Citation
Life cycle assessment of transit systems in the U.S. and India

Material Information

Title:
Life cycle assessment of transit systems in the U.S. and India implications for a carbon-constrained future
Creator:
Whitaker, Michael ( Michael Bryce )
Publication Date:
Language:
English
Physical Description:
xx, 198 leaves : illustrations ; 28 cm

Subjects

Subjects / Keywords:
Product life cycle -- India -- Chennai ( lcsh )
Product life cycle -- Colorado -- Denver ( lcsh )
Local transit -- India -- Chennai ( lcsh )
Local transit -- Colorado -- Denver ( lcsh )
Automobiles -- Motors -- Exhaust gas -- India -- Chennai ( lcsh )
Automobiles -- Motors -- Exhaust gas -- Colorado -- Denver ( lcsh )
Greenhouse gases -- Colorado -- Denver ( lcsh )
Greenhouse gases -- India -- Chennai ( lcsh )
Automobiles -- Motors -- Exhaust gas ( fast )
Greenhouse gases ( fast )
Local transit ( fast )
Product life cycle ( fast )
Colorado -- Denver ( fast )
India -- Chennai ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 188-198).
General Note:
Department of Civil Engineering
Statement of Responsibility:
by Michael Bryce Whitaker.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
166326840 ( OCLC )
ocn166326840
Classification:
LD1193.E53 2007d W44 ( lcc )

Full Text
LIFE CYCLE ASSESSMENT OF TRANSIT SYSTEMS IN THE U.S. AND
INDIA: IMPLICATIONS FOR A CARBON-CONSTRAINED FUTURE
By
Michael Bryce Whitaker
B.S., Stanford University, 2001
M.S., Stanford University, 2001
A thesis submitted to the
University of Colorado at Denver
and Health Sciences Center
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Civil Engineering
2007


This thesis for the Doctor of Philosophy
degree by
Michael Bryce Whitaker
has been approved
by
Anuradha Ramaswami
Paul Komor

Date


Whitaker, Michael B (Ph.D., Civil Engineering)
Life Cycle Assessment of Transit Systems in the U.S. and India: Implications for a
Carbon-Constrained Future
Thesis directed by Associate Professor Anuradha Ramaswami
ABSTRACT
Parallel life cycle assessments (LCAs) of energy use and greenhouse gas (GHG)
emissions of baseline mass transit systems in the U.S. and India are conducted. The
LCAs focus on bus and electrified urban rail transit systems in Denver, USA and
Chennai, India. The life cycle GHG analysis is then applied to evaluate GHG
mitigation in the transportation sector in Denver. A major difficulty in using life
cycle data for transport policy decision-making is that baseline LCAs of buses and
electrified urban rail have not been conducted in either India or the U.S. The
contribution of this thesis is to begin filling this important data gap on the life cycle
impacts of transit systems in each country, providing for the first time a side-by-side
comparison of life cycle energy use and GHG emissions of mass transit systems in
the developed and developing world. The analysis is relevant for GHG mitigation
for climate actions as well as urban infrastructure planning for a carbon-constrained
future.


The analysis found that established life cycle inventory databases from industrialized
nations cannot be used for developing country vehicle manufacturing and operation
analyses without explicitly accounting for significant process differences. For buses,
Indian buses have an energy and GHG advantage over U.S. buses [11 vs 24 MJ/v-
km; 0.9 vs 1.8 kgCC^e/v-km] due to lighter vehicles, fewer auxiliaries, better fuel
economy, and higher ridership. For electrified urban rail, the results are reversed
with U.S. systems being lighter and having greater energy efficiency [12 vs 26 MJ/v-
km; 3.0 vs 6.4 kgCChe/v-km]. Incorporating ridership amplifies these results on a
per passenger basis yielding kgCC^e/p-km factors of 0.02, 0.2, 0.4 and 0.2 for India
bus, U.S. bus, India rail, and U.S. rail respectively. GHG emission factors of electric
grids are shown to be up to three times greater than for diesel fuel, comparatively
disadvantaging electrified urban rail systems. Finally, analysis of GHG reduction
options for the transportation sector in Denver shows that mass transit system supply
must be coupled with market-based and City policies to reduce travel demand, along
with state-level initiatives to improve fuel-propulsion systems, to achieve even
moderate GHG reductions.
This abstract accurately represents the content of the candidate^ thesis. I
recommend its publication
Signed
Anuradha Ramaswami


DEDICATION
I dedicate this thesis to my family and friends who have supported me throughout
my life. In particular, I could not have completed this work without the loving
support of my parents, John and Kathie, and sister Shawna.


ACKNOWLEDGEMENT
I would like to first thank my advisor, Anu Ramaswami, for encouraging me
throughout the development of this thesis. 1 have gained many new insights through
her approach to the analysis of sustainability and through the exposure to the Indian
culture that I obtained during the time that she arranged for me to live in Chennai. I
would also like to thank the members of my committee for serving on this project.
Moreover, the research could not have been completed without funding from the
Department of Educations Graduate Assistance in Areas of National Need
(GAANN) program and through funding and policy development experience with
the City and County of Denver. Finally, I will always value the friendships that Ive
developed with my fellow USIEP colleagues and will remember fondly the
conversations of B:30.


TABLE OF CONTENTS
Figures...................................................................xv
Tables.................................................................xviii
Chapter
1. Introduction........................................................1
1.1 Mass Transit Trends in India........................................5
1.2 Mass Transit Trends in the U.S......................................8
1.3 Life Cycle Assessment..............................................10
1.3.1 Background.........................................................10
1.3.2 Wells-to-Wheels Framework..........................................11
1.3.3 Life Cycle Assessment Methodology..................................12
1.4 Thesis Objectives and Chapters.....................................14
2. Life Cycle Assessment of Energy Use and Greenhouse Gas Emissions
for Buses in Chennai, India and Denver, USA........................16
2.1 Goal and Objectives................................................16
2.2 Environmental Priorities in Chennai, India.........................17
2.2.1 Survey Methods.....................................................19
2.2.2 Survey Questions...................................................22
2.2.3 Survey Results.....................................................23
2.3 Life Cycle Assessment Framework....................................27
2.4 Raw Data and Methods...............................................30


2.4.1 Fuel Production..........................................................32
2.4.1.1 Data Sources............................................................32
2.4.1.2 Raw Energy Data........................................................33
2.4.2 Bus Materials Extraction and Processing................................35
2.4.2.1 Data Sources............................................................35
2.4.2.2 Raw Data...............................................................37
2.4.3 Bus Assembly and Major Components........................................40
2.4.3.1 Data Sources............................................................40
2.4.3.2 Raw Data...............................................................41
2.4.4 Bus Operations...........................................................43
2.4.4.1 Data Sources............................................................43
2.4.4.2 Raw Data...............................................................43
2.4.5 Bus Ridership............................................................44
2.4.5.1 Data Sources............................................................45
2.4.5.2 Raw Data...............................................................46
2.4.6 Greenhouse Gas Emission Factors..........................................47
2.5 Results and Analysis.....................................................48
2.5.1 Bus Materials Production.................................................48
2.5.2 Wells-to-Wheels Fuel Analysis............................................51
2.5.3 Bus Life Cycle Assessment................................................52
viii


2.5.3.1 Life Cycle Stage Comparison..........................................52
2.5.3.2 Bus Assembly Impacts.................................................56
2.5.3.3 Benchmarking Material Impact Calculation.............................57
2.5.3.4 Bus Operation Impacts................................................59
2.6 Discussion............................................................61
2.6.1 Data Availability and Uncertainty.....................................61
2.6.2 Sensitivity Analysis..................................................62
2.6.2.1 General Fuel Economy Impacts.........................................62
2.6.2.2 Impact of Vehicle Weight and Air Conditioning on
Bus Fuel Economy......................................................63
2.6.2.3 Environmental Factors................................................65
2.6.2.4 Drive Cycle Impacts on Fuel Economy..................................66
2.6.3 Alternative Fuels Impact on GHG Emissions.............................67
2.6.4 Applicability of GaBi 4 to Developing Country LCAs....................69
2.7 Conclusions...........................................................70
3. A Comparison of Greenhouse Gas Emissions and
Energy Use for Electrified Urban Rail and Buses in
India and the U.S.....................................................72
3.1 Objectives............................................................72
3.2 Raw Data & Sources.................................................. 74
3.2.1 Chennai Mass Rapid Transit System.....................................75
3.2.1.1 System Description...................................................75


3.2.1.2 Data Sources
76
3.2.2 RTD Denver Light Rail System.......................................77
3.2.2.1 System Description................................................77
3.2.2.2 Data Sources......................................................78
3.3 Methodology........................................................79
3.3.1 Track Construction.................................................81
3.3.2 Vehicle Manufacturing..............................................81
3.3.3 Vehicle Operations & Ridership.....................................77
3.4 Results & Analysis.................................................83
3.4.1 Vehicle Manufacturing..............................................83
3.4.2 Vehicle Operations and Ridership...................................84
3.4.3 Life Cycle Energy Use and Greenhouse Gas Emissions.................87
3.5 Discussion.........................................................90
3.5.1 Impact of Fuel Emission Factors....................................91
3.5.2 Ridership..........................................................93
3.5.3 Mass Transit System Energy and Greenhouse Gas Comparison...........94
3.5.4 Mass Transit Operation Costs and Greenhouse Gas Emissions..........96
3.5.4.1 Overview..........................................................96
3.5.4.2 Greenhouse Gas versus Operation Cost Transit System Analysis......97
3.5.4.3 Significance of the Results......................................101
3.5.4.4 Comparison of Chennai and Denver Systems to U.S. Averages........101
x


3.5.4.5 Local Air Pollution...................................................103
3.5.4.6 Life Cycle Costs of Electrified Urban Rail and Bus Systems............104
3.5.4.7 Data Gap Analysis.....................................................104
3.6 Conclusions and Future Work............................................105
4. The Role of Mass Transit in Climate Change Mitigation
for the Transport Sector in Denver, CO...............................108
4.1 Introduction...........................................................108
4.2 Objectives.............................................................110
4.3 Denver Greenhouse Gas Inventory........................................Ill
4.3.1 Inventory Methodology..................................................Ill
4.3.2 Transportation Sector Greenhouse Gas Profile...........................113
4.3.3 Denver Transportation Sector Greenhouse Gas Mitigation Goal..........115
4.4 Methodology for Transportation Sector Action Option Selection.........116
4.4.1 Literature Search......................................................116
4.4.2 Policy Mechanism Categories........................................... 120
4.4.3 Criteria for Evaluation................................................120
4.4.3.1 Greenhouse Gas Reduction Potential....................................121
4.4.3.2 Greenhouse Gas Cost-effectiveness.....................................122
4.4.3.3 Political Feasibility.................................................123
4.4.3.4 Ancillary Benefits....................................................123
4.5 Transportation Action Options..........................................124
xi


4.5.1 Strategies Designed to Increase Supply/Demand of Mass Transit/Multi-
Modal Transport Options..............................................124
4.5.1.1 Denvers FasTracks Expansion Project................................125
4.5.1.2 Best Workplace for Commuter Programs................................127
4.5.1.3 Individualized Travel Marketing Program.............................129
4.5.1.4 Density Focus for New Population....................................130
4.5.1.5 Mass Transit Supply-Demand Summary................................. 132
4.5.2 City-scale Market Mechanisms to Dis-incentivize Private
Vehicle Travel......................................................134
4.5.2.1 Pay-As-You-Drive Auto Insurance and Vehicle Registration Fees......134
4.5.2.2 Showcase Car-Share in Appropriately Dense Neighborhoods............ 136
4.5.2.3 Market Mechanism Summary...........................................137
4.5.3 Programs for Reducing Vehicle Fuel/Propulsion System Impacts.....138
4.5.3.1 Adopt Californias Clean Car Program...............................139
4.5.3.2 Increased Renewable Fuels Standard for Colorado.................... 140
4.5.3.3 Low Rolling Resistance Tires.......................................142
4.5.3.4 Western Region Fee-bate System..................................... 143
4.5.3.5 Pass a Renewable Fuels Ordinance for All Large Fleets in Denver....144
4.5.3.6 Vehicle Fuel/Propulsion Systems Summary............................145
4.5.4 Certified Carbon Offset Purchase Programs...........................145
4.5.4.1 Certified Carbon Offsets Programs for Airline and Auto Travel......147
xii


4.5.4.2 Carbon-Neutral City Government Fleet through Car Share...............148
4.5.4.3 Carbon Offset Summary................................................149
4.5.4.4 Cumulative Summary.................................................. 150
4.5.5 More Complex Transportation Action Options............................152
4.5.5.1 Raise DMV Registration Fees to Purchase Certified Carbon Offsets... 152
4.5.5.2 Enplaned Passenger Fee at Denver International Airport...............153
4.5.5.3 Annual Vehicle Attribute-based Taxes and Fees........................154
4.5.5.4 Congestion Pricing...................................................154
4.5.5.5 High Occupancy Only Highways and Slug Lines..........................155
4.5.5.6 Full Externality-Based Road Pricing..................................156
4.5.5.7 More Complex Transport Options Summary...............................156
4.6 Ancillary Benefits....................................................158
4.6.1 Overview..............................................................158
4.6.2 Local-scale Ancillary Benefits........................................161
4.7 Conclusions & Recommendations.........................................163
5. Conclusions, Contributions, and Future Work...........................166
5.1 Conclusions.......................................................... 166
5.1.1 General Life Cycle Assessment Conclusions.............................166
5.1.2 Transit Buses........................................................ 167
5.1.3 Electrified urban rail................................................167
5.1.4 Transit Bus vs. Electrified Urban Rail................................168
xiii


5.1.5 Denver Transport Sector Greenhouse Gas Mitigation Policies.........169
5.2 Major Contributions of the Thesis..................................170
5.3 Future Work........................................................173
Appendix
A. Survey Approval Letter.............................................174
B. English Version of the Survey......................................175
C. Sample of Translated Tamil Survey..................................183
D. Hazardous Waste Inventory Reporting Form for India.................184
E. Land Use Densification Option......................................185
References................................................................188
xiv


LIST OF FIGURES
Figure
1.1 Framework for Conducting Environmental Life Cycle Assessment......... 13
2.1 Author Michael Whitaker Presenting on Life Cycle Assessment at the
Confederation of Indian Industries Workshop in Chennai, India.........21
2.2 Relative Impact Category Rankings.....................................23
2.3 Specific Impact Indicator Rankings....................................25
2.4 Viking 222 Bus........................................................28
2.5 Orion ZF Bus..........................................................28
2.6 Four Major Stages of a Bus Life Cycle Assessment......................29
2.7 Diesel Fuel Refining..................................................32
2.8 Steel Production......................................................35
2.9 Aluminum Production...................................................36
2.10 High Bus Ridership in Chennai.........................................45
2.11 Greenhouse Gas Emissions Per Vehicle-Km by Life Cycle Stage
for India (left) and the U.S. (right).................................55
2.12 Energy Use Per Vehicle-Km by Life Cycle Stage for India (left)
and the U.S. (right)..................................................55
2.13 Bus Chassis...........................................................56
2.14 Life Cycle Transit Bus Greenhouse Gas Emission Intensities
Per V-Km and Per P-Km.................................................60
2.15 Bus Air Conditioning Unit.............................................65
2.16 Biodiesel Bus.........................................................67
xv


3.1 Chennai MRTS EMU....................................................75
3.2 Chennai Suburban and MRTS Rail System............................76
3.3 Denver Light Rail System............................................77
3.4 Greenhouse Gas Emissions Per Vehicle-Km by Life Cycle
Stage for India (left) and U.S. (right)............................88
3.5 Energy Use Per Vehicle-Km by Life Cycle Stage for
India (left) and U.S. (right)......................................89
3.6 Life Cycle Electrified urban rail Greenhouse Gas Emission
Intensities Per V-km and Per P-km..................................83
3.7 Life Cycle Energy Use Comparison of Mass Transit Systems
in Chennai, India and Denver, U.S..................................94
3.8 Life Cycle Greenhouse Gas Comparison of Mass Transit
Systems in Chennai, India and Denver, U.S..........................95
3.9 Greenhouse Gas Emissions and Operating Costs of
Transit Systems on a Per Vehicle-Kilometer Basis...................100
4.1 GHG Emissions Summary by Energy Source and
Activity for Denver in 2005 (mtC02e)...............................113
4.2 Transportation Sector W2W Greenhouse Gas Emissions by
Mode of Transport..................................................114
4.3 Denver Light Rail Vehicle..........................................125
4.4 Commuting in Denver................................................127
4.5 South Perth, Australia.............................................129
4.6 Density vs. Private Vehicle Commuting..............................130
4.7 RTD Denver Bus.....................................................133
I
XVI


4.8 Pay-As-You-Drive Auto Insurance................................... 135
4.9 Car Share Vehicle..................................................136
4.10 Hybrid Vehicles Are a Component of the California
Clean Car Program..................................................139
4.11 Alternative Fuel Pumps.............................................141
4.12 Low Rolling Resistance Tires...................................... 142
4.13 Renewable Fuel Requirements for Fleet Vehicles.....................144
4.14 Airport Kiosk......................................................147
4.15 Denver Snow Plow...................................................149
4.16 Greenhouse Gas Reduction Potential of
Transportation Sector Options..................................... 151
4.17 London Congestion Pricing......................................... 154
xvii


LIST OF TABLES
Table
1.1 Comparison of Denver, Colorado, USA and Chennai,
Tamil Nadu, India.....................................................4
2.1 India survey respondents............................................21
2.2 Detailed breakdown of interview responses...........................21
2.3 Bus life cycle assessment availability & sources in Chennia, India
& Denver, USA........................................................31
2.4 Energy used in stage divided by energy content of diesel fuel.......33
2.5 India crude oil to diesel pathway details...........................34
2.6 Wells-to-pump energy efficiency and greenhouse gas comparison.......34
2.7 Data for steel production calculations..............................38
2.8 Data for aluminum production calculations...........................39
2.9 Energy intensive primary bus materials in India and the U.S.........41
2.10 Fuel consumed in bus chassis assembly and for assembling/
producing first level components in India............................41
2.11 Major bus components comparison Orion ZF (USA) and
Viking 222 (India)...................................................42
2.12 Fuel economy for transit buses in Chennai and Denver................44
2.13 Average bus ridership for Chennai, India and Denver, USA............46
2.14 Greenhouse gas emission factors.....................................47
2.15 Material sector greenhouse gas emission intensity...................48
xviii


2.16 Energy and greenhouse gas emissions for diesel production (W2P),
vehicle operation (P2W), and overall (W2W)..............................51
2.17 Bus life cycle greenhouse gas emissions and energy use...................54
2.18 Comparison of upstream GHG emission methodologies........................58
2.19 Potential impact of drive cycle changes on W2W GHG emissions.............67
2.20 Wells-to-wheels GHG reductions for fuel switching (mtCChe)...............68
3.1 Electrified urban rail vehicle dimensions for
Chennai, India and Denver, USA..........................................74
3.2 Light rail energy use comparison for Denver..............................79
3.3 Raw data summary with sources for analysis of
electrified urban rail in India & U.S...................................80
3.4 Relative greenhouse gas emission contributions of
vehicle manufacturing to the life cycle of electrified urban rail
from Gabi 4 and steel weight calculations...............................83
3.5 Ridership comparison for electrified urban rail and
transit bus systems.....................................................85
3.6 Greenhouse gas comparison for mass transit system operations.............85
3.7 Life cycle energy use and greenhouse gas emissions
of electrified urban rail in India and the U.S..........................87
3.8 GHG emission factors of mass transit fuels...............................92
3.9 Mass transit system ridership numbers....................................93
3.10 Operating greenhouse gas emissions and cost for
mass transit systems in India and the U.S...............................97
3.11 Transit systems included in the analysis of average
GHG emissions vs. vehicle operating costs for the U.S...................99
xix


4.1 Key data related to Denvers GHG reduction goal..........................115
4.2 Greenhouse gas emissions per vehicle mile traveled (VMT)............119
4.3 City options for promoting land-use densification........................131
4.4 Strategies designed to directly promote mass transit ridership...........133
4.5 City-scale market mechanisms designed to
decrease private vehicle travel........................................138
4.6 Programs for reducing vehicle fuel/propulsion system impacts.............146
4.7 Certified carbon offset purchase programs................................150
4.8 More complex transport action options summary............................157
4.9 Estimates of local-scale ancillary benefits of
greenhouse gas reductions policies.....................................161
xx


1. Introduction
The consumption of vast quantities of petroleum-derived fuels in the transportation
sector poses a threat not only to the health of human beings and global ecosystems,
but also to the stability and sustainability of modem industrial economies that have
become increasingly dependent on imported crude oil to meet their energy needs.
World transport systems are presently 96% dependent on oil (McAuley 2003). By
2025, world oil demand is expected to increase to 119 million barrels per day with
75% of this increase coming from the transportation sector (U.S. Energy Information
Administration 2003). Transportation is responsible for approximately 67% of
Americas current petroleum use, 55% of which is imported. The import percentage
is expected to increase to 68% by 2025 without major changes in petroleum
consumption (U.S. Energy Information Administration 2004).
In the U.S., 28% of total greenhouse gas emissions (GHG) are emitted from the
transportation sector (U.S. Environmental Protection Agency 2006). In India,
approximately 10% of GHG emissions come from the transportation sector (Shukla,
Nair et al. 2003). This lower number reflects the lesser reliance of the Indian
population on personal vehicle travel. However, countries like India and China (that
contain more than one-third of the worlds population), continue to move on a fast-
1


track toward rapid industrialization and urbanization. According to an article in the
Economic Times, India and China are recording annual growth rates in private
vehicle ownership ranging from 15-75% (2005). The increases in private vehicle
ownerships correspond to rises in oil demand in these nations and will lead to greater
transportation sector contributions to national GHG emissions.
Even without including the health and environmental impacts of rising petroleum
use, it is clear that major changes to the transportation sectors world-wide will be
required to deal with scarce oil resources. One change that is being proposed
throughout the world is an increase in the use of mass transit, particularly buses
along with electrified urban rail in some localities. In order to make better decisions
to reduce transportation-related GHG impacts in a carbon-constrained future, life
cycle assessments (LCAs) should be applied that look not only at the environmental
impacts from the emissions of a vehicle but also examine the production of the fuel,
fuel distribution, vehicle manufacturing, and end-of-life impacts. By expanding the
boundaries, a picture of cradle-to-grave impacts can be developed to better guide
decision-makers on pollutants such as GHGs that have a global reach.
The focus of this thesis is on the life cycle energy use and greenhouse gas emission
impacts of currently available mass transit systems in the U.S. and India and on
2


applying such analysis to evaluate city-scale policies that can be implemented to
reduce GHG emissions in the transportation sector.
Life cycle GHG assessments are conducted to evaluate bus and electrified urban rail
mass transit systems in Denver, USA and Chennai, India. These two cities provide
for an interesting comparison. Denver is a mid-sized metropolitan area in a
developed nation that has been aggressive in attempting to address the environmental
impacts of its bus fleet. Chennai is a larger metropolitan area in a developing
country that will be faced with significant choices regarding how to reduce pollution
from its bus fleet. The cities have similar average commute times with Chennai at
35 minutes (PayScale 2007) and Denver at 28 minutes (Denver Regional Council of
Governments 2001). Table 1.1 gives key statistics for the two cities and for the
transit fleets operating in each.
3


Table 1.1 Comparison of Denver, Colorado, USA and Chennai, Tamil Nadu, India
Denver Chennai
Population (million) 2 8
Land Area (km2) 400 174
Elevation (m) 1,650 6
Average Commute Time (min) 28 35
Bus System Operator Denver's Regional Transportation District Chennai's Metropolitan Transit Corporation
Bus Fleet Size 1,071 2,773
Scheduled Daily Bus Service (km) 260,000 685,000
Bus Ridership (passenger-km/vehicle-km) 9.4 38.8
Number of Bus Routes 174 484
Average Age of Fleet (yrs) 6.0 7.6
Average Bus Fuel Economy (km/l) 1.88 4.23
Rail System Operator Denvers Regional Transportation District Southern Railway
Rail Ridership (passenger-km/vehicle-km) 12.7 15.0
Approx. Route Length (km) 35 9
Average Rail Fuel Economy (kWh/v-km) 3 6
(note: Denver encompasses the Denver-Aurora
Metropolitan Area)
Sources: http://www.rtd-denver.com,
http://www.mtcbus.org
The life cycle GHG analysis is then applied to evaluate GHG mitigation in the
transport sector in Denver. A major difficulty in using life cycle data for transport
policy decision-making is that baseline LCAs of buses and electrified urban rail
systems have not been conducted for either India or the U.S. The contribution of this
thesis is to begin filling this important data gap on the life cycle impacts of mass
transit systems in India and the U.S. An overview of mass transit systems in each
country is provided next as an introduction.
4


1.1 Mass Transit Trends in India
In India, despite increasing bus fleets, urban commuters have continued to show a
preference for private vehicles with efficiency, commute time, and reliability being
commonly cited reasons. An Indian car consumes nearly five times more energy per
passenger-kilometer than a 52-seater bus while two-wheelers consume 2.6 times
more energy on a passenger-km basis (Dhavse 2003). In addition to increasing
transportation energy demands, private vehicle use also takes up significantly more
space on the roads per passenger kilometer. Two-wheelers occupy 54 times more
road space and cars 38 times more road space than the 52-seater bus per passenger-
kilometer. The stress on transportation infrastructure caused by the increasing use of
private vehicles can be seen in the snarled traffic patterns of major Indian cities.
However, current buses in India are overcrowded and highly polluting. Changes
being made to buses to address these challenges include increased use of private
buses, switching to compressed natural gas (CNG) or biodiesel, and in some cities,
switching to electrified urban rail.
Due to low income levels and highly populous cities, public transit plays a vital role
in the functioning of Indian cities. However, despite the necessity, supplies of public
transit and the services provided are woefully insufficient. Only 17 out of the 35
cities in India with more than one million people have dedicated city bus transit
5


services and only four cities (Mumbai, Delhi, Kolkata, and Chennai) have public rail
transit systems (Singh 2005). According to Indias Ministry of Road Transport &
Highways, there was a 50% increase in the total number of registered motor vehicles
in Chennai during the period 1995-2000 with total vehicles rising from 768,000 in
1995 to 1,150,000 in 2000. Over 90% of these registered vehicles are personal
vehicles (cars, jeeps, and two-wheelers) (Singh 2005).
The Ministry of Urban Development (MUD) tracks the typical modal split for Indian
cities as a percentage of the total number of trips. For a city the size of Chennai (>5
million people), the split is walking (28.4%), mass transport (62.8%), taxis / three-
wheeler rickshaws (7.0%), car (6.1%), two-wheeler (14.8%), and bicycle (9.4%).
According to MUD, the desirable modal split for a city that size is mass transit (70-
85%), bicycle (15-20%), and other modes (10-15%) (Singh 2005). The desired
modal split for mass transit increases with city size suggesting that as the rapid
growth of Indian cities continues, the demand for mass transit will also rise. The
average large Indian city is already falling several percentage points below the
desired modal share for mass transit, and this gap is likely to widen in the future.
The lack of adequate mass transit modal share could be a result of inherent design
limitations in the systems. Most buses use truck engines and chasses that are not
optimized for use in urban driving conditions. Moreover, mass transit systems are
6


often overcrowded; buses designed to carry 50 passengers frequently service twice
that amount during peak hours, forcing some passengers to ride on the outside of the
bus (Singh 2005) and face roads that are severely congested without rights-of-way
for buses. The congested roads encourage the use of smaller vehicles (two-wheelers
or rickshaws) that can better move through the streets. The congestion on Indian
streets comes down principally to too many vehicles on an inadequate road system.
In the U.S., cities with populations greater than 100,000 people have an average of
28.2% of their total developed area dedicated to roads and streets (note that negative
impacts such as decreased water infiltration and heat island effects are associated
with increased road coverage). In India, this number drops to 16.1% (Singh 2005)
and is in practice lower due to encroachment onto roadways by vendors, pedestrians,
animal-drawn carts, and bicyclists. In many large cities, average speeds for urban
buses are as slow as 6-10 km/hr (Pucher, Korattyswaroopam et al. 2004).
Urban and suburban rail systems are increasingly being used in attempts to expand
mass transit ridership in Indian cities. According to Indian Railways, during the
1990s the number of passenger-kilometers served by suburban rail systems increased
nearly 50% (Pucher, Korattyswaroopam et al. 2004). Similar national statistics for
bus transport do not exist due to the decentralized control of bus systems throughout
India. However, during the same 10-year period, the bus fleet in Chennai did grow
by 54% (Pucher, Korattyswaroopam et al. 2004). A choice between increased
7


investment in bus rapid transit versus electrified urban rail for mass transit is
emerging in many Indian cities as they plan for increased urban populations. LCA-
based analysis of the two systems can provide useful information on the energy and
greenhouse gas savings that can be expected from these competing mass transit
strategies in India.
1.2 Mass Transit Trends in the U.S.
Mass transit use in the U.S. peaked during the 1940s with systems averaging 40
passenger-kilometers per vehicle-kilometers (p-km/v-km) in 1945. By the time
governments began owning and operating the public transit systems, the number of
passenger-kilometers per vehicle-kilometer had fallen to 21.8. In 1990, p-km/v-km
hit a low of 14.4 and has slowly climbed over the last 15 years to 14.8 in 2004.
Moreover, public transport boardings per capita (including motorbus, metro, light
rail, trolley bus & regional rail) have climbed from 27.8 in 1995 to 30.6 in 2002
(Wendell Cox Consultancy 2006). The American Public Transit Association also
claims that due in part to spikes in gasoline prices, more than 25 transit systems in
the U.S. showed double digit increases in ridership between November 2004 and
November 2005 including Tulsa, OK (22%), Salt Lake City, UT (17.7%), and
Houston and Dallas, TX (14.9%) (Associated Press 2006). However, despite this
rise in ridership, the public transport market share has declined slightly from a value
8


of 1.71% in 1995 to 1.57% in 2004. These numbers suggest that although the
existing public transport is being utilized more by the surrounding population, other
modes of transport (like personal vehicle use) are outpacing public transport
expansion.
As gas prices rise and the world looks for ways to wean itself off of oil dependency,
attention has turned towards expanding public transit services as one option.
However, some have questioned whether public transit actually saves energy when
compared to personal vehicle travel. Lawyer contends that although mass transit was
about twice as energy efficient as personal vehicle travel in the 1970s, current
systems cannot claim the same advantage. The loss of energy efficiency advantage
is due to a combination of personal vehicles becoming 50% more energy efficient
than they were in 70s while mass transit systems are 1/3 less efficient (2003)
(combining for an overall change equaling a factor of two). Moreover, data collected
by Cox (2002) contends that the average automobile uses 3,635 Btu / passenger mile
while buses use 4802 Btu / passenger mile and electrified urban rail uses 3,168 Btu /
passenger mile. This data suggests society only gains an energy advantage if
automobile trips are replaced by electrified urban rail and not by buses. Chapters 2
and 3 will examine this assertion by comparing the life cycle energy use of transit
buses and electrified urban rail in India and the U.S. on a per passenger-km basis.
As will be shown during this LCA, energy use calculations for transit systems are
9


highly dependent on assumptions and on system boundaries. Full LCAs on both
buses and rail systems are required so that governments seeking to expand their
public transit systems can be well-informed as to the true life cycle environmental
impacts before decisions are made. This study is a first, preliminary attempt at a
comparative LCA of transit bus and rail systems in both the developed and
developing worlds.
1.3 Life Cycle Assessment
1.3.1 Background
Environmental life cycle assessment (LCA) is an analysis procedure that attempts to
identify and quantify the environmental impacts of a product or process throughout
its cradle-to-grave life cycle (U.S. Environmental Protection Agency 2007). The
cradle represents the beginning of a products life when the raw materials needed for
its construction are extracted from the ground. The grave represents the end of a
products useful life when it is disposed of or recycled. LCA attempts to capture the
life cycle environmental impacts that occur between these two endpoints of a
product. For example, when looking at the impacts of a new vehicle introduced into
the marketplace, an LCA examines not only fuel use and tail pipe emissions but also
examines the impacts of manufacturing the vehicle, extracting the raw materials,
10


producing the fuel, distributing the fuel, and disposing of the vehicle at the end of its
life.
1.3.2 Wells-to-Wheels Framework
Life cycle vehicle impacts can be divided into two primary categories: (1) the
vehicle cycle including the production and use of motor vehicles and (2) the fuel
cycle (also called wells-to-wheels (or W2W)) including the production and use of
motor fuels. W2W impacts are split into two main stages: wells-to-pump (W2P)
and pump-to-wheels (P2W). W2P impacts are the result of the upstream processing
and transportation of the fuel demanded by the vehicle beginning with the raw
material extraction at the well and ending when the fuel is ready to be distributed at
the pump. P2W impacts are directly related to the operation of the vehicle including
all emissions generated by combusting the fuel in the vehicle. Together, the W2P
and P2W impacts comprise the W2W analysis that gives a complete picture of the
environmental impacts of operating a vehicle, particularly in the context of global-
scale emissions.


1.3.3 Life Cycle Assessment Methodology
This study examines the life cycle W2W energy use and GHG emissions of transit
buses and rail systems in Chennai, India and Denver, USA. The four steps of a life
cycle assessment are shown in Figure 1.1 and include:
1) . Goal and scope definition: stating the purpose of the study and identifying the
appropriate boundaries.
2) . Life cycle inventory analysis (LCD: quantifying the material and energy use
along with the environmental emissions associated with the life cycle of the product
or process
3) . Life cycle impact assessment (LCIA): interpreting the results of the analysis to
determine impacts on human health and the environment.
4) . Interpretation: evaluating opportunities to improve the environmental
performance of the process or product by reducing material and energy requirements
or environmental impacts.
12


Figure 1.1 Framework for Conducting Environmental Life Cycle Assessments (U.S.
Environmental Protection Agency 2007)
This thesis focuses on steps 1, 2, and 4. For transit systems in the U.S. and India,
this thesis delineates the boundary (step 1), conducts a life cycle inventory of the
respective systems (step 2) and interprets the data consistent with environmental
priorities in the U.S. and India (step 4). The life cycle impact assessment (step 3) is
omitted as extrapolations between inventoried emissions and aggregate human and
environmental impacts are controversial even in the U.S., as discussed during the
development of EPAs TRACI model (2007), and therefore may not be suitable to
translation to India. To the greatest extent possible, this thesis uses parallel
methodologies in the U.S. and India to examine data availability and to look for
13


broad stroke comparisons between the vehicle manufacturing, fuel production, and
vehicle operation stages. The baseline LCAs can then be used for comparisons with
established life cycle inventory databases and to identify the most critical life cycle
stages to target for emission reductions. The study will provide, for the first time, a
side-by-side comparison of LCAs in developed and developing country mass transit
applications.
1.6 Thesis Objectives and Chapters
The following chapters will delve more deeply into the life cycle greenhouse gas and
energy impacts of transit systems in the U.S. and India and into options for city-scale
actions to reduce transportation sector GHG emissions in Denver. The overarching
objectives of the thesis are:
1) To conduct life cycle assessments of energy use and greenhouse gas
emissions of bus and electrified urban rail systems in India and the U.S.
2) To analyze the life cycle inventory results in order to make broad stroke
environmental impact comparisons between electrified urban rail and bus
systems and between mass transit systems in India and the U.S.
3) To use the case study of Denvers GHG mitigation project to evaluate the
relative effectiveness of mass transit systems in meeting city-level GHG
reduction goals compared to other transportation sector action options.
14


The thesis is organized into the following chapters.
Chapter 2: Life Cycle Assessment of Energy Use and Greenhouse Gas Emissions
for Buses in Chennai, India and Denver, USA
Chapter 3: A Comparison of Greenhouse Gas Emissions and Energy Use for
Electrified Urban Rail and Buses in India and the U.S.
Chapter 4: The Role of Mass Transit in Climate Change Mitigation for the Transport
Sector in Denver, CO
Chapter 5: Conclusions, Contributions, and Future Work
15


2. Life Cycle Assessment of Energy Use and Greenhouse Gas Emissions for
Buses in Chennai, India and Denver, USA
2.1 Goal and Objectives
The overall goal of this paper is to conduct a first life cycle assessment (LCA) of
mass transit systems in India along with broad comparisons using similar U.S.
analyses. The specific objectives of this life cycle assessment of greenhouse gas
(GHG) emissions and energy use for buses in Chennai, India and Denver, USA are
the following:
1) Evaluate the environmental priorities of the citizens of Tamil Nadu, India
through the use of pilot life cycle impact indicator surveys.
2) Conduct life cycle assessments of transit buses in Chennai and Denver
using parallel methodologies to the greatest extent possible.
3) Gather fleet specific fuel economy and ridership data from bus operators in
each city along with real-time measurements of ridership.
16


4) Calculate the relative impact of bus assembly and materials manufacturing
on the bus life cycle impacts compared to wells-to-wheels impacts.
5) Identify the most likely reasons for differences in the life cycle impacts of
the buses in the two cities.
6) Benchmark the results against an established life cycle inventory database
from Europe to check transferability to developing world LCAs.
7) Run sensitivity analyses to determine the impact of changes in fuel
economy and alternative fuels on the life cycle impacts of the buses.
Greenhouse gas emissions are measured in terms of metric tonnes or kilograms of
carbon dioxide equivalent emissions (mtCC^e or kgCC^e).
2.2 Environmental Priorities in Chennai, India
The first life cycle assessments performed in this thesis were designed, to the
greatest extent possible, to be parallel analyses of transit buses in the U.S. and India.
However, before gathering data for a life cycle inventory, it is important to
understand the priorities of the studys target audience of decision-makers and
17


citizens. The primary indicators used in LCAs were originally developed based on
analyses of the most important environmental priorities in Western Europe and North
America. Key studies for the development of these indicators included the Dobris
Assessment (Stanners and Bordeau 1995) and the development of EPAs TRACI
program (2007). Common indicators representing developed world priorities include
climate change, non-renewable energy use, nutrification potential, eco-toxicity,
water and land use, acidification potential, human toxicity, and stratospheric ozone
depletion. For a baseline LCA to be relevant in India, the results should provide
information about impacts and indicators that are considered to be important by
decision-makers and citizens in the developing world.
At present, the primary environmental indicators used in LCAs are based on the
perceptions and priorities of developed nations. Only one recent study, the United
Nations Environmental Program (UNEP) indicator survey (Jolliet 2003), attempted
to identify the environmental priorities of developing nations and to determine the
needs of LCA users around the world. The study assessed the needs for conducting
both life cycle inventories and life cycle impact assessments by surveying 91 LCA
experts including 67 from Europe, 12 from North America, 8 from Asia and the
Pacific, 5 from Africa, and 5 from Latin America / the Caribbean. Despite efforts to
reach out to the developing world, the spread of survey respondents shows a heavy
bias (87%) towards the European and North American continents that have already
18


been well studied, and moreover, the study only obtained results from respondents
who are already LCA users or are actively considering conducting an LCA. The
views of ordinary citizens in developing countries should also be examined to
determine if the outcomes of LCAs are relevant to them.
2.2.1 Survey Methods
As part of this thesis, a pilot life cycle impact indicator priority survey was
conducted of 81 residents of Tamil Nadu, India using a direct interview method to
begin the process of assessing developing world LCA priorities. Survey questions
were translated into the local language Tamil as needed and responses were
anonymous with only demographic information being collected. Respondents were
approximately evenly split between rural/urban, male/female, and educated/less
educated. Less educated was defined as having middle school or less education
while educated included all respondents who had attended high school or college.
An identical survey was also conducted online using English as the language. The
online survey sought responses from businesses in Tamil Nadu to give an alternative
perspective to that of individual citizens. However, few respondents felt comfortable
speaking for the environmental priorities of their company and requested to answer
the survey from their personal perspective. Therefore, the 11 responses of the online
19


survey are added to the on ground interview responses to give a total Indian response
pool of 92 respondents.
Survey theory as outlined by Rea and Parker (1992) states that a sample population
of 97 respondents is required to get a sample mean that is within +/- 10% of the true,
large population mean (with a confidence of 95%). While the number of responses
(92) is slightly lower than the 97 required for a 95% confidence interval at 10%
margin of error, the number of responses is significantly higher than the number
gathered from this part of the world by the previous UNEP study. The pool of
responses is also much broader, expanding beyond the community of LCA experts
and users to include the general public. While conclusive statistical answers may not
be able to be drawn from this pilot survey, the responses do show trends that should
be useful in producing relevant life cycle indicators. The breakdown of survey
respondents is contained in Tables 2.1 and 2.2 below.
20


Table 2.1 India survey respondents
Main Categories Number
Total Number of Interview Responses 81
Male 40
Female 41
Urban 40
Rural 41
Educated 40
Uneducated 41

Total Number of Online Responses 11
Grand Total 92
Table 2.2 Detailed breakdown
of interview responses
Sub-categories from Interviews Number
Male-Educated-Urban 10
Male-Uneducated-Urban 9
Female-Educated-Urban 11
Female-Uneducated-Urban 10
Male-Educated-Rural 9
Male-Uneducated-Rural 12
Female-Educated-Rural 10
Female-Uneducated Rural 10
The online portion of the survey was conducted by the author using contacts
obtained while working at Ashok Leyland and from contacts made while presenting
on life cycle assessment at a Confederation of Indian Industries workshop.
21


The interview portion of the survey was administered with the aid of Professor
Ramesh Mahadevan. Prof. Mahadevan translated the survey from English into the
local language Tamil and then used his Chennai-based students to administer the
survey using a personal interview format. The full English version of the survey is
included in Appendix B along with a sample of the translated Tamil version in
Appendix C.
2.2.2 Survey Questions
Survey respondents were first asked to allocate a total of 100 points between five
broad categories of environmental impacts: (1) Global impacts, (2) Resource use
impacts, (3) Local health impacts, (4) Local ecosystem impacts, and (5) Local
cultural/visual impacts. Next, survey respondents were asked to provide a score for
individual environmental impact indicators within each of the broader categories
with a score of 1 meaning that the impact has no impact on their life or would have
no influence on their decision-making and a score of 5 meaning that the indicator is
critically important in their lives or to their decision-making.
22


2.2.3 Survey Results
The relative category scores from all 92 respondents were averaged and are
presented in Figure 2.2 along with their standard deviations shown as error bars. An
initial hypothesis going into the study was that global climate change impacts may
not be a high priority at the local level in India. However, overall survey results
indicated that while resource use impacts were most important across almost all
categories of respondents, global impacts were considered at least as important as
local health impacts, coming in the second tier of priority rankings. Local ecosystem
and local cultural/visual impacts were considered least important.
Relative Category Rankings
Error bars represent +/-1 std. deviation
Impacts Impacts
Figure 2.2 Relative Impact Category Rankings
23


A two-tailed, paired Ttest was used to evaluate the results. The results suggest three
tiers of broad environmental priorities that can be separated at a 98% level of
confidence. Categories within Tiers 2 & 3 were not statistically different (80% level
of confidence).
Tier 1: Resource Use Impacts
Tier 2: Global Impacts and Local Health Impacts
Tier 3: Local Ecosystem Impacts and Local Cultural/Visual Impacts
Next, the specific indicators were examined to determine their relevance to the
community and also if the rankings of the indicators corresponding to the Tier 1 and
Tier 2 categories tended to be greater than the Tier 3 indicators. Respondents were
not told which indicators fit into each broad category but were instead allowed to
rank each indicator individually based on the relevance it would have to their
decision-making or the impact it would have on their life. The average scores on the
1 (low) to 5 (high) scale and standard deviations for each indicator were calculated
and are presented in descending rank order in Figure 2.3
24


Specific Indicator Scores
Score (1-5)
Figure 2.3: Specific Impact Indicator Rankings
Consistent with the broad category emphasis on global and resource use impacts, the
top five specific indicators that were identified as most important across the
categories of respondents were:
Climate Change Impacts
Changes in Rainfall / Vegetative Cover
Inefficiencies in water use
Sea level rise
Stratospheric zone depletion
25


While the sample size is not large enough to draw definitive conclusions on life
cycle indicator priorities, a few general trends were apparent:
Respondents were both aware of and concerned about how global impacts
such as climate change may impact their lives.
Water use was the most significant resource category of concern as compared
to land, energy, or minerals use (> 99% confidence).
Rural respondents were most concerned with any impacts that might affect
agriculture
In general, indicators that could be directly tied to an impact on a
respondents livelihood were deemed most important and tended to outweigh
impacts on local health.
This survey revealed that global climate change and its associated potential impacts
such as changes in rainfall patterns and sea level rise are of concern to both urban
and rural citizens of Tamil Nadu, India. In addition, water use was revealed as an
important environmental priority consistent with already occurring and increasingly
severe water scarcity issues projected for the developing world (Meinzen-Dick and
Rosegrant 2001). As a result, this study initially focused on water use, energy use,
and associated greenhouse gas (GHG) emissions. However, water use data was not
available from India on a consistent basis.
26


Likewise, data on toxic releases that would impact local health were not recorded in
a usable form in India. Appendix D shows a typical hazardous waste reporting form
for India with releases aggregated into categories such as ETP sludge and acid
residue as opposed to being listed by individual substance (such as specific heavy
metals, dioxins, etc) as is done in the U.S. through databases such as those found at
Scorecard.org. The aggregated Indian categories are more difficult to map directly
to health impacts or to compare to similar emissions in the U.S.
As a result of the unavailability of the water and hazardous waste data in usable
forms in India, the LCA focused primarily on energy use and GHG emissions over
the life cycle of a bus in Chennai, India compared with a bus in Denver, USA.
2.3 Life Cycle Assessment Framework
The life cycle assessment diagram in Figure 2.6 outlines the basic framework for the
analysis of transit buses in the U.S. and India. The major stages include (1) fuel
production, (2) bus material extraction and production, (3) bus assembly and major
components, and (4) bus operation and ridership. Bus end-of-life (5) was omitted
due to the lack of reliable data in India. To ensure parallel analysis, the U.S. and
Indian buses were evaluated starting from bus operation and moving upstream from
that point, using the identical lifetimes of 12 years and 800,000 km per bus that are
27


assumed by the Federal Transit Administration (2006). Ashok Leylands Viking 222
pictured in Figure 2.4 is used to represent the Chennai MTC bus. The Orion ZF bus
pictured in Figure 2.5 is used to represent the Denver RTD bus.
Figure 2.5 Orion ZF Bus
28


Figure 2.6: Four Major Stages of a Bus Life Cycle Assessment
(End-of-life stage is not considered due to lack of data)
29


2.4 Raw Data and Methods
The following sections describe the methodologies used for gathering data for (1)
fuel production, (2) bus materials extraction & production, (3) bus assembly & major
components, and (4) bus operation & ridership. The availability of data and data
sources in India and the U.S. are summarized in Table 2.3 for the various bus LCA
stages shown in Figure 2.6. Complete citations for the Indian and U.S. data sources
listed in Table 2.3 are contained in the reference section. The raw data displayed in
the remainder of this section is drawn from the sources listed in Table 2.3.
30


Table 2.3: Bus life cycle assessment data availability & sources in Chennai, India & Denver, USA
1) Fuel & Electricity Production Indian Data Sources U.S. Data Sources
Crude Recovery Cleveland 2006 Argonne National Laboratory GREET 1.7, 2006 default assumptions
Crude Transportation & Storage Ecobilan TEAM 2007 (ocean tankers), India specific distance estimates
Conventional Diesel Refining Chennai Petroleum Corporation Limited Annual Report 2004-2005
Diesel Transportation, Storage, & Distribution Ashok Leyland (India) fuel estimates and personal communications
Wells-to-Pump Energy Efficiency Calculated from above data
Electricity Generation Emission Factors The Energy and Resources Institute (TERI) India and the Center for Clean Air Policy (CCAP) India GHG Mitigation Report 2006 U.S. Energy Information Administration "Updated State-level GHG ..." 2002
Electricity Transmission & Distribution Losses GREET 1.7 model assumption
Coal Extraction Singareni Collieries Company Limited Presentation, India U.S. Department of Energy, EERE Coal Mining Report 2002
2) Bus Materials Extraction & Production Indian Data Sources U.S. Data Sources
Iron Ore Extraction U.S. Department of Energy, EERE Steel Sector Report 2000
Steel / Cast Iron Production Tata Steel 2005 & Steel Authority India Limited 2005 Reports U.S. Steel Corporation Report 2005
Bauxite Mining International Aluminum Institute 2004 U.S. Department of Energy, EERE Aluminum Sector Report 1997
Alumina Refining National Aluminum Company Limited Report 2006
Aluminum Production
3) Bus Assembly & Major Components Indian Data Sources U.S. Data Sources
Engine Internal Energy Use Data Provided by Ashok Leyland, India No Data Provided by U.S. Manufacturers
Chassis Assembly
Frame Assembly
Gear Box
Front Axle
Rear Axle
Batteries Excide Batteries Personal Communication (India)
Tires Apollo Tyres Report for Indian Bureau of Energy Efficiency (BEE) 2006
Brakes Brakes India Limited Report for BEE 2006
Part Descriptions Ashok Leyland Viking 222 Spec Sheet Orion ZF Spec Sheet
Gross Vehicle Weight
Seating Capacity
Bus Material Weight Ashok Leyland Estimate North American Bus Industry Expert Estimate
4) Bus Operation & Rfdership Indian Data Sources U.S. Data Sources
Fuel Economy Ashok Leyland & Chennai Metropolitan Transport Corporation (CMTC) Denver Rapid Transit District Supplied Data, Federal Transit Administration National Transit Database (2005)
Engine Oil Use Ashok Leyland
Ridership CMTC and Private Study
31


2.4.1 Fuel Production
2.4.1.1 Data Sources
GREET 1.7, a public domain model developed by Argonne National Laboratory
(2007), was used to calculate energy use and greenhouse gas emissions from diesel
fuel production in the U.S using 2006 default values. The same process was
replicated in the Indian fuel sector using publicly
available data. For the Indian diesel pathway, the
crude oil was tracked from extraction in the United
Arab Emirates, through shipping by Suez Max sea
tankers to the Port of Chennai, to the Chennai
Petroleum Limited Corporation (CPCL) refinery in
Manali near Chennai. Energy use at the refinery
was calculated using CPCL documents that showed
both fuel use at the refinery and fuel loss during the
processing and were allocated based on the mass of diesel refined compared to other
refined petroleum products. The last step calculated energy use for transporting the
finished diesel fuel from Manali to gas stations in the city center.
Figure 2.7 Diesel Fuel Refining
32


2.4.1.2 Raw Energy Data
Table 2.4 shows a side-by-side comparison of the efficiency calculations for four of
the major stages and for the overall wells-to-pump (W2P) process. The efficiency is
defined as energy out (energy content of fuel) divided by the total energy input of the
stage (energy content of fuel plus process energy).
n=Ef/(Ef + Ep) (2.1)
r) = Efficiency of the stage
Ef = Energy content of the fuel
Ep = Energy used in the process
Table 2.4 Step-wise and wells-to-pump energy efficiency for diesel production
Stage India Efficiency GREET Efficiency
Crude Recovery 95% 98%
Crude T&S 100% 99%
Diesel Refining 91% 89%
Conventional Diesel T&S&D 100% 99%
Wells-To-Pump Aggregate Total 87% 85%
Values are rounded to the nearest percent.
As shown in Table 2.4, the overall W2P efficiency calculated for India is slightly
greater than in the U.S. Table 2.5 shows the details for the India pathway
calculations corresponding to the 87% W2P energy efficiency shown in Table 2.4.
Data sources for these calculations are in Table 2.3.
33


Table 2.5 India crude oil to diesel pathway details
Category Value (MJ/L)
Energy Used for Crude Oil Extraction 1.9
Tanker Transport from Middle East to Chennai Port 0.06
Transport from Chennai Port to Manali Refinery 0.005
Refinery
Electricity 0.016
Furnace Oil 0.014
FCCU Coke 0.001
Fuel Gas 0.003
Refinery Total 3.9
Transportation from Manali to Chennai 0.012
Total Well-to-Pump Energy Input 6.0
Energy Content of Final Diesel Fuel (energy out) 38.6
Well-to-Pump Efficiency (energy out / total energy input) 87%
To check the reasonableness of the Indian calculations, the values were benchmarked
against two other studies: a fuel cycle analysis conducted using Gabi 4 in Northern
Europe (Silva, Goncalves et al. 2006) and a fuel cycle analysis conducted in
Australia (Beer, Grant et al. 2000). Table 2.6 shows that GREET energy and GHG
emission estimates are markedly higher than the other studies.
Table 2.6: Wells-to-pump energy efficiency and greenhouse gas comparison
(data sources are shown in Table 2.3)
Study Energy Consumption (MJ/L) GHG Emissions (mtC02e/mt diesel)
India Fuel Cycle LCA 6.0 0.53
GREET 1.7 (U.S.) 7.1 0.74
Europe Gabi 4 Analysis 6.1 0.65
Australia Fuel Cycle Analysis 6.0 0.58
The GHG intensity estimate for India is 28% lower than the U.S. largely due to
lower energy consumption estimates. However, the Indian value for energy
34


consumption is on par with the Northern Europe and Australian studies, and the
GHG emission intensity estimate is within 10-20%. The Indian GHG estimate
contained in Table 2.6 is used for the Indian bus calculations in this study. U.S.
energy use and GHG emissions may be higher due to stricter refining standards and
longer transportation distances. The impact on life cycle GHG emissions of applying
the higher GREET value to Indian diesel processing is discussed in Section 2.6.1.
2.4.2 Bus Materials Extraction and Processing
2.4.2.1 Data Sources
Steel, cast iron, and aluminum compose the majority of the
weight in U.S. and Indian buses. Cast iron is produced
during the integrated steel and iron making process and
requires about 60% of the energy needed to produce steel.
The steel and aluminum sectors were evaluated in detail to
determine differences in energy intensity for material
production starting from raw material extraction. Similar
calculations were performed to note differences in diesel
fuel combustion emission factors and upstream diesel
Figure 2.8 Steel Production
processing factors. In the U.S., energy intensity data is
tracked by the Department of Energys Office of Energy Efficiency and Renewable
35


Energy (Energetics Incorporated 2000). In India, similar energy intensity data is
tracked by the Bureau of Energy Efficiency. The steel sector in India was evaluated
using data from Tata Steel (2005) and Steel Authority India Limited (2005), two of
the industry leaders in India with a combined market share of 48% (Steel Authority
India Limited 2005). In the U.S., U.S. Steel Corporation (2005), the 7th largest steel
producer in the world (largest U.S. based), was taken as representative for producing
steel from virgin materials. Both analyses considered the extended life cycle of iron
and steel making including iron ore mining.
For the aluminum sector analysis, the National Aluminum Company Limited
(NALCO) (2006), with 30% of Indias market share, was studied for the Indian
sector representative while Alcoa (2005) and Alcan (2005) (the 2nd and 3rd largest
worldwide aluminum producers), and a sector wide report prepared for the U.S.
Department of Energy (Energetics
Incorporated 1997) were used for the U.S.
data. All data in this analysis was obtained
from publicly available documents and was
tracked from bauxite mining (International
Aluminum Institute 2004) through final
aluminum production, although the comparisons were complicated by data being
presented in different formats and in differing levels of detail.
Figure 2.9 Aluminum Production
36


2.4.2.2 Raw Data
The raw data used for the calculations of GHG impacts of steel and aluminum
production are contained in Tables 2.7 and 2.8 respectively. Assumptions common
to calculations for both the U.S. and India are contained in table footnotes. Any
energy use that could not be determined to come from a specific fuel type was
assigned to electricity. Table 2.7 highlights the specific energy advantage that steel
production in the U.S. has over steel production in India. The U.S. steel sector also
is able to produce a higher percentage of recycled steel than India. Table 2.8 shows
that U.S. aluminum data was available as aggregated specific energy for the overall
process and as GHG emissions for each step while Indian GHG emissions were
calculated from the use of individual fuels in each stage of the aluminum production
process. Indian aluminum energy intensity was computed at 67 GJ/tcs, greater than
the worldwide average of 56 GJ/tcs.
37


Table 2.7 Energy & greenhouse gas for steel production calculations
(data sources shown in Table 2.3)
Production Stage Production Process Units India U.S.
Iron Ore Extraction See footnotes A,B,C - - -
Energy Use Required for Cast Iron Production
Steel Processing Energy Required for Coal Extraction for Coke kWh/mt coal 18 12
Iron Ore Required for Steel Production mt/mt steel 1.4 1.2
Electricity Demanded During Steel Production kWh/mt steel 537 1528
Coking Coal Demanded during Steel Production kg/mt steel 746 400
Non-Coking Coal Demanded during Steel Production kg/mt steel 194 -
Petro Fuel Demanded during Steel Production kg/mt steel 6 -
Summary Comparison Specific Energy Required for Virgin Steel Production0,0 GJ/tcs (mt crude steel) 27.6 16.3
Average Percentage of Steel Produced Using Recycled Scrap % 3 42
Greenhouse Gas Emission Intensity mtC02e/mt steel 3.1 2.6
A. 92 kg of C02e per mt iron ore are released during iron ore extraction
based on 99 kWh of electricity and 15 m3 of natural gas used per mt of iron ore produced
B. Cast iron is assumed to require 60% of the energy required for steel
production
C. Producing recycled steel uses 29% of the energy required for producing virgin steel (Tata
Steel 2005)
D. The worldwide average specific energy required for steel production is 19.1 GJ/tcs (Tata
Steel 2005)
E. Tata Steel's specific energy consumption is 24.7 GJ/tcs while Steel Authority India Limited's is 30.5 GJ/tcs
38


Table 2.8 Energy & greenhouse gas emissions from aluminum production
calculations (data sources are shown in Table 2.3, U.S. data only available in GHG emissions)
Production Stage Production Process Units India U.S.
General Assumptions See Footnotes A,B,C,D,E - - -
Bauxite Mining Total Energy Required for Bauxite Mining MJ/mt bauxite 102 -
Diesel Fuel Used for Bauxite Mining L/mt bauxite 1.8 -
Fuel Oil Used for Bauxite Mining L/mt bauxite 1.7 -
Coal Used for Bauxite Mining kg/mt bauxite 0.2 -
GHG Emissions from Bauxite Mining kgC02e/mt bauxite - 16.7
Alumina Processing Electricity Demanded for Alumina Processing kWh/mt alumina 337 -
Fuel Oil Used for Calcination kg/mt alumina 77.4 -
Coal Used for Steam kg/mt alumina 636 -
Oil used for Steam kg/mt alumina 3.3 -
GHG Emissions from Alumina Processing1" kg C02e/mt alumina - 1620
Aluminum Production Electricity Demanded for Aluminum Reduction kWh/mt aluminum 14728 -
Fuel Oil Demanded for Aluminum Reduction L/mt aluminum 84 -
CP Coke Demanded for Aluminum Reduction kg/mt aluminum 392 -
GHG Emissions from Energy Use during Aluminum Production kg C02e/mt aluminum - 5310
Summary Comparison Specific Energy Demand for Aluminum Reduction6" GJ/mt aluminum 67 56
Greenhouse Gas Emission Intensity mtC02e/mt aluminum 33.7 9.3
A. 2.5 mt bauxite is required per mt alumina
B. 2 mt alumina is required per mt aluminum
C. Recycled aluminum production requires 6% of the energy required by primary aluminum
D. GHG emissions from carbon anode production equals 120 kg C02e/mt aluminum
E. GHG emissions from the carbon anode effect equals 2200 kg C02e/mt aluminum
F. In the U.S., the energy demand for all aluminum raw materials is 8200 kWh/mt aluminum
G. Hindalco (another Indian company) operates at a specific energy of 57 GJ/mt aluminum
H. The worldwide average for aluminum production specific energy is 55.6 GJ/mt aluminum
European LCA estimates equal 55.3 GJ/mt aluminum and Alcan and Alcoa equal 48-54.
39


2.4.3 Bus Assembly and Major Components
2.4.3.1 Data Sources
India-specific bus assembly energy use data was acquired from Ashok Leyland, a
major bus manufacturer in India that supplies over 80% of Chennais bus fleet.
Equivalent data in the U.S. was not released from any of the four major U.S. bus
manufacturers who were contacted (Gillig Corporation, North American Bus
Industries, Orion Bus Industries, Motor Coach Industries). Comprehensive lists of
the materials used to make the major components for U.S. and Indian buses were not
available, however a comparison was still attempted for the Orion ZF transit bus
(Orion Bus Industries 2007) in Denver and the Viking 222 (Ashok Leyland 2007)
transit bus in Chennai. The buses were compared in terms of gross volumetric
weight (GVW) and major material categories. Table 2.9 represents the best
approximations based on expert estimates of materials for North American buses
from James Prendergast of the Canadian Vehicle Technology Centre and based on
Indian bus material estimates provided by Ashok Leyland employees.
40


2.4.3.2 Raw Data
Table 2.9 highlights a three metric tonne (mt) weight difference between the GVWs
of the U.S. and Indian buses.
Table 2.9: Energy intensive primary bus materials in India and the U.S.
Bus Weight (mt) India Viking 222 U.S. Orion ZF
GVW 15.4* 18.4*
Steel/Cast Iron 3.8 5.8
Aluminum 1.9 2.5
Tires 0.7 0.7
Plastics/Other Fitments 1.7 2.9
* Data from spec sheets. Material distribution estimate from country experts
In addition to supplying estimates of the overall material use in the Indian bus,
Ashok Leyland also supplied fuel specific energy data for the production and
assembly of the engine, chassis, frame, gear box, front axle and rear axle as shown in
Table 2.10. Data for the tires, batteries, and brakes were taken from manufacturer
reports as described in Table 2.3.
Table 2.10 Fuel consumed in bus chassis assembly and for assembling/producing
first level components in India
Bus Assembly Stage Electricity (kWh) Petroleum Fuels (L) Coal (kg)
Engine 299 8.1 0.03
Chassis 483 0.0 0.00
Brakes 110 0.0 0.00
Tires 480 0.0 0.00
Frame 5 0.7 0.00
Batteries 50 1.5 0.01
Gear Box 315 1.4 0.00
Front Axle 2 0.3 0.00
Rear Axle 6 0.8 0.00
Total 1752 12.7 0.04
41


The total amount of each material used is the most uncertain data in this study,
however, the GVW obtained from the spec sheets for each bus (representing the
most reliable data) shows that the U.S. bus is approximately three metric tonnes
(20%) heavier than its Indian equivalent. As will be shown later, primary material
production comprises less than 10% of life cycle greenhouse gas impacts for each
bus. For comparison, the embodied energy of manufacturing cars has been found to
represent a similar 10% of the vehicles life cycle greenhouse gas impacts over a
shorter lifetime of 150,000 km (Beer, Grant et al. 2000). Table 2.11 lists the specific
component information that was provided to the author by U.S. and Indian sources or
from vehicle specification sheets.
Table 2.11 Major bus components comparison Orion ZF (USA) and Viking 222
(India)______________________________________________________________
Component Weight (kg) Part Description
Engine
Orion ZF 706 Cummins ISL
Viking 222 510 H Series 6ETI 6 cylinder
Transmission/Clutch/Gear Box
Orion ZF 310 ZF Ecomat 2 HP 592C
Viking 222 24 Single Plate Dry Type Clutch
Viking 222 105 5-speed synchromesh gear box
Front Axle
Orion ZF 168 Meritor Solid Beam: 6620 kg capacity FG-941
Viking 222 360 Forge I section reverse Elliot type: 5,600 kg capacity
Rear Axle
Orion ZF Meritor Conventional: 26,000 lbs RC-26-720
Viking 222 500 Fully floating, single speed, spiral bevel gear
Gross Vehicle Weight
Orion ZF 18,413 Seating capacity = 45 persons
Viking 222 15,430 Seating capacity = 48 persons
Primary Materials Steel, aluminum, cast iron
42


2.4.4 Bus Operations
2.4.4.1 Data Sources
Denver RTD provided detailed fuel economy data for 234 Orion ZF buses over a
three month period. At the end of each service day, the miles driven and fuel
required to fill the tank were recorded. Buses were rotated throughout various RTD
bus routes making it impossible to assess the sensitivity of fuel economy to drive
cycles in Denver. Indian bus fuel economy was taken from two sources. The first
source consisted of data tracked directly by Ashok Leyland (AL). AL tracked 21
buses on 7 routes in Chennai over the course of 7 months. This data was used to
calculate the Viking average fuel economy and standard deviation. Then data from
the website of the Chennai Metropolitan Transit Corporation (MTC) was used to
determine how the Viking 222 bus fuel economy compared to the Chennai MTC
fleet average.
2.4.4.2 Raw Data
Table 2.12 summarizes the fuel economy data for transit buses in Chennai and
Denver calculated from data gathered by fleet operators. Table 2.12 shows that the
43


fuel economy for buses in Chennai is more than 125% greater than for buses in
Denver. Further analysis showed little difference across routes with a relative
standard deviation for fuel economy in Chennai of less than 5%. While route-
specific fuel economy was unavailable in Denver, the overall standard deviation for
fuel economy equaled approximately 5%.
Table 2.12 Fuel economy for transit buses in Chennai and Denver
Fuel Economy (km/l) Standard Deviation (km/l)
Denver RTD 1.88 0.11

Ashok Leyland Viking 4.23 0.15
Chennai MTC 3.65 -
2.4.5 Bus Ridership
Mass transit systems are ideally designed to efficiently move people, not vehicles.
While greenhouse gas emissions are typically measured per vehicle-km (v-km), a
more appropriate measure of system efficiency may be GHG emissions per
passenger-km (p-km). While v-km calculations give lower GHG emissions to
systems with highly efficient, technically advanced vehicles, p-km calculations more
directly reflect the number of people moved by a mass transit system. Ridership
volumes and occupancy data are needed for such computations.
44


R = P-KM / V-KM (2.2)
R = Ridership
P-KM = Kilometers traveled by riders on the vehicle
V-KM = Revenue kilometers traveled by the vehicle
2.4.5.1 Data Sources
Ridership data for India was initially
obtained from the Chennai Metropolitan
Transport Corporation website, and then
an independent study was commissioned
by the author to evaluate Chennai bus
ridership numbers. Local high school
students were hired to ride four bus routes within Chennai, both during rush hour and
during daytime, off-peak hours. Each route was ridden on two different weekdays,
two rush and two non-rush rides per day for a total of 32 data sets. Ridership
numbers were recorded ten times during each ride.
For the U.S., the Federal Transit Administration collects ridership data from transit
agencies around the country in their National Transit Database (Federal Transit
Administration 2007). Ridership data for Denver RTD was examined for the year
2005. Denvers average bus capacity figures were calculated by taking the total
Figure 2.10 High Bus Ridership in Chennai
45


number of passenger miles and dividing by revenue vehicle miles. RTD rotates its
buses throughout the system meaning that buses are equally likely to run on high
ridership or low ridership routes. Therefore, the system average should fairly
represent the p-km / v-km for RTD buses.
2.4.5.2 Raw Data
Table 2.13 shows the ridership data for Chennai and Denver in terms of passenger-
km / vehicle-km. The buses have similar maximum seated capacities.
Table 2.13 Average bus ridership for Chennai, India and Denver, USA
Bus Capacity (# of passengers) Passenger-km (p-km)/ Vehicle-km (v-km) Average % Occupancy
Chennai MTC 48 38.8 81%
Denver RTD 45 9.4 21%
The Chennai ridership is in line with current and forecasted ridership for Delhi of 36-
40 p-km/v-km (Bose, Nesamani et al. 2001). The author-initiated independent
survey measured ridership for the four Chennai bus routes as averaging 55.0 p-km /
v-km, far more than the average Chennai capacity estimates, however, when off-
peak, evening buses and weekend buses are included in the mix, the average
ridership will likely approach the reported average of 38.8 p-km / v-km. The
appropriateness of the Denver ridership numbers was confirmed by the author riding
RTD buses and recording the number of passengers at several points during the
46


route, mimicking the study conducted in Chennai. Denver ridership data is also in
line with Australian ridership estimates of 10.5 p-km/v-km (Beer, Grant et al. 2000).
2.4.6 Greenhouse Gas Emission Factors
Once the energy use in the various LCA stages was determined from a variety of
sources described in the earlier sections, the raw data was converted to greenhouse
gas emissions for the life cycle assessment of transit buses in the U.S. and India. The
conversion factors used for this analysis are contained in Table 2.14.
Table 2.14 Greenhouse gas emission factors
Fuel Units lndiaA U.S. B
Electricity0 kg C02e/kWh 0.98 0.66
Diesel mtC02e/mt combusted 3.3 3.2
Coal mtC02e/mt combusted 2.3 2.4
Natural Gas kg C02e/mA3 combusted 2.2 2.0
A. India emission factors and T&D loss data taken from the TERI GHG Report (TERI 2006)
B. U.S. emission factors and T&D loss data taken from Energy Information Administration reports
C. Electricity emission factors include T&D losses of 26% in India (TERI) and 8% in the U.S (GREET).
The small differences in emission factors for the fuels are based on the assumptions
made in Indian (The Energy and Resource Institute and The Center for Clean Air
Policy 2006) and U.S. (U.S. Energy Information Administration 2002) reports
focused on developing country-specific emission factors for India and the U.S. The
electricity factors are based on the national grid averages for each country, including
expected transmission and distribution (T&D) losses of 26% in India (The Energy
47


and Resource Institute and The Center for Clean Air Policy 2006) and 8% in the U.S
(Argonne National Laboratory 2007).
2.5 Results and Analysis
2.5.1 Bus Materials Production
The greenhouse gas analysis begins by examining the material sector emission
intensity differences between the U.S. and India. Table 2.15 shows the GHG gas
emission intensity of each sector in terms of metric tonnes of CC^e released per
metric tonne of material used.
Table 2.15 Material sector greenhouse gas emission intensity
(mt CChe/mt material used)
Sector India US
Aluminum 33.7 9.3
Steel 3.1 2.6
Diesel (P2W) 3.3 3.2
Diesel (W2P) 0.5 0.7
Aluminum production is a highly energy intensive process that relies on large
quantities of electricity. Therefore, the source of electricity plays a significant factor
in the total GHG emissions for aluminum production. In India, the GHG intensity
factor (mtCChe/mt material) is approximately 33.7 while in the U.S., the factor is
much lower at only 9.3. The U.S. number is significantly lower because the U.S.
48


aluminum industry relies largely on hydropower with a majority of processing plants
located in the Pacific Northwest. This reliance on hydropower lowers the emission
factor for the U.S. electricity used in the aluminum process, thereby reducing total
sector intensity. If all of the electricity in the U.S. was assumed to come from coal
instead of a large percentage from hydropower, the U.S. sector intensity would
increase to 20.5 mt C02e/tonne of aluminum, much closer to the India sector value
of 33.7.
The Indian aluminum sector is also negatively influenced by electricity T&D losses
assumed at 26% compared to only 8% in the U.S. as stated in section 2.4.6. Note
that if the Indian aluminum plants are located next to generation facilities and
therefore suffer much lower losses, Indian aluminum intensity would drop to 28.4
mtC02e/mt aluminum (assuming no loss). This change would result in material
manufacturing accounting for 9% of life cycle GHG emissions for Indian buses as
opposed to 11%, not changing the conclusion that material manufacturing in India
accounts for approximately 10% of the total life cycle bus GHG emissions. Energy
use can be reduced by up to 95% if recycled aluminum is used instead of virgin
aluminum. Both LCAs assume the use of virgin aluminum.
The Indian steel sector registered a GHG intensity of 3.1 mtCChe/mt steel while the
U.S. intensity measured 2.6. US Steel Corporation uses less energy than the
49


worldwide industry average while both Tata Steel and Steel Authority Limited are
well above worldwide industry average specific energy consumption as shown in
Table 2.7. Some reasons for higher average energy use in India include smaller
scales of operation, poorer quality raw materials, and slower updating to modern
steel making technologies. The production of recycled steel is a much less energy
intensive process than the production of primary steel, using up to 70% less energy.
Worldwide, approximately 42% of steel is recycled. In contrast, Tata Steel in India
uses less than 4% recycled steel. This discrepancy is largely due to the lack of scrap
steel available in the Indian market. The bus LCAs assumed that steel for both buses
was produced from virgin materials. However, the U.S. bus is much more likely to
use recycled steel, giving it a potential materials GHG advantage over the Indian bus.
Diesel fuel combustion factors for Indian diesel and U.S. diesel were similar with
Indian diesel measuring 3.3 mtCC^e/mt diesel and US diesel measuring 3.2. The
Indian LCA analysis yielded an emission intensity of 0.5 for upstream diesel
processing in India. In comparison, the full life cycle analysis from GREET
estimated an emission factor of 0.7 for the U.S.
50


2.5.2 Wells-to-Wheels Fuel Analysis
Using the energy use data for the four main life cycle stages and the above GHG
conversion factors, the wells-to-wheels (W2W) energy use and GHG emissions for
Indian and U.S. buses were calculated as shown in Table 2.16. While W2P energy
use in the U.S. and India are similar (6.0 MJ/L), India has a distinct advantage in
energy use per vehicle kilometer due to the higher fuel economy of Indian buses.
Indian buses show similar expected advantages during the bus operation stage due to
higher ridership.
Table 2.16: Energy and greenhouse gas emissions for diesel production (W2P),
vehicle operation (P2W), and overall (W2W)
India U.S.
Energy (MJ/km) GHG (kg C02e/km) Energy (MJ/km) GHG (kg C02e/km)
Wells-to-Pump (W2P)ab 1.4 0.1 3.8 0.3
Pump-to-Wheels (P2W)c,d 9.1 0.7 20.6 1.4
Wells-to-Wheels (W2W) 10.5 0.7 23.8 1.8
* Numbers in the table are rounded
A Refining data from Chennai Petroleum Corporation Limited (2004-2005),
transportation data from Ashok Leyland, crude oil extraction energy use
based on Energy Information Agency estimates, electricity data from Indian Ministry of Power.
B Results from GREET 1.7, scenario for wells-to pump diesel production in U.S.
C Data from Metropolitan Transit Corporation Chennai, February 2005-August 2005.
D Data from Denver Rapid Transit District, March 2005-April 2006
51


W2P = Wells-to-pump energy use
Edp = Diesel processing energy (MJ/1)
FE = Fuel economy (km/1)
(2.3)
2P = E(
dp
/FE
P2W = Edc / FE (2.4)
P2W = Pump-to-wheels energy use
Edc = Energy content of the diesel fuel (MJ/1)
FE = Fuel economy (km/1)
2.5.3 Bus Life Cycle Assessment
Using the vehicle weights and materials in Table 2.9 and the material GHG
intensities in Table 2.15, the overall bus LCA energy and GHG emissions are
calculated in Table 2.17. The energy and GHG impacts of the transit buses were
broken down by life cycle stage in order to identify where the major opportunities
are for reductions.
2.5.3.1 Life Cycle Stage Comparison
The life cycle stage impact comparison in Table 2.17 clearly shows that the wells-to-
wheels emissions of the bus comprise the vast majority of the bus greenhouse gas
impacts. In both the U.S. and India, bus operation and diesel fuel processing account
for over 90% of total life cycle GHG emissions. The wells-to-wheels impacts of
52


buses will tend to be relatively greater than other vehicles such as automobiles due to
the longer assumed lifetimes of 12 years and 800,000 km. The primary driver for
wells-to-wheels GHG emissions is the fuel economy of the bus, more than diesel fuel
processing or combustion emission factor differences. This comparison highlights
that while there are differences in the material and manufacturing impacts of Indian
and U.S. buses, the majority of attention should be focused on wells-to-wheels
impacts in looking for opportunities for significantly decreasing life time bus GHG
emissions. The percentage breakdowns of the GHG emissions and energy use by life
cycle stage are shown in Figures 2.11 and 2.12 respectively.
Bus operation clearly dominates for both GHG emissions and energy use with
upstream fuel processing having the second greatest impact. These two categories
combine to form the wells-to-wheels (W2W) impacts that comprise the vast majority
of life cycle GHG and energy impacts for buses in both India and the U.S.
53


Table 2.17 Bus life cycle greenhouse gas emissions and energy use
India u.s.
Life Cycle GHG Emissions Life Cycle Energy Use Life Cycle GHG Emissions Life Cycle Energy Use
mtC02e GJ mtC02e GJ
1. Fuel Processing 85 1135 265 2553
2. Bus Materials Extraction & Processing 72 217 38 239
3. Bus Assembly & Major Components 2 7
4. Bus Operation 535 7300 1141 16426
Total for Bus Lifetime 694 8659 1445 19218
Impacts Per Unit of Vehicle Travel
GHG (kg C02e) Energy (MJ) GHG (kg (C02e) Energy (MJ)
Per Vehicle-km 0.9 11 1.8 24
Per Passenger- km 0.02 0.3 0.19 2.6
A. Assume a bus lifetime of 800,000 km
B. Assume 9.4 p-km / v-km for the U S. bus and 38.8 p-km / v-km for the Indian bus
54


India Bus GHG Emission by Life Cycle Stage
(Total Emissions 0.9 kgC02e/v*km)
1. Fuel Processing
S%
U.S. Bus GHG Emissions by Life Cycle Stage
(Total Emissions 1.8 kgC02e/v-km)
Figure 2.11 Greenhouse Gas Emissions Per Vehicle-Km by Life Cycle Stage for
India (left) and the U.S. (right).
India Bus Energy Use by Life Cycle Stage
(Total Energy Use = 11 MJ/v-km)
3. Bus Assembly &
Major Components
0.1%
1. Fuel Processing
U.S. Bus Energy Use by Life Cycle Stage
(Total Energy Use = 24 MJ/v-km)
1. Fuel Processing
Figure 2.12 Energy Use Per Vehicle-Km by Life Cycle Stage for India (left) and the
U.S. (right)
55


2.5.3.2 Bus Assembly Impacts
The LCA in India studied the primary
stages of bus assembly that were
identified as the production of the
engine, brakes, tires, batteries, gear box,
front axle, and rear axle, along with the
assembly of the chassis and frame.
GHG emission estimates were calculated
based on the electricity, coal, and diesel fuel used during each bus assembly stage.
The total GHG emissions for the nine bus assembly stages were estimated at less
than 0.2% of the total life cycle GHG emissions of an Indian bus. The total life cycle
GHG emissions for Indian buses are about 50% lower than life cycle GHG buses for
U.S. buses. Major bus manufacturers in North America were contacted to attempt to
gather detailed bus assembly energy use data for the U.S., however no bus
manufacturers were willing to provide this data. However, bus assembly energy use
in the U.S. and India should be on a similar scale. Based on the higher overall GHG
emissions for a U.S. bus life cycle, the expected impact of bus assembly in the U.S.
is less than 0.5%. This value is below the significance threshold for a typical LCA
and therefore detailed calculations have been omitted.

Figure 2.13 Bus Chassis
56


2.5.3.3 Benchmarking Material Impact Calculation
The Indian and U.S. bus material life cycle assessments were benchmarked against
an established life cycle inventory database to determine the potential impact of the
material data uncertainty on the overall analysis. Gabi 4 is a software system that is
designed for building life cycle balances for products and processes (PE International
2007). The GaBi 4 software can provide valuable data for greenhouse gas
accounting, life cycle assessment, and total cost accounting. Gabi 4 data sets are
taken from industry-specific analyses as well as reviews of patents and technical
literature. Data sets cover a wide range of industries including primary metal
production, fuel processing, electricity generation, and final product production. The
GaBi 4 Ecoinvent 1.2 database contains life cycle inventory data for a typical Swiss
transit bus. The database results were scaled from the European weights in the
database to the weight of the Indian and Denver buses for comparison as shown in
Table 2.18.
57


Table 2.18 Comparison of upstream GHG emission methodologies
Table 2.18: Comparison of Upstream GHG Emission Methodologies
Bus Weight (mt) GHG Materials / Manufacturing Total GHG Emissions (mtC02e)
U.S Bus8 11.9 38 1445
Gabi Estimate 11.9 37 1443

p India Bus 8.0 72 694
Gabi Estimate 8.0 25 647
A. Bus weight taken from sum of materials in Table 2.9 not from GVW
B. U S. bus measurements from this study
C. India bus measurements from this study
Table 2.18 suggests the following:
- The Gabi 4 upstream manufacturing estimates are very similar to U.S.
calculations leading to predicted splits for GHG emissions by stage that are
virtually the same.
Gabi 4 underestimates the upstream emissions from materials and
manufacturing in India likely because the aluminum and steel sectors in India
are more energy intensive, use less efficient fuel, and suffer from high
electricity T&D losses compared to the U.S. and Europe.
Indian buses have significantly lower life cycle GHG emissions than U.S.
buses due largely to the fuel economy advantage from reduced vehicle
weights, regardless of the upstream material calculation method.
In all cases, W2W emissions account for approximately 90% or more of the
vehicles total life cycle GHG emissions.
58


2.5.3.4 Bus Operation Impacts
As shown in Table 2.17, Indian buses have much lower life cycle GHG emission and
energy use intensities on both a per v-km and per p-km basis. Chennai buses
average a GHG emission intensity 0.9 kg C02e / v-km compared with 1.8 kg CC^e /
v-km for Denver buses. In addition to this advantage, Chennai MTC buses also
average greater ridership numbers than Denver RTD buses. Chennai buses average
38.8 p-km / v-km while Denver buses average only 9.4 p-km / v-km. The greater
Indian bus ridership increases the GHG intensity advantage for buses in Chennai
when calculated on a per p-km basis. Chennai buses emit 0.02 kg C02e / p-km
versus 0.19 kg C02e / p-km for Denver buses, a per p-km advantage of 88% for the
Indian buses versus a per v-km advantage of only 50%.
The initial analysis based on v-km shows that Indian buses operate at double the
efficiency of U.S. buses, largely due to less vehicle weight, resulting in the emission
of fewer life cycle GHG emissions per v-km. Additionally, Indian buses enjoy more
than a four times advantage in terms of ridership. Therefore, the already more
efficient Indian buses have an even greater advantage over U.S. buses when GHG
calculations are done in terms of p-km. Figure 2.14 details the GHG emissions of
Indian and U.S. buses on a per unit kilometer traveled basis.
59


Life Cycle Transit Bus Greenhouse Gas Emissions
Per Vehicle-km Per Passenger-km
Figure 2.14: Life Cycle Transit Bus Greenhouse Gas Emission Intensities Per V-
Km and Per P-Km
In terms of bus maintenance, Ashok Leyland buses get an average of 1,799 km/1 of
engine oil used while Denver RTD buses get 1,828 km/1, a difference of less than
0.2% (data supplied by the transit agencies). Moreover, inclusion of the well-to-
wheels impacts of producing and consuming engine oil would increase the life cycle
GHG emissions by less than 0.2%. Detailed data on other maintenance practices,
including energy use in the facilities was unavailable and therefore was omitted from
the analysis. However, the inclusion of maintenance data is unlikely to change the
conclusions of this study as the potential impacts are small compared with the actual
operation of the vehicle. The more likely impact of vehicle maintenance is to
improve fuel efficiency, thereby reducing GHG emissions. A study by TERI (The
Energy and Resources Institute, India) estimates that improved vehicle maintenance
60


in India could reduce operational GHG emissions by 10-20%, while in the U.S., the
potential GHG reductions are estimated at only approximately 5% (Deb 2004).
2.6 Discussion
2.6.1 Data Availability and Uncertainty
Conducting parallel life cycle assessment in India and the U.S. proved difficult due
to great variances in data availability and quality. Data for the vehicle operation
stage and bus ridership was made available by the respective transit corporation and
is considered high quality. The data on the energy and GHG intensities of steel and
aluminum production is also considered complete based on publicly available
reports. Bus assembly data for India was obtained through the author working at the
headquarters of Ashok Leyland in India for two months and is not considered widely
available. The author was unable to obtain similar data for the U.S., although the
expected contribution to life cycle GHG impacts is less than 0.5%.
Two pieces of data are considered highly uncertain. The first is the amount of
materials used in the production of each bus. Materials were estimated from
consultations with industry experts in India and the U.S. but neither Ashok Leyland
61


nor Orion Bus Industries were willing or able to supply a comprehensive list of the
weights of each material used in their buses. The impact of the material uncertainty
on the overall study results is minimized by the total contribution of material
production and bus assembly being less than 10% and is discussed further in Section
3.5.4.7.
The second uncertain data point is the GHG intensity value calculated for diesel fuel
processing in India. If the higher GREET GHG intensity value was used in all
calculations, the total life cycle GHG emissions for the Indian bus would have
increased from 694 mtCChe to 727 tritCC^e, an increase of less than 5%. The GHG
intensity of the Indian bus on a per v-km basis remains at 0.9 and the GHG intensity
on a p-km basis remains at 0.02, indicating that the results are unchanged to one
significant figure.
2.6.2 Sensitivity Analysis
2.6.2.1 General Fuel Economy Impacts
Fuel economy directly impacts both the GHG emissions from diesel fuel combusted
in the bus engine and GHG emissions from the upstream processing/refining of the
diesel fuel. For the Denver RTD average fuel economy of 1.88 km/1, GHG
62


emissions during operation and upstream processing equaled 1141 mtCChe and 265
mtC02e respectively for a wells-to-wheels (W2W) fuel emission total of 1406
mtCChe. This value represents 97% of bus GHG emissions including bus operation,
fuel processing, and primary material production. Fuel economy changes were then
evaluated to determine how significantly W2W GHG emissions would be impacted
relative to the Denver RTD baseline. Raising the fuel economy of the Denver bus by
one standard deviation (0.11 km/1) would decrease W2W emissions by 5% while
decreasing the fuel economy by one standard deviation would increase W2W
emissions by 6%.
2.6.2.2 Impact of Vehicle Weight and Air Conditioning on Bus Fuel Economy
Numerous factors could be leading to the difference in fuel economy between Indian
and U.S. buses, biasing the fuel economy in each direction. Typical Indian buses do
not have passenger windows which would increase the drag on the bus. Also, many
would assume that the traffic in Indian cities, like Chennai, would decrease fuel
economy significantly. However, although chaotic at times, Chennai has far fewer
stoplights than Denver and smaller vehicles tend to cede right-of-way to Indian
buses. As a result, the difference between the Chennai and Denver drive cycles may
have only a secondary impact on fuel economy. Two primary drivers of the
difference may be the differences in vehicle weight and the use of air conditioning.
63


All other factors being equal, increased vehicle weight will decrease fuel economy.
According to a recent study by Jacobson (2006), fuel economy decreases by 0.03%
for each pound of vehicle weight increase.
FEd = 0.03 VWj (2.4)
FEd, Fuel economy decrease (%)
VWj Vehicle weight increase (lb)
The U.S. bus is approximately three metric tonnes heavier than the Indian bus. This
should lead to a 3.4 km/1 decrease in fuel economy for U.S. buses vs. Indian buses.
The observed difference for this study is approximately 2.4 km/1. The observed
difference may be lower than expected due to lower ridership figures in the U.S.
decreasing the actual difference in vehicle operating weights to less than the
theoretical three metric tonnes expected from the difference in GVWs. The
difference in average ridership would eliminate 2/3rds of the Indian bus weight
advantage.
The presence of air condition/heating systems on U.S. buses may be another
significant influence on fuel economy. The air conditioning impact can come via
two methods: (1) the air/heating system can add 1,000-2,000 lbs to the vehicle
weight and (2) power drawn for operating the air/heating system directly reduces
fuel economy. According to a study by the National Renewable Energy Laboratory
64


(NREL) (Rugh, Hovland et al. 2004), fuel
economy decreases by 18% in an
automobile and 14% in a truck while the
air conditioning is running. Specific bus
numbers are not readily available but
similar drops in the range of 10-20%
Figure 2.15 Bus Air Conditioning Unit
would be in line with these estimates. The
reduction from A/C power draw could equal 0.2-0.4 km/I with an additional decrease
of 0.4-0.8 km/1 from the increased vehicle weight due to the system. Indian transit
buses would be expected to suffer similar fuel economy losses as electric doors and
air conditioning systems are added to improve local bus service quality (The
Economic Times 2006).
2.6.2.3 Environmental Factors
Fuel economy may also be influenced by environmental factors. Denver sits at an
altitude of approximately one mile high while Chennai is at sea level. Increased
altitude decreases drag and requires less fuel to achieve a proper fuel/air mixture in
the engine, but results in a decrease in engine power. Fuel economy impacts are
determined based on whether the engine is tuned for high altitude operation
(Goodyear 2007). Also, Chennai is a significantly warmer city than Denver with an
65


average annual temperature of 82 F vs. 50 F for Denver. In general, fuel economy
increases with rises in ambient air temperature as both tire resistance and
aerodynamic drag decrease. These fuel economy advantages are slightly offset by
less efficient engine operation at higher temperatures, but the net trend is for
increased fuel economy with increased air temperature (Goodyear 2007). It is
unclear to what extent environmental conditions account for the difference in fuel
economy between Denver and Chennai. Denvers buses compare favorably with bus
fleets in the U.S. city fleets listed in Table 3.11 (1.88 km/1 vs 1.6 km/1). Therefore,
environmental conditions are unlikely to account for the full difference in fuel
economy between Indian and U.S. buses.
2.6.2.4 Drive Cycle Impacts on Fuel Economy
The average fuel economy of Denver buses is greater than the traditional bus drive
cycles tested in the U.S. (central business district, NY bus, and Manhattan)
suggesting that there may be limited opportunities for improving the fuel economy of
Denver buses through drive cycle improvements. As congestion in the metro area
increases, Denver bus fuel economies will likely decrease, thereby increasing W2W
GHG emissions.
66


The W2W GHG emissions released for the baseline RTD bus were 1,406 mtC02e.
The influence of Denvers drive cycle becoming more congested can be seen in
Table 2.19 with life cycle emissions potentially increasing 2-3 times if Denvers
drive cycle becomes more like that found in New York City.
Table 2.19 Potential impact of drive cycle changes on W2W GHG emissions
Drive Cycle Fuel Economy (km/l) Total W2W GHG Emissions Released (mtC02e)
Current RTD Bus 1.88 1406
U.S Bus CBD 1.7 1555
U.S Bus NY Bus 0.6 4407
U.S. Bus Manhattan 1.0 2698
2.6.3 Alternative Fuels Impact on GHG Emissions
Alternative fuels are often discussed as a way to reduce the environmental impacts of
mass transit systems. In the U.S., the primary alternative fuels for bus transportation
are biodiesel and compressed natural gas (CNG). Biodiesel is generally combined
with conventional diesel in a 20% by volume
blend (B20).
The alternative fuel analysis for buses was
conducted using GREET 1.7 and an
Figure 2.16 Biodiesel Bus
67


evaluation of fuel economy of alternative-fueled buses from several studies. These
studies found that the fuel economy of a typical bus on the central business district
drive cycle operating on CNG (15 studies) averages 23% less than the fuel economy
of a bus operating on conventional diesel (13 studies). Additionally, an NREL study
(Proc and Bamitt 2005) found that a Denver RTD bus operating in Boulder, CO
suffers a 3% decrease in fuel economy while operating on B20 as compared to
conventional diesel.
GREET was then used to calculate the expected upstream GHG emissions for
conventional diesel, CNG, and B20 buses. The upstream values were combined with
expected in-use emissions to determine the potential for W2W GHG emission
reductions from fuel switching. The calculations show that switching to B20 will
reduce GHG emissions by 10% from the Denver RTD base case while a switch to
CNG only improves emissions by 1%. In comparison, the Indian bus that also runs
on conventional diesel but has a significantly higher fuel economy has W2W GHG
emissions that are 54% lower than the RTD base case. The potential changes in
W2W GHG emissions from fuel switching are summarized in Table 2.20.
Table 2.20 Wells-to-wheels GHG reductions for fuel switching (mtCChe)
Bus Operation Upstream Total Reduction from RTD Base Case
RTD Base Case 1141 265 1406 0%
B20 1146 118 1264 10%
CNG 1048 343 1391 1%
68


While GHG emission benefits from fuel switching may be small, much larger
benefits can be realized through reductions in criteria air pollutants such as CO,
NOx, SO2, and PM leading to significant human health and smog reduction benefits.
2.6.4 Transferability of GaBi 4 to Developing Country LCAs
While the GaBi 4 estimate of bus manufacturing matched closely with U.S.
estimates, the predicted GHG values for the Indian bus were 67% less than the
values calculated in this study as shown in Table 2.18. This discrepancy suggests
that life cycle inventory databases created for developed countries may not be
directly applicable to developing countries without the addition of process specific
conversion factors. In India, the steel and aluminum production sectors have GHG
intensities that are higher than counterpart sectors in the U.S. and worldwide. This
difference may be the leading cause in the GHG emissions calculated for the
manufacture of the India bus being greater than the GaBi 4 estimates. However, as a
reminder, the bus material data in this study was subject to the least availability and
greatest degree of uncertainty. Therefore, the further analysis of if, and in what
capacity, developed country life cycle inventory databases can be applied to
developing country processes would be a much needed area of future research.
69


2.7 Conclusions
The following conclusions can be drawn from this study:
1) Resource use impacts, particularly water, are of the greatest concern to the
citizens of Tamil Nadu, India with global impacts and local health impacts
falling in the next tier. Citizens are both aware of and concerned with
impacts that may occur as a result of global climate change.
2) Bus assembly accounts for less than 0.5% of the total life cycle GHG impacts
for a transit bus.
3) Bus material manufacturing accounts for about 10% of life cycle GHG in
India compared with 3% in the U.S. The higher Indian percentage is a result
of more efficient buses reducing the wells-to-wheels GHG impacts of Indian
buses and higher GHG intensities for the production of primary materials in
the U.S. compared to India.
4) India specific analysis of diesel processing appeared to yield a lower GHG
intensity in India than in the U.S. Using the higher GHG intensity value from
the U.S. for Indian calculations would have increased life cycle GHG
emission for the Indian bus by less than 5%.
70


5) Chennai buses have fuel economies that are more than double the fuel
economy of their Denver counterparts. The fuel economy difference is likely
due to Indian buses being lighter weight and running without air
conditioning. As a result, even with a higher contribution from material
manufacturing, Indian buses release about 50% fewer greenhouse gases than
U.S. buses per v-km.
6) Bus ridership in Chennai is more than 4 times greater than in the U.S. As a
result, bus GHG emissions per p-km are almost 90% less in India than in the
U.S.
7) Switching to alternative fuels will lead to small W2W GHG reductions of 1%
for CNG and 10% for B20. However, emission reductions of other criteria
pollutants associated with significant health impacts are likely to be much
greater.
8) More analysis is needed to determine if life cycle inventory databases
modeled on developed country processes can be applied to developing
country inventories. This work indicates that established life cycle
inventories are not transferable to developing countries without considering
process differences.
71


3. A Comparison of Greenhouse Gas Emissions and Energy Use for Electrified
Urban Rail and Buses in India and the U.S.
3.1 Objectives
Chapter 2 establishes a baseline for life cycle energy use and greenhouse gas
emissions for bus transport in India and the U.S. Moreover, it examines the impacts
of major material production and evaluates the potential impacts of strategies, such
as fuel-switching, that are designed to reduce the environmental impacts of mass
transit systems. However, city planners face decisions beyond what fuels to use in
buses or how to expand the current bus system to promote additional ridership.
Planners are often faced with the fundamental decision of whether to allocate limited
public transportation funding to the future expansion and development of current bus
systems or towards the introduction or capacity addition of electrified urban rail.
Several cities have invested vast sums of money over the past decade on electrified
urban rail or are considering near term rail expansion projects including Denver,
Austin, Baltimore, Boston, Dallas, Houston, Los Angeles, Portland, St Louis, Seattle,
and Mexico City (Light Rail Central 2007) Additionally, numerous cities have also
promoted bus rapid transit as a means of increasing mass transit ridership and
decreasing GHG emissions including Boston, Seattle, Miami, Pittsburgh, Los
72


Angeles, Brisbane, Sydney, Leeds, Bogota, Curitiba, Quito and Sao Paulo among
others (Transportation Research Board 2007). Chapter 3 provides for a comparison
of the life cycle energy use and greenhouse gas emissions for bus and electrified
urban rail systems as a starting point for city planners attempting to make this
decision.
The objectives of Chapter 3 are:
1) To compare the anticipated operational energy use and greenhouse emissions
for electrified urban rail and bus systems in India and the U.S including the
impact of ridership.
2) To evaluate the potential operating costs of each system based on expected
fuel consumption
3) To use a broad stroke analysis to explore the relative contribution of vehicle
assembly and track construction on the life cycle impacts of electrified urban
rail systems.
4) To outline the relative energy, operational cost, and greenhouse gas (GHG)
emission impacts of each city for easy comparison.
73


3.2 Raw Data & Sources
The electrified urban rail data for this chapter focuses on two systems: (1) the
Chennai electrified urban rail Mass Rapid Transit System (MRTS) and (2) the RTD
Denver Light Rail system (DLRS). Descriptions of both systems and details on data
sources are contained in the following sections. Table 3.3 at the end of the section
summarizes the key raw data along with the data sources for both India and the U.S.
Table 3.1 gives a brief summary of the key electrified urban rail vehicle dimensions
in each country.
Table 3.1 Electrified urban rail vehicle dimensions for Chennai, India and Denver,
USA
Vehicle Dimensions Units India U.S.
Weight mt 124 44
Length meters 18.2 24.5
Height meters 3.8 2.7
Width meters 3.7 3.8
Source: Indian Railways Fan Club 2007 and Regional Transportation District Denver 2006
74


3.2.1 Chennai Mass Rapid Transit System
3.2.1.1 System Description
The Chennai Mass Rapid Transit System is an
electrified urban rail system designed to serve
commuters within the city of Chennai, India.
Phase I of the four phase system runs for 8.6 km
from Chennai Beach to Tirumailai, primarily
along the Buckingham Canal near the
Coromandel Coast on the southeastern coast of India. The trains run on double-
track, broad gauge lines. The trains primarily consist of three linked Electric
Multiple Units (called a three car rake) manufactured at the Integral Coach Factory.
The MRTS system is designed to handle nine car rakes, but low ridership demand
has resulted in a decision to use smaller rakes that run more frequently (Indian
Railways Fan Club 2007).
The MRTS system has 8 stations and is partly elevated (5.8 km out of 8.6 km). The
system runs approximately 100 trains per day with a maximum capacity of 600,000
passengers. However, only an approximately 9,000 passengers ride the trains each
day. Some reasons cited for the low ridership include parallel bus lines that are less
Figure 3.1 Chennai MRTS EMU from
www.thehindubusinessline.com
75


expensive and poor locating in a lower density area of
town (Mitric and Chatterton 2005). The ridership
equates to approximately 15 p-km / v-km which is 5
times lower than the Chennai suburban rail system
average of 75 p-km / v-km. Over the course of two
months, the author rode the full length of the MRTS
system from Tirumailai to Chennai Beach as part of
the daily work commute.
3.2.1.2 Data Sources
The Chennai MRTS is a well studied and analyzed electrified urban rail system by
both fans of Indian railways and by official agencies and researchers. An excellent
starting point for the study of any Indian rail system is the Indian Railways Fan Club
(www.irfca.oru). The site provides a comprehensive overview along with technical
details for rail systems throughout India, including the MRTS. The site also contains
more detailed information on the Electric Multiple Units (EMUs) that comprise the
MRTS trains. The MRTS EMUs are manufactured by the Integral Coach Factory
(www.icf.gov.in) whose website provided additional technical details on the
vehicles.



Figure 3.2 Chennai Suburban
and MRTS Rail System
76


Additional information on the construction and operation of the system was taken
from a critical report by the Comptroller and Auditor General of India (2006).
Moreover, ridership and system performance data was taken from a report written by
Mitric and Chatterton (2005) for the World Bank that included a comprehensive,
detailed view of urban transport in Chennai, including the MRTS. Finally, vehicle
operation energy use, while not available specifically for the Chennai MRTS system,
was calculated using numbers from the official Indian Railways 2004-2005 yearbook
(Indian Railways 2005) for the operation of energy demand of both EMUs as
individual vehicles and for the operation of electric locomotives on a 1,000 gross
tonne kilometer basis. The numbers corresponded well (within 10%) and the
average of the two values was used for subsequent calculations.
3.2.2 RTD Denver Light Rail System
3.2.2.1 System Description
The RTD Denver Light Rail System (DLRS) is
designed to serve the transit needs of the residents of
the Denver, CO metropolitan area. With the Figure 3.3 Denver Light Rail System
from www.rtd-denver.com
77


recent completion of the TREX light rail expansion in the southeast corridor, the
DLRS currently operates 36 stations through 3 major transit corridors with
approximately 35 miles of track. The DLRS runs with a vehicle fleet of 104 light
rail vehicles, primarily SD-160s manufactured by Siemens. The DLRS system
averages 34,273 boardings per weekday and over 10 million boardings per year
(Regional Transportation District Denver 2007).
3.2.2.2 Data Sources
For the Denver RTD Light Rail system, an abundance of documents are available
through the general RTD website (www .rtd-denver.com). The most important portal
for light rail specific information and general system descriptions is located at
http://www.rtd-denver.com/LightRail/index.html. From this portal, one can link to
comprehensive design documents that detail the specifications for both light rail
vehicles and the light rail track that were used to calculate values in this study.
DLRS ridership data was gathered from the same National Transit Database (2007)
system that was used to obtain Denver bus ridership figures. While energy use for
the operation of Denver light rail vehicles is not publicly available, the data used in
the study was acquired through personal communication with John Shonsey at RTD
Denver who back calculated the value from total light rail electricity use and
aggregate light rail vehicle kilometers traveled. The DLRS energy use estimates are
78


in line with estimates from the Calgary light rail system and are lower than the
national system average.
Table 3.2 Light rail energy use comparison for Denver
Light Rail Energy Use
System kWh/veh-km
Denver 3
Calgary 3.2
National Average 5.1
The most important raw data for the calculation of electrified urban rail system
impacts in the U.S. and India are contained in Table 3.3 along with the original
source of the data. Calculation results and analysis are presented in Section 3.3
following Table 3.3.
3.3 Methodology
The calculations for this report focus not only on comparing electrified urban rail
systems in Chennai, India and Denver, USA but also on placing the impacts of these
systems in the context of the bus analysis from the previous chapter. This side-by-
side comparison is intended to allow city planners to make better informed choices
regarding the future development of mass transit systems to meet the needs of the
citizenry while also accounting for environmental impacts. For the electrified urban
rail systems, three primary life cycle stages were examined: (1) track construction,
(2) vehicle manufacturing, (3) vehicle operation.
79


Table 3.3 Raw data summary with sources for analysis of electrified urban rail in
India & U.S.
Data Category Units Chennai India India Source Denver U.S.A. U.S. Source .
Electrified urban rail Ridership p-km / v-km 15.0 Mitric and Chatterton (2005) World Bank Report 12.7 U.S. National Transit Database
Bus Ridership p-km / v-km 38.8 Chennai Metropolitan Transport Corporation Website 9.4 U.S. National Transit Database
Electrified urban rail Vehicle Weight mt 124 Indian Railway Fan Club Website EMU Page 44 RTD Denver Light Rail Vehicle Specification Guide
Track Steel Weight mt 240 Comptroller and Auditor General Of India MRTS Report 114 RTD Denver Light Rail Track Specification Guide
Electrified urban rail Operation Energy Use kWh/ km 6 Indian Railways Yearbook 2004-2005 3 Personal Communication with RTD Denver
Electricity Emission Factor kgC02e / kWh 0.98 See Table 2.14 0.96 U.S. Energy Information Administration Colorado Emission Factors and T&D Loss
Diesel Emission Factor kgC02e / kWh 0.32 Calculated from Table 2.15 0.32 Calculated from Table 2.15
Assumed LifetimeA km 800k Assumption from Section 2.3 800k Assumption from Section 2.3
Steel Production Emission Factor mtC02e / mt steel 3.1 See Table 2.15 2.6 See Table 2.15
GaBi 4 Electrified urban rail Manufacturing Based on Country- specific Weight mtC02e 354 Gabi 4 Ecoinvent 1.2 Database 126 Gabi 4 Ecoinvent 1.2 Database
Electrified urban rail Manufacturing Estimate from Steel mtC02e 384 Calculated from values in this table 114 Calculated from values in this table
A Assumed similar life time to buses for ease of comparison
B Assumed entire weight of vehicle is steel for first cut calculation
80