION FACTOR
DEVELOPMENT AND IMPACT DETERMINATION
D. AP-42 SECTION CONTAINING EMISSION FACTORS FOR
LOCOMOTIVES, CONSTRUCTION EQUIPMENT AND
MARINE VESSELS
E. LIST OF PERSONS CONTACTED
v
-------
LIST OF FIGURES
Figure Page
II-1 Average Cost of Railroad Diesel Fuel by Year 7
II-2 Locomotive Deliveries by Year 9
II-3 Age Distribution of U.S. Locomotive Fleet as of 1981 10
IV-1 Bunker Prices, London 40
IV-2 Type of Propulsion System in Active U.S. Flag Vessels
by Year Launched 43
IV-3 Emission Levels from Marine Diesel Engines with Design
Speeds Under 600 RPM 48
IV-4 HC Emissions from Four Medium-Speed Marine Diesel
Engines 51
IV-5 CO Emissions from Four Medium-Speed Marine Diesel
Engines 52
IV-6 NOx Emissions from Four Medium-Speed Marine Diesel
Engines 53
IV-7 Typical Brake Specific NOx Map—Four-Stroke Cycle
Turbocharged Marine Engine 58
IV-8 Typical Brake Specific NOx Map-Two-Stroke Cycle Turbo-
Charged Marine Engine 59
IV-9 Hypothetical WHEC Pitch Schedule, Propulsion Systems,
Inc.,® Propeller 61
vii
-------
LIST OF TABLES
Table Page
II-l Large Diesel Electric Locomotives Produced in the
United States 11
11-2 Distribution of Locomotives by Manufacturer and
Horsepower 12
II-3 Changes in Diesel Electric Locomotives - 1972 to 1981 13
II-4 Current Locomotive Duty Cycles Used for Line Haul and
Switching Locomotives 14
11-5 Revised Line Haul Locomotive Duty Cycle 15
II-6 Revised Switching Locomotive Duty Cycle 16
11-7 Published Studies of Measured Locomotive Emissions 17
II-8 Chronological List of Locomotive Emission Studies Used 18
II-9 Measured Locomotive Emissions from Published Studies
Expressed in Grams per Horsepower Hour on Line Haul
Cycles 20
11-10 Line Haul Locomotive Emissions Based on Current Duty
Cycles Emission Factors 22
11-11 Recommended Line Haul Locomotive Emission Factors
Based on SwRl 1983 Operating Cycle 23
11-12 Recommended Switching Locomotive Emission Factors
Based on SwRl 1983 Operating Cycle 24
11-13 Distribution of EMD Engines by Year of Manufacture,
Type of Engine and Engine Model 25
11-14 Distribution of G.E. Engines by Year of Manufacturer 26
11-15 National Average Locomotive Emission Factors 26
11-16 National Impact of Locomotive Emissions 27
11-17 Annual Locomotive Emissions in Several Regions 29
11-18 Locomotive Contributions to Area Total Annual
Emissions 30
11-19 Miles of Railroad Track Per Thousand Square Miles of
Area for Several States 32
viii
-------
LIST OF TABLES (CONT'D)
Table Page
III-l Sales Weighted Diesel Emission Factors for
Construction and Industrial Equipment from CAL/ERT
Study 35
III-2 Comparison of Present AP-42 Construction
Equipment Diesel Emission Factors with CAL/ERT
Study Diesel Emission Factors 36
III-3 Impact of Construction Equipment on Air Quality in
California 36
III-4 Estimated National Impact of Construction
Equipment Emissions 37
IV-1 Marine Fuel Terminology 41
IV-2 Diesel Engines Having Measured Emissions in Draft of
Report EPA 450/4-84-001 45
IV-3 Recommended Emission Factors for Marine Diesel
Engines with Design Speed Below 600 RPM 47
IV-4 Relationships of Typical Tonnage, Vessel Type, and
Propulsion Unit Capacity 50
IV-5 Marine Vessels in U.S. Waters Classed for Air
Pollution Studies 56
IV-6 Marine Emissions Impact on Three Areas 62
i
I
ix
-------
SUMMARY
This study examined three categories of off-highway mobile emission
sources to determine current emission factors of hydrocarbons (HC), carbon
monoxide (CO), and oxides of nitrogen (NOx), together with the national and
regional impact of these sources. The three sources examined were
locomotives, construction equipment, and marine vessels. The emission factors
for these sources, as listed in the EPA publication, "Compilation of Emission
Factors" (generally referred to by its original report number, AP-42), are
approximately ten years old. A literature search was conducted to identify
changes in engine design and operation over the past ten years that would cause
changes in emission factors. Additional measured emission data were also
sought to broaden the data bases on which emission factors for these sources
were based.
Locomotives
For locomotives, it was found that rising fuel prices had led to
improvements in engine design as well as changes in railroad operation. The
principal operating change was to curtail the practice of running locomotives at
idle when not in use. With locomotives being shut down more often when not in
use, the duty cycles on which the locomotive emission factors were based
needed to be changed. From a calculated estimate of line haul and switch
engine operating time, new locomotive duty cycles were developed. The
literature search found additional locomotive emission test data not used in the
current listing of locomotive emission factors. Using the new duty cycles and
the additional emissions test data, new locomotive emission factors were
developed. A comparison of the new national average emission factors and the
current AP-42 factors in terms of pounds of pollutant per 1000 gallons of fuel is
shown below.
Locomotive National Average
HC
CO
NOy
Revised locomotive estimate
41.3
187
533
Current AP-42 locomotive factors
94
130
370
The revised factors are recommended as replacements for the current AP-42
locomotive emission factors.
The new national locomotive emission factors, together with the total
railroad fuel usage in 1981, were used to calculate the national annual
locomotive emissions of HC, CO, and NOx. These emissions in terms of
percentages of emissions from all sources for each pollutant, and in terms of
the percentage of emissions from mobile sources are:
xi
-------
Annual Locomotive Emissions As;
Percent of Percent of
All Sources Mobile Sources
HC 0.3 0.9
CO OA 0.5
NOx 4.6 11
Three previous studies of locomotive emissions in several regions of the
country were examined. In general, the HC and CO emissions from locomotives
were well below two percent of HC or CO emissions from all sources in the
regions examined. Locomotive NOx emissions were higher percentages of NOx
emissions from all sources in each area. The Chicago area had one of the
highest locomotive contributions to total NOx, ranging from 3.6 to 8.1 percent
of total NOx emissions, depending on the method used to determine the
locomotive emissions. As stationary and on-highway NOx controls become
more effective in the future, these percentages could increase.
Construction Equipment
Early in the literature search for information on construction equipment,
it was learned that the State of California had recently completed a study that
included construction equipment. In response to that study, and as an input to
the study, a consortium of industry groups sponsored a study of its own. This
industry study, known as the CAL/ERT study, produced a set of construction
equipment diesel emission factors based on the most comprehensive survey of
construction equipment emissions to date. These diesel emissions factors do
not differ greatly from the construction equipment emission factors in AP-42.
Nevertheless, the new factors are recommended as replacements to the current
AP-42 factors, because of the more extensive data base used. Each type of
construction equipment (wheeled tractor, crane, etc.) has its own emission
factor. However, an unweighted average was calculated for both the new
recommended factors and the AP-42 factors. The comparisons of the averages
and the ranges of HC, CO, and NOx between the new factors and the AP-42
factors are shown below.
Diesel Emission Factors, g/bhp hr
HC CO NOy
Current AP-42
Const. Equip. Avg. 0.83 2.70 11.7
CAL/ERT
Const. Equip. Avg. 0.90 3.80 10.3
Current AP-42
Const. Equip. Range 0.36 to 1.39 1.80 to HA0 6.6 to 15.7
CAL/ERT
Const. Equip. Range 0.36 to 1.80 1.54 to 7.80 6.6 to 14.7
xii
-------
The CARB estimates of construction equipment annual emissions in
California were scaled to provide an estimate of nationwide construction
equipment annual emissions, as shown below.
Construction Equipment Nationwide Annual Emission As A:
Percent of All Sources Percent of Mobile Sources
HC 0.2 0.6
CO 0.6 0.7
NOx 2.0 4.7
Several areas in California were examined in the CAL/ERT study as well as by
the California Air Resources Board (CARB) for regional impact of construction
equipment emissions. In the South Coast Air Basin (Los Angeles area) and
Fresno County the impact were:
Construction Equipment
Emissions as Percent
of Total Species Emissions
HC CO NOy
0.08 0.20 0.91
CAL/ERT Estimate,
South Coast Air Basin
CARB Estimate,
South Coast Air Basin 0.25 0.64 2.45
CAL/ERT Estimate,
Fresno County 0.18 0.60 2.20
Marine Vessels
Marine vessels have also been affected by the rise in fuel costs over the
past ten years. One of the more significant changes caused by this fuel cost
change was the increased use of diesel powered vessels by the U.S. merchant
fleet. Marine diesel engine manufacturers are also working to enable their
engines to operate on poorer grades of fuel. The marine emission factors in
AP-42 are based primarily on emission tests conducted in 1972 on 13 Coast
Guard Cutters. There are few emission studies in the literature since that
time. None of the literature surveyed provided sufficient valid information to
determine new marine diesel engine emission factors. In addition there is
insufficient usable information available to define either the numbers of various
classes of marine vessels or the number of the various makes and models of
diesel engines.
An estimate of national impact of marine vessels was obtained from
railroad impact using the fact that inland waterway traffic carries about 42
percent of the freight railroads carry, and requires approximately 81 percent of
xiii
-------
the fuel per revenue ton-mile. The coastal and foreign shipping contributions
were then included, to provide the estimate of marine vessels emission
contribution to national emissions shown below. On a regional basis, marine
vessels were estimated to contribute 1.5 percent of the total NOx in the New
York area, and possibly as much as 2.5 percent of the total NOx in the Houston
area.
National Annual Marine Vessel Emissions As:
Percent of Percent of
All Sources Mobile Sources
HC 0.2 0.6
CO 0.2 0.2
NOx 2.1 5.0
Conclusions and Recommendations
The three off-highway emission sources investigated in this study are not
yet major sources of gaseous pollutants. However, as more on-highway vehicles
meet the current and future emission standards, uncontrolled off-highway
emission sources will increase in importance. At the present time, the
information required to accurately assess the pollution contribution of these
off- highway sources is not available. While marine vessels require the most
research to satisfy these information needs, railroad locomotives and
construction equipment also require additional research. In fact, this study has
shown that while marine emissions are based on the weakest data, emissions
factors from the other two categories are not much better.
For locomotives, the most important information need is for a definition
of locomotive duty cycles (in terms of hours per day in each throttle notch).
The duty cycles should be based on a twenty-four hour day and include the time
the engine is shut down. For construction equipment, where emission factors
are based on new equipment, there is a need for in-service emissions
measurements. These in-service measurements would allow the emission
factors for new equipment to be adjusted to reflect actual emissions in the
field.
It is recommended that a survey be made to determine how marine vessels
should be classed for emission studies. The population of each class in U.S.
waters should be obtained. In addition, the diesel engine population by
manufacturer, model, and design horsepower should be obtained. Typical duty
cycles in terms of power indication on the vessel (1/3, standard, full, etc.)
should be defined from data collected during actual in-harbor operation of each
class of vessel and the more popular engines. Finally, measurement of actual
marine diesei exhaust emissions and fuel consumption is required at the engine
operating conditions defined by the duty cycles.
xiv
-------
I. INTRODUCTION
Reliable information on air pollutant emissions from all sources is
required by Federal, state, and local government environmental officials for air
quality planning, standard setting, and evaluation of control requirements and
strategies. Such information is also required by various public and private
sector personnel performing site-specifc environmental impact analyses related
to proposed construction projects. Information on emissions of pollutants from
sources is usually in the form of emission factors. An emission factor is the
measure of the rate at which a single source entity emits pollutants while
performing its intended work function. Emission factors from almost all
conceivable sources have been assembled in the EPA document, "Compilation of
Air Pollutant Emission Factors" (AP-42).^' This document has become the
standard reference for individuals and organizations requiring pollutant
emission information.
In the category of Mobile Sources, on-highway vehicles, because of their
great numbers, have received the most attention. There are currently national
gaseous emissions standards for passenger cars, motorcycles, light trucks, and
heavy duty truck engines. Off-highway mobile sources have received less
attention, since past studies have shown them to be less important air pollutant
sources. Currently, of all off-highway sources, only aircraft engines have
regulations for gaseous emissions.
Studies are now projecting that off-highway sources will become a larger
part of mobile source emissions, especially NOx emissions, in the future. In
counties which do not meet one or more of the gaseous pollutant ambient air
quality standards, off-highway mobile source emissions may require closer
scrutiny. The latest edition (March 1981) of AP-42 (as the emission factor
document has traditionally been called) did not include revision of the off-
highway mobile source emission factors. Obviously, if off-highway mobile
sources become a more significant part of the total air pollution burden,
accurate off-highway emission factors are required. This project was
performed to review the AP-42 gaseous emission factors for several off-
highway mobile source categories.
Objective
The objective of this study was to determine if any changes have occurred
in gaseous emission levels from several off-highway mobile source categories
since the last listing of their emission factors in AP-^2. Only the regulated
gaseous emissions, HC, CO and NOx, were included in this study. If new
emission factors were needed, they were recommended. The magnitude of each
source category's contribution to the total pollutant emissions in representative
regions was also determined. Recommendations were also made as to what, if
any, additional testing, data collection or modeling is needed for these major
off-highway sources.
~Numbers in parentheses designate references at the end of the report.
-------
Selection of Source Categories
There are numerous categories of off-highway mobile sources, including:
• locomotives
• aircraft (military and commercial)
• marine vessels (military, shipping and fishing)
• farm equipment
• construction equipment
• industrial equipment
• lawn and garden implements
• transport refrigeration units
• off-highway recreational vehicles
• recreational aircraft (general aviation)
• inboard recreational marine craft
• outboard marine craft
• snowmobiles
• helicopters
The time and resources available to this project did not permit an
investigation of emission factors from all categories of off-highway mobile
sources. It was decided to select three of the major off-highway mobile source
categories for study. After discussions with the EPA Branch Technical
Representative, locomotives, construction equipment and marine vessels were
selected for study. Each of these categories is discussed in a separate section
of the body of this report.
Off-Highway Mobile Sources and Photochemical Oxidants
At this point, a brief comment is in order to point out that the effect of
the off-highway sources upon ambient photochemical chemical oxident levels is
not covered in the report, and to explain the reason these effects are not
covered. Two of the gaseous emissions from off-highway mobile sources HC
and NOx contribute to the formation of photochemical oxidants in the ambient
air. The amount of ozone(03) in the air is taken as a measure of the
photochemical oxidants level. There is currently an ambient air quality
standard for ozone, but none for HC. It is due to the relationship between HC
and ozone that there are HC emission standards for on-highway mobile sources.
Hydrocarbons and NOx interact in the formation of ozone, but the interaction is
complex and dependent on a number of variables other than the concentrations
of HC and NOx. In general, increasing the ambient HC concentration at
constant NOx concentration either has no effect, or increases O3
concentrations. The effect of NOx and O3 formation varies with HC
concentration. For a constant high ambient HC concentration, an NOx
concentration increase is associated with an O3 concentration increase. At low
ambient HC concentrations, an NOx concentration increase can be associated
with an O3 concentration increase, no change in O3, or an O3 decrease.
Determining the effect of HC and NOx emission changes on ozone
concentration requires the use of a computer model. One popular model is the
EPA area-specific EKMA (Empirical Kinetic Modeling Approach) model. Plots
from the EKMA program showing the relationship between HC, NOx and O3 for
2
-------
Fresno County and Sacramento, California, are included in Appendix A as
examples of the complex relationship between the three pollutants. The
estimation of ozone concentrations for an area in the future requires modeling
that area with assumptions about all possible air pollution sources. Such a task
is beyond the effort allotted for this study. Therefore, in examining each of the
off-highway mobile sources, the effect on ozone concentrations due to changes
in emission levels from these sources will not be discussed. It should be pointed
out, however, that since most of the off-highway mobile sources investigated in
this project are powered by diesel engines, the NOx emissions are far larger
than the HC and CO emissions. In addition, while in 1982 there were only 11
counties in the country that exceeded the ambient NOx air quality standard,
there were ^77 counties where all or part of the county exceeded the ambient
O3 air quality standard.
3
-------
II. LOCOMOTIVES
While railroads are not presently the extensive mode of transportation
they were 75 years ago, they still carry over 35 percent of the nation's freight,
making them an important off-highway emission source. Since the early 1960's
railroad locomotives in the U.S. have been almost exclusively diesel-powered.
Over the past 20 years, there have been only a few models of locomotive diesel
engines used in the United States. Operationally, the locomotive diesel is
controlled by a throttle with eight distinct positions or "notches", in addition to
idle and dynamic braking settings. These facts make the evaluation of
locomotive emissions somewhat less complicated than other off-highway
sources, since emission measurements are required at only ten different
throttle settings of a few models of engines.
The time spent in each of these throttle notches determines the total
amount of gaseous pollutants put into the air on a daily basis. This time-
throttle position schedule is referred to as a duty cycle, and is comparable to
the vehicle speed-time driving cycles used for automobile emissions. The
determination of average locomotive duty cycles for line haul and switch
engines is an important and integral part of determining locomotive emission
factors.
After reviewing the locomotive emission factors in AP-42, a thorough
search was conducted for published literature on locomotive emissions, changes
in locomotive engines since 1972, changes in railroad operating practices, and
statistics on number, size and manufacturer of locomotives in service. The
search started with work done at SwRI, including literature collected by various
individuals at SwRI working on locomotive engine studies. As part of this
effort, a computerized search was conducted on the following data bases: SAE
reports, AS ME reports, NTIS, Doctoral Dissertations, and Engineering Index.
The computerized search resulted in 583 abstracts, of which 28 appeared to be
useful. Of the 28, 20 were on hand. The remaining eight were obtained for this
project. The NTIS published search entitled "Emission Factors" was also
consulted. The annual review and forecast issues of "Railway Age" magazine
since 1970 were consulted for changes in locomotives, locomotive engines, and
railway operations.
As part of the literature search, the EPA library at Research Triangle
Park, which has a complete collection of OAQPS reports, was visited for a
stack search of any reports not found by other means. In addition,
representatives, of the U.S. suppliers of locomotives were contacted for
information on their products. The Association of American Railroads (AAR)
was contacted for emissions information and statistics. The Federal Railway
Administration was also contacted for any information it might have.
AP-^2 Emission Factors
The locomotive emission factors in AP-^2 were last updated in 1973.
Locomotive emission factors in AP-^2 are presented for a national average
locomotive mix. Emission factors are also presented for two-stroke blower
scavenged engines on both switching and road operating cycles, two-stroke
5
-------
turbocharged engines on road cycles, and four-stroke engines on both switching
and road cycles. These emissions are presented in terms of pounds per 1000
gallons of fuel, kilograms per 1000 liters of fuel, grams per horsepower hour,
and grams per metric horsepower hour. The HC, CO, and NOx emission factors
were apparently derived from a study of three locomotive engines done at SwRI
in 1972/2) jn addition, and approximate load factors of 0.4 was given for the
road cycle, and a load factor of 0.06 for the switching cycle.
Recent Trends in Railroad Operations and Locomotive Engines
The changes in the railroad industry that affect emissions are in two
areas: railroad operations and locomotive engines. The literature and verbal
information resulting from the information search indicated that there had been
a great deal of change in the railroad industry over the past ten years. To begin
with, there were fewer railroad companies in 1982 than there in 1972, mostly as
a result of mergers. Railroads are divided in several classes by the Interstate
Commerce Commission, depending on operating revenue. In 1965 Class I
railroads had operating revenues above 5 million dollars. In 1976 the threshold
for Class I railroads was raised to 10 million dollars, and in 1978 to 50 million
dollars. The 1983 operating revenue threshold for Class I railroads was
approximately 82 million dollars. In 1970 there were 71 Class I line haul
railroad companies and 273 Class II railroad companies.^' In 1979 there were
t+0 Class I and 23 Class II railroad companies.'^) By 3une 1982, the Class I
railroad companies had dropped to 33 companies.'^'
The miles of track have also decreased from 331,129 in 1972 to 290,000 in
1980. Yet the revenue ton-miles of freight hauled increased from 771,168x10^
in 1970 to 926,000x10^ in 1981. The revenue passenger miles during the same
period increased slightly from 10,903x10^ to 11,800x10^.^
Over the past ten years, the railroads have made a serious effort to
reduce their fuel consumption. The reason for this effort is the sharp increase
in fuel price, as shown in Figure II-1. In 1981, the average cost of a gallon of
railroad diesel fuel was over nine times that for 1971J5) The success of these
fuel-saving efforts is shown by the fact that Class I railroads hauled 22 oercent
more ton-miles of freight per gallon of fuel in 1981 than they did in 1971.(5'
From an operational standpoint, the railroads instituted a number of
changes to reduce fuel spillage and waste/6) The change that affects emissions
most is a change in the practice of allowing locomotives to idle when not in use.
When fuel was cheap, it had been the practice of railroads never to shut down a
locomotive in service. This practice was the result of design factors of
locomotive diesels, which make start-up and warm-up a labor-intensive, time-
consuming process. However, as fuel costs rose, the economies of having
engines idling when not in use were examined. Many railroads instituted a
policy of shutting down their locomotives if possible. While the guidelines vary
from railroad to railroad, locomotives are in general shut down if they will not
be needed within a following time period varying from one half to four hours.
This practice is used as long as the ambient temperature is above some
minimum, usually between 40° and 50°F.
The number of locomotives has remained relatively constant over the past
ten years. In 1971 there were 27,189 locomotives in service, of which 26,897
6
-------
90
80
70
60
50
40
30
20
10
0
i
"I » l I I 1 I I L
160
1965
1 I I I I I
1970 1975
Year
J I I I 1 I I I I I
1980
1985
Figure II-l. Average cost of railroad diesel fuel by year
-------
were diesel electric units.^ In 1981, there were 28,067 locomotives in service,
of which 27,981 were diesel electric units. While the number of locomotives
has increased only slightly, the total horsepower available has increased from
54.2 million in 1971 to 65 million in 1981.'*) The number of new locomotives
delivered each year since 1972 is shown in Figure II-2. The age distribution is
shown in Figure II-3. Note that 25 percent of the locomotives have been built
since 1975, while approximately 28 percent of the locomotives in service were
built before 1960.
There are currently only two manufacturers of locomotives in the United
States, General Electric (G.E.) and Electro Motive Division of GM (EMD), and
one in Canada (Bombardier). This has been the situation since 1969, when Alco
ceased production in the U.S. Alco engines have continued to be used in
Canada, first by MLW, then by its successor, Bombardier. Occasionally over
the past ten years MLW/Bombardier has sold a few engines to U.S. Railroads;
but all the remaining locomotives are either EMD or G.E., with EMD having the
larger share of the market. Table II-1 lists the locomotives models currently
sold by EMD and G.E.
Ten years ago it was possible to obtain statistics of the number of
locomotives by size and manufacturer from the periodical, "Railway
Locomotives and Cars." This magazine ceased publication in 1974. Since that
time, it has been difficult to obtain statistics on the manufacturer and
horsepower of locomotives in service. In a telephone conversation with Mr.
Richard Cataldi of the Association of American Railroads (AAT), it was learned
that sometime in 1984 the AAR will have a computerized data base of
locomotives entitled "ULMER LOCOMOTIVE".^) This data will be able to
provide a breakdown of locomotives by manufacturer and horsepower.
However, it will not be available for this project.
As a possible source of this information, Mr. Cataldi suggested the
publication "Diesel Locomotive Rosters, U.S., Canada and Mexico".'**) This
booklet is published for railroad hobbyists, and was available from a local hobby
shop. The booklet contains a detailed roster of locomotives by railroad for over
70 railroads. The locomotives rosters were entered into a computer file, and
processed to obtain a listing by horsepower and manufacturer using the SPSS
statistical program. This distribution is shown in Table II-2. Note that over 80
percent of the locomotives are powered by EMD engines, and that EMD and
G.E. account for approximately 95 percent of all locomotives. The total
number of locomotives given in Table II-2 is greater than the total number of
locomotives given in the text above. The number in Table 11-2 includes
railroads other than Class I railroads, while the other total reflected only
locomotives in service on Class I railroads.
Over the past ten years, both U.S. manufacturers have worked to improve
not only the efficiency of their diesel engines, but also the efficiency of their
total drive system, including controls and drive wheel traction. A list of some
of the changes made by both manufacturers from 1972 to 1982 is shown in Table
II-3. It was reported that some of these changes are available as retrofit kits
for older locomotives, but exactly what hardware is available for retrofit was
not able to be determined.
8
-------
2000
1800
1600
1400
03
Q)
> 1200
-P
£
g 1000
i-3
iw
\o o
0)
800
2 600
400
200
Rebuilt
New
0 l-
iy/3 1974 1975 1976 1977 1978 1979 1980 1981
Calendar Year
Figure I1-2. Locomotive deliveries by year
-------
30
m
u
•rf
>
u
OJ
w
c
•H
(/!
(U
>
-H
4-1
0
u
-------
TABLE II-l. LARGE DIESEL ELECTRIC LOCOMOTIVES PRODUCED IN THE UNITED STATES
Model
SL80
SL110
SL144
B18-7
B23-7
B30-7A
C30-7A
B36-7
C36-7
Use
Wheel
Arrangement
Switching
Switching
Switching
General
General
General
General
General
General
Purpose
Purpose
Purpose
Purpose
Purpose
Purpose
B-B
B-B
B-B
B-B
B-B
B-B
C-C
B-B
C~C
(b)
(c)
No. of
Engines
2
2
2
1
1
1
1
1
1
Engine
Model
Number of
Cylinders
General Electric
.(a)
(a)
NT855L4
NT855L4
KTA-1150L
FDL-8
FDL-12
FDL-12
FDL-12
FDL-16
FDL-16
(a)
6
6
6
8
12
12
12
16
16
HP per
Engine
300
300
550
1800
2250
3000
3000
3600
3600
Turbo
Charged
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Max.
Locomotive
Weight, lbs
160,000
220,000
288,000
268,000
280,000
280,000
420,000
280,000
420,000
Max.
Speed, mph
21
21
35
70
70
70
70
70
70
EMD
SW1001
MP15
GP38-2
GP40-2
GP50
SD40-2
SD50
Switching
Switching
General
General
General
General
General
Purpose
Purpose
Purpose
Purpose
Purpose
B-B
B-B
B-B
C-C
C-C
1
1
1
1
1
1
1
8-645E
12-645E
16-645E
16-645E3B
16-645F3
16-645E3B
16-645F3
8
12
16
16
16
16
16
1100
1650
2200
3300
3800
3300
3800
No
No
Yes
Yes
Yes
Yes
Yes
230,000
248,000
250,000
256,000
260,000
368,000
368,000
65
65
65
65
70
65
70
(a)
„ jManufactured by Cummins
. . B—B is two sets of two axles
(c)
C-C is two sets of three axles
-------
Horsepower
Range
< 1000
1000-1499
1500-1999
2000-2499
2500-2999
3000-3999
4000-4999
5000-5999
6000-6999
Unknown
Total
TABLE II-2. DISTRIBUTION OF LOCOMOTIVES BY
MANUFACTURER AND HORSEPOWER(a)
Manufacturer
EMD
481
2644
7459
4648
1491
8555
5
9
45
22
25359
GE
8
0
116
895
649
2451
44
22
0
25
4210
Other
and
Alco. Bombardier Unknown
8
67
169
141
86
32
3
106
699
5 86
2 5
1
13
19
522
195
1108
Total
500
2817
844 3
5775
2233
11039
62
31
45
261
31206
(a)
as of 1981
12
-------
TABLE II-3. CHANGES IN DIESEL ELECTRIC LOCOMOTIVES - 1972 to 1981
Year
1972
G.E.
EMD
1974
1978
1979
1980
1981
Fluid amplifier water cooling
system. New engine water filling
and draining system. New wheel-
slip detection and correction
arrangement.
Modification of speed-throttle
notch schedule reduced smoke
by 50%.
Higher-efficiency turbocharger
on 12-cylinder C28-7 locomotive.
New two engine switch lomocotive
(one engine can be shut down at
idle), including more efficient
traction motors.
Improvements in turbocharger,
turbo seal system, optimization
of engine speed schedule, modi-
fication in radiator fan and
dynamic braking system. Claims
a 7.2% fuel saving on B36-7
locomotive over the earlier
U36B locomotive.
Production of the B36-7. Claims
8% less fuel consumption. Also
28 to 44% more adhesion, based
on a new adhesion system.
Production of C36-7 locomotive
(six axle version of the
B36-7)
Introduced "Dash 2" series
of locomotive with major
changes in electrical
controls
Changes in cylinder liners,
pistons and fuel injectors
for less smoke (1972-1974)
Production of the B version
of the 645-E engine with
improvements to give 3.3%
fuel savings per locomotive.
Changes: reduced idle speed,
greater thermal efficiency,
modifications in operation
and control of dynamic braking
and cooling fans. Use of
automatic engine purge
control.
Production version of the
GP50 locomotive. Claims
3% less fuel consumption,
17% more power from same
engine. Also 33% more
adhesion based on a new
adhesion system. Using the
new 645F series engine.
Production of the SD50
(six axle version of the
GP50)
13
-------
Many of the changes that resulted in fuel economy improvements have the
potential to change the locomotive emission factors as currently listed in AP-
42, due to changes in the locomotive duty cycle and changes in engine
efficiency. The fuel economy improvements may result in changes to national
and regional locomotive emission impact estimates as well, due to a reduction
in hours of operation per ton-mile as well as changes in emission factors. Thus
it is apparent that both the locomotive duty cycles and the emission factors
given in AP-42 require revision.
Revised Duty Cycles
Before new emission factors can be developed, the problem of how
changes in railroad operations over the past ten years have affected the
locomotive duty cycles must be resolved. As explained earlier, railroads no
longer allow locomotives to idle for long periods of time when not in use. Since
the duty cycles developed in the past were based on throttle clocks that ran
when the engine was running, the standby idle time was included in the duty
cycle. However, the duty cycles did not necessarily represent 24 hours-a-day
operation, since the time the engine was not running was not included in the
duty cycle. While some locomotives may have run for days at a time, on a
yearly basis there were undoubtedly periods of engine shutdown for
maintenance and inspection. The currently-used line haul and switching cycles
are shown in Table II-4. Duty cycles reflecting current operating practices
should show less time at idle. Any new duty cycle should also include a new
category of "engine off", to reflect a 2k hour day status of the locomotive.
TABLE II-4. CURRENT LOCOMOTIVE DUTY CYCLES USED
FOR LINE HAUL AND SWITCHING LOCOMOTIVES
Percent Time in Mode
G.E. Engines
EMD Engines
G.E. Avg.
EMD Line
G.E. Sc EMD Engines
Throttle Setting
1 Cycle
Haul (1972)
ATSF Switching
Notch 1
5
3
10
Notch 2
2.5
3
5
Notch 3
2.0
3
4
Notch 4
5.0
3
2
Notch 5
2.0
3
1
Notch 6
2.0
3
1
Notch 7
2.5
3
0
Notch 8
21
30
0
Idle
54
41
77
Dynamic Braking
4
8
0
The problem of defining new operational duty cycles is twofold: one, to
define the time fraction for the various throttle notches during engine
operation; and two, to define how many hours a day (on a yearly average) the
locomotive operates on this duty cycle. In discussing this problem with
representatives of the AAR and the two engine manufacturers, it appears that
14
-------
no duty cycle studies are available that reflect the current operational patterns
of railroad locomotives. The lack of current duty cycles and information on
percent of total time locomotives are in operation are important deficiencies in
the information needed to determine accurate emission factors for locomotives.
Since the duty cycle is so critical to usable locomotive emission factors, it was
decided to attempt to develop revised switching and line haul duty cycles from
available information.
A statistical summary published by the Association of American Railroads
(AAR) provided the information necessary to calculate the average number of
hours per day a locomotive is in actual use.(^) For 1980, the latest statistics
available, an average freight locomotive was in actual operation 9.2 hours per
day. Appendix B contains the details of the calculations for this value.
Currently published and utilized duty cycles were developed from throttle
clocks which did not account for the time a locomotive engine was not running.
It is assumed that locomotives are available approximately 96 percent of the
time. Then on the average, a locomotive is available 23 hours per day. The
1980 freight locomotive operation (non-idle) time of 9.2 hours, is then
approximately 40 percent of 23 hours. The "G.E. Average 1" (G.E. Al) duty
cycle, with 46 percent non-idle time, comes closer to having 40 percent non-
idle time than the EMD line haul cycle, with 59 percent non-idle time. Thus,
the "G.E. Al" cycle will be used as the basis for a new line-haul duty cycle.
To adjust the G.E. Al cycle to reflect current railroad operating
practices, it is necessary to make some estimate of the decrease in average
daily idle time that has resulted from shutting engines down when not needed
(at ambient temperatures above 50°F). Using the "Climatic Atlas of the United
States,"(10) was estimated that, considering the entire country, conditions
allow locomotive engine shutdown in approximately 55 percent of the
occurrences. The basis of this estimate is shown in Appendix B. A revised 24-
hour line haul operating cycle based on the G.E. Al cycle, 96 percent
locomotive availability (23 hours per day), and idle time equal to 45% of the
original 12.42 hours of idle (54 percent of 23 hours) is presented in Table II-5.
TABLE II-5. REVISED LINE HAUL LOCOMOTIVE DUTY CYCLE
Percent of
Hours of a
Engine Condition
Total Time
24 Hour Day
Engine Off
32.6
7.83
Idle
23.3
5.59
Dynamic Braking
3.8
0.92
Notch 1
4.8
1.15
Notch 2
2.4
0.58
Notch 3
1.9
0.46
Notch 4
4.8
1.15
Notch 5
1.9
0.46
Notch 6
1.9
0.46
Notch 7
2.4
0.58
Notch 8
20.1
4.83
15
-------
A similar, but simplified, analysis can be performed to give a new
switching cycle. For the switching cycle, it is assumed that the ATSF switching
cycle is correct, and that switch engines are also available 96 percent of the
time. For a 23-hour day, the ATSF switching cycle has 17.7 hours at idle and
6.3 hours non-idle operation. From the AAR statistics an average switch engine
travels 83.4 miles/day (see Appendix B). This gives an average switch engine
speed of approximately 13 miles per hour while operating. If it is again
assumed that idle time has been reduced by 55 percent, then the revised 24-
hour switching cycle is as shown in Table II-6.
TABLE II-6. REVISED SWITCHING LOCOMOTIVE DUTY CYCLE
Percent of
Hours of a
Engine Condition
Total Time
24 Hour Day
Engine Off
44.7
10.74
Idle
33.2
7.97
Dynamic Braking
0.0
0.00
Notch I
9.6
2.30
Notch 2
4.8
1.15
Notch 3
3.8
0.92
Notch 4
1.9
0.46
Notch 5
1.0
0.23
Notch 6
1.0
0.23
Notch 7
0.0
0.00
Notch 8
0.0
0.00
Revised Emission Factors
The literature search revealed only a small amount of published
information on emission factors. The locomotive emission factors in AP-42 are
apparently based solely on the three engines tested at SwRI in 1972.(2) Previous
to the SwRI work, some emission tests were done under sponsorship of the AAR
and three railroad companies.^ 1) Subsequent to the 1972 work, SwRI measured
emissions on a limited number of railroad engines for a variety of organizations
in the 1972 to 1975 time period. Most of this work has been published as
contract final reports and ASME papers.'^,13) There are also two papers in
which EMD reported some results of their own emission tests.^M5 ) Table II-7
lists the published studies and the engines tested. The AAR has recently
completed an emission study of 40 locomotives, both before and after
rebuilding. The data analysis is not completed, and a published final report on
that study is not expected until late 1984.
The published studies listed in Table 11-7 which presented actual emissions
measurements were reviewed. There are only six useful studies which present
emission measurements, but some of the studies have resulted in more than one
publication. For purposes of this project, these studies have been numbered
chronologically and referred to as Study I through Study 6. The studies are
listed in Table II-8, which contains the study number, report references, date of
study, and engines on which emissions were measured.
16
-------
TABLE II-7. PUBLISHED STUDIES OF MEASURED LOCOMOTIVE EMISSIONS
Report Title
"Report on Exhaust Emission
of Selected Railroad
Diesel Locomotives"
Study Sponsor Report Date
AAR and AT&SF, March 19 72
SP, and UP
Railroads
Report
Number
Locomotive Engines Tested
one G.E. FDL-16
Two EMD 645E3 (turbo)
One EMD 645E1 (roots blown)
"Status Report on Locomotives
as Sources of Air Pollution"
by Max Ephraim Jr.
EMD
1972
SAE 720604 Unknown number of EMD 645E
"Exhaust Emissions fron Un-
controlled Vehicles and
Related Equipment Using
Internal Combustion Engines.
Part I - Locomotive Diesel
Engines and Marine Counter-
parts" by Charles Hare and
Karl Springer, Southwest
Research Institute
EPA
October 1972
One EMD 12-567
One EMD 16-645E3
One G.E. 7FDL-16
"Locomotive Exhaust Emissions
and Their Impact" by Charles
Hare, Karl Springer, and
Thomas Huls
EPA
April 1974
ASME
74-DGP-3
(same tests as Oct. 1972 study)
November 19 74 ASME
74-WA/RT-l
"NOx Studies with EMD 2-567 DOT and EPA April 1974 ASME
Diesel Engine" by John Storment, 74-DGP-14
Karl Springer, and K. Hegenrother
"Exhaust Emissions of Selected AAR and AT&SF,
Railroad Diesel Locomotives" by SP, and UP
A.H. Bryant and T.A. Tennyson Railroads
"Four Cycle Diesel Electric General April 1975 ASME
Locomotive Exhaust Emissions: Electric and 15-DGP-10
A Field Study" by J.G. Hoffman, SP Railroad
Jr., Karl Springer, and T.A.
Tennyson
One EMD 2-567
(same tests as March 1982 study)
Seven G.E. 7FDL-16
-------
TABLE II-8. CHRONOLOGICAL LIST OF LOCOMOTIVE EMISSION
STUDIES USED
Study
1
5
6
Test Dates
Locomotives Tested
March, April 1971
Report Date 1972
April 1972
Nov. 72 to March 73
1972 to 1974
Report Date 1983
two EMD 645E3 turbo, 20 cyl.
one EMD 645E blown, 16 cyl.
one G.E. FDL turbo, 16 cyl.
unknown number of 1972 EMD 645E
engines both turbo and blown
one EMD 567 blown, 12 cyl.
one EMD 645E3 turbo, 16 cyl.
one G.E. 7FDL turbo, 16 cyl.
one EMD 567 blown, 2 cyl. with
three different injectors
seven G.E. 7FDL turbo, 16 cyl.
unknown number of EMD 645F3B
turbocharged engines
References
11, 16, 17
If
2, 18
12, 19
13, 20
15
Results from tests of nine G.E. locomotives are available. Test results
from two EMD 567 engines and four EMD 645E engines (3 turbocharged and 1
roots-blown) are available. The average emissions from an unknown number of
EMD 645E engines (turbocharged and roots-blown calculated separately) as
measured by EMD are also available. Finally, the cycle emissions from an EMD
645F engine tested on the UIC/ORE duty cycle are available. The tests of all
these engines were performed between 1971 and 1982, and include the SwRI
tests on which the current AP-42 emission factors are based. As far as can be
ascertained, this is a complete collection of all the measured locomotive
emission published in the open literature. While these data can be used to
enlarge the AP-42 data base, there is little published information on which to
base emission factors of the new fuel-efficient engines being produced by both
G.E. and EMD. A request was made for unpublished data from the locomotive
engine manufacturers but none were obtained.
Considerable effort was expended to try to reconcile and aggregate these
results into on cohesive set of emission factors. This effort was complicated by
results being presented using different duty cycles, different measurement
techniques for NOx and the fact that some NOx values were corrected for
humidity and some were not. Some studies also presented throttle notch-by-
throttle notch emissions; others presented only composite cycle emissions.
To include as much data as possible, it was decided to initially calculate
composite cycle emission factors based on two different, but similar current
duty cycles. The cycle called the "G.E. average 1" cycle was used for the G.E.
18
-------
engines, and the cycle called the "EMD line haul" cycle was used for the EMD
engines. The percent time spent in each throttle notch for these two cycles is
shown previously in Table II-4. After the emission factors for the various
engine models were determined using the current duty cycles, they were
adjusted for the revised duty cycles.
It was decided that the chemiluminescent technique for NOx would be
used as the reference measurement method of NOx. In study 1, where a
chemiluminescent NOx analyzer was not used, the NDIR NO readings were
multiplied by 0.95 to approximate chemiluminescent NOx values. For study 2,
the NOx reading was multiplied by 0.89, to reflect EMD in-house experience.
All NOx values were corrected for humidity by multiplying the NOx value by a
humidity correction factor, k, defined as:
1
k = l-0.0025(H-75)
Where H is the specific humidity in grams per pound of dry air.
Table II-9 presents a summary of all the emissions data available by
engine type, expressed in terms of g/bhp-hr on the current line haul cycles.
These data are not the recommended emission factors. Rather, these measured
emissions require evaluation and alteration based on demonstrated relationships
between engine types and conversion to the revised duty cycles to arrive at
final emissions factors.
The G.E. engines were examined first. G.E. changed the throttle stop-
engine speed schedule about 1974. Fortunately, Studies 3 and 5 tested G.E.
engines using both speed schedules. There are emissions data available from
five G.E. engines using the new speed schedule. The average of these engines
was taken as the basis for all G.E. engine emissions. Two other configurations
of G.E. engines were calculated. One configuration used the original speed
schedule; the other configuration was with the new speed schedule and a low
sac injector. There were four G.E. locomotives tested with the old speed
schedule.
Study 3 tested one locomotive on both speed schedules, and Study 5 tested
two locomotives on both schedules. The differences in the average HC and CO
emissions between the four engines with the old speed schedule and the five
engines with the new speed schedule agree with the changes seen on the
individual engines in Studies 3 and 5. However, the difference in the average
NOx value is less than the difference seen in the individual engine tests.
Studies 3 and 5 both showed a k.5 percent decrease in NOx emissions when an
engine was changed from the old speed schedule to the new speed schedule.
Since the new speed schedule values are serving as the basis for the G.E. engine
emission values, the recommended NOx emission factor for the old speed
schedule is k.5 percent higher than the new speed schedule value, or 14.8 g/bhp
hr. Study 5 presented data on one engine with low sac injectors. These data
are used to calculate the emission factors for new G.E. engines.
19
-------
TABLE II-9. MEASURED LOCOMOTIVE EMISSIONS FROM PUBLISHED
STUDIES EXPRESSED IN GRAMS PER HORSEPOWER HOUR
ON LINE HAUL CYCLES
Engine Description
Number
Tested
Avg. Emissions, g/bhp-hr, (range)
HC
CO
NOv
FDL, old speed schedule
(1957 to approx. 1974)
FDL, new speed schedule
(approx. 1973 to present)
FDL, new speed sch. low sac
injectors
EMD 567 spherical injectors
pre 1959
EMD 567 needle injectors
1959 to 1966
EMD 567 low sac injectors
(retrofit after 1972)
EMD 645E blown, needle injector
(1966 to 1972)
EMD 645E turbo, needle injectors
(1966 to 1972)
EMD 645F3B turbo, low sac injectors unknown
General Electric Engines
2.2 (1.7 to 2.5)
2.3 (2.0 to 2.6)
0.6
Electromotive Division
2.7
1.2
0.7
unknown
1.1
0.8 (0.7 to 0.9)
0.4(d)
4.2 (3.6 to 4.5)
1.8
6.4
10.7
7.4
10.8
3.2 (2.5 to 4.0
0.6(a)
14.0 (10.9 to 18.9)
2.5 (2.0 to 3.0) 14.2 (10.4 to 19.7)
10.7
12.1
9.8
13.0
12.5
11.6 (8.7 to 11.6)
14.1(a)
(a)on UIC/ORE cycle
-------
The analysis of the EMD engine emissions is more involved. SAE paper
720604 (Study 2) presents the average emissions results from a number of new
645E engines as tested by EMD. These engines were tested in 1972.
Apparently, they were equipped with the standard (not low sac) needle
injectors. In a telephone discussion of EMD emissions measurements with Mr.
Hugh Williams of EMD(20) it was learned that during this time period, EMD
used NDIR instrumentation for NO measurement and UV for NO?. EMD has
since converted to measurement of NOx by chemiluminescence (CL). Their
experience is that CL NOx measurements are 11 percent lower than NOx levels
measured using NDIR and UV. Applying this 11 percent to the NOx values in
Study 2 gives NOx values of 12 g/bhp hr for the turbocharged, and 13 g/bhp hr
for blown version of the EMD 645 engine. Since this study appears to use the
largest number of engines, it will be used as the basis for the EMD emission
factors.
The HC emissions from one of the 645E turbocharged engines tested in
Study 1 and the 645 turbo engine tested in Study 3 agree with the HC emission
level of Study 2, giving a range of HC values from 0.7 to 0.9 g/bhp-hr. The NOx
levels from all three 645E turbo engines tested in Studies 1 and 3, while lower
than the Study 2 average, show fair agreement with the average NOx (adjusted
to CL NOx) values from Study 2. The CO levels from one of the 645E turbo
engines from Study 1 and the CO level from the 645 turbo engine tested in
Study 3 agree with the Study 2 values, giving a CO range from 2.5 to 4.0 g/bhp-
hr. The HC and CO emissions from one 645E turbo engine tested in Study 1
(locomotive SP8803) did not agree well with HC and CO emission levels from
the other studies. Therefore, these emissions were not used in the final analysis
of the EMD 645 emission factors.
The emission factors for both the G.E. and EMD line haul engines using
current duty cycles are presented in Table II-10. Since the emission factors,
which represent fleet averages, are based on very few engine tests, there is no
justification for presenting them to more than two significant figures. The
purpose in this analysis was to obtain the emission relationships between the
various engine models using as many data as possible. Once these trends were
established, the emission factors using the revised duty cycles could be
calculated.
Where emissions data were available by individual notch setting, revised
emission factors were obtained using the revised duty cycle percent time-in-
notch as revised weighting factors. Where only composite emission factors
were available, revised composite emission factors were developed by changing
the original ones in proportion to change made in the notch-by-notch factors.
The resulting revised emission factors, recommended as replacements for the
current AP-42 emission factors, are shown in Tables 11-11 and 11-12.
These emission factors are in terms of mass of emissions per power
output. To use these factors for regional impact, total horsepower hours for
the region by line haul and switch engines must be known. This information is
generally not readily available. What is normally available is the number of
locomotives in a region. From the data in Table II-2 and some assumptions
about which locomotives are line haul and which are switchers, it should be
possible to determine an average horsepower for line haul and switch engines.
21
-------
TABLE 11-10. LINE HAUL LOCOMOTIVE BASED ON CURRENT DUTY CYCLES EMISSION FACTORS
Line Haul Emission Factor,
g/bhp hr
_____ Engine Description HC CO NOy
General Electric Locomotives
FDL, old speed schedule 2.2 4.2 14
(1957-1974)
FDL, new speed schedule 2.3 2.5 14
(1974 to approx. 1980)
FDL, after approx. 1980 0.6 1.8 14
Electromotive Division
567 spherical injectors (pre 1959) 2.7 6.4 12
567 needle injectors (1959 to 1966) 1.2 11 10
645E needle injectors, blown (1966 to 1972) 1.1 11 13
645E needle injectors, turbo (1966 to 1972) 0.8 3.2 12
567 low sac injectors (retrofit after 1972) 0.7 7.4 13
645E low sac, blown (1972 - ) 0.6^ 7.6^ 17 ^
645E low sac, turbo (1972 - ) 0.5^a' 2.2^ 16
64 5F low sac, turbo (1982 - ) 0.4 0,6 14
estimated from 567 trends
22
-------
K3
u>
TABLE 11-11. RECOMMENDED LINE HAUL LOCOMOTIVE EMISSION FACTORS
BASED ON SWRI 19 83 OPERATING CYCLE
Line Haul Emission Factor Cycle
g/bhp-hr Load Cycle
Engine Description HC CO NOx Factor BSFC
General Electric Locomotives
FDL, old speed schedule (1957-1974) 2.1 4.2 14 0.275 0.364
FDL, new speed schedule (19 74 to
approximately 1980) 2.2 2.4 14 0.279 0.344
FDL, after approximately 1980 0.6 1.6 14 0.279 0.344
Electromotive Division
567 spherical injectors (pre 1959) 2.7 6.1 12 0.268 0.516
567 needle injectors (1959 to 1966) 1.2 10 10 0.269 0.507
645E needle injectors, blown (1966 to 1972) 1.1 11 13 0.270 0.409
645E needle injectors, turbo (1966 to 1972) 0.8 3.2 11 0.277 0.379
567 low sac injectors (retrofit after 1972) 0.6 7.1 13 0.273 0.469
645E low sac, blown (1972 - ) 0.6(a) 7.6(a) 17(a) 0.274(a) 0.378(a)
645E low sac, turbo (1972 - ) 0.5^a^ 2.2^a^ 15^a^ 0.281^ 0.351^
(b) (c)
645F low sac, turbo (1982 - ) 0.4 0.6 14 0.281 0.336
|a!Estimated from 567 trends
, ,Estimated from 645E
(c)
Estimated from 645E and Reference 1.
-------
TABLE 11-12. RECOMMENDED SWITCHING LOCOMOTIVE EMISSION FACTORS
BASED ON SWRI 1983 OPERATING CYCLE
Switching Emission Factor,
g/bhp-hr
Engine Description
HC
General Electric Locomotives
FDL, old speed schedule (1957-1974) 3.0
FDL, new speed schedule (19 74 to approx. 1980) 3.5
FDL, after approximately 19 80 0.8
CO
8.4
4.6
3.3
NOx
19
19
14
Load
Factor
0.043
0.049
0.049
Cycle
BSFC
0.464
0.432
0.432
(b)
Electromotive Division
567 spherical injectors (pre 1959) 5.9
567 needle injectors (1959 to 1966) 3.3
645E needle injectors, blown (1966 to 1972) 2.2
645E needle injectors, turbo (1966 to 19 72) 2.2
567 low sac injectors (retrofit after 1972) 1.1
645E low sac, blown (1972 - ) 0.7
645E low sac, turbo (1972 - ) 0.7
645F low sac, turbo (1982 - ) 0.9
(a)
(a)
2.7
3.0
7„1
7.8
1.8
4.3
4.7
1.6
(a)
(a)
15
15
29
13
17
33
15'
17
0.033
0o 0 35
0.036
0.040
0.924
0. 815
0.520
0.043 0.482
0.616
(a' 0.041(a) 0.393(a)
0.049(a) 0.364(a)
0.049(b) 0.349(c)
(a)
(b)
(c)
Estimated from 567 trends
As s ume d val ue
Estimated from 645E and Reference 1,
-------
If a weighted percent horsepower (load factor) is known with the brake specific
emission factor for each cycle, then emissions in grams per hour and per year
can be calculated from available data. Sometimes fuel consumption is the best
information available. In these cases, a cycle BSFC would allow the use of the
g/bhp-hr emission factors. To facilitate these calculations, cycle load factors
and cycle BSFC values are included in Tables 11-11 and 11-12.
These detailed emission factors are useful for studies where detailed
information about locomotives is available. This availability is not generally
the situation. The remaining task was to combine the factors, using
information on age distribution of each manufacturers locomotives and the
percentage of the locomotive population represented by each manufacturer, to
obtain a set of revised national average emission factors in terms of pounds per
thousand gallons of fuel. These factors can be compared to the national
average factors currently in AP-42. The data put on computer file from "Diesel
Locomotive Rosters"^' were used to determine the fraction of total
locomotives in each of the year brackets shown in Tables 11-11 and 11-12 for
both manufacturers, with line haul and switch engines calculated separately.
These values are shown in Tables 11-13 and 11-14.
TABLE 11-13. DISTRIBUTION OF EMD ENGINES BY YEAR OF
MANUFACTURER, TYPE OF ENGINE AND ENGINE MODEL
1965 or Earlier 1966-1971 1972-1981
Percent of EMD Switch Engines
567 with needle injectors
567 with low sac injectors
33.7
33.6
645 blown w/ needle injectors
645 turbo w/ needle injectors
645 blown w/ low sac injectors
645 turbo w/ low sac injectors
3.9
3.9
3.9
3.9
6.9
10.3
Percent of EMD Line Haul Engines
567 with needle injectors
567 with low sac injectors
16.0
16.0
645 blown w/ needle injectors
645 turbo w/ needle injectors
645 blown w/ low sac injectors
645 turbo w/ low sac injectors
6.6
6.6
6.6
6.6
16.7
25.0
25
-------
TABLE 11-14. DISTRIBUTION OF G.E. ENGINES BY YEAR
OF MANUFACTURER
Percent of G.E. Engines by Type
1973 or Earlier 1974-1980 1981 or Later Total
Switching 100 0 0 100
Line Haul 65.4 34.3 0.3 100
National average line haul and switch engine emission factors in terms of
g/bhp hr were then calculated using the national fractions as weighting factors
for the emissions values and BSFC from Tables II-11 and 11-12. The emission
factors were converted to lbs per 1000 gallons of fuel using the national
average weighted BSFC. These national average line haul and switching
emission factors are shown in Table 11-15. Line haul engines represent 80.1
percent of the engines in service, with switch engines the remaining 19.9
percent.^) Using these percentages, total national average locomotive
emission factors were calculated. These factors are also shown in Table 11-15.
TABLE 11-15. NATIONAL AVERAGE LOCOMOTIVE EMISSION FACTORS
Emission Factors, pounds/1000 gallons of fuel
HC
CO
NOy
Switch Engines
47.4
86.8
468
Line Haul Engines
38.9
226
558
All Engines
41.3
187
533
Current AP-42 Factor
94
130
370
National and Regional Impact
Using the national average emission factors shown in Table 11-15 and the
total nationwide railroad fuel used in 1981 from Reference 5, the nationwide
annual railroad emissions in 1981 were determined. The total nationwide
emissions for each gaseous pollutant from all sources were obtained from the
EPA summary of 1981 emissions.^) From the nationwide emissions from all
sources, the locomotive emissions as percentages of emissions from all sources
and from mobile sources were calculated. The results of the calculation are
shown in Table 11-16.
26
-------
TABLE 11-16. NATIONAL IMPACT OF LOCOMOTIVE EMISSION
Nationwide
HC
CO
NO^
All sources, metric tons/year
21 x 106 91 x 106 19.5 x 106
All mobile sources, metric tons/year 7.5 x 10^ 70 x 10^ 8.2 x 10^
Locomotives, metric tons/year
70,371 318,631 908,184
Locomotives, percent of all sources
0.3
0.4
4.6
Locomotives, percent of mobile sources 0.9
0.5
11
Three studies were obtained that examined the impact of locomotive
emissions on a regional basis^2,23,24)> Qne of the reports (reference 24) was a
draft of the EPA "Report to Congress on Railroad Emissions." This draft report
presented the results of an examination of railroad emissions impact or five
different areas: Philadelphia, Chicago, St. Louis, Kansas City and Los Angeles.
Reference 22 reports on the study done by Walden Research Division of Abcor,
Inc., to determine locomotive emissions in the St. Louis area for the RAPS
(Regional Air Pollution Study) program. The remaining study, done by the
University of Michigan (UofM) School of Public Health under an EPA grant,
examined the impact of locomotive emissions in the Chicago area.
The draft Report to Congress used three different methods to determine
the railroad emissions in each of the five areas studied. The first method used
an estimate of the number of locomotives in the area together with duty cycles
in terms of number of hours per day in various throttle notches and emissions in
grams per hour to obtain the total locomotive emissions in each area. The
second method used gross ton miles of freight hauled and pounds of fuel burned
per ton-mile of freight to get non-idle fuel consumption. Fuel consumption at
idle and for switch engines was estimated from non-idle line haul fuel
consumption. The third method used county-by-county annual fuel consumption
(estimated from statewide annual fuel consumption) and the national average
emission factors in AP-42 in terms of pounds of emissions per thousand pounds
of fuel.
The Walden Research study used the most detailed methodology with the
most measured data. It allocated railroad activity by maps of track location
and a complete, one-day inventory of railroad activity including:
• Routing, routine and locomotive information for each train in the
study area
• Total active and idle hours and locomotive information for each rail
yard in the AQCR
• Interyard transfer routing and run time
27
-------
The annual horsepower hours of operation in each grid were determined from
the railroad activity information, data from the manufacturers, and the SwRI
locomotive study for EPA.(2) The fuel used in each grid was computed from the
annual horsepower hours of operation in each grid and fuel consumption derived
from AP-^2. The grid emissions were then calculated using the emission
factors for each engine type in terms of pounds per thousand pounds of fuel as
given in AP-42. Total annual emissions in the AQCR were obtained by summing
the emissions from each grid.
The Chicago study by UofM determined the annual emissions using an
average engine horsepower and load factor and annual hours of locomotive
operation in the AQCR, together with emission factors in grams per horsepower
hour from the SwRI locomotive study. The calculation of hours of locomotive
operation was begun by determining the miles of track in the AQCR, then
expressing this track mileage as a fraction of total national track mileage. The
hours of line haul operation and switchyard activity in the Chicago AQCR were
both assumed to be this same fraction of the total national line haul and
switchyard hours of activity.
The locomotive emissions in tons per year from each of the studies are
presented in Table 11-17. As can be seen from the table, the individual
estimates for each area vary widely. Except for the values of HC, the St. Louis
estimates made by the EPA study do not differ too greatly from the St. Louis
estimates made in the Walden Research study. As stated above, the Walden
Research study appears to be based on the greatest amount of measured or
sampled data, and so is considered to have the best credibility. Comparing the
St. Louis estimates with the Chicago estimates, it is obvious that the UofM
study underestimates the locomotive emissions in the Chicago area. It would be
reasonable to expect the Chicago emissions to be much larger than the St.
Louis emissions, yet the Chicago emissions estimates from the UofM study are
nearly equal to the St. Louis estimates from the Walden Research study.
Because the Chicago estimates from the UofM study are apparently too low,
they will be dropped from this discussion.
The locomotive emissions as percentages of the total emissions in the
AQCR are presented in Table 11-18. These percentages are based on the 1977
National Emissions Report. Note that regardless of the method or study,
locomotive HC emissions are always below two percent of total HC emissions,
and are often less than one percent. Except for the Method 1 CO estimates in
Chicago and Kansas City, locomotive CO emissions are always below one half
percent of the total CO emissions. The locomotive NOx emissions constitute
the highest percentage of the total area emissions of any of the three pollutants
studied.
Locomotive NOx emissions as a percentage of total NOx emissions also
vary widely from area to area; and within an area, from method to method.
The cities most affected by locomotive NOx emissions appear to be Chicago
and Kansas City. Depending on the methodology used, the locomotive NOx
emissions vary from approximately four to eight percent of the total NOx
emissions in either city.
28
-------
TABLE 11-17. ANNUAL LOCOMOTIVE
EMISSIONS IN SEVERAL REGIONS
Chicago
Ref 24 Ml
M2
M3
Ref 23
St. Louis
Ref 24 Ml
M2
M3
Ref 22
Philadelphia
Ref 24 Ml
M2
M3
Kansas City
Ref 24 Ml
M2
M3
Los Angeles
Ref 24 Ml
M2
M3
Emissions, tons/year
HC CO NOx
14,856
11,839
9,264
3,900
34,633
8, 395
12,813
4,380
61 f260
27,193
36,463
11,300
2,341
2,947
1,936
4,220
5,442
2 ,738
2,679
4,35 0
10,221
8,977
7,627
11,935
1,833
1,915
2,935
4 ,266
1,564
4 ,056
7,778
5 ,111
11,548
2,549
2 ,511
2 ,201
5,936
1,968
3,044
10,733
6,448
8,662
2,261
2 ,443
6,151
5,261
2,228
8,508
9,700
7,266
24,214
29
-------
TABLE 11-18. LOCOMOTIVE CONTRIBUTIONS
TO AREA TOTAL ANNUAL EMISSIONS
Percent of Total Annual Emissions
(a)
HC CO NOy Revised NOx(b)
Chicago
Ref 24 Ml 1.65 1.04 8.12
M2 1.32 0.25 3.61
M3 1.03 0.38 4.84 6.9
St. Louis
Ref 24 Ml 0.74 0.39 2.36
M2 0.93 0.20 2.08
M3 0.61 0.19 1.76 2.5
Ref 22 1.33 0.31 2.75
Philadelphia
Ref 24 Ml 0.26 0.18 2.24
M2 0.27 0.07 1.47
M3 0.41 0.17 3.33 4.8
Kansas City
Ref 24 Ml 1.03 0.79 7.49
M2 1.01 0.26 4.50
M3 0.89 0.40 6.05 8.6
Los Angeles
Ref 24 Ml 0.15 0.09 1.71
M2 0.17 0.04 1.28
M3 0.42 0.14 4.28 6.1
TaT
Based on 1977 emissions.
using the revised national average N0X emission factor from this
study. Applicable to reference 24, method 3 only.
30
-------
These emissions impact estimates were calculated using essentially the
1972 time frame emissions, either from AP-42 or from studies on which the AP-
42 factors were based. The revised national average emission factors
calculated in this study (and shown in Table 11-15) are lower for HC and higher
for CO and NOx than the present AP-k2 national average factors. The method
1 and 2 of reference 2k were not calculated using emission factors that are
directly comparable to the emission factors in Table 11-15. Method 3 did use
emission factors comparable to these revised emission factors. Since NOx is
the emission of most interest, the locomotive NOx emissions as a percent of
annual total NOx emissions for the five areas were recalculated by multiplying
the method 3 percentages by the ratio of the revised emission factors to the
Reference 2k emission factor. These revised NOx emissions as percents of the
total NOx emissions in the area are shown in Table 11-18.
There are other factors that will also tend to change the impact presented
in Table 11-17. The EPA now uses counties rather than AQCR's to evaluate
compliance with air quality standards. How this will affect the locomotive
emissions in percentages of total area emissions in unknown. It is also assumed
that the State Implementation Plans are working, so that in most areas the
total tons of the three gaseous emissions have been reduced, which would tend
to increase the locomotive share.
Considering all these factors, annual locomotive NOx emissions as high as
ten percent of the total NOx emissions in a county are conceivable. This level
must be put in prospective, however, by considering how many counties fail to
meet the current NOx ambient standard. A computer listing of those counties
not in compliance with ambient air standards as of February 1983 was obtained
from the EPA Office of Air Quality Planning and Standards.^' Only 11
counties in the country do not meet the NOx ambient air standards. Of these
eleven, six are in Colorado, four in California, and one in Illinois. The county in
Illinois is Cook County, which contains Chicago. The four counties in California
are all in Southern California and include Los Angeles and three adjacent
counties. The Colorado counties are Denver and surrounding counties.
While Denver was not one of the areas studied, it is not likely that
locomotives are a large part of the emissions in the Denver area. Colorado as a
whole has a fewer miles of track per square mile of area than any of the States
which had study areas, as can be seen from the figures in Table 11-19. The
locomotive contribution to NOx in the Los Angeles AQCR has already been
shown to be small.(See Table 11-18) This leaves Chicago as the only area
exceeding the NOx ambient standard where locomotive NOx emissions might be
important. A more thorough study of locomotive emissions in the Chicago area
and their contribution to ambient concentrations (rather than just tons per year
of emissions) is recommended.
31
-------
TABLE 11-19. MILES OF RAILROAD TRAKC PER THOUSAND
SQUARE MILES OF AREA FOR SEVERAL STATES
Miles of Track per
State 1000 Sq. Miles of Area
California 41.15
Colorado 32.7*1
Illinois 179.84
Kansas 87.54
Missouri 84.02
Ohio 173.21
Pennsylvania 153.55
32
-------
III. CONSTRUCTION EQUIPMENT
Construction equipment includes machines that dig, move, grade and
compact soil; machines that compound, lay, and compact paving materials;
machines that lift and move structual components and materials; and machines
that generate electricity, pump water and compress air. The equipment is
powered primarily by diesel engines, with a small fraction of it powered by
gasoline engines. Not all construction equipment is self-propelled; some is
mobile in the sense that it can be, and is, moved from place to place.
Construction equipment was chosen for investigation in the project because a
previous project at SwRI^5) hac| indicated that the national emissions impact of
construction equipment was approximately the same as the locomotive
emissions impact.
At the beginning of the literature search, after review of the AP-^2
emission factor and source documents, the EPA Branch Technical
Representative provided information that the State of California had just
completed a study of farm, construction and industrial equipment in California.
From contact with State of California personnel, it was learned that not only
had the state done a study of construction equipment impact, but a consortium
of industry organizations had also commissioned a study of farm, construction
and industrial equipment emission factors. Reports on the two studies were
obtained. The reports were so comprehensive that no other literature was
deemed necessary.
Current AP-42 Emission Factors
The current AP-42 construction equipment emission factors were last
updated in 1975. These factors, listed under "heavy duty construction
equipment," were taken from measurements of construction equipment
emissions at SwRI during the early 1970's. The AP-42 construction equipment
emission factors are included in Appendix D for reference. Gaseous emission
factors are listed in AP-42 for ten categories of diesel-powered equipment and
five categories of gasoline-powered equipment. In addition, the estimated
annual hours of operation are included for ten categories on construction
equipment.
Revised Emission Factors
As mentioned above, two recent reports were obtained on the subject of
construction equipment emissions. One report was done by Environmental
Research and Technology, Inc/26) This report was prepared under sponsorship
of the Farm and Industrial Equipment Institute (FIEI), Engine Manufacturers
Association (EMA) and Contruction Industry Manufacturers Association (CIMA).
It is referred to as the CAL/ERT Report. The other report was by the
California Air Resources Board (CARB)/27)
The report by the industry groups is the most comprehensive study of
emission factors ever done on construction equipment. The procedure used to
obtain the emission factors was to have each manufacturer submit measured
emissions data to the public accouting firm of Ernst and Whinney. This firm
33
-------
collected and aggregated the data, and presented only the aggregated data to
the study contractor, Environmental Research and Technology. In this way,
manufacturers' confidential information could be used without compromise.
The CARB also used these emission factors in their staff analysis of
construction equipment.(27)
The sales-weighted emission factors for diesel construction and industrial
equipment from the CAL/ERT study are shown in Table III-1. These factors
were compiled from data taken by 13 engine manufacturers, and they represent
391 models of construction equipment. Since these emission factors are based
on comprehensive, up-to-date emission measurements, it is recommended that
they be used in place of the current AP-42 emission factors, simply because of
their more defensible data base. Construction equipment gasoline engine
emission factors were not given in the CAL/ERT report. According to the
report, less than five percent of sales use gasoline engines, and the trend is
toward complete dieselization. If gasoline engine emission factors are needed,
the present factors in AP-42 should be sufficient. The new diesel factors will
have little effect on any studies that have been or will be done on construction
equipment contribution to air pollution, because the emission factors are in
good agreement with the factors currently in AP-42 as can be seen by the
comparison shown in Table III-2.
Regional and National Impact
Both the CARB and the construction equipment industry analyzed the
impact of construction equipment on regional air quality using the gaseous
emission factors shown in Table III-l. The CARB analyzed seven regions of
California for air quality impact from construction/industrial equipment, as
well as presenting tons per day from construction/industrial equipment for the
entire state. The CAL/ERT study analyzed three regions of the state, as well
as presenting statewide tons per year from construction/industrial equipment.
While both reports contain statewide estimates, the only common region in the
two reports is the South Coast Air Basin (SCAB).
A comparison of the construction equipment impact, in terms of tons per
year of HC, CO, and NOx, from the two reports for the whole state and the
SCAB is presented in Table III-3. Comparing the estimates presented in Table
I1I-3, the CARB statewide tons per year estimates are approximately twice the
industry estimates, while the CARB estimates for the SCAB are approximately
three times the industry estimates. The CAL/ERT estimate of Fresno County
area had the largest fraction of total emissions from construction equipment;
2.2 percent for NOx. If it is assumed that the CARB estimate for this area
would also be three times the industry estimate, then the NOx emissions from
construction equipment in Fresno County could be as high as six percent of the
total NOx emissions. For these estimates mobile industrial equipment, such as
mining and forestry equipment, was included with construction equipment in the
impact analysis.
In discussing the differences between the two emission impact estimates
with the CARB staff^S) it was learned that a series of meetings is planned in
198^ between the CARB and the industry trade group consortium. These
meeting were planned in an attempt to develop a single set of emission impact
estimates upon which both groups can agree.
34
-------
TABLE III-l. SALES WEIGHTED DIESEL EMISSION FACTORS FOR CONSTRUCTION
AND INDUSTRIAL EQUIPMENT FROM CAL/ERT STUDY
Emissions Factors
(gm/bhp-hr)
Equipment Type
HC
CO
N0X
Track Type Tractor, 90+ HP
0.37
1.65
6.60
Wheel Load i'2-l/2 cu. yd.
0.60
2.07
8.31
Ind. Wheel Tractor
1.76
7.34
11.91
Wheel Tractor Scraper
0.55
2.45
7.46
Log Skidder
0.61
3.18
9.82
( cl )
Off-Highway Trucks
0.37
2.28
8.15
Motor Graders
0.36
1.54
7.14
(c)
Hydraulic Excavator, Crawler
1.22
3.18
11.01
Trencher
Concrete Paver
Compact Loader
1.10
4.57
10.02
Wheel Loader <2-1/2 cu. yd.
1.29
3.26
9. 24
Track Type Loader, 90+ HP
0.47
1.56
7.76
Track Type Tractor 20.89 HP
1.33
2.91
9.63
Track Type Loader, 20-89 HP
1.80
3.02
10.97
Roller Compactor, Static
0.88
5.33
11.84
(d)
Crane Lattice Boom, Wheel & Crawler
0.59
4.99
12.45
Hydraulic, Wheel, One Station
0.80
7.80
14.69
(c)
Hydraulic Excavator, Wheel
1.22
3.18
11.01
Roller Compactor, Vibratory
1.06
6.72
14.27
Crane, Hyd, Wheel, Multi-Station
0.68
3.71
12.47
Bituminous Paver
0.99
5.19
11.18
includes pavement cold planer, wheel dozer
^includes generator sets, contractor's engine-driven pumps, road wideners
(c)
same as hydraulic excavator wheel
^ includes pipe layers
35
-------
TABLE III-2. COMPARISON OF PRESENT AP-42 CONSTRUCTION
EQUIPMENT DIESEL EMISSION FACTORS WITH CAL/ERT
STUDY DIESEL EMISSION FACTORS
HC
AP-42 avg.
CAL/ERT avg.
AP-42 range
CAL/ERT range
Emission Factors g/bhp-hr
CO
0.83
0.90
0.26 to 1.39
0.36 to 1.80
2.70
3.80
1.80 to 4.40
1.54 to7.80
NO,
11.7
10.3
6.6 to 15.7
6.6 to 14.7
TABLE III-3. IMPACT OF CONSTRUCTION EQUIPMENT ON
AIR QUALITY IN CALIFORNIA
1.
2.
3.
Calendar year
Tons/year, Statewide
HC
CO
NOx
Tons/year, South Coast Air Basin
HC
CO
NOv
CAL/ERT
Report^3)
1979
2158
21471
20083
500
5068
4527
4.
Construction equipment emissions as a percent of
total tons/year from all sources.
a. South Coast Air Basin
HC
CO
NOx
b. Fresno County
HC
CO
NOx
c. Sacramento Modeling Grid
HC
CO
NOv
0.08%
0.20%
0.91%
0.18%
0.60%
2.20%
0.24%
0.36%
1.70%
CARB Report^
1979
4991
52755
40332
L535
16196^
12391(c)
0.25^)
0.64^)
2.45(d)
ta)Reference 6
(b)Reference 5
(c^Back calculated to 1979 from 1982 data in report using CARB estimate of
1.5% annual growth rate
(d)Calculated from CARB data in 3 and total emissions calculated from
CAL/ERT percentage adjusted for larger construction fraction.
36
-------
It should be pointed out that a high percentage contribution to total
emissions says nothing about whether the area has an air pollution problem. For
example, Fresno county has the highest portion of NOx emissions from
construction and industrial equipment (perhaps as high as six percent) of any
California area examined. However, Fresno County does not exceed the
ambient NOx standards. The SCAB, which includes part of all four California
counties that do not meet the ambient NOx standard, has between
approximately 0.9 and 2.5 percent of total NOx emissions (depending on which
estimate is used) from construction and industrial NOx emissions.
The national impact of construction equipment on HC, CO, and NOx
emissions was estimated using the CARB estimate of statewide construction
equipment emissions. The assumption was made that construction equipment
emissions are proportional to population. Thus, Callifornia with 10.45 percent
of the national population was assumed to have 10.45 percent of the
construction equipment emissions. Using this factor, the national construction
equipment emissions were calculated in terms of percent of nationwide
emissions from all sources and percent of nationwide mobile source emissions,
and are shown in Table III-4. These estimates are based on the 1981 nationwide
emissions from Reference 21. This technique may somewhat overestimate the
national emissions from construction equipment. Because California is a
rapidly growing state it's construction equipment emissions are probably greater
than ten percent of the total.
TABLE III-4. ESTIMATED NATIONAL IMPACT OF CONSTRUCTION
EQUIPMENT EMISSIONS
Construction Equipment Nationwide Annual Emissions as a:
Percent of All Sources Percent of Mobile Sources
HC
0.2
0.6
CO
0.6
0.7
NOx
2.0
4.7
37
-------
IV. MARINE VESSELS
The final class of off-highway mobile emission sources investigated was
marine vessels. This study excluded outboard engine-powered vessels, since
traditionally outboards have been considered as a separate class of off-highway
mobile sources. The marine emission factors presented in AP-42, together with
the source studies from which the factors were derived, were examined in
detail. Trends in propulsion technology for marine vessels were reviewed to
determine if there had been changes which would require new emission factors.
A study of marine vessel emissions factors just being completed by the EPA
Office of Air Quality Planning and Standards (OAQPS), for which a draft final
was available, was also reviewed in detail. Finally, several air quality impact
studies for marine vessels were reviewed.
Current Marine Emission Factors in AP-^2
The latest AP-^2 marine vessel emission factors were compiled in 1975.
Average emission factors in terms of pounds per 1000 gallons fuel and kilograms
per 1000 liters of fuel are presented by waterway classification (river, great
lakes, and coastal) and by propulsion system (steam boiler and diesel). A set of
emission factors is also presented for diesel-powered electrical generators in
marine vessels. The narratives accompanying the emission factor tables, as
well as table footnotes, indicate that the factors are based on limited data. In
addition, the marine diesel emission factor table has an incorrect entry at 1550
horsepower. The emissions given at 1550 horsepower are for a Coast Guard
steam boiler propulsion system, not a diesel engine. These facts argue for an
update of the AP-^2 marine emission factors, even without consideration of
changes in marine emissions over the last ten years. The marine vessel section
from AP-^2 is included in Appendix D for reference.
Trends in Marine Propulsion
As with locomotives, fuel costs have been the driving factor in marine
propulsion changes during the past ten years. The cost of marine fuel over the
past twenty years is shown graphically in Figure IV-1. Note that the 1983 costs
of both marine diesel and marine fuel oil were approximately seven times their
1973 costs. At this point, a short digression is necessary to examine the terms
used to describe marine fuels. Apparently the same marine fuel is called by a
variety of common names. Table IV-1 lists some of these names. Thus, the
Marine Diesel Fuel and Marine Fuel Oil referred to in Figure IV-1 correspond to
No. 4D and No. 6 fuel oil, respectively.
The increase in fuel prices has resulted in several changes in marine
propulsion systems. One of the most basic changes is the use of diesel engines
in U.S. merchant shipping. Historically, the U.S. merchant fleet (and U.S.
Navy) have been powered by steam boiler propulsion systems, while most of the
rest of the world's ocean shipping has been powered by diesel engines. Diesel
engines, in general, have better thermal efficiency than marine steam
propulsion systems.^) The use of steam propulsion by U.S. ships results from
legislative requirements that cargo shipped between U.S. ports be carried in
U.S. flag vessels, and that U.S. flag vessels be built in U.S. shipyards using U.S.
39
-------
350
300
/
i ^
/
\
\
250
200
150
100
50
Marine
Diesel Fuel
Marine
Fuel Oil
Source: Reference 29
X
1960
1965
1970 1975
Year
1980
1985
Figure IV-1. Bunker Prices, London.
40
-------
TABLE IV-1. MARINE FUEL TERMINOLOGY
TABLE I: Marine fuel terminology
Description Color
Common Names
Approximate SRI Approximate SUS
Viscosity Viscosity
Range (SlOCF Range @100°F
Heavy or
Residual
Black,
Dark Brown
Marine Fuel Oil (MFO)
Bunker Fuel Oil (BFO)
Heavy Fuel Oil (HFO)
No. 6 Fuel
Bunker C
3500-6000 +
4000-9000-t
Blends of Black, Intermediate Fuel Oil (IFO)
Heavy and Dark Brown (Heavy Oil, sometimes)
Distillate Intermediate Bunker Fuel (IBF)
Light Fuel Oil (LFO)
Light (Residual)
Blended (Diesel Fuel)
Light or Medium
Marine Fuel Oil
No. 4, No. 5 Light, No. 5 Heavy
Thin Fuel Oil (TFO)
200-3000
250-3200
Distillate
Dark Colored
(Some
Residual
Content)
Light Colored
(No Residual
Content)
Marine Diesel Fuel (MDF)
Marine Diesel Heavy
Marine Diesel 35-110
No. 4D
B1, B2
Marine Diesel Light
Gas Oil
Diesel Fuel 31-42
Diesel Gas Oil
No. 2
2-D, A2
38-125
33-47
From reference 31
41
-------
equipment. Until recently, very large slow speed diesels were not
manufactured in the U.S. Recently however, these engines have begun to be
manufactured in the U.S., mostly under license to European companies. This
trend to diesel engines can be seen in Figure IV-2. In this figure, the numbers
of active U.S. flag oceangoing vessels of 2000 gross tons and over in 1983 are
shown by type of propulsion system. Note the rise of diesel power in ships built
in the 1970's and the dominance of diesel in ships built in the 1980's.
Gas turbines were touted highly in 1970's, but their generally lower
thermal efficiency (than diesels) has made them uncompetitive as fuel costs
have risen. In fact, several U.S. flag container vessels have had their gas
turbines removed and replaced with diesel engines. While the first coal fueled
ship (and consequently steam powered) built in the U.S. in nearly 50 years was
recently launched,(32) problems of boiler stoking and fly ash removal will
probably keep the coal/steam propulsion system from being popular.^3)
Worldwide, diesel engines are expected to remain the overwhelmingly
predominant marine power source for the forseeable future. In a recent listing
of ships of 1,000 gross tons^a' or over being built in shipyards worldwide
(excluding military ships), only 10 ships of the 1398 ships listed were not diesel
powered.^) Thus, in 1983, greater than 99 percent of the ships of 1000 gross
tons or more under construction or on order were diesel-powered. The
horsepower of these diesel ships ranged from 1550 horsepower for a 1500 dead
weight ton (DWT)W ferry to 44,560 horsepower for a 57,800 DWT container
carrier.
A second trend in marine propulsion is the attempt to use cheaper, less
refined, more viscous fuels in diesel engines. Since the late 19Ws the very
slow speed marine diesel engines (under 200 rpm) have been able to operate on
residual fuels. Experimental work has been done recently on engines in the
1000 rpm range to develop their capability to burn residual or mixtures of
residual and distillate fuels. (3^,35,36)
The last trend observed in marine propulsion is the development of more
efficient diesel engines. All diesel engine manufacturers are working to
increase the fuel efficiency of their engines in response to rising fuel prices.
The increases obtained are not breakthroughs, but rather careful engineering
changes that generally result in fractional percent increases in fuel economy.
When the changes are added together, some engines have improved fuel
economy between five and ten percent.'"'
These trends in marine propulsion point to the increasing importance of
the marine diesel emissions factor estimate. With declining importance of
steam boiler propulsion systems, the boiler emission factors currently given in
AP-^2 are probably sufficient, but the diesel emission factors definitely require
updating both to enlarge the data base and to account for the changes in diesel
propulsion systems listed above.
aGross tons is a volume measurement. One gross ton =100 ft^ of enclosed
space.
bDead weight tons is a measure of the weight of a ship in long tons (22W lbs) at
maximum draft.
42
-------
160
140
120
Q, 100
•H
CO
u
Q)
rO
% 60
40
20
0
JB
~
Steam turbine
Diesel
Gas turbine
Unknown
Source: Reference 32
pre 1950
1950 to 1959 1960 to 1969 1970 to 1979 1980 to 1983
Unknown
Year Launched
Figure IV-2. Type of propulsion system in active U.S. flag vessels by year launched
-------
Review of OAQPS Report
During the search for information on marine vessel emissions, it was
learned from EPA's Office of Air Quality Planning and Standards (OAQPS) that
they had recently completed a study of marine vessel emissions. The study was
conducted by Engineering-Science Corp. (E-S) for OAQPS. A draft of the final
report entitled "Emission Factor Documentation for AP-42: Section 3.2.3,
Inboard Powered Vessels" (EPA 450/4-84-001) was available.
Since the need for an update of the marine emission factors had been
demonstrated, it was considered likely that this study would provide the needed
update. The latest version of that draft final report was obtained and reviewed.
The study did an excellent job of collecting the available data on marine vessel
emissions. However, the methodology used to aggregate the data into emission
factors appears flawed. For steam power plants, this methodology did not
create a large problem. The boiler emission factors from the E-S study do not
vary greatly from those currently in AP-42. In addition, except for the cities of
San Diego and Norfolk, which are homeports for a large number of U.S. Navy
ships, steam-powered vessels will not be a large fraction of marine traffic in
the future.
The gasoline fueled pleasure craft emission factors are based on outboard
engine studies done at SwRI during the early 1970's. Almost all outboard
motors are two-stroke engines, whereas inboard gasoline engines are almost all
four-stroke automotive derivative engines. Since two-stroke and four-stroke
gasoline engines have greatly different emissions, the inboard gasoline emission
factors in the E-S report are not recommended for use. The current AP-42
emission factors for inboard gasoline engines are based on precontrolled
automotive engines. These AP-42 emission factors for gasoline inboard engines
should be retained as the best available estimate, until some actual
measurement studies are conducted. The diesel-powered pleasure boat emission
factors in the E-S study are the same as those currently in AP-42.
As mentioned earlier in the section, the diesel engine emission factors for
marine vessels have become the most important set of marine emission factors.
The diesel propulsion engines for which measured emissions were obtained for
the E-S report are shown in Table IV-2. Conspicuous by their absence are
emission values from EMD engines, which constitute the bulk of towboat
engines. Note however, that there are some engines in each category of speed,
size, and usage.
While the data collection was excellent, analysis of the diesel emissions
data presented in the report did not adequately consider a number of factors
and did not combine individual engine data points in a useful manner. The
greatest problem was the way the individual engine data were aggregated to
produce emission factors for various horsepower classes. Because of the
sparcity of data, large increments of percent power were combined (for
example 0 to 15 percent power, and 15 to 45 percent). All data points available
in a specific power interval were then averaged together. For example, if two
engines A, and B, had data in the power internal between 0 and 15 percent, with
engine A having 6 data points at idle and engine B having one data point at 10
percent power,all seven data points were averaged to give the emission levels in
44
-------
TABLE IV-2.
DIESEL ENGINES HAVING MEASURED EMISSIONS IN DRAFT OF REPORT EPA 450/4-84-001
Manufacturer
Engine Series
No. of
Cylinders
Max RPM
HP Range
Typical Usage
Low Speed Engines (less than 600
rpm)
M.A.N.
KSZ (2-stroke)
4 to 12
110-145
7840 to 44,040
large tankers and liners
M.A.N.
I<5cV 52/55
6 to 9
450
7200 to 21,600
smaller tankers and liners
Pielstick
(4-stroke)
12 to 18
520
6000 to 11,700
smaller tankers and liners
Union (built in 1942)
06
—
350
300
Coast Guard buoy tender
Medium Speed Engines (600 to 1300 rpm)
Fairbanks-Morse
38D8-1/8,
38TD8-1/8 (4-stroke)
4 to 12
6, 9, 12
750/900
750/900
708 to 2760
1750 to 4200
tow boats, work boats, etc.
Caterpillar
D379, D398, D399
(4-stroke)
8, 12, 16
1300
640 to 1380
fishing boats, pleasure craft
Waukesha
6LRDCSM
—
1200
500
Coast Guard Harbour tug
Ingersoll-Rand
(built in 1943)
Type S
—
720
600
Coast Guard Harbour tug
Cooper-Bessemer
(built in 1942, 1944)
GN-8, GND-8
—
600-700
600-700
Coast Guard buoy tender
Cooper-Bessemer
FVBM-12-T
—
600
1580
Coast Guard cutter
Alco
16-251-B
—
1000
2500
Coast Guard cutter
M.A.N.
L&V 20/27, 25/30
4 to 15
750-1000
545 to 9000
tow boats, work boats, etc.
High Speed Engines (greater than 1300 rpm)
Caterpillar
3200, 3300 <5c 3400
(4-stroke)
4 to 12
2100-2800
100 to 650
fishing boats, pleasure craft
Cummins
(4-stroke)
6, 8
1800-3000
195 to 520
fishing boats, pleasure craft
Cummins
VT-12-900M
12
2300
900
Coast Guard patrol boat
G.M.
6071-A
—
2000
200
Coast Guard utility boats
-------
the 0 to 15 percent power range. Note that this gives the emissions from
engine A six times the weight of the emissions from engine B. In the 15 to 40
percent power range, if engine A had one data point at 20 percent power and
engine B had six data points at 40 percent power, all seven data points were
again averaged to produce the average emission level in the 15 to 45 percent
range. In this interval engine B emissions are given six times the weight of the
engine A emissions.
This method makes the "average" emission factor in a given power
interval a function of the emissions from the engine with the most data points
in that interval. However, the engine with the most data points can change with
power interval. The proper way to utilize the data would be to develop curves
of emissions as functions of power for each engine. Then, based on the in-use
population of each engine, population-weighted average emissions at specific
percent power levels could be computed.
Another problem with the methodology is that while the emission factors
are presented on both a fuel specific and a power specific basis, a different
data base was used for the fuel specific emissions than was used for the power
specific emissions. Thus, the two sets of emission factors are not really
equivalent.
Other problems with the factors have to do with the data themselves.
Almost all of the part-power data come from three sources: an in-use study of
Coast Guard cutter and small boat emissions, an engine dynamometer study of
Coast Guard small boat engines emissions, and part-power rated speed data
from 13-mode emission tests of Cummins engines. The Cummins data as
presented in the draft are not correct. A check with Cummins personnel
revealed that Cummins inadvertently supplied E-S with the wrong values. Even
if the data were correct, the Cummins data should not have been used because
they do not represent the usual operating conditions for propulsion diesels. The
Cummins 13 mode data are for various part power conditions at rated speed.
Most marine propulsion engines in the Cummins size range run on an operating
line through the speed-power map, so that part-power operation occurs at
various speeds lower than rated speed. Thus, part power emissions at rated
speed are not representative of emissions during the usual operation of these
engines.
A review of the reports from which the in-use Coast Guard data were
taken reveals that four of the 14 engine models tested were built during World
War II, and are not representative of the majority of diesels now in service.
These engines should not be included in the data base. In addition, a G.M.
engine installed in a 40 foot launch was tested while tied to a pier. The
resulting power points were at part speed, but several were at full throttle for
that speed. Again this is not a usual engine operating condition. Diesel
generator emissions from this Coast Guard study were also aggregated with
propulsion engine emissions in the E-S study. Many engine models are used for
both generators and main propulsion engines. However, generator rated power
is generally less than main propulsion rated power for the same engine. All of
the generators in the Coast Guard study were also constant speed units. Thus,
their part power conditions are not representative of propulsion unit part power
operation, as explained above.
46
-------
An attempt was made to reanalyze the data contained in the E-S report
to provide a more useful set of marine diesel emission factors. The diesel
engines were divided by rated speed into slow speed (less than 600 rpm),
medium speed (600 to 1300 rpm), and high speed (greater than 1300 rpm)
engines. To some extent rated power is inversely proportional to speed; the slow
speed engines in general have the highest rated power, while the high speed
engines have the lowest rated power. It was decided to exclude the three Coast
Guard cutter engines built in the 1940's as not representative of currently
installed marine diesel engines.
For slow speed engines, data were available from three different engine
series representing two manufacturers. The data consisted of two full-power
points from one model M.A.N, four-stroke engine (L or V 52/55), two part power
points from each of four different model M.A.N, two-stroke engines, a full
power point from another model M.A.N, two-stroke engine, and a full power
point from Colt-Pielstick four-stroke engine, for a total of 12 data points. The
M.A.N, engines were tested on residual fuel, and the Pielstick engine on D2
diesel fuel. Obviously, there are hardly sufficient data to determine emission
factors with any confidence. Nevertheless, they are all that is available, and
they are plotted in Figure IV-3.
As can be seen from the figure, a trend with power from 60 to 100% can
not be established. The only reasonable use of these data is to average all the
data points separately for each emission to arrive at single factors for 60 to 100
percent power for each emission type. This process was done for all three
emissions, except that the lowest NOx value at 100 percent power was omitted.
The resulting emission factors are shown in Table IV-3. From data contained in
the E-S report and our own experience, it appears that there is little difference
in gaseous emission levels from diesel engines using different fuels, as long as
the engine is running on the fuel for which it was designed. Therefore, the
values in Table IV-3 can be used for engines designed for either distillate or
residual fuel.
TABLE IV-3. RECOMMENDED EMISSION FACTORS FOR MARINE DIESEL
ENGINES WITH DESIGN SPEEDS BELOW 600 RPM
Nominal Power Range: Above 3000 horsepower
Application: a) all foreign ships above 10,000 dead weight tons
b) Some newer U.S. registry ships above 10,000 dead weight tons
c) For both distillate and residual fuel
Emission Factor, g/kW-hr (g/hp-hr)
Percent Load HC CO NOv
60 to 100 0.48(0.35) 1.6(1.2) 13(9.4)
As mentioned above, the marine diesel engine power range is generally
inversely proportional to the rpm range. Thus, the slow speed engines tend to
represent large power outputs. The M.A.N, and Pielstick engines in the data
47
-------
O M.A.N. 2-stroke engines, residual fuel
~ M.A.N. 4-stroke engines, residual fuel
A Colt-Pielstick 2-stroke engine, D2 fuel
1.0
M
&
k °-5
cn
O
K
~
o
6
-
-®)
0
&
=8
A
0
iiit
40 50 60 70 80
Percent Load
90
~
100
Figure IV-3. Emission levels from marine diesel engines
with design speeds under 600 RPM
48
-------
base range from 7,200 to *<4,000 horsepower. While several Japanese firms
manufacture slow speed dieseis with maximum output under 3,000 horsepower,
most other slow speed diesel engines have design outputs above 3,000
horsepower.
The information in Table IV-4, taken from the E-S draft report, indicates
that engines of greater than 3,000 horsepower are generally used in ships above
10,000 dead weight tons. This size vessel is normally used for ocean shipping.
Recall that almost without exception, non-U.S. registry merchant ships use
diesel engines for propulsive power, while only some of the newer U.S. registry
ships use dieseis. Thus, the emission factors given in Table V-3 could be used
for all foreign ships in U.S. waters and for a fraction of the US. registry
merchant ships, whether distillate or residual fueled.
Emissions from medium-speed engines were also investigated. Of the
eight medium-speed engine types listed in Table IV-2 the Ingersoll-Rand and
Cooper-Bessemer engines built in the 19^0's were not used. Of the six
remaining engine models with emissions data, four had data from the in-use
study of Coast Guard vessels, and two had summary data from the
manufacturer. In the course of investigating the emissions from the medium-
speed engines, a number of errors were found in the Caterpillar data presented
in the draft E-S report. In a phone conversation with Caterpillar it was learned
that these errors apparently originated at the manufacturer. For this study,
the corrected Caterpillar data for the D379, D398 and D399 were obtained from
the manufacturer.
Of the six medium-speed engine models for which there were emissions
data available, only four had part-power emissions. Engine power output was
not available for all four of these engines, but fuel consumption was. Emission
levels were therefore calculated in terms of grams per pound of fuel consumed.
These emissions data were supplemented with EMD locomotive emissions data
from SwRI studies. Plots of HC, CO, and NOx emissions as a function of
percent power for these four engines are shown in Figure IV-4, IV-5 and IV-6.
Percent power levels for the Waukesha engine were not available, so its
emissions were not used.
As can be seen from examination of the figures, none of the engines has
the same pattern of emissions with power level. There are many possible
reasons for this difference. Except for the EMD engine, all part power engine
data were from the in-use Coast Guard study, for which percent load was not
measured, but rather assigned by vessel speed mode (slow, 1/3, 2/3, standard,
full or flank). Thus it is probable that the percent load chosen for each engine
was not exactly correct. In fact, for these curves the percent power values for
Coast Guard data were adjusted slightly (up to 5 percent) to attempt to provide
some pattern of emissions with power. This adjustment was still not sufficient
to bring emission values from each engine into a common pattern. Apparently,
variations in emissions at the same speed and load for various engine models, as
well as variation in engine speed at the same percent load for various engine
models, contributed to this lack of a common emissions pattern with percent
power. In addition, for turbocharged engines, the turbocharger schedule varies
with engine model. The use of variable pitch propellers on certain marine
vessels can also cause operating point and emission differences from
49
-------
TABLE IV-4. RELATIONSHIPS OF TYPICAL TONNAGE, VESSEL TYPE,
AND PROPULSION UNIT CAPACITY3
Vessel dead
weight tonnage
General carrier
103 kw(103 HP)
Bulk carrier
103 kw(103 HP)
Tanker
103 kw(103 HP)
5,000
1.8 - 3.2
1.2 - 1.6
0.8 - 1.6
(2.4 - 4.3)
(1.6 - 2.2)
(1.1 - 2.1)
10,000
3.6 - 6.3
2.4 - 3.3
1.6 - 3.1
(4.8 - 8.5)
(3.2 - 4.4)
(2.2 - 4.2)
20,000
7.2 - 12.7
4.8 - 6.6
3.3 - 6.4
(9.6 - 17.0)
(6.4 - 8.8)
(4.4 - 8.4)
30,000
10.7 - 19.0
7.2 - 9.8
4.9 - 9.4
(14.4 - 25.5)
(9.6 - 13.2)
(6.6 - 12.6)
40,000
14.3 - 25.4
9.6 - 13.1
6.6 - 12.5
(19.2 - 34.0)
(12.8 - 17.6)
(8.8 - 16.8)
50,000
17.9 - 31.7
11.9 - 16.4
8.2 - 15.7
(24.0 - 42.5)
(16.0 - 22.0)
(11.0 - 21.0)
75,000
b
b
12.3 - 23.5
(16.5 - 31.5)
100,000
b
b
16.4 - 31.3
(22.0 - 42.0)
a,I'ypical for operating speeds of 10-18 knots. Specific vessel tonnage
vs propulsion unit capacity may differ, especially at higher speeds.
Ranges are a general guide to propulsion unit capacity, for use when
^actual capacities are not known, kw = kilowatt, HP - horsepower
Not available
50
-------
O Fairbanks Morse 38TD-8-1/8
V ALCO 16-251-B
0 Cooper-Bessemer FVBM-12T
X EMD 645 E3
Percent Power
Figure IV-4. HC emissions from four medium-speed marine diesel engines
51
-------
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
0
O Fairbanks Morse 38TD-8-1/8
V &LCO 16-251-B
(3 Cooper-Bessemer FVBM-12T
X EMD 645 E3
I I I I I 1 I I I I
10 20 30 40 50 60 70 80 90 100
Percent Power
r-5, CO emissions from four medium-speed marine diesel engines
52
-------
40
38
36
34
32
30
28
26
24
22
20
18
16
14
12
10
re
O Fairbanks Morse 38TD-8-1/8
V ALCO 16-251-B
O Cooper-Bessemer FVBM-12T
X EMD 645 E3
1 I I 1 I I I I I I I
0 10 20 30 40 50 60 70 80 90 100
Percent Power
V-6. N0X emissions from four medium-speed marine diesel engines
53
-------
installation-to-installation for the same engine model. The result of these
considerations is that marine engines may operate at very different points on
the engine performance map, with very different emissions, and still be at the
same percentage of rated power.
These considerations also indicate that useful marine emission factors
require a more detailed analysis than available information allows. Without
additional information on the duty cycles of the various classes of shipping,
information on the percentage of total engines in service by manufacturer, and
additional emission data for the more popular engines, adequate marine vessel
emission factors can not be developed.
The diesel emission factors presented in the E-S report are based heavily
on the same study of in-use Coast Guard vessels on which the current AP-42
diesel emission factors are based. In addition, as explained above, there are a
number of methodology and analytical problems in the E-S study. Therefore it
is recommended that the present AP-42 diesel emission factors continue to be
used with two changes. The first change is the deletion of the 1550 horsepower
entry (this entry is really for a steam boiler, not a diesel engine). The second
change is the addition of the high horsepower engine emission factors given in
Table IV-3 of this section.
Information Needed for Marine Diesel Emission Factors
Since it was determined that more information was needed to determine
useful marine diesel emission factors, an attempt was made to obtain that
information; or if the information was not available, to define what is needed
for the benefit of future studies. Information is first required on the kinds of
vessels in service by vessel usage, together with the numbers of vessels in each
usage category. Typical or average operating cycles of each usage category are
needed. Then a determination of propulsion system type and the in-use
population of the various engine models is required. Finally, emission
measurements on the more popular engine propulsion systems are needed to
complete the calculation of the marine diesel emission factors.
In this discussion, as in AP-42, a distinction is made between marine
propulsion engines and marine electrical generator engines. For a variety of
reasons, these two uses of diesel engines can be expected to have somewhat
different emission factors. Only marine diesel propulsion engines will be
discussed here.
Sources for statistics on marine vessels in the United States are diverse
and generally not in a form that is usable for environmental impact
determinations. A number of published sources were consulted.(38-^0) In
addition, telephone contact was made with a number of persons in shipping,
shipbuilding and marine industry associations.^ 1-44)
United States marine traffic can be defined by location (great lakes,
river, coastal, or transocean) or by type of vessel (bulk carrier, liner, towboat,
workboat, fishing boat. etc). Some locations have predominately one type of
vessel traffic. For instance, river traffic is predominantly towboats pushing
54
-------
barges. However, most locations will have a large variety of types of vessels.
Because of this diversity of vessels in most locations, the emission factor needs
are best served by considering vessels by type, regardless of where they are
located.
For this project, marine vessels were divided in 10 types, as shown in
Table IV-5. This table does not show outboard motor-powered boats, since they
have traditionally been considered separately in AP-42. The vessels have been
separated into various categories because their different use results in different
main propulsion systems, different design power to displacement ratios and
different operating cycles. Consequently, the different types can be expected
to have different emission rates. Note that many of the column entries are
blank, indicating that the information was not available in the sources located.
Estimates of the number of vessels have been obtained from a number of
sources. For consideration of air pollution impact, the important population
estimate is the number of vessels in U.S. waters per day. While the definition
of the water area where vessel emissions would affect regional air pollution
varies from region to region, a suggested starting point for coastal areas would
be the waters where U.S. inland rules-of-the-road apply. The estimated number
of vessels in the U.S. waters for each type of vessel is also shown in Table IV-5.
Except where all vessels are always within U.S. waters, little information was
available on which to base an estimate. For foreign shipping, the number was
based on yearly foreign ships visiting New York, Los Angeles, and Houston(^)
then extended to the whole country. For the U.S. flag vessels, including Navy
ships, it was assumed that one-third of the fleet was in U.S. waters at any one
time. For U.S. Coast Guard and fishing boats, it was assumed that 90 percent
were within U.S. water each day. The total estimated inboard-powered marine
vessels, excluding pleasure boats, in U.S. waters on an average day is
approximately 29,000 vessels.
In Table IV-5 the main propulsion units are listed as either steam or
diesel. Except for U.S. flag merchant ships and the U.S. Navy, diesel propulsion
is used almost exclusively. The diesel engines for the vessels listed in Table IV-5
range from under 100 horsepower to greater than 30,000 horsepower per engine.
The smallest engines are marine versions of truck engines, and have design
speeds up to 2,800 rpm. The largest engines have design speeds of 100 rpm or
less. Within these limits, engine rpm is somewhat inversely proportional to
design horsepower. An article in Marine Engineering/Log'^ indicated that for
ships built in 1976 with diesel engines over 10,000 horsepower, approximately 79
percent had slow-speed, two stroke engines.
Marine diesel powerplants are produced by a large number of companies
worldwide. The share of installations for each manufacturer is not known. It
appears that worldwide Burmeister and Wain (BicW), M.A.N, and Sultzer
account for a large portion of the two-stroke slow-speed installations, and that
M.A.N., Sultzer, and Pielstick are popular four-stroke engines.^7,^8) in the
U.S., EMD is apparently the most popular diesel engine manufacturer for river
towboat engines.'^)
A possible source has been found for most of the missing marine diesel
information needed; however, the information could be costly. The American
55
-------
TABLE IV-5. MARINE VESSELS IN U.S. WATERS CLASSED FOR AIR POLLUTION STUDIES
Type of Vessel
L/i
U.S. Flag Tankers and bulk
carriers
U.S. Flag Liners
a. ocean
b. river
c. Great Lakes
Oceangoing Foreign Tankers
and bulk carriers
Foreign Liners
Towboats
Oceangoing Workboats and
tugs
Ferries
a. general
b. railroad
U.S. Navy <5c Coast Guard
a. combatant
b. auxiliary
c. Coast Guard
Fishing Vessels
Inboard Pleasure Craft
Total Number^3)
333(c)
r
L
840(c)
1226^
26
23,000^
4890(c)
2833^)
78(c)
405(e)
169(e)
252(e)
19,500({)
700,000(g)
No. Within U.S.
Waters(^) Per Day
111
277
1226
261
1000 1
J
4890
2833
78
134
56
226
17,550
700,000
Type of
Main Propulsion
80% steam, 20% diesel
80% steam, 20% diesel
diesel
80% steam, 20% diesel
diesel
diesel
diesel
diesel
p 70% steam, 25%
|_nuclear, 5% othecJ
98% diesel, 2% steam
diesel
diesel, gasoline
Average
HP (Range)
14380
6806
1556
3070
1554
(100-18,000)
(^unless otherwise noted, this is the number of vessels homeported in the U.S.
(b)total number of ships worldwide. Source: World Almanac
(c)Source: "Summary of U.S. Flag Passenger and Cargo Vessels." Corps of Engineers,
Waterborne Commerce Statistics Center, New Orleans, LA.
(d)Source: "Marine Engineering/Log." Dune 15, 1983
(e)Source: Jane's Fighting Ships 1982-1983
(f)Source: Telecon with M. Kinter, American Waterway Operators
(g)Source: AP-42
(^Estimated. See test for explanation.
-------
Bureau of Shipping (ABS) publishes a ship registry that includes information on
over 50,000 vessels (apparently most of the ships in the world above 500 gross
tons). Part of the information includes main propulsion type, horsepower and
manufacturer. This information is on a computerized data base. Custom
searches of the data base are available through ABS Computers, Inc., a
subsidiary of ABS. In theory it should be possible to obtain statistical
descriptions (mean, mode, median, standard deviation and range) for vessel size
and horsepower as well as number of engine installations by engine
manufacturer for the types of vessels listed in Table IV-5. The cost of this
information would probably be on the order of several thousand dollars.'^)
Since this expenditure was not planned for this project, funds are not available
for such a data base search. Nevertheless, the service offers an opportunity to
obtain information which is absolutely necessary before reliable marine diesel
emission factors can be developed. It is mentioned here for consideration in
future work concerning marine vessels.
The only information on vessel operating cycles found was for Coast
Guard vessels.^5*) Operating cycles are required if emission factors are to be
expressed in terms most useful to air quality planners, such as grams of
emissions per mile or hour of operation, or grams of emissions per ton mile or
ton-hour of operation. This area also requires a great deal more research, but
it is necessary if realistic marine diesel emission factors are to be developed.
As was pointed out in the review of the E-S report on marine emission
factors, usable diesel emission factors require much more than the compilation
of a few part-power points. There is likely to be a wide variation in emissions
when all marine vessels are considered. Some of the emissions variation is due
to vessel type and operating cycle as mentioned above. Even within a given
vessel type, however, there are likely to be wide variations in diesel emissions
because of differences in propulsion systems.
To understand why knowledge of the vessel propulsion system is
important, it is necessary to understand how the total propulsion system can
affect emissions. First of all, there are two different systems for using diesel
engines in marine propulsion. One system is the direct-drive system, in which
the engine is coupled directly to the propeller (either with or without reduction
gears). With this system, the engine operates at various speeds and loads. The
second system is a diesel-electric system in which the diesel engine drives a
generator and the propeller is driven by an electric motor. With this system,
the diesel engine most often operates at one speed and varying loads. While the
installed percentage of the two system types are not known, the direct drive
appears to be more popular/32)
Typical engine performance maps for a four-stroke, turbocharged,
medium-speed marine diesel engine and a comparable two-stroke turbocharged
engine are presented in Figure IV-7 and IV-8, respectively. Shown on the maps
are lines of constant BSNOx and typical operating curves for a direct drive
application of each engine. These maps are taken from a SwRI study conducted
for the U.S. Coast Guard.^2) ^s can bee seen from the figures, each engine
type produces a different pattern of BSNOx with power level. These maps are
considered typical, but emission patterns and levels can vary considerably from
57
-------
Max. —
415
49 8
580
664
747
830
913
1000
Figure IV-/. Typical brake specific N0X map — four-stroke
cycle turbocharged marine engine
58
-------
T
T
1 r
Estimated
Behavior
Constant
k. BSNO.,
r Operating
\ Carve
453 574 755 315 900
Figure IV-8. Typical brake specific N0X map — two-stroke
cycle turbocharged marine engine
59
-------
one engine model to another. In addition, if the operating line changes, the
pattern and magnitude of the BSNOx values can change.
The two main parameters that can change the operating curve are: the
reduction gear ratio (if reduction gears are used) and the propellers pitch. For
a given vessel, it takes a certain amount of power to move the vessel through
the water at a given speed. Reduction gear ratios are fixed for each
installation; but if the reduction gear ratio were changed, the engine speed at
which a given vessel speed is achieved would change, just as shifting gears in an
automobile changes the engine speed for a given vehicle speed. Since there is
no transmission in a maine vessel propulsion system, different reduction gear
ratios in different installations will produce different engine operating curves
for a given engine or engine type.
The propeller pitch will also affect the shaft speed at which a given
horsepower is attained. Some propellers are fixed (or constant) pitch. Other
propellers are equipped to change pitch while in operation. To show the effects
of propeller pitch, a variable pitch propeller map is shown in Figure IV-9. This
map is for a Coast Guard WHEC type cutter, and is taken from the recently
completed study by SwRI for the Coast Guard.'^l) The figure shows the
relationships between propeller speed, horsepower, propeller pitch, and vessel
speed. For this discussion, the important observation is that for a given vessel
speed, the propeller rpm, and hence, the engine speed, changes as the propeller
pitch (proportional to the pitch ratio "P/D" in the figure) is changed.
The object of this discussion is to demonstrate the importance of having a
good compilation of the in-use marine diesel propulsion systems before
attempting to determine emission factors for marine diesel engines. Once the
population percentages of the various in-use systems have been ascertained,
emission measurements on representative systems can be made along the engine
operating curve to produce truly representative emission factors. While such an
undertaking is beyond the scope of this project, it is recommended for a future
study.
Regional and National Impact
Only a few studies of the regional impact of marine vessels were found in
the iiterature.'23,.53-.56) three most recent studies'23,55,56)^ a|j completed
in 1976 to 1978 time period, were reviewed. Each study examined a different
region. The regions examined were: St. Louis, The Port of New York, and
Houston. Somewhat similar methods were used in the New York and the St.
Louis studies, while a different methodology was used in the Houston study. In
the process of comparing the results from these studies, it was discovered that
there were errors in the calculation of diesel emission factors in both the Port
of New York and Houston studies. These errors were corrected, and the annual
marine emissions recomputed for both studies.
The recalculated emissions (in tons per year) from non-military vessels
determined by each of the studies shown in Table IV-6. Also listed in the table
are the percentage contributions of marine vessels to total regional emissions
of each pollutant. To enable a comparison between regions, the 1970 commerce
in tons of cargo for each port (from reference 58) is presented in the table.
60
-------
9000
8000 _
7000 _
6000
SHIP SPEED
V=20 KNOTS
or
LU
2
O
Cl.
LU
C/">
ad
O
O
Llj
a:
5000 _
PITCH RATIO
P/D=l .4 1.31
V =3205 LT
PITCH
SCHEDUL
HANDLE
POSITION
URRENT
PITCH
SCHEDULE
LEGEND
V=10 KNOTS
SHIP SPEED
4000 _
3000 _
2000 _
1000
= PROPOSED
3 PITCH SCHEDULE
25 CURRENT
PITCH SCHEDULE
70 80 90 100 110 120 130 140 150 160 170 180
PROPELLER RPM
Figure IV-9. Hypothetical WHEC pitch schedule, propulsion Systems, Inc., Propeller
61
-------
Since St. Louis is a river port, the cargo figure includes not only cargo with
origin or destination in St. Louis, but also cargo that "passes by" on the river.
The cargo tonnage estimate for St. Louis was obtained from reference 55.
TABLE IV-6. MARINE EMISSIONS IMPACT OF THREE AREAS
Value by Port Location
Data Item
Emissions, tons/year
HC
CO
NOx
Percent of Total Area Emissions
HC
CO
NOx
Cargo, millions of tons per year
Cargo specific emissions,
tons/million tons of cargo
HC
CO
NOv
Port of New York
St. Louis New Jersey Houston
939 3289 364
2103 7683 782
3292 15694 2664
0.32 0.31 0.10
0.06 0.18 0.09
0.76 1.56 1.10
50.00 174.01 64.65
18.78 18.90 6.61
42.06 44.15 18.40
65.84 90.19 56.59
While the vessel emissions from each region should not necessarily be
exactly proportional to the cargo level, the cargo specific emissions in terms of
tons of emissions per million tons of cargo could all be expected to be of the
same order of magnitude. This comparison holds for New York and St. Louis,
whose cargo specific emissions are in good agreement, but not when comparing
Houston with either New York or St. Louis. Part of the reason for these
differences appears to be due to the propulsion system mix in the various ports,
St. Louis was assumed to be all diesels, as would be expected in an area where
almost all waterborne vessels are towboats. The New York Harbor study
concluded that 95 percent of the operational hours in New York Harbor were
accounted for by diesel-powered vessels. The Houston study indicated that only
65 percent of the cargo tonnage in the f5art of Houston was carried by diesel-
powered vessels. This difference in percent of diesel vessels accounts for some
of the difference between New York and Houston marine emissions on a cargo
specific basis.
After a review of the calculational methods used in both the Houston
study and the New York study, it is considered likely that the Houston study
probably underestimates the marine emissions. The key parameter of the
methodology used in the Houston study is the fuel consumption per thousand
tons of cargo carried for various classes of vessels. The variables used to
62
-------
calculate this parameter appear to be more suited to open water cruising than
maneuvering in a harbor. Therefore, the values of fuel consumption per
thousand tons of cargo used in the Houston study were probably too low, causing
the total marine emissions per year to be too low. If the New York emissions
per million tons of cargo are used for Houston, the annual Houston marine
emissions would be:
HC CO NOv
Estimated Houston emission (tons/year) 1,222 2,854 5,819
Percent of total Houston emissions 0.33 0.32 2.5
Even with these revised emissions, the marine contribution in the Houston
area is less than one percent for HC and CO, and less than 3 percent for NOx.
Thus for all three regions investigated, marine emissions were less than one
percent of the total HC and CO emissions, and less than three percent of the
total NOx emissions.
As with all determination of total emissions in an area, this analysis does
not address the effect of marine vessel emissions on microscale situations.
There are likely to be areas around docks and marine terminals where marine
vessels are the predominate sources of ambient air pollution. However, no
situations have been found in the literature in which marine vessel emissions
have caused an area to exceed ambient air standards.
Emissions estimates for the South Coast Air Basin (SCAB) have been
obtained for both locomotives and construction equipment. It was therefore
decided to also estimate the marine vessel NOx emission in the SCAB to
provide one region with the NOx emissions impact from all three off-highway
emission sources studied in this project. Marine vessel NOx emissions in the
SCAB were estimated using the total cargo tons handled in the ports of Los
Angeles and Long Beach in 1981 (75.06 million tons, from reference 58)
together with the cargo specific NOx emission factor from the Port of New
York (see Table IV-6). Using these figures, and the total SCAB NOx emissions
from reference 26, it is estimated that marine vessel emissions contributed
6771 tons of NOx, or 1.5 percent of the total SCAB annual NOx emissions.
Estimate of emissions based on cargo specific emission factors can only
be made in areas where the majority of the ships are being loaded or unloaded
since it is based on cargo handled. To obtain a nationwide estimate of marine
vessel emissions another method is required, since a large quantity of the
emissions will occur in the river networks. The method used was to estimate
the national marine emission percentages from the railroad percentages using
the fact that inland waterway traffic carries about 42 percent of the freight
railroads carry, but requires only approximately 81 percent of the fuel per
revenue ton-mile. The detailed calculations for this estimate are shown in
Appendix C. The resulting estimate of marine vessel contributions to
nationwide emissions are shown on the following page.
63
-------
National Annual Marine Vessel Emissions As:
Percent of Percent of
All Sources Mobile Sources
HC 0.2 0.6
CO 0.2 0.2
N Ox 2.1 5.0
64
-------
V. DISCUSSION, CONCLUSIONS, AND RECOMMENDATIONS
This study has examined emissions from three off-highway mobile source
categories. New emission factors were recommended for two of the categories,
and national and regional emissions were estimated for each category
individually. The study indicated that for the sources investigated, the largest
air pollution impact was from NOx emissions. On a nationwide basis, it is
estimated that in 1981 there were 19.5 million tons of NOx emitted into the
atmosphere(21). Mobile sources were estimated to contribute 8.2 million tons
of this NOx, or about 42 percent of the total^D. The contribution to
nationwide NOx emissions from the three sources investigated in this study are
shown below.
Percent of 1981 NOv Emissions from
Source Category All Sources Mobile Source
Railroads
4.6
11.0
Construction/equipment
2.0
4.7
Marine vessels
2.1
5.0
Total of 3 categories 8.7 20.7
Since on-highway mobile source NOx emissions standards will tend to reduce
mobile source NOx emissions as a whole, and because of projected increases in
the off-highway NOx due to increased economic activity (more goods shipped by
rail and water, and more construction activity), the percentages shown above
are likely to increase somewhat in the future.
Since emissions from the three categories are not uniform over the whole
country, it is more useful to examine emission impact by region. The NOx
emissions in tons per year from all three off-highway categories were
determined for the South Coast Air Basin (SCAB) in California. The SCAB
includes a major part of Los Angeles County, all of Orange County, and parts of
Riverside and San Bernardino Counties. As of early 1983, this area exceeded
the ambient air quality standard for ozone,1having the highest ozone
readings in the country'21). There were only 11 counties in the whole U.S. that
exceeded the NOx air quality standard(^). This area contains all or part of
four of those counties. The contributions from the three off-highway sources to
the 1979 (1977 for railroads) SCAB NOx emissions are shown below.
Percent of Total SCAB NOv Emissions
Category Low Estimate High Estimate
Railroads 1.7 6.1
Construction equipment 0.9 2.4
Marine vessels L5 L5
Total 4.1 10.0
65
-------
A low and a high estimate are shown because different methods and sources
produced different estimates. The railroad percentage estimates are for the
entire four counties, not just the portions of the counties in the SCAB. There
are other regions where the contributions for railroad emissions were higher. In
Kansas City and Chicago, for example, the high estimates of railroad
contributions NOx emissions in 1977 were approximately 9 and 8 percent,
respectively.
Conclusions
The following conclusions were derived from this study of emissions from
locomotives, construction equipment, and marine vessels.
Locomotives:
1. Over the past 10 years, a number of changes have occurred in railroad
operating practices and in locomotive engines themselves, due to
rapidly increasing fuel cost. These changes, together with the
availability of additional locomotive emissions data, ail indicated that
new locomotive emission factors were required.
2. The locomotive emission factors developed in this study in general
have lower HC and higher CO and NOx values than those in AP-42. As
an example, for the national average emission factors, HC is 44
percent, CO is 138 percent and NOx is 143 percent of the AP-42
values. The revised factors can be used to provide improved estimates
of the contribution of locomotives to local, regional, and national air
pollution totals.
3. A recently completed AAR study of 40 locomotives should provide a
far superior locomotive emissions data base, when the study results
are released.
4. There is a need for additional information on current locomotive
operating cycles.
5. Locomotive contributions to HC and CO levels, both on a national and
regional basis, are under two percent of total HC or CO Emissions.
Locomotive contributions to NOx emissions in 1981 were
approximately 4.6 percent of the national total. In the future, in some
regions, the locomotive contribution may exceed 10 percent of total
regional NOx emissions.
Construction Equipment:
1. A recent study, called the CAL/ERT study, sponsored by a consortium
of industry groups, has produced the most comprehensive investigation
of construction equipment emissions done to date.
66
-------
2. The CAL/ERT study emission factors are recommended as
replacements for the current construction equipment emission factors
in AP-42. These new factors do not differ greatly from those
currently in AP-42. The average of all new HC emission factors is
eight percent higher, CO is 41 percent higher and NOx is 12 percent
lower, than the AP-42 averages.
3. A deficiency in the CAL/ERT report is that the emissions factors are
not necessarily based on in-service units.
4. Construction equipment HC and CO are less than one percent of total
national HC and CO emissions. Construction equipment contributed
approximately two percent of the national NOx emissions for 1981,
and about 4.7 percent of the 1981 national mobile source NOx
emissions. As an example of regional impact, construction equipment
in the South Coast Air Basin of California may contribute as much as
2.4 percent of the total NOx emissions in the region.
Marine Vessels:
1. Over the past ten years, rising fuel costs have caused several changes
in marine vessel powerplants. One of the most significant is the
increased use of diesel engines in the U.S. merchant fleet. In addition,
marine diesel engine manufacturers are striving to increase the
efficiency of the engine while permitting operation with poorer grades
of fuel.
2. The data available to calculate marine vessel emissions and emissions
impact are extremely scarce.
3. Marine boiler emissions from either AP-42 or the draft report of EPA-
450/4-84-001 are adequate for present use.
4. Extending previous regional studies to a national estimate on the basis
of cargo tonnage, it appears that marine vessels contribute
approximately two percent of the total national NOx emissions, and
approximately five percent of the national mobile source NOx
emissions. Regionally, marine vessels can contribute as much as two
and one half percent of the NOx emissions. The fact that the marine
vessel contribution is higher at the national level than at the regional
level runs counter to the intuitive notion that the regional impact
should be higher. A possible reason for this difference is that the
majority of the ton-miles of cargo for this estimate are accumulated
on the U.S. river systems outside of urban areas.
Recommendations
The off-highway mobile emission sources investigated in this study are not
major sources of air pollution at this time. However, as more on-highway
vehicles meeting current and future emission regulations are put in service, off-
highway mobile sources will increase in importance, with air quality planners
becoming increasingly interested in them. At the present time, much of the
67
-------
information required to accurately assess the contribution of these off-highway
mobile sources is not available. To provide this information, a list of
recommended research work, in order of the priority perceived from this study,
is presented below.
1. A survey of marine vessels is required to determine how vessels should
be classed for emissions studies, and the population of each class in
U.S. waters. In addition, diesel engine population by manufacturer and
design horsepower is required. This latter information should be
obtainable from the computer data base of the American Bureau of
Shipping.
2. Once the popular makes and models of marine diesel engines and the
classes of vessels are defined, cycles for typical in-harbor operation of
each class of vessel and the more popular engines should be defined.
3. A project is required to measure actual marine diesel engine exhaust
emissions and fuel consumption at engine operating conditions defined
by the duty cycles. These measurements should be made on a number
of the most popular engine makers and models.
It is recommended that a study be conducted to define locomotive
duty cycles representing current operational practices. The duty
cycles should be based on a twenty-four hour day, and include the time
the engine is shut down.
5. While locomotive emission factor estimates are considered to be
better than those for construction equipment or marine vessels, the
emission results of the AAR 40-locomotive study just completed would
no doubt add greatly to the accuracy of locomotive emission factors.
It is recommended that these data from the AAR study, when they
become available, be incorporated into the locomotive emissions data
base, and new emission factors calculated.
6. The AAR should soon have a computer data base of locomotives in
service. It is recommended that this data base be used as soon as
available, to update information on number of engines by
manufacturer, model, and horsepower, as well as information on
number of line-haul and switch engines.
7. The construction equipment emission factors recommended in this
report are assumed to be for new engines. It is recommended that a
selection of in-use construction equipment be tested for emissions to
define the relationships between new and in-use emissions.
8. The negotiations between the state of California and the construction
industry group should be followed to determine what construction
equipment impact estimate is finally agreed upon for the California
regions.
68
-------
REFERENCES
1. "Compilation of Air Pollutant Emission Factors," EPA Report No. AP-^2.
Third Edition, Aug 1977. U.S. Environmental Protection Agency, Office of
Air Quality Planning and Standards, Research Triangle Park, N.C.
2. Hare, Charles T. and Springer, Karl J., "Exhaust Emissions From Controlled
Vehicles and Related Equipment Using Internal Combustion Engines. Part
1. Locomotive Diesel Engines and Marine Counterparts." Final Report by
Southwest Research Institute to the Environmental Protection Agency,
October, 1972.
3. U.S. Bureau of the Census, "Statistical Abstract of the United States:
1981." (102nd Edition) Washington, D.C. 1981.
k. Railway Age magazine, October 25, 1982, page 3, Simmons, Boardman
Publishing Corporation, Bristol, Conn.
5. "Yearbook of Railroad Facts - 1983." Association of American Railroads,
Washington, D.C.
6. "Getting More Ton-Miles to the Gallon", Railway Age, September 3, 1982,
page 23. Simmons - Boardman Publishing Corporation, Bristol, Conn.
7. Telephone Conversation between Mr. Richard Cataldi of the Association of
American Railroads and Mr. Melvin Ingalls of Southwest Research
Institute, October 1983.
8. McDonald, Charles W., "Diesel Locomotive Rosters: United States,
Canada, Mexico" Kalmbach Publishing Co., Milwaukee, Wisconsin, 1982.
9. "Statistics of Railroads of Class I in the United States, Years 1971 to
1981," Statistical Summary Number 66. Association of American
Railroads, Washington, D.C., December 1982.
10. "Climatic Atlas of the United States." National Oceanic and Atmospheric
Administration, National Climatic Center, Ashville, N.C., 1979.
11. "Report on Exhaust Emissions of Selected Railroad Diesel Locomotives"
jointly Funded by the Association of American Railroads, Atchison, Topeka
and Santa Fe Railway Co., Southern Pacific Transportation Co., and Union
Pacific Railroad Co. Report prepared by Southern Pacific Transportation
Co., San Francisco, Calif., March 1972.
12. Storment, 3ohn O. and Springer, Karl 3., "Assessment of Control
Techniques for Reducing Emissions from Locomotive Engines." Task VI
Final Report by Southwest Research Institute to the Environmental
Protection Agency under Contract EHS 70-108, April, 1973.
13. Hoffman, 3.G. 3r., Springer, K.J., and Tennyson, T.O., "Four Cycle Diesel
Electric Locomotive Exhaust Emissions: A Field Study" ASME Paper 75-
DGP-10, April, 1975.
69
-------
14. Ephraim, Max Jr., "Status Report on Locomotives as Source of Air
Pollution," SAE Paper 720604.
15. Kotlin, 3.3., Dunteman, N.R., William, H.A. 3r., and Scott, D.I., "The First
of a Series of High Efficiency, High BMEP Turbocharged Two-Stroke Cycle '
Diesel Engines - The General Motors EMD 645B Engine" ASME Paper 83-
DGP-7.
16. "Exhaust Emissions from Diesel Locomotives" Sub-council report, National
Industrial Pollution Control Council, Dept. of Commerce. U.S.
Government Printing Office, April, 1973.
17. Bryant, A. H. and Tennyson, T.O., "Exhaust Emissions of Selected Railroad
Diesel Locomotives" ASME Paper 74-WA/RT-l, November, 1974.
18. Hare, C.T., Springer, K.J., and Huls, T.A., "Locomotive Exhaust Emissions
and Their Impact" ASME Paper 74-DGP-3, April, 1974.
19. Storment, 3. O., Springer, K. 3., AND Hergenrother, K. M., "NOx Studies
with EM 2-567 Diesel Engine" ASME Paper 74-DGP-14, April 1974.
20. Hoffman, 3. G. 3r., et al, "Four Cycles Diesel Electric Locomotive Exhaust
Emissions: A Field Study," ASME Paper 75-DGP-10.
21. "National Air Quality and Emissions Trends Report, 1981," Report No. EPA
450/4-83-011, U.S. Environmental Protection Agency, Office of Air Quality
Planning and Standards, Research Triangle Park, N.C.
22. Wiltsee, Kenneth, W., "Assessment of Railroad Fuel Use and Emissions for
the Regional Air Pollution Study." EPA Report 450/3-77-025, April, 1977.
23. Zinger, D. E. and Hecker, L. H., "Gaseous Emissions from Unregulated
Mobile Sources." Final Report by the University of Michigan School of
Public health to EPA under Grant No. R-803568-01-0.
24. "Report to Congress on Railroad Emissions - A Study Based on Existing
Data." Draft report prepared by Emission Control Technology Division of
the U.S. EPA.
25. Hare, C. T. and Springer, K. 3., "Exhaust Emissions from Uncontrolled
Vehicles and Related Equipment Using Internal Combustion Engines. Final
Report, Part 5, Heavy-Duty Farm Construction and Industrial Engines."
Final Report to the Environmental Protection Agency by Southwest
Research Institute under contract EHS 70-108, October, 1973.
26. "Feasibility, Cost and Air Quality Impact of Potential Emission Control
Requirements on Farm, Construction, and Industrial Equipment in
California" Prepared by Environmental Research and Technology, Inc,
under sponsorship of the Farm and Industrial Equipment Institute, Engine
Manufacturers Association, and Construction Industry Manufacturers
Association. Document PA841, May, 1982.
70
-------
27. "Status Report: Emissions Inventory on Non-Farm (MS-1), Farm (MS-2) and
and Lawn and Garden (Utility) (MS-3) Equipment" Prepared by the Mobile
Source Control Division, California Air Resources Board, El Monte,
California, July, 1983.
28. Telephone conversation between Mr. Rod Summerfield of the California
Air Resources Board, El Monte, California, and Mr. Melvin Ingalls of
Southwest Research Institute, San Antonio, Texas on January 6, 1984.
29. Williams, R. E., et al, "Future Marine Fuels - Prediction and Alleviation of
Potential Combustion and Lubrication Problems," ASME Paper 84-DGP-ll.
American Society of Mechanical Engineers, New York, N.Y.
30. Gillmer, T. C., Modern Ship Design, U.S. Naval Institute Press, Annapolis,
Maryland.
31. Marine Engineering/Log Magazine, January, 1983, page 74.
32. Marine Engineering/Log, Year Book Issue, June 15, 1983.
33. Telephone conversation between Melvin Ingalls, Southwest Research
Institute and Stuart Dattner, Texas Air Control Board, Austin, Texas,
November 23, 1983.
34. Marine Engineering/Log, January 1983, pages 74-76.
35. Marine Engineering/Log, August 1983, page 60.
36. Marine Engineering/Log, June, 1981, pages 43 to 46.
37. "645 EC is EMD's Most Fuel Efficient Marine Diesel Yet," Marine
Engineering/Log, January, 1983, pages 69-72.
38. "Marine Engineering/Log," various issues.
39. "Marine Reporter and Engineering News," various issues.
40. "Summary of U.S. Flag Passenger and Cargo Vessels Operating or Available
for Operation on 1 May 1982." U.S. Corp of Engineers, Waterborne
Commerce Statistics Center, New Orleans, LA.
41. Telephone conversation between Melvin Ingalls, Southwest Research
Institute and Marcie Kinter, American Waterways Operators Association,
November 17, 1983.
42. Telephone conversation between Melvin Ingalls, Southwest Research
Institute and Mr. Anders, Corp. of Engineers, Waterbone Commerce
Statistics Center, New Orleans, LA., November 17, 1983.
43. Telephone conversation between Melvin Ingalls, Southwest Research
Institute and Mr. Ron Olander, Dravo Marine Equipment Company,
Pittsburgh, PA., November 18, 1983.
71
-------
44. Telephone conversation between Melvin Ingalls, Southwest Research
Institute and Mr. Jim Manika, Marketing Research, Electromotive Division,
La Grange, IL, November 21, 1983.
45. "World Hub of Maritime Activities, New York - New Jersey," Marine
Engineering/Log, June, 1981, page 74.
46. "Cutting Propulsion Costs" Marine Engineering/Log, August, 1979.
47. Marine Engineering/Log, August, 1983, page 59.
48. Marine Engineering/Log, June, 1982, page 15.
49. "The 645 and Blends," Marine Engineering/Log, January, 1983.
50. Personal Communication dated 22 November, 1983, with Ms. M. J. Miller,
ABS Computers, Inc., New York, New York.
51. Baker, Q. A. and Storment, J. O., "Coast Guard Cutter Duty Cycle and
Propellers/Diesel Engine Efficiency Study," Report No. CG-D-26-80. Final
Report by Southwest Research Institute to U.S. Dept. of Transportation
under Contract DOT-TSC-920, January, 1981.
52. Storment, J. O., Wood, C. D., and Mathis, R. J., "A Study of Fuel Economy
and Emission Reduction Methods for Marine and Locomotive Diesel
Engines," Report No. DOT-TSC-OST-75-41 and CG-D-124-75. Interim
report by Southwest Research Institute to U.S. Dept. of Transportation
under Contract DOT-TSC-920, September, 1975.
53. Schueneman, J. J., "Some Aspects of Marine Air Problems on the Great
Lakes," J. Air. Pol. Control Assoc. 14:23-29, September, 1964.
54. Pearson, J. R. "Ships as Sources of Emissions," Puget Sound Air Pollution
Control Agency, Seattle, Wash., (Presented at the Annual Meeting of the
Pacific Northwest International Section of the Air Pollution Control
Association, Portland, Ore., November, 1969.)
55. Sturm, Joseph C., "An Estimation of River Towboat Air Pollution in St.
Louis, Missouri," Report No. DOT-TSC-OST-75-42, NTIS No. PB251 711.
U.S. Dept. of Transportation, February, 1976.
56. Dattner, S. L., Ledbetter, J. O., and Miksod, R. W., "A method for Rapid
Calculation of Merchant Vessel Combustion Emissions," Journal of the Air
Pollution Control Assoc., Vol. 30, No. 3, March, 1980, pages 305 to 309.
57. Computer Printout titled, "Counties Not Meeting the National Ambient Air
Quality Standard," dated 2/3/83. From the EPA Control Program
Development Division, Office of Air Quality Planning and Standards,
Research Triangle Park, N.C.
58. "Waterborne Commerce of the United States, calendar year 1981. Part 5,
National Summaries," report number WRSC-WCUS-81-5. U.S. Army Corp.
of Engineers, Waterboone Commerce Statistics Center, New Orleans, LA.
72
-------
APPENDIX A
EXAMPLES OF PLOTS FROM THE EKMA COMPUTER PROGRAM SHOWING
THE RELATIONSHIP BETWEEN HC, NOx AND O3
-------
>
I
K>
2,0 2.5
NMHC, PPMC
«X SCUD ** JULY 13. 1970
-------
o
O)
o
. 120 °
U)
L'J
CO
(\i
C\J
o
X
CT>
iJ
/
tO
IO
CJ
o
0
trt
01
u:
O
O
c<
mm SACRAMENTO ** JULY 21. 1970
-------
APPENDIX B
SUPPORTING MATERIAL FOR LOCOMOTIVE EMISSION FACTOR
DEVELOPMENT AND IMPACT DETERMINATION
-------
Calculation of Average Daily Locomotive
Operating Hours^'
A. Freight Locomotives:
annual diesel locomotive miles = 1.316 x 10^
diesel locomotive average speed = 18.2 miles/hour
annual diesel locomotive hours = (1.316 x 10^)/(18.2) = 72320384
number of diesel locomotives in freight service = 21592
annual operating hours per locomotive = 72320384/21592 = 3349
daily operating hours per locomotive = 3344/365 = 9.2
B. Passenger Locomotives:
annual diesel locomotive miles = 48.423 x 10^
diesel locomotive average speed = 31.3 miles/hour
C. Switching Locomotives:
total annual diesel yard switching miles = 167.734 x 10^
number of diesel switching locomotive = 5511
annual diesel yard switching miles per locomotive = (167,734 x 10*>)/5511
= 30436
daily switching miles per locomotive = 30436/365 = 83.387
fo^Based on 1980 Statistics as given in Statistical Summary Numbers 66.
B-2
-------
ESTIMATE OF PERCENT OF OPERATING TIME THAT AMBIENT
TEMPERATURE IS ABOVE 50 F
1. From "Climatic Atlas of the United States" the areas of the country
where the average minimum temperature is above 50 F by month.
Month Areas Where Avg. Minimum Temp, is 50°F
January All of Florida; Brownsville, TX.
February All of Florida; Brownsville, TX.
March Florida, South Texas, Louisiana
April FL, GA, SC, AL, MS, LA, AR, TX, Desert Southwest, Southern
California
May All states east of Colorado and South of Missouri, California,
most of Arizona and Northwest
June All of country except high Rockies
July All of country except high Rockies
August All of country except high Rockies
September All of country east of Colorado and South of Minnesota and
Vermont. Also CA, AZ, southern New Mexico.
October Same as April
November Same as March
December Same as January
B-3
-------
2. Calculation of Fraction of Time Available
Months
Area
Fraction of
Annual hr
Fraction
of Countya
Product
November through
February
Florida and
Brownsville, TX
0.330
0.02
0.0066
March
FL, South TX, LA
0.083
0.05
0.0042
April and May
Southern U.S.
0.167
0.50
0.0835
June through
August
All of U.S.
0.250
1.00
0.2500
September
2/3 of U.S.
0.083
0.67
0.1680
October
Southern U.S.
0.083
0.50
Total
0.0420
0.5500
aFraction of railroad miles used with small area for November through
February
and for March
3. Thus, approximately 55 percent of the time weather conditions will permit
engine shut down.
B-4
-------
APPENDIX C
SUPPORTING MATERIAL FOR MARINE EMISSION FACTORS
DEVELOPMENT AND IMPACT DETERMINATION
-------
Calculation of National Marine
Vessel Emissions Impact
1. The marine vessel emissions are divided into two parts: the river
contribution and the ocean harbor contribution. The river contribution is
calculated by comparison to the railroad contribution. The ocean harbor
contribution directly from cargo specific emission factor and total tons of
coastal cargo.
2. River Contribution
(a) From "Railroad Facts, 1983(1) in 1981, Great Lakes and river traffic
accounted for 16 percent of the total freight shipped in the U.S.
Railroads accounted for 37.9 percent. Therefore, inland waterborne
traffic accounted for:
inland water = 16/37.9 = 0.422
(b) Inland vessels use 3.3 gal/1000 ton-miles^). Railroads in 1981 used
4.08 gal/1000 revenue ton-miles'^ Thus inland vessels would use
3.3/4.08 = 0.81 of fuel railroads use to carry the same ton-miles of
cargo
(c) Inland vessels in 1981 use the following percent of railroad fuel:
0.81 x .422 = 0.34 or 34 percent of the
fuel used by the railroads
(d) Assuming inland vessels (which are mostly powered by EMD engines)
have the same fuel specific emissions as railroads the inland vessel
contribution to national emissions is 34 percent of the railroads
contribution.
Railroad Inland Vessels ( RRx.34)
Percent of Percent of Percent of Percent of
All Sources Mobile Sources All Sources Mobile Sources
HC 0.34 0.94 0.12 0.32
CO 0.35 0.46 0.12 0.16
NOx 4.6 11 1.6 3.7
3. Coastal Contribution
(a) In 1981 foreign and coastal cargo handled amounted to 1179.1 x 106
tons(2)
C-2
-------
(b) From text cargo specific emission factors for coastal areas are
represented by the New York harbor cargo specific emission factors.
The tons of emissions per year ares
HC = (19 tons/106 tons)xl 179.1xl06 tons = 22,403 tons/year
CO = (44 tons/106 tons)xll79.1xl06 tons = 51,880 tons/year
NOx = (90 tons/106 tons)xl 179.1xl06 tons = 106,119 tons/year
(c) As a percent of total nationwide emissions using the 1981 Emissions
Report^);
Percent of Percent of
All Sources Mobile Sources
HC (22,403/21.0xl06)xl00 = 0.11 (22,403/7.5 x 106) = 0.30
CO (51,880/91xl06)xl00 = 0.06 (51,880/70xl06xl00) = 0.07
NOx (106,119/19.5xl06)xl00 = 0.54 (106,119/8.2xl06)xl00 = 1.29
4. Total Marine Impact
Emissions Percent of Total
HC CO NOy
All
Sources
Mobile
Sources
Oil
Sources
Mobile
Sources
All
Sources
Mobile
Sources
River
0.12
0.32
0.12
0.16
1.6
3.7
Coastal
0.11
0.30
0.06
0.07
0.5
hi
Total
0.23
0.62
0.18
0.23
2.1
5.0
References
1. "Yearbook of Railroad Facts - 1982." Association of American
Railroads, Washington, D.C.
2. Dattner, S.L., Ledbetter, J. O., and Miksod, R. W., "A Method for Rapid
Calculation of Merchant Vessel Combustion Emissions," Journal of the
Air Pollution Control Assoc., Vol. 30, No. 3, March, 1980, pages 305 to
309.
3. "National Air Quality and Emissions Trends Report, 1981, "Report No.
EPA 450/4-83-011, U.S. Environmental Protection Agency, Office Air
Quality Planning and Standards, Research Triangle Park, NC.
C-3
-------
APPENDIX D
AP-42 SECTIONS CONTAINING EMISSION FACTORS FOR LOCOMOTIVES,
CONSTRUCTION EQUIPMENT AND MARINE VESSELS
-------
3.2.2 Locomotives
by David S. Kircher
3.2.2.1 General - Railroad locomotives generally follow one of two use patterns: railyard switching or road-haul
service. Locomotives can be classified on the basis of engine configuration and use pattern into five categories:
2-stroke switch locomotive (supercharged), 4-stroke switch locomotive, 2-stroke road service locomotive
(supercharged), 2-stroke road service locomotive (turbocharged), and 4-stroke road service locomotive.
The engine duty cycle of locomotives is much simpler than many other applications involving diesel internal
combustion engines because locomotives usually have only eight throttle positions in addition to idle and
dynamic brake. Emission testing is made easier and the results are probably quite accurate because of the
simplicity of the locomotive duty cycle.
3.2.2.2 Emissions - Emissions from railroad locomotives are presented two ways in this section. Table 3.2.2-1
contains average factors based on the nationwide locomotive population breakdown by category. Table 3.2.2-2
gives emission factors by locomotive category on the basis of fuel consumption and on the basis of work output
(horsepower hour).
The calculation of emissions using fuel-based emission factors is straightforward. Emissions are simply the
product of the fuel usage and the emission factor. In order to apply the work output emission factor, however, an
Table 3.2.2-1. AVERAGE LOCOMOTIVE
EMISSION FACTORS BASED
ON NATIONWIDE STATISTICS*
Pollutant
Average
emissions'5
lb/103 gal
kg/103 liter
Particulates0
25
3.0
Sulfur oxidesd
57
6.8
(SOx as SO2)
Carbon monoxide
130
16
Hydrocarbons
94
11
Nitrogen oxides
370
44
(NOx as N02)
Aldehydes
5.5
0.66
(as HCHO)
Organic acids0
7
0.84
8 Reference 1.
b Bated on emission data contained in Table 3.2.2-2
and the breakdown of locomotive use by engine
category in the United States in Reference 1.
c Data based on highway diesel data from Reference
2. No actual locomotive particulate test data are
available.
^ Based on a fuel sulfur content of 0.4 percent from
Reference 3.
4/73 Internal Combustion Engine Sources 3.2.2-1
D-2
-------
Table 3.2.2-2. EMISSION FACTORS BY LOCOMOTIVE ENGINE
CATEGORY8
EMISSION FACTOR RATING: B
Engine category
2-Stroke
2-Stroke
2-Stroke
supercharged
4-Stroke
supercharged
turbocharged
4-Stroke
Pollutant
switch
switch
road
road
road
Carbon monoxide
lb/103 gal
84
380
66
160
180
kg/103 liter
10
46
7.9
19
22
g/hphr
3.9
13
1.8
4.0
4.1
g/metric hphr
3.9
13
1.8
4.0
4.1
Hydrocarbon
lb/103 gal
190
146
148
28
99
kg/103 liter
23
17
18
3.4
12
g/hphr
8.9
5.0
4.0
0.70
2.2
g/metric hphr
8.9
5.0
4.0
0.70
2.2
Nitrogen oxides
(NOx as N02)
lb/103 gal
250
490
350
330
470
key'lO3 liter
30
59
42
40
56
g/hphr
11
17
9.4
8.2
10
g/metric hphr
11
17
9.4
8.2
10
a Use average factors (Table 3.2.2-1) for pollutants not listed in this table.
additional calculation is necessary. Horsepower hours can be obtained using the following equation:
w=lph
where: w = Work output (horsepower hour)
l = Load factor (average power produced during operation divided by available power)
p = Available horsepower
h = Hours of usage at load factor (1)
After the work output has been determined, emissions are simply the product of the work output and the
emission factor. An approximate load factor for a line-haul locomotive (road service) is 0.4; a typical switch
engine load factor is approximately 0.06.1
References for Section 3.2.2
1. Hare, C.T. and K.J. Springer. Exhaust Emissions from Uncontrolled Vehicles and Related Equipment Using
Internal Combustion Engines. Part 1. Locomotive Diesel Engines and Marine Counterparts. Final Report.
Southwest Research Institute. San Antonio, Texas Prepared for the Environmental Protection Agency,
Research Triangle Park, N.C., under Contract Number EHA 70-108. October 1972.
2. Young, T.C. Unpublished Data from the Engine Manufacturers Association. Chicago, 111. May 1970.
3. Hanley, G.P. Exhaust Emission Information on Electro-Motive Railroad Locomotives and Diesel Engines.
General Motors Corp. Warren, Mich. October 1971.
3.2.2-2 EMISSION FACTORS 4/73
D-3
-------
3.2.7 Heavy-Duty Construction Equipment
by David S. Kircher
3.2.7.1 General - Because few sales, population, or usage data are available for construction equipment, a number
of assumptions were necessary in formulating the emission factors presented in this section.* The useful life of
construction equipment is fairly short because of the frequent and severe usage it must endure. The annual usage of
the various categories of equipment considered here ranges from 740 hours (wheeled tractors and rollers) to 2000
hours (scrapers and off-highway trucks). This high level of use results in average vehicle lifetimes of only 6 to 16
years. The equipment categories in this section include: tracklaying tractors, tracklaying shovel loaders, motor
graders, scrapers, off-highway trucks, wheeled loaders, wheeled tractors, rollers, wheeled dozers, and miscellaneous
machines. The latter category contains a vast array of less numerous mobile and semi-mobile machines used in
construction, such as, belt loaders, cranes, pumps, mixers, and generators. With the exception of rollers, the
majority of the equipment within each category is diesel-powered.
3.2.7.2 Emissions — Emission factors for heavy-duty construction equipment are reported in Table 3.2.7-1 for
diesel engines and in Table 3.2.7-2 for gasoline engines. The factors are reported in three different forms—on the
basis of running time, fuel consumed, and power consumed. In order to estimate emissions from time-based
emission factors, annual equipment usage in hours must be estimated. The following estimates of use for the
equipment listed in the tables should permit reasonable emission calculations.
Category
Annual operation, hours/year
Tracklaying tractors
1050
Tracklaying shovel loaders
1100
Motor graders
830
Scrapers
2000
Off-highway trucks
2000
Wheeled loadecs
1140
Wheeled tractors
740
Rollers
740
Wheeled dozers
2000
Miscellaneous
1000
The best method for calculating emissions, however, is on the basis of "brake specific" emission factors (g/kWh
or g/hphr). Emissions are calculated by taking the product of the brake specific emission factor, the usage in hours,
the power available (that is, rated power), and the load factor (the power actually used divided by the power
available).
References for Section 3.2.7
1. Hare, C. T. and K. J. Springer. Exhaust Emissions from Uncontrolled Vehicles and Related Equipment Using
Internal Combustion Engines - Final Report. Part 5: Heavy-Duty Farm, Construction, and Industrial Engines.
Southwest Research Institute, San Antonio, Tex. Prepared for Environmental Protection Agency, Research
Triangle Park, N.C., under Contract No. EHS 70-108. October 1973. 105 p.
2. Hare, C. T. Letter to C. C. Masser of Environmental Protection Agency, Research Triangle Park, N.C.,
concerning fuel-based emission rates for farm, construction, and industrial engines. San Antonio, Tex. January
14,1974.4 p.
1/75 Internal Combustion Engine Sources 3.2.7-1
D-4
-------
Table 3.2.7-1. EMISSION FACTORS FOR HEAVY-DUTY, DIESEL-POWERED CONSTRUCTION
EQUIPMENT3
EMISSION FACTOR RATING: C
Tracklaying
Wheeled
Wheeled
Motor
Pollutant
tractor
tractor
dozer
Scraper
grader
Carbon monoxide
g/hr
175.
973.
335.
660.
97.7
Ib/hr
0.386
2.15
0.739
1.46
0.215
g/kWh
3.21
5.90
2.45
3.81
2.94
g/hphr
2.39
4.40
1.83
2.84
2.19
kg/103 liter
10.5
19.3
7.90
11.8
9.35
lb/103 gal
87.5
161.
65.9
98.3
78.0
Exhaust hydrocarbons
g/hr
50.1
67.2
106.
284.
24.7
Ib/hr
0.110
0.148
0.234
0.626
0.054
g/kWh
0.919
1.86
0.772
1.64
0.656
g/hphr
0.685
1.39
0.576
1.22
0.489
kg/103 liter
3.01
6.10
2.48
5.06
2.09
lb/103 gal
25.1
50.9
20.7
42.2
17.4
Nitrogen oxides
(NOx as N02)
478.
g/hr
665.
451.
2290.
2820.
Ib/hr
1.47
0.994
5.05
6.22
1.05
g/kWh
12.2
12.5
16.8
16.2
14.1
g/hphr
9.08
9.35
12.5
12.1
10.5
kg/103 liter
39.8
41.0
53.9
50.2
44.8
lb/103 gal
332.
342.
450.
419.
374.
Aldehydes
(RCHO as HCHO)
g/hr
12.4
13.5
29.5
65.
5.54
Ib/hr
0.027
0.030
0.065
0.143
0.012
g/kWh
0.228
0.378
0.215
0.375
0.162
g/hphr
0.170
0.282
0.160
0.280
0.121
kg/103 liter
0.745
1.23
0.690
1.16
0.517
lb/103 gal
6.22
10.3
5.76
9.69
4.31
Sulfur oxides
(SOv as SO2)
g?hr
62.3
40.9
158.
210.
39.0
Ib/hr
0.137
0.090
0.348
0.463
0.086
g/kWh
1.14
1.14
1.16
1.21
1.17
g/hphr
0.851
0.851
0.867
0.901
0.874
kg/103 liter
3.73
3.73
3.74
3.74
3.73
lb/103 gal
31.1
31.1
31.2
31.2
31.1
Particulate
g/hr
50.7
61.5
75.
184.
27.7
Ib/hr
0.112
0.136
0.165
0.406
0.061
g/kWh
0.928
1.70
0.551
1.06
0.838
g/hphr
0.692
1.27
0.411
0.789
0.625
kg/103 liter
3.03
5.57
1.77
3.27
2.66
lb/103 gal
25.3
46.5
14.8
27.3
22.2
aReferences 1 and 2.
3.2.7-2
EMISSION FACTORS
D-5
1/75
-------
Table 3.2.7-1 (continued). EMISSION FACTORS FOR HEAVY-DUTY, DIESEL-POWERED
CONSTRUCTION EQUIPMENT3
EMISSION FACTOR RATING: C
Off-
Wheeled
Tracklaying
Highway
Miscel-
Pollutant
loader
loader
truck
Roller
laneous
Carbon monoxide
g/hr
251.
72.5
610.
83.5
188.
Ib/hr
0.553
0.160
1.34
0.184
0.414
g/kWh
3.51
2.41
3.51
4.89
3.78
g/hphr
2.62
1.80
2.62
3.65
2.82
kg/103 liter
11.4
7.90
11.0
13.7
11.3
lb/103 gal
95.4
65.9
92.2
114.
94.2
Exhaust hydrocarbons
g/hr
84.7
14.5
198.
24.7
71.4
Ib/hr
0.187
0.032
0.437
0.054
0.157
g/kWh
1.19
0.485
1.14
1.05
1.39
g/hphr
0.888
0.362
0.853
0.781
1.04
kg/103 liter
3.87
1.58
3.60
2.91
4.16
lb/103 gal
32.3
13.2
30.0
24.3
34.7
Nitrogen oxides
(NOx as NO2)
g/hr
1090.
265.
3460.
474.
1030.
Ib/hr
2.40
0.584
7.63
1.04
2.27
g/kWh
15.0
8.80
20.0
21.1
19.8
g/hphr
11.2
6.56
14.9
15.7
14.8
kg/103 liter
48.9
28.8
62.8
58.5
59.2
lb/103 gal
408.
240.
524.
488.
494.
Aldehydes
(RCHO as HCHO)
g/hr
18.8
4.00
51.0
7.43
13.9
Ib/hr "
0.041
0.009
0.112
0.016
0.031
g/kWh
0.264
0.134
0.295
0.263
0.272
g/hphr
0.197
0.100
0.220
0.196
0.203
kg/103 liter
0.859
0.439
0.928
0.731
0.813
lb/103 gal
7.17
3.66
7.74
6.10
6.78
Sulfur oxides
(SOx as SO2)
g/hr
82.5
34.4
206.
30.5
64.7
Ib/hr
0.182
0.076
0.454
0.067
0.143
g/kWh
1.15
1.14
1.19
1.34
1.25
g/hphr
0.857
0.853
0.887
1.00
0.932
kg/103 liter
3.74
3.74
3.74
3.73
3.73
lb/103 gal
31.2
31.2
31.2
31.1
31.1
Particulate
g/hr
77.9
26.4
116.
22.7
63.2
Ib/hr
0.172
0.058
0.256
0.050
0.139
g/kWh
1.08
0.878
0.673
1.04
1.21
g/hphr
0.805
0.655
0.502
0.778
0.902
kg/103 liter
3.51
2.88
2.12
2.90
3.61
lb/103 gal
29.3
24.0
17.7
24.2
30.1
'References 1 and 2.
1/75 Internal Combustion Engine Sources 3.2.7-3
D-6
-------
Table 3.2.7-2. EMISSION FACTORS FOR HEAVY-DUTY GASOLINE-POWERED
CONSTRUCTION EQUIPMENT3
EMISSION FACTOR RATING: C
Wheeled
Motor
Wheeled
Miscel-
Pollutant
tractor
grader
loader
Roller
laneous
Carbon monoxide
g/hr
4320.
5490.
7060.
6080.
7720.
Ib/hr
9.52
12.1
15.6
13.4
17.0
g/kWh
190.
251.
219.
271.
266.
g/hphr
142.
187.
163.
202.
198.
kg/103 liter
389.
469.
435.
460.
475.
lb/103 gal
3250.
3910.
3630.
3840.
3960.
Exhaust hydrocarbons
g/hr
164.
186.
241.
277.
254.
Ib/hr
0.362
0.410
0.531
0.611
0.560
g/kWh
7.16
8.48
7.46
12.40
8.70
g/hphr
5.34
6.32
5.56
9.25
6.49
kg/103 liter
14.6
15.8
14.9
21.1
15.6
lb/103 gal
122.
132.
124.
176.
130.
Evaporative
hydrocarbons'5
g/hr
30.9
30.0
29.7
28.2
25.4
Ib/hr
0.0681
0.0661
0.0655
0.0622
0.0560
Crankcase
hydrocarbons^
g/hr
32.6
37.1
48.2
55.5
50.7
Ib/hr
0.0719
0.0818
0.106
0.122
0.112
Nitrogen oxides
(NOx as N02>
187.
g/hr
195.
145.
235.
164.
Ib/hr
0.430
0.320
0.518
0.362
0.412
g/kWh
8.54
6.57
7.27
7.08
6.42
g/hphr
6.37
4.90
5.42
5.28
4.79
kg/103 liter
17.5
12.2
14.5
12.0
11.5
lb/103 gal
146.
102.
121.
100.
95.8
Aldehydes
(RCHO as HCHO)
g/hr
7.97
8.80
9.65
7.57
9.00
Ib/hr
0.0176
0.0194
0.0213
0.0167
0.0T98
g/kWh
0.341
0.386
0.298
0.343
0.298
g/hphr
0.254
0.288
0.222
0.256
0.222
kg/103 liter
0.697
0.721
0.593
0.582
0.532
lb/103 gal
5.82
6.02
4.95
4.86
4.44
Sulfur oxides
(SOx as SO2)
10.6
g/hr
7.03
7.59
10.6
8.38
Ib/hr
0.0155
0.0167
0.0234
0.0185
0.0234
g/kWh
0.304
0.341
0.319
0.373
0.354
g/hphr
0.227
0.254
0.238
0.278
0.264
kg/103 liter
0.623
0.636
0.636
0.633
0.633
lb/103 gal
5.20
5.31
5.31
5.28
5.28
3.2.7-4
EMISSION FACTORS
D-7
1/75
-------
Table 3.2.7-2. (continued). EMISSION FACTORS FOR HEAVY-DUTY GASOLINE-POWERED
CONSTRUCTION EQUIPMENT3
EMISSION FACTOR RATING: C
Wheeled
Motor
Wheeled
Miscel-
Pollutant
tractor
grader
loader
Roller
laneous
Particulate
g/hr
10.9
9.40
13.5
11.8
11.7
Ib/hr
0.0240
0.0207
0.0298
0.0260
0.0258
g/kWh
0.484
0.440
0.421
0.527
0.406
g/hphr
0.361
0.328
0.314
0.393
0.303
kg/103 liter
0.991
0.822
0.839
0.895
0.726
lb/103 gal
8.27
6.86
7.00
7.47
6.06
aReferences 1 and 2.
^Evaporative and crankcase hydrocarbons based on operating time only (Reference 1).
1/75
Internal Combustion Engine Sources
D-8
3.2.7-5
-------
3.2.3 Inboard-Powered Vessels
Revised by David S. Kircher
3.2.3.1 General - Vessels classified on the basis of use will generally fall into one of three categories: commercial,
pleasure, or military. Although usage and population data on vessels are, as a rule, relatively scarce, information on
commercial and military vessels is more readily available than data on pleasure craft. Information on military
vessels is available in several study reports,1"5 but data on pleasure craft are limited to sales-related facts and
figures.6-10
Commercial vessel population and usage data have been further subdivided by a number of industrial and
governmental researchers into waterway classifications'1-16 (for example, Great Lakes vessels, river vessels, and
coastal vessels). The vessels operating in each of these waterway classes have similar characteristics such as size,
weight, speed, commodities transported, engine design (external or internal combustion), fuel used, and distance
traveled. The wide variation between classes, however, necessitates the separate assessment of each of the waterway
classes with respect to air pollution.
Information on military vessels is available from both the U.S. Navy and the U.S. Coast Guard as a result of
studies completed recently. The U.S. Navy has released several reports that summarize its air pollution assessment
work.3"5 Emission data have been collected in addition to vessel population and usage information. Extensive
study of the air pollutant emissions from U.S. Coast Guard watercraft has been completed by the U.S. Department
of Transportation. The results of this study are summarized in two reports.1"2 The first report takes an in-depth
look at population/usage of Coast Guard vessels. The second report, dealing with emission test results, forms the
basis for the emission factors presented in this section for Coast Guard vessels as well as for non-military diesel
vessels.
Although a large portion of the pleasure craft in the U.S. are powered by gasoline outboard motors (see section
3.2.4 of this document), there are numerous larger pleasure craft that use inboard power either with or without
"out-drive" (an outboard-like lower unit). Vessels falling into the inboard pleasure craft category utilize either Otto
cycle (gasoline) or diesel cycle internal combustion engines. Engine horsepower varies appreciably from the small
"auxiliary" engine used in sailboats to the larger diesels used in yachts.
3.2.3.2 Emissions
Commercial vessels. Commercial vessels may emit air pollutants under two major modes of operation:
underway and at dockside (auxiliary power).
Emissions underway are influenced by a great variety of factors including power source (steam or diesel). engine
size (in kilowatts or horsepower), fuel used (coal, residual oil, or diesel oil), and operating speed and load.
Commercial vessels operating within or near the geographic boundaries of the United States fall into one of the
three categories of use discussed above (Great Lakes, rivers, coastline). Tables 3.2.3-1 and 3.2.3-2 contain emission
information on commercial vessels falling irrto these three categories. Table 3.2.3-3 presents emission factors tor
diesel marine engines at various operating modes on the basis of horsepower. These data are applicable to any vessel
having a similar size engine, not just to commercial vessels.
Unless a ship receives auxiliary steam from dockside facilities, goes immediately into drydock, or is out of
operation after arrival in port, she continues her emissions at dockside. Power must be made available for the ship's
lighting, heating, pumps, refrigeration, ventilation, etc. A few steam ships use auxiliary engines (diesel) to supply
power, but they generally operate one or more main boilers under reduced draft and lowered fuel rates-a very
inefficient process. Motorships (ships powered by internal combustion engines) normally use diesel-powered
generators to furnish auxiliary power.17 Emissions from these diesel-powered generators may also be a source of
underway emissions if they are used away from port. Emissions from auxiliary power systems, in terms of the
1/75 Internal Combustion Engine Sources 3.2.3-1
D-9
-------
Table 3.2.3-1. AVERAGE EMISSION FACTORS FOR
COMMERCIAL MOTORSHIPS BY WATERWAY
CLASSIFICATION
EMISSION FACTOR RATING: C
Class0
Emissions3
River
Great Lakes
Coastal
Sulfur oxides'3
(SOx as SO2)
kg/103 liter
lb/103 gal
3.2
27
3.2
27
3.2
27
Carbon monoxide
kg/103 liter
lb/103 gal
12
100
13
110
13
110
Hydrocarbons
kg/103 liter
lb/103 gal
6.0
50
7.0
59
6.0
50
Nitrogen oxides
(NOx as NO2)
kg/103 liter
lb/103 gal
33
280
31
260
32
270
aExpressed as function of fuel consumed (based on emission data from
Reference 2 and population/usage data from References 11 through 16.
^Calculated, not measured. Based on 0.20 percent sulfur content fuel
and density of 0.854 kg/liter (7.12 lb/gal) from Reference 17.
cVery approximate particulate emission factors from Reference 2 are
470 g/hr (1.04 Ib/hr). The reference does not contain sufficient
information to calculate fuel-based factors.
quantity of fuel consumed, are presented in Table 3.2.3-4. In some instances, fuel quantities used may not be
available, so calculation of emissions based on kilowatt hours (kWh ) produced may be necessary. For operating
loads in excess of zero percent, the mass emissions (e^) in kilograms p$r hour (pounds per hour) are given by:
ej = kief (1)
where: k = a constant that relates fuel consumption to kilowatt hours,2
that is, 3.63 xlO"4 1000 liters fuel/kWh
or
9.59 xlO"5 1000 gal fuel/kWh
1 = the load, kW
ef = the fuel-specific emission factor from Table 3.2.3-4, kg/103 liter (lb/103 gal)
3.2.3-2 EMISSION FACTORS 1/75
D-] 0
-------
Table 3.2.3-2. EMISSION FACTORS FOR COMMERCIAL STEAMSHIPS-ALL GEOGRAPHIC AREAS
EMISSION FACTOR RATING: D
Fuel and operating mode3
Residual oilb
Distillate oil'3
Hoteling
Cruise
Full
Hoteling
Cruise
Full
Pollutant
kg/103
liter
lb/103
gal
kg/103
liter
lb/103
gal
kg/103
liter
lb/103
gal
kg/103
liter
lb/103
gal
kg/103
liter
lb/103
gal
kg/103
liter
lb/103
gal
Particulates0
1.20d
10.0d
2.40
20.0
6.78
56.5
1.8
15
1.78
15
1.78
15
Sulfur oxides
jo
I>j
«
-------
Table 3.2.3-3. DIESEL VESSEL EMISSION FACTORS BY OPERATING MODE3
EMISSION FACTOR RATING: C
Emissions'3
Nitrogen oxides
Carbon monoxide
Hydrocarbons
(NOx as N02)
lb/103
kg/103
lb/103
kg/103
lb/103
kg/103
Horsepower
Mode
gal
liter
gal
liter
gal
liter
200
Idle
210.3
25.2
391.2
46.9
6.4
0.8
Slow
145.4
17.4
103.2
12.4
207.8
25.0
Cruise
126.3
15.1
170.2
20.4
422.9
50.7
Full
142.1
17.0
60.0
7.2
255.0
30.6
300
Slow
59.0
7.1
56.7
6.8
337.5
40.4
Cruise
47.3
5.7
51.1
6.1
389.3
46.7
Full
58.5
7.0
21.0
2.5
275.1
33.0
500
Idle
282.5
33.8
118.1
14.1
99.4
11.9
Cruise
99.7
11.9
44.5
5.3
338.6
40.6
Full
84.2
10.1
22.8
2.7
269.2
32.3
600
Idle
171.7
20.6
68.0
8.2
307.1
36.8
Slow
50.8
6.1
16.6
2.0
251.5
30.1
Cruise
77.6
9.3
24.1
2.9
349.2
41.8
700
Idle
293.2
35.1
95.8
11.5
246.0
29.5
Cruise
36.0
4.3
8.8
1.1
452.8
54.2
900
Idle
223.7
26.8
249.1
29.8
107.5
12.9
2/3
62.2
7.5
16.8
2.0
167.2
20.0
Cruise
80.9
9.7
17.1
2.1
360.0
43.1
1550
Idle
12.2
1.5
—
39.9
4.8
Cruise
3.3
0.4
0.64 '
0.1
36.2
4.3
Full
7.0
0.8
1.64
0.2
37.4
4.5
1580
Slow
122.4
14.7
—
-
371.3
44.5
Cruise
44.6
5.3
—
623.1
74.6
Full
237.7
28.5
16.8
2.0
472.0
5.7
2500
Slow
59.8
7.2
22.6
2.7
419.6
50.3
2/3
126.5
15.2
14.7
1.8
326.2
39.1
Cruise
78.3
9.4
16.8
2.0
391.7
46.9
Full
95.9
11.5
21.3
2.6
399.6
47.9
3600
Slow
148.5
17.8
60.0
7.2
367.0
44.0
2/3
28.1
3.4
25.4
3.0
358.6
43.0
Cruise
41.4
5.0
32.8
4.0
339.6
40.7
Full
62.4
7.5
29.5
3.5
307.0
36.8
^Reference 2.
Particulate arid sulfur oxides data are not available.
3.2.3-4
EMISSION FACTORS
D-12
1/75
-------
Table 3.2.3-4. AVERAGE EMISSION FACTORS FOR DIESEL-POWERED ELECTRICAL
GENERATORS IN VESSELSa
EMISSION FACTOR RATING: C
Emissions
Sulfur oxides
Carbon
Hydro-
Nitrogen oxides
Rated
Load,0
(SOx as S02)d
monoxide
carbons
(NOx as NO2)
output, t>
% rated
lb/103
kg/103
lb/103 .
kg/103
lb/103
kg/103
lb/103
kg/103
kW
output
gal
liter
gal
liter
gal
liter
gal
liter
20
0
27
3.2
150
18.0
263
31.5
434
52.0
25
27
3.2
79.7
9.55
204
24.4
444
53.2
50
27
3.2
53.4
6.40
144
17.3
477
57.2
75
27
3.2
28.5
3.42
84.7
10.2
495
59.3
40
0
27
3.2
153
18.3
584
70.0
214
25.6
25
27
3.2
89.0
10.7
370
44.3
219
26.2
50
27
3.2
67.6
8.10
285
34.2
226
27.1
75
27
3.2
64.1
7.68
231
27.7
233
27.9
200
0
27
3.2
134
16.1
135
16.2
142
17.0
25
27
3.2
97.9
11.7
33.5
4.01
141
16.9
50
27
3.2
62.3
7.47
17.8
2.13
140
16.8
75
27
3.2
26.7
3.20
17.5
2.10
137
16.4
500
0
27
3.2
58.4
7.00
209
25.0
153
18.3
25
27
3.2
53.4
6.40
109
13.0
222
26.6
50
27
3.2
48.1
5.76
81.9
9.8
293
35.1
75
27
3.2
43.7
5.24
59.1
7.08
364
43.6
aReference 2.
^Maximum rated output of the diesel-powered generator.
°Generator electrical output (for example, a 20 kW generator at 50 percent load equals 10 kW output).
^Calculated, not measured, based on 0.20 percent fuel sulfur content and density of 0.854 kg/liter (7.12 lb/gal) from Reference 17.
At zero load conditions, mass emission rates (ej) may be approximated in terms of kg/hr (lb/hr) using the
following relationship:
el = ^ratedef (2)
where: k = a constant that relates rated output and fuel consumption,
that is, 6.93 x 10'5 1000 liters fuel/kW
or
1.83 xlO'5 1000 gal fuel/kW
1 rated = the rated output, kW
ef = the fuel-specific emission factor from Table 3.2.3-4, kg/103 liter (lb/103 gal)
Pleasure craft. Many of the engine designs used in inboard pleasure craft are also used either in military vessels
(diesel) or in highway vehicles (gasoline). Out of a total of 700,000 inboard pleasure craft registered in the United
States in 1972, nearly 300,000 were inboard/outdrive. According to sales data, 60 to 70 percent of these
1/75 Internal Combustion Engine Sources 3.2.3-5
D-l 3
-------
inboard/outdrive craft used gasoline-powered automotive engines rated at more than 130 horsepower. The
remaining 400,000 pleasure craft used conventional inboard drives that were powered by a variety of powerplants,
both gasoline and diesel. Because emission data are not available for pleasure craft, Coast Guard and automotive
data2,19 are used to characterize emission factors for this class of vessels in Table 3.2.3-5.
Military vessels. Military vessels are powered by a wide variety of both diesel and steam power plants. Many of the
emission data used in this section are the result of emission testing programs conducted by the U.S. Navy and the
U.S. Coast Guard.1'3'5 A separate table containing data on military vessels is not provided here, but the included
tables should be sufficient to calculate approximate military vessel emissions.
TABLE 3.2.3.-5. AVERAGE EMISSION FACTORS FOR INBOARD PLEASURE CRAFT8
EMISSION FACTOR RATING: D
Based on fuel consumption
Diesel engine'3
Gasoline engine0
Based on operating time
kg/103
lb/103
kg/103
lb/103
Diesel engineb
Gasoline enginec
Pollutant
liter
gal
liter
gal
kg/hr
Ib/hr
kg/hr
Ib/hr
Sulfur oxides^
(SOx as SO2)
3.2
27
0.77
6.4
-
-
0.008
0.019
Carbon monoxide
17
140
149
1240
-
-
1.69
3.73
Hydrocarbons
22
180
10.3
86
-
-
0.117
0.258
Nitrogen oxides
(NOx as N02)
41
340
15.7
131
-
-
0.179
0.394
aAverage emission factors are based on the duty cycle developed for large outboards (> 48 kilowatts or > 65 horsepower) from Refer-
ence 7. The above factors take into account the impact of water scrubbing of underwater gasoline engine exhaust, alto from Reference
7. All values given are for single engine craft and must be modified for multiple engine vessels.
^Based on tests of diesel engines in Coast Guard vessels. Reference 2.
cBased on tests of automotive engines, Reference 19. Fuel consumption of 11.4 liter/hr (3 gal/hr) assumed. The resulting factors are
¦only rough estimates.
^Based on fuel sulfur content of 0.20 percent for diesel fuel and 0.043 percent tor gasoline from References 7 and 17. Calculated using
fuel density of 0.740 kg/liter (6.17 lb/gal) for gasoline and 0.854 kg/liter (7.12 lb/gal) for diesel fuel.
References for Section 3.2.3
1. Walter, R. A., A. J. Broderick, J. C. Sturm, and E. C. Klaubert. USCG Pollution Abatement Program: A
Preliminary Study of Vessel and Boat Exhaust Emissions. U.S. Department of Transportation, Transportation
Systems Center. Cambridge, Mass. Prepared for the United States Coast Guard, Washington, D.C. Report No.
DOT-TSC-USCG-72-3. November 1971. 119 p.
3.2.3-6 EMISSION FACTORS 1/75
D- 14
-------
2. Souza, A. F. A Study of Emissions from Coast Guard Cutters. Final Report. Scott Research Laboratories, Inc.
Plumsteadville, Pa. Prepared for the Department of Transportation, Transportation Systems Center,
Cambridge, Mass., under Contract No. DOT-TSC-429. February 1973.
3. Wallace, B. L. Evaluation of Developed Methodology for Shipboard Steam Generator Systems. Department of
the Navy. Naval Ship Research and Development Center. Materials Department. Annapolis, Md. Report No.
28463. March 1973. 18 p.
4. Waldron, A. L. Sampling of Emission Products from Ships' Boiler Stacks. Department of the Navy. Naval Ship
Research and Development Center. Annapolis, Md. Report No. 28-169. April 1972. 7 p.
5. Foernsler, R. 0. Naval Ship. Systems Air Contamination Control and Environmental Data Base Programs;
Progress Report. Department of the Navy. Naval Ship Research and Development Center. Annapolis, Md.
Report No. 28443. February 1973. 9 p.
6. The Boating Business 1972. The Boating Industry Magazine. Chicago, 111. 1973.
7. Hare, C. T. and K. J. Springer. Exhaust Emissions from Uncontrolled Vehicles and Related Equipment Using
Internal Combustion Engines. Final Report Part 2. Outboard Motors. Southwest Research Institute. San
Antonio, Tex. Prepared for the Environmental Protection Agency, Research Triangle Park, N.C., under
Contract No. EHS 70-108. January 1973. 57 p.
8. Hurst, J. W. 1974 Chrysler Gasoline Marine Engines. Chrysler Corporation. Detroit, Mich.
9. Mercruiser Sterndrives/ Inboards 73. Mercury Marine, Division of the Brunswick Corporation. Fond du Lac,
Wise. 1972.
10. Boating 1972. Marex. Chicago, Illinois, and the National Association of Engine and Boat Manufacturers.
Greenwich, Conn. 1972. 8 p.
11. Transportation Lines on the Great Lakes System 1970. Transportation Series 3. Corps of Engineers, United
States Army, Waterborne Commerce Statistics Center. New Orleans, La. 1970. 26 p.
12. Transportation Lines on the Mississippi and the Gulf Intracoastal Waterway 1970. Transportation Series 4.
Corps of Engineers, United States Army, Waterborne Commerce Statistics Center. New Orleans, La. 1970. 232
P-
13. Transportation Lines on the Atlantic, Gulf and Pacific Coasts 1970. Transportation Series 5. Corps of
Engineers. United States Army. Waterborne Commerce Statistics Center. New Orleans, La. 1970. 201 p.
14. Schueneman, J. J. Some Aspects of Marine Air Pollution Problems on the Great Lakes. J. Air Pol. Control
Assoc. 14:23-29, September 1964.
15. 1971 Inland Waterborne Commerce Statistics. The American Waterways Operations, Inc. Washington, D.C.
October 1972. 38 p.
16. Horsepower on the Inland Waterways. List No. 23. The Waterways Journal. St. Louis, Mo. 1972. 2 p.
17. Hare, C. T. and K. J. Springer. Exhaust Emissions from Uncontrolled Vehicles and Related Equipment Using
Internal Combustion Engines. Part 1. Locomotive Diesel Engines and Marine Counterparts. Southwest
Research Institute. San Antonio, Tex. Prepared for the Environmental Protection Agency, Research Triangle
Park, N.C., under Contract No. EHS 70-108. October 1972. 39 p.
18. Pearson, J. R. Ships as Sources of Emissions. Puget Sound Air Pollution Control Agency. Seattle, Wash.
(Presented at the Annual Meeting of the Pacific Northwest International Section of the Air Pollution Control
Association. Portland, Ore. November 1969.)
19. Study of Emissions from Light-Duty Vehicles in Six Cities. Automotive Environmental Systems, Inc. San
Bernardino, Calif. Prepared for the Environmental Protection Agency, Research Triangle Park, N.C., under
Contract No. 68-04-0042. June 1971.
1/75 Internal Combustion Engine Sources 3.2.3-7
D-15
-------
APPENDIX E
LIST OF PERSONS CONTACTED
-------
PERSONS CONTACTED
Locomotives
Mr. Richard Cataldi
Association of American Railroads
Washington, DC
(202)835-9182
Mr. Arnie Bang
Federal Railroad Administration
Washington, D.C.
(202)426-4000
Mr. Jack Hoffman
General Electric Co.
Erie, PA
(814) 875-3172
Mr. Hugh Williams
Electromotive Division
McCook, IL
(312)387-6736
Mr. Peter Hutchins
Emission Control Technology Division
U.S. EPA
Ann Arbor, MI
(313) 668-4340
Construction Equipment
Mr. Charles Hudson
International Harvester
Melrose Park,IL
(312) 865-3717
Mr. Rod Summerfield
California Air Resources Board
El Monte, CA
(213) 575-6844
Marine Vessels
Mr. William Lamason
Office of Air Quality Planning and Standards
U.S. EPA
Research Triangle Park, NC
(919) 541-5585
E-2
-------
Mr. Anders
Corp. of Engineers
Waterborne Commerce Statistics Center
New Orleans, LA
(504)885-6807
Ms. Marcie Kinter
American Waterways Operators
Arlington, VA
(703)841-9300
Mr. Ron Olander
Dravo Marine Equipment Co.
Pittsburg, PA
(412)777-5448
Ms. Mary Jane Miller
American Bureau of Shipping
New York, NY
(212)440-0300
Mr. Jim Manika
Marketing Research, Electromotive Division
McCook, IL
(312)387-5500
Mr. Stuart Dattner
Texas Air Control Board
Austin, TX
(512) 451-5711
E-3
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
460/3-85-004
3. RECIPIENT'S ACCESSION»NO.
4. TITLE AND SUBTITLE
RECOMMENDED REVISIONS TO GASEOUS EMISSION FACTORS
FROM SEVERAL CLASSES OF OFF-HIGHWAY MOBILE SOURCES
5. REPORT DATE
March 1985
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Melvin N. Ingalls
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORG \NIZATION NAME AND ADDRESS
Southwest Research Institute
6220 Culebra Road
San Antonio, Texas 78284
10. PROGRAM ELEMENT NO.
Work Assignment No. 8
11. CONTRACT/GRANT NO.
68-03-3162
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Protection Agency
2565 Plymouth Road
Ann Arbor, Michigan 48105
13. TYPE OF REPORT AND PERIOD COVERED
Final Report (8/83 to 9/84)
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This study examined three categories of off-highway mobile emission sources to
determine current emission factors of hydrocarbons (HC), carbon monoxide (CO),
and oxides of nitrogen (NOx). The three categories examined were locomotives,
marine vessels, and farm and construction equipment. National and regional
impact of these emission sources were also examined. For locomotives, it was
found that rising fuel prices had led to engine improvements as well as changes
in locomotive operation. Additional measured emission data were also found in
the literature. Using new duty cycles and the additional emissions test data,
new locomotive emission factors were developed. In U.S. waters, marine emissions
are changing as diesel engines are used in a larger portion of the U.S. Merchant
Fleet. None of the literature surveyed provided sufficient valid information to
determine new marine diesel engine emission factors. New farm and construction
equipment emission factors were recommended based on a recent study found in
the literature.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air Pollution
Exhaust Emissions
Diesel Locomotives
Marine Engines
Agricultural Equipment
Construction Equipment
Locomotive Emissions
Marine Emissions
Off-Highway Emissions
Farm Equipment Emissions
Construction Equipment
Emissions
13. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
106
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |