OPEN ACCESS

Environmental assessment of passengertransportation should include infrastructure andsupply chainsTo cite this article Mikhail V Chester and Arpad Horvath 2009 Environ Res Lett 4 024008

View the article online for updates and enhancements

You may also likeEnergy and air emission implications of adecentralized wastewater systemArman Shehabi Jennifer R Stokes andArpad Horvath

-

Life cycle assessment of energy and CO2emissions for residential buildings inJakarta IndonesiaU Surahman T Kubota and A Wijaya

-

Fuel Cycle Environmental Assessment forElectric Vehicles in ChinaShuhua Li Qinxing Cao Jun Li et al

-

Recent citationsEnvironmental impact analysis for theconstruction of subway stationsComparison between open-excavationand underground-excavation schemeMinghui Liu et al

-

A comparative life cycle assessment ofelectric compressed natural gas anddiesel buses in ThailandNatalie R Gabriel et al

-

Developing prescriptive maintenancestrategies in the aviation industry based ona discrete-event simulation framework forpost-prognostics decision makingRobert Meissner et al

-

This content was downloaded from IP address 6521228167 on 07112021 at 1729

IOP PUBLISHING ENVIRONMENTAL RESEARCH LETTERS

Environ Res Lett 4 (2009) 024008 (8pp) doi1010881748-932642024008

Environmental assessment of passengertransportation should includeinfrastructure and supply chainsMikhail V Chester1 and Arpad Horvath

Department of Civil and Environmental Engineering University of California 760 Davis HallBerkeley CA 94720 USA

E-mail mchestercalberkeleyedu and horvathceberkeleyedu

Received 6 January 2009Accepted for publication 5 May 2009Published 8 June 2009Online at stacksioporgERL4024008

AbstractTo appropriately mitigate environmental impacts from transportation it is necessary fordecision makers to consider the life-cycle energy use and emissions Most currentdecision-making relies on analysis at the tailpipe ignoring vehicle production infrastructureprovision and fuel production required for support We present results of a comprehensivelife-cycle energy greenhouse gas emissions and selected criteria air pollutant emissionsinventory for automobiles buses trains and airplanes in the US including vehiclesinfrastructure fuel production and supply chains We find that total life-cycle energy inputs andgreenhouse gas emissions contribute an additional 63 for onroad 155 for rail and 31 forair systems over vehicle tailpipe operation Inventorying criteria air pollutants shows thatvehicle non-operational components often dominate total emissions Life-cycle criteria airpollutant emissions are between 11 and 800 times larger than vehicle operation Ranges inpassenger occupancy can easily change the relative performance of modes

Keywords passenger transportation life-cycle assessment cars autos buses trains railaircraft planes energy fuel emissions greenhouse gas criteria air pollutants

S Supplementary data are available from stacksioporgERL4024008

1 Background

Passenger transportationrsquos energy requirements and emissionsare receiving more and more scrutiny as concern for energysecurity global warming and human health impacts growsPassenger transportation is responsible for 20 of US energyconsumption (approximately 5 of global consumption) andcombustion emissions are strongly positively correlated [1]The potentially massive impacts of securing petroleumresources climate change human health and equity issuesassociated with transportation emissions have accelerateddiscussions about transportation environmental policy

Governmental policy has historically relied on energy andemission analysis of automobiles buses trains and aircraft attheir tailpipe ignoring vehicle production and maintenance

1 Author to whom any correspondence should be addressed

infrastructure provision and fuel production requirements tosupport these modes Such is the case with CAFE and aircraftemission standards which target vehicle operation only [2 3]Recently decision-making bodies have started to look to life-cycle assessments (LCA) for critical inputs typically relatedto transportation fuels [4 5] In order to effectively mitigateenvironmental impacts from transportation modes life-cycleenvironmental performance should be considered includingboth the direct and indirect processes and services requiredto operate the vehicle This includes raw materials extractionmanufacturing construction operation maintenance and endof life of vehicles infrastructure and fuels Decisions shouldnot be made based on partial data acting as indicators for wholesystem performance

To date a comprehensive LCA of passenger transportationin the US has not been completed Several studies and

1748-932609024008+08$3000 copy 2009 IOP Publishing Ltd Printed in the UK1

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

models analyze a single mode particular externalities orspecific phases but none have performed a complete LCAof multiple modes including vehicle infrastructure and fuelinventories for energy consumption greenhouse gas emissionsand criteria air pollutant emissions incorporating supplychains [6ndash9] The automobile has received the greatestattention while buses rail and air have received little focusA review of environmental literature related to the three modalcategories is shown in table S1 of the supporting information(SI) (available at stacksioporgERL4024008)

2 Methodology

Onroad rail and air travel are inventoried to determine energyconsumption greenhouse gas (GHG) emissions and criteriaair pollutant (CAP) emissions (excluding PM lead and ozonedue to lack of data) The onroad systems include threeautomobiles and two urban buses (off-peak and peak) A sedan(2005 Toyota Camry) SUV (2005 Chevrolet Trailblazer)and pickup (2005 Ford F-150) are chosen to represent therange in the US automobile fleet and critical performancecharacteristics [10ndash12] 83 of rail passenger kilometersare performed by metropolitan systems (with Amtrak servingthe remaining) [1] The generalized rail modes (heavyrail electric metro heavy rail diesel commuter transit andlight rail transit (LRT)) are chosen to capture the gamutof physical size fuel input and service niche The metroand commuter rail are modeled after the San Francisco BayArearsquos (SFBA) Bay Area Rapid Transit and Caltrain whilethe LRT modes are modeled after San Franciscorsquos (SF)Muni Metro and the Boston Green Line Air modes areevaluated by small (Embraer 145) midsize (Boeing 737) andlarge (Boeing 747) aircraft to represent the range of impactsfrom aircraft sizes passenger occupancy and short to longhaul segment performance [13] An extended discussionof the characteristics and representativeness of the modesselected is found in the SI US average data are used for allonroad and air mode components and particular geographicoperating conditions are not captured [14 15] Rail operationalperformance is determined from specific systems [15ndash18]

A hybrid LCA model was employed for this analysis [19]The use of this LCA approach is discussed in the SI anddetailed extensively in [20] The life-cycle phases includedare shown in table 1 The components are evaluated from thematerials extraction through the use phase including supplychains For example the manufacturing of an automobileincludes the energy and emissions from extraction of rawmaterials such as iron ore for steel through the assembly of thatsteel in the vehicle End-of-life phases are not included dueto the complexities of evaluating waste management optionsand material reuse Indirect impacts are included ie theenergy and emissions resulting from the support infrastructureof a process or product such as electricity generation forautomobile manufacturing

For each component in the modersquos life cycle environ-mental performance is calculated and then normalized perpassenger-kilometer-traveled (PKT) The energy inputs andemissions from that component may have occurred annually(such as from electricity generation for train propulsion) or

over the componentrsquos lifetime (such as train station construc-tion) and are normalized appropriately Detailed analyses anddata used for normalization are found in [20] including mode-specific adjustments (such as the removal of freight and mailattributions from passenger air travel) Equation (1) providesthe generalized formula for determining component energy oremissions

EM =Csum

c

EFMc times UMc(t)

PKTM(t)(1)

where EM is total energy or emissions per PKT formode M M is the set of modes sedan train aircraft etcc is vehicle infrastructure or fuel life-cycle componentEF is environmental (energy or emission) factor forcomponent cU is activity resulting in EF for component cPKT is PKT performed by mode M during time t forcomponent c

The fundamental environmental factors used for deter-mining a componentrsquos energy and emissions come from avariety of sources They are detailed in SI tables S2ndashS4(available at stacksioporgERL4024008) Further eachcomponentrsquos modeling details are discussed in [20] whichprovides the specific mathematical framework used as well asextensive documentation of data sources and other parameters(such as component lifetimes and mode vehicle and passengerkilometers traveled) Parameter uncertainty is also evaluated inthe SI

Results for modal average occupancy per-PKT perfor-mance are reported While understanding of marginal perfor-mance is necessary for transportation planners to evaluate theadditional cost of a PKT given a vested infrastructure and theassumption that many public transit trips will occur regardlessthe average performance characteristics allow for the totalenvironmental inventorying of a system over its lifetime

3 Results and component comparisons

With 79 components evaluated across the modes the groupingsin table 1 are used to report and discuss inventory results

31 Energy

The energy inputs for the different systems range from directfossil fuel use such as gasoline diesel and jet fuel to indirectfossil fuel use in electricity generation The non-operationalvehicle phases use a combination of energy inputs for directand indirect requirements For example the construction ofan airport runway requires direct energy to transport and placethe concrete and indirect energy to extract and process the rawmaterials Figure 1 shows total energy inputs for each mode

While tailpipe components account for a large portionof modal life-cycle energy consumption auto and bus non-operational components have non-negligible results Activeoperation accounts for 65ndash74 of onroad 24ndash39 of railand 69ndash79 of air travel life-cycle energy Inactive operationaccounts for 3 of bus 7ndash21 of rail and 2ndash14 of air

2

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

Table 1 Analysis components (for each component energy inputs and emissions are determined The components are shown by generalizedmode but evaluated independently for each system)

Grouping Automobiles and buses Rail Air

Vehicles

Operational components

Active operation bull Runningbull Cold start

bull Running bull Take offbull Climb outbull Cruisebull Approachbull Landing

Inactive operation bull Idling bull Idlingbull Auxiliaries (HVAC and lighting)

bull Auxiliary power unit operationbull Startupbull Taxi outbull Taxi in

Non-operational components

Manufacturing (facilityconstruction excluded)

bull Vehicle manufacturingbull Engine manufacturing

bull Train manufacturingbull Propulsion systemmanufacturing

bull Aircraft manufacturingbull Engine manufacturing

Maintenance bull Vehicle maintenancebull Tire replacement

bull Train maintenancebull Train cleaningbull Flooring replacement

bull Aircraft maintenancebull Engine maintenance

Insurance bull Vehicle liability bull Crew health and benefitsbull Train liability

bull Crew health and benefitsbull Aircraft liability

Infrastructure

Construction bull Roadway construction bull Station constructionbull Track construction

bull Airport constructionbull Runwaytaxiwaytarmacconstruction

Operation bull Roadway lightingbull Herbicide sprayingbull Roadway salting

bull Station lightingbull Escalatorsbull Train controlbull Station parking lightingbull Station miscellaneous(eg other electrical equipment)

bull Runway lightingbull Deicing fluid productionbull Ground support equipmentoperation

Maintenance bull Roadway maintenance bull Station maintenancebull Station cleaning

bull Airport maintenance

Parking bull Roadside surface lot andparking garage parking

bull Station parking bull Airport parking

Insurance bull Non-crew health insurance andbenefitsbull Infrastructure liability insurance

bull Non-crew health and benefitsbull Infrastructure liability

Fuels

Production bull Gasoline and diesel fuelrefining and distribution (includesthrough fuel truck deliverystopping at fuel station Servicestation construction andoperation is excluded)

bull Train electricity generationbull Train diesel fuel refining anddistribution (Caltrain)bull Train electricity transmission anddistribution lossesbull Infrastructure electricityproductionbull Infrastructure electricitytransmission and distribution losses

bull Jet fuel refining and distribution

modes The automobile and bus non-operational componentsare dominated by electricity production steel production andtruck and air transport of materials in vehicle manufacturingand maintenance [20] The construction of the US roadand highway infrastructure has large energy implications (inmaterial extraction material production and constructionoperations) between 03 and 04 MJPKT for autos [21ndash23]

Rail modes have the smallest fraction of operational tototal energy due to their low electricity requirements per

PKT relative to their large supporting infrastructures [20]The construction and operation of rail mode infrastructureresults in total energy requirements about twice that ofoperational

Aircraft have the largest operational to total life-cycleenergy ratios due to their large fuel requirements per PKTand relatively small infrastructure The active and inactiveoperational groupings include several components (table 1) andenergy consumption is dominated by the cruise phase [24 25]

3

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

Figure 1 Energy consumption and GHG emissions per PKT (The vehicle operation components are shown with gray patterns Other vehiclecomponents are shown in shades of blue Infrastructure components are shown in shades of red and orange The fuel production component isshown in green All components appear in the order they are shown in the legend)

32 Greenhouse gases

The energy inputs described are heavily dominated by fossilfuels resulting in a strong positive correlation with GHGemissions The life-cycle component contributions are roughlythe same as the GHG contributions and produce 14ndash16 timeslarger life-cycle factors for onroad 18ndash25 times for rail and12ndash13 times for air than the operational components Totalemissions for each mode are shown in figure 1

While the energy input to GHG emissions correlationholds for almost all modes there is a more pronounced effectbetween the California (CA) and Massachusetts (MA) LRTsystems The San Francisco Bay Arearsquos electricity is 49fossil fuel-based and Massachusettsrsquos is 82 [26 27] Theresult is that the Massachusetts LRT which is the lowestoperational energy user and roughly equivalent in life-cycleenergy use to the other rail modes is the largest GHG emitter

33 Criteria air pollutants

Figure 2 shows SO2 NOX and CO emissions for eachlife-cycle component The inclusion of non-operationalcomponents can lead to an order of magnitude larger emissionfactor for total emissions relative to operational emissions

331 SO2 contributors Electricity generation SO2

emissions dominate life-cycle component contributions for allmodes While electric rail modes have large contributionsfrom vehicle operation components this is not the case forautos buses and commuter rail due to the removal of sulfurfrom gasoline and diesel fuels Low sulfur levels in fuelsresult in low SO2 emissions from fuel combustion compared tothe relatively large SO2 emissions from electricity generationin other components Total automobile SO2 emissions are19ndash26 times larger than operational emissions and are due tovehicle manufacturing and maintenance roadway constructionand operation (particularly lighting) parking construction andgasoline production The electricity requirements in vehiclemanufacturing vehicle maintenance roadway lighting roadmaterial production and fuel production (as well as off-gasing)result in significant SO2 contributions [20 21 26 28] Busemissions are dominated by vehicle manufacturing roadwaymaintenance [21] and fuel production Vehicle manufacturinginfrastructure construction infrastructure operation parkinginsurance and fuel production produce emission factorsfor rail modes that are 2ndash800 times (assuming Tier 2standards) larger than operational components The majority ofvehicle manufacturing emissions result from direct electricity

4

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

Figure 2 Criteria air pollutant emissions in mg per PKT (The vehicle operation components are shown with gray patterns Other vehiclecomponents are shown in shades of blue Infrastructure components are shown in shades of red and orange The fuel production component isshown in green All components appear in the order they are shown in the legend)

requirements in assembling the parts as well as the energyrequirements to produce steel and aluminum for trainsTotal aircraft SO2 emissions are composed of 64ndash71 non-operational emissions and are attributed mostly to the directelectricity requirements in aircraft manufacturing and indirectelectricity requirements in the extraction and refinement ofcopper and aluminum [20]

332 NOX contributors Life-cycle NOX emissions areoften dominated by tailpipe components however autos andelectric rail modes show non-negligible contributions fromother components Non-operational NOX emissions are dueto several common components from the supply chains ofall the modes direct electricity use indirect electricity usefor material production and processes and truck and railtransportation With onroad modes electricity requirementsfor vehicle manufacturing and maintenance as well as truckand rail material transport are large contributors [20] The

transport of materials for asphalt surfaces is the primary culpritin roadway and parking construction [21] Fuel refineryelectricity and diesel equipment use in oil extraction add tothe componentrsquos contribution to total emissions [20] Forrail the dependence on concrete in infrastructure (resulting inlarge electricity requirements for cement manufacturing anddiesel equipment use in placement) impacts the contributionfrom construction and maintenance increasing total NOX

emissions by 24ndash12 times for the electric modes and 11times for commuter rail Aircraft manufacturing infrastructureoperation and fuel production produce emissions from aircraftthat are 12 times larger than operational emissions The directelectricity requirements and truck and rail transport are the keycomponents in aircraft manufacturing

333 CO contributors While automobile CO emissionsare dominated by the vehicle operation phase this is not thecase for bus rail and air modes Automobile CO emissions

5

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

are approximately 110 and 40 times larger per PKT thanrail and aircraft respectively due to a roughly equivalentper vehicle-kilometers-traveled (VKT) emission factor butvastly different occupancy rates The largest non-operationalcomponent is vehicle manufacturing which accounts for about3 and 28 of total automobile and bus emissions due mainlyto truck transport of materials and parts The productionof cement for concrete in stations and truck transport ofsupplies for insurance operations are the underlying non-operational causes for rail CO emissions Large concreterequirements result in large CO emissions during cementproduction for station construction and maintenance [20]Rail infrastructure emissions (140ndash260 mgPKT) are 42ndash76 of life-cycle emissions (270ndash430 mgPKT) Trucktransport in aircraft manufacturing airport ground supportequipment (GSE) operation and jet fuel production producelife-cycle emissions that are 26ndash85 times larger than operation(30ndash180 mgPKT) [24 25] The use of diesel trucks tomove parts and materials needed for aircraft manufacturingcontributes strongly to the component (20ndash90 mgPKT) [20]The emissions from airport operation are dominated by GSEoperations Particularly the use of gasoline baggage tractorscontributes to roughly half of all GSE emissions [25 29]

4 Sensitivity to passenger occupancy

While the per-VKT performance of any mode can potentiallybe improved through technological advancements the per-PKT performance which captures the energy and emissionsintensity of moving passengers is the result of occupancyrates An evaluation of these occupancy rates with realistic lowand high ridership illustrates both the potential environmentalperformance of the mode as well as the passenger conditionswhen modes are equivalent

Figure 3 highlights these ranges showing average occu-pancy life-cycle performance and the ranges of performancefrom low and high ridership (low ridership captures the largestenergy consumption and emissions per PKT at the worstperforming times while high ridership captures the modersquosbest performance) Auto low occupancy is specified as onepassenger and the high as the number of seats Bus lowoccupancy is specified as five passengers and the high as60 passengers (including standing passengers) Rail lowoccupancy is specified as 25 of the number of seats andthe high as 110 of seats (to capture standing passengers)Aircraft low occupancy is 50 and the high is 100 of thenumber of seats The occupancy ranges are detailed in SI tableS5 (available at stacksioporgERL4024008) Discussion ofthe environmental performance of transit modes often focuseson the ranking of vehicles assuming average occupancy Thisapproach does not acknowledge that there are many conditionsunder which modes can perform equally For example anSUV (which is one of the worst energy performers) with 2passengers (giving 35 MJPKT) is equivalent to a bus with8 passengers Similarly CA HRT with 120 passengers (27occupancy giving 18 MJPKT) is equivalent to a midsizeaircraft with 105 passengers (75 occupancy) Similarlycommuter rail (with one of the highest average per-PKT

Figure 3 Occupancy sensitivity (Average occupancy and life-cycleperformance is shown as the blue (autos) purple (bus) red (trains)and green (aircraft) bars The maroon-colored line captures the rangein per-PKT energy consumption and emissions at low and highoccupancy)

NOX emission rates) at 34 occupancy (147 passengers) isequivalent to a bus with 13 passengers or a sedan with onepassenger Focusing on occupancy improvements does notacknowledge the sensitivity of performance to technologicalchanges For example holding occupancy at the averageelectric rail modes would have to decrease SO2 per-PKTemissions between 24 and 85 to compete with onroad modesan effort that would have to focus on electricity fuel inputs andscrubbers at power plants

6

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

Figure 1 Energy consumption and GHG emissions per PKT (The vehicle operation components are shown with gray patterns Other vehiclecomponents are shown in shades of blue Infrastructure components are shown in shades of red and orange The fuel production component isshown in green All components appear in the order they are shown in the legend)

32 Greenhouse gases

The energy inputs described are heavily dominated by fossilfuels resulting in a strong positive correlation with GHGemissions The life-cycle component contributions are roughlythe same as the GHG contributions and produce 14ndash16 timeslarger life-cycle factors for onroad 18ndash25 times for rail and12ndash13 times for air than the operational components Totalemissions for each mode are shown in figure 1

While the energy input to GHG emissions correlationholds for almost all modes there is a more pronounced effectbetween the California (CA) and Massachusetts (MA) LRTsystems The San Francisco Bay Arearsquos electricity is 49fossil fuel-based and Massachusettsrsquos is 82 [26 27] Theresult is that the Massachusetts LRT which is the lowestoperational energy user and roughly equivalent in life-cycleenergy use to the other rail modes is the largest GHG emitter

33 Criteria air pollutants

Figure 2 shows SO2 NOX and CO emissions for eachlife-cycle component The inclusion of non-operationalcomponents can lead to an order of magnitude larger emissionfactor for total emissions relative to operational emissions

331 SO2 contributors Electricity generation SO2

emissions dominate life-cycle component contributions for allmodes While electric rail modes have large contributionsfrom vehicle operation components this is not the case forautos buses and commuter rail due to the removal of sulfurfrom gasoline and diesel fuels Low sulfur levels in fuelsresult in low SO2 emissions from fuel combustion compared tothe relatively large SO2 emissions from electricity generationin other components Total automobile SO2 emissions are19ndash26 times larger than operational emissions and are due tovehicle manufacturing and maintenance roadway constructionand operation (particularly lighting) parking construction andgasoline production The electricity requirements in vehiclemanufacturing vehicle maintenance roadway lighting roadmaterial production and fuel production (as well as off-gasing)result in significant SO2 contributions [20 21 26 28] Busemissions are dominated by vehicle manufacturing roadwaymaintenance [21] and fuel production Vehicle manufacturinginfrastructure construction infrastructure operation parkinginsurance and fuel production produce emission factorsfor rail modes that are 2ndash800 times (assuming Tier 2standards) larger than operational components The majority ofvehicle manufacturing emissions result from direct electricity

4

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

Figure 2 Criteria air pollutant emissions in mg per PKT (The vehicle operation components are shown with gray patterns Other vehiclecomponents are shown in shades of blue Infrastructure components are shown in shades of red and orange The fuel production component isshown in green All components appear in the order they are shown in the legend)

requirements in assembling the parts as well as the energyrequirements to produce steel and aluminum for trainsTotal aircraft SO2 emissions are composed of 64ndash71 non-operational emissions and are attributed mostly to the directelectricity requirements in aircraft manufacturing and indirectelectricity requirements in the extraction and refinement ofcopper and aluminum [20]

332 NOX contributors Life-cycle NOX emissions areoften dominated by tailpipe components however autos andelectric rail modes show non-negligible contributions fromother components Non-operational NOX emissions are dueto several common components from the supply chains ofall the modes direct electricity use indirect electricity usefor material production and processes and truck and railtransportation With onroad modes electricity requirementsfor vehicle manufacturing and maintenance as well as truckand rail material transport are large contributors [20] The

transport of materials for asphalt surfaces is the primary culpritin roadway and parking construction [21] Fuel refineryelectricity and diesel equipment use in oil extraction add tothe componentrsquos contribution to total emissions [20] Forrail the dependence on concrete in infrastructure (resulting inlarge electricity requirements for cement manufacturing anddiesel equipment use in placement) impacts the contributionfrom construction and maintenance increasing total NOX

emissions by 24ndash12 times for the electric modes and 11times for commuter rail Aircraft manufacturing infrastructureoperation and fuel production produce emissions from aircraftthat are 12 times larger than operational emissions The directelectricity requirements and truck and rail transport are the keycomponents in aircraft manufacturing

333 CO contributors While automobile CO emissionsare dominated by the vehicle operation phase this is not thecase for bus rail and air modes Automobile CO emissions

5

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

are approximately 110 and 40 times larger per PKT thanrail and aircraft respectively due to a roughly equivalentper vehicle-kilometers-traveled (VKT) emission factor butvastly different occupancy rates The largest non-operationalcomponent is vehicle manufacturing which accounts for about3 and 28 of total automobile and bus emissions due mainlyto truck transport of materials and parts The productionof cement for concrete in stations and truck transport ofsupplies for insurance operations are the underlying non-operational causes for rail CO emissions Large concreterequirements result in large CO emissions during cementproduction for station construction and maintenance [20]Rail infrastructure emissions (140ndash260 mgPKT) are 42ndash76 of life-cycle emissions (270ndash430 mgPKT) Trucktransport in aircraft manufacturing airport ground supportequipment (GSE) operation and jet fuel production producelife-cycle emissions that are 26ndash85 times larger than operation(30ndash180 mgPKT) [24 25] The use of diesel trucks tomove parts and materials needed for aircraft manufacturingcontributes strongly to the component (20ndash90 mgPKT) [20]The emissions from airport operation are dominated by GSEoperations Particularly the use of gasoline baggage tractorscontributes to roughly half of all GSE emissions [25 29]

4 Sensitivity to passenger occupancy

While the per-VKT performance of any mode can potentiallybe improved through technological advancements the per-PKT performance which captures the energy and emissionsintensity of moving passengers is the result of occupancyrates An evaluation of these occupancy rates with realistic lowand high ridership illustrates both the potential environmentalperformance of the mode as well as the passenger conditionswhen modes are equivalent

Figure 3 highlights these ranges showing average occu-pancy life-cycle performance and the ranges of performancefrom low and high ridership (low ridership captures the largestenergy consumption and emissions per PKT at the worstperforming times while high ridership captures the modersquosbest performance) Auto low occupancy is specified as onepassenger and the high as the number of seats Bus lowoccupancy is specified as five passengers and the high as60 passengers (including standing passengers) Rail lowoccupancy is specified as 25 of the number of seats andthe high as 110 of seats (to capture standing passengers)Aircraft low occupancy is 50 and the high is 100 of thenumber of seats The occupancy ranges are detailed in SI tableS5 (available at stacksioporgERL4024008) Discussion ofthe environmental performance of transit modes often focuseson the ranking of vehicles assuming average occupancy Thisapproach does not acknowledge that there are many conditionsunder which modes can perform equally For example anSUV (which is one of the worst energy performers) with 2passengers (giving 35 MJPKT) is equivalent to a bus with8 passengers Similarly CA HRT with 120 passengers (27occupancy giving 18 MJPKT) is equivalent to a midsizeaircraft with 105 passengers (75 occupancy) Similarlycommuter rail (with one of the highest average per-PKT

Figure 3 Occupancy sensitivity (Average occupancy and life-cycleperformance is shown as the blue (autos) purple (bus) red (trains)and green (aircraft) bars The maroon-colored line captures the rangein per-PKT energy consumption and emissions at low and highoccupancy)

NOX emission rates) at 34 occupancy (147 passengers) isequivalent to a bus with 13 passengers or a sedan with onepassenger Focusing on occupancy improvements does notacknowledge the sensitivity of performance to technologicalchanges For example holding occupancy at the averageelectric rail modes would have to decrease SO2 per-PKTemissions between 24 and 85 to compete with onroad modesan effort that would have to focus on electricity fuel inputs andscrubbers at power plants

6

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

Figure 2 Criteria air pollutant emissions in mg per PKT (The vehicle operation components are shown with gray patterns Other vehiclecomponents are shown in shades of blue Infrastructure components are shown in shades of red and orange The fuel production component isshown in green All components appear in the order they are shown in the legend)

requirements in assembling the parts as well as the energyrequirements to produce steel and aluminum for trainsTotal aircraft SO2 emissions are composed of 64ndash71 non-operational emissions and are attributed mostly to the directelectricity requirements in aircraft manufacturing and indirectelectricity requirements in the extraction and refinement ofcopper and aluminum [20]

332 NOX contributors Life-cycle NOX emissions areoften dominated by tailpipe components however autos andelectric rail modes show non-negligible contributions fromother components Non-operational NOX emissions are dueto several common components from the supply chains ofall the modes direct electricity use indirect electricity usefor material production and processes and truck and railtransportation With onroad modes electricity requirementsfor vehicle manufacturing and maintenance as well as truckand rail material transport are large contributors [20] The

transport of materials for asphalt surfaces is the primary culpritin roadway and parking construction [21] Fuel refineryelectricity and diesel equipment use in oil extraction add tothe componentrsquos contribution to total emissions [20] Forrail the dependence on concrete in infrastructure (resulting inlarge electricity requirements for cement manufacturing anddiesel equipment use in placement) impacts the contributionfrom construction and maintenance increasing total NOX

emissions by 24ndash12 times for the electric modes and 11times for commuter rail Aircraft manufacturing infrastructureoperation and fuel production produce emissions from aircraftthat are 12 times larger than operational emissions The directelectricity requirements and truck and rail transport are the keycomponents in aircraft manufacturing

333 CO contributors While automobile CO emissionsare dominated by the vehicle operation phase this is not thecase for bus rail and air modes Automobile CO emissions

5

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

are approximately 110 and 40 times larger per PKT thanrail and aircraft respectively due to a roughly equivalentper vehicle-kilometers-traveled (VKT) emission factor butvastly different occupancy rates The largest non-operationalcomponent is vehicle manufacturing which accounts for about3 and 28 of total automobile and bus emissions due mainlyto truck transport of materials and parts The productionof cement for concrete in stations and truck transport ofsupplies for insurance operations are the underlying non-operational causes for rail CO emissions Large concreterequirements result in large CO emissions during cementproduction for station construction and maintenance [20]Rail infrastructure emissions (140ndash260 mgPKT) are 42ndash76 of life-cycle emissions (270ndash430 mgPKT) Trucktransport in aircraft manufacturing airport ground supportequipment (GSE) operation and jet fuel production producelife-cycle emissions that are 26ndash85 times larger than operation(30ndash180 mgPKT) [24 25] The use of diesel trucks tomove parts and materials needed for aircraft manufacturingcontributes strongly to the component (20ndash90 mgPKT) [20]The emissions from airport operation are dominated by GSEoperations Particularly the use of gasoline baggage tractorscontributes to roughly half of all GSE emissions [25 29]

4 Sensitivity to passenger occupancy

While the per-VKT performance of any mode can potentiallybe improved through technological advancements the per-PKT performance which captures the energy and emissionsintensity of moving passengers is the result of occupancyrates An evaluation of these occupancy rates with realistic lowand high ridership illustrates both the potential environmentalperformance of the mode as well as the passenger conditionswhen modes are equivalent

Figure 3 highlights these ranges showing average occu-pancy life-cycle performance and the ranges of performancefrom low and high ridership (low ridership captures the largestenergy consumption and emissions per PKT at the worstperforming times while high ridership captures the modersquosbest performance) Auto low occupancy is specified as onepassenger and the high as the number of seats Bus lowoccupancy is specified as five passengers and the high as60 passengers (including standing passengers) Rail lowoccupancy is specified as 25 of the number of seats andthe high as 110 of seats (to capture standing passengers)Aircraft low occupancy is 50 and the high is 100 of thenumber of seats The occupancy ranges are detailed in SI tableS5 (available at stacksioporgERL4024008) Discussion ofthe environmental performance of transit modes often focuseson the ranking of vehicles assuming average occupancy Thisapproach does not acknowledge that there are many conditionsunder which modes can perform equally For example anSUV (which is one of the worst energy performers) with 2passengers (giving 35 MJPKT) is equivalent to a bus with8 passengers Similarly CA HRT with 120 passengers (27occupancy giving 18 MJPKT) is equivalent to a midsizeaircraft with 105 passengers (75 occupancy) Similarlycommuter rail (with one of the highest average per-PKT

Figure 3 Occupancy sensitivity (Average occupancy and life-cycleperformance is shown as the blue (autos) purple (bus) red (trains)and green (aircraft) bars The maroon-colored line captures the rangein per-PKT energy consumption and emissions at low and highoccupancy)

NOX emission rates) at 34 occupancy (147 passengers) isequivalent to a bus with 13 passengers or a sedan with onepassenger Focusing on occupancy improvements does notacknowledge the sensitivity of performance to technologicalchanges For example holding occupancy at the averageelectric rail modes would have to decrease SO2 per-PKTemissions between 24 and 85 to compete with onroad modesan effort that would have to focus on electricity fuel inputs andscrubbers at power plants

6

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

are approximately 110 and 40 times larger per PKT thanrail and aircraft respectively due to a roughly equivalentper vehicle-kilometers-traveled (VKT) emission factor butvastly different occupancy rates The largest non-operationalcomponent is vehicle manufacturing which accounts for about3 and 28 of total automobile and bus emissions due mainlyto truck transport of materials and parts The productionof cement for concrete in stations and truck transport ofsupplies for insurance operations are the underlying non-operational causes for rail CO emissions Large concreterequirements result in large CO emissions during cementproduction for station construction and maintenance [20]Rail infrastructure emissions (140ndash260 mgPKT) are 42ndash76 of life-cycle emissions (270ndash430 mgPKT) Trucktransport in aircraft manufacturing airport ground supportequipment (GSE) operation and jet fuel production producelife-cycle emissions that are 26ndash85 times larger than operation(30ndash180 mgPKT) [24 25] The use of diesel trucks tomove parts and materials needed for aircraft manufacturingcontributes strongly to the component (20ndash90 mgPKT) [20]The emissions from airport operation are dominated by GSEoperations Particularly the use of gasoline baggage tractorscontributes to roughly half of all GSE emissions [25 29]

4 Sensitivity to passenger occupancy

While the per-VKT performance of any mode can potentiallybe improved through technological advancements the per-PKT performance which captures the energy and emissionsintensity of moving passengers is the result of occupancyrates An evaluation of these occupancy rates with realistic lowand high ridership illustrates both the potential environmentalperformance of the mode as well as the passenger conditionswhen modes are equivalent

Figure 3 highlights these ranges showing average occu-pancy life-cycle performance and the ranges of performancefrom low and high ridership (low ridership captures the largestenergy consumption and emissions per PKT at the worstperforming times while high ridership captures the modersquosbest performance) Auto low occupancy is specified as onepassenger and the high as the number of seats Bus lowoccupancy is specified as five passengers and the high as60 passengers (including standing passengers) Rail lowoccupancy is specified as 25 of the number of seats andthe high as 110 of seats (to capture standing passengers)Aircraft low occupancy is 50 and the high is 100 of thenumber of seats The occupancy ranges are detailed in SI tableS5 (available at stacksioporgERL4024008) Discussion ofthe environmental performance of transit modes often focuseson the ranking of vehicles assuming average occupancy Thisapproach does not acknowledge that there are many conditionsunder which modes can perform equally For example anSUV (which is one of the worst energy performers) with 2passengers (giving 35 MJPKT) is equivalent to a bus with8 passengers Similarly CA HRT with 120 passengers (27occupancy giving 18 MJPKT) is equivalent to a midsizeaircraft with 105 passengers (75 occupancy) Similarlycommuter rail (with one of the highest average per-PKT

Figure 3 Occupancy sensitivity (Average occupancy and life-cycleperformance is shown as the blue (autos) purple (bus) red (trains)and green (aircraft) bars The maroon-colored line captures the rangein per-PKT energy consumption and emissions at low and highoccupancy)

NOX emission rates) at 34 occupancy (147 passengers) isequivalent to a bus with 13 passengers or a sedan with onepassenger Focusing on occupancy improvements does notacknowledge the sensitivity of performance to technologicalchanges For example holding occupancy at the averageelectric rail modes would have to decrease SO2 per-PKTemissions between 24 and 85 to compete with onroad modesan effort that would have to focus on electricity fuel inputs andscrubbers at power plants

6

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

5 Appropriate emission reduction targets

The dominant contributions to energy consumption and GHGemissions for onroad and air modes are from operationalcomponents This suggests that technological advancementsto improve fuel economy and switches to lower fossil carbonfuels are the most effective for improving environmentalperformance Railrsquos energy consumption and GHG emissionsare more strongly influenced by non-operational componentsthan onroad and air While energy efficiency improvementsare still warranted coupled with lower fossil carbon fuelsin electricity generation reductions in station constructionenergy use and infrastructure operation could have notableeffects Particularly the reduction in concrete use orswitching to lower energy input and GHG-intensity materialswould improve infrastructure construction performance whilereduced electricity consumption and cleaner fuels forelectricity generation would improve infrastructure operationUtilizing higher percentages of electricity from hydro and otherrenewable sources for rail operations could result in significantGHG reductions over fossil-based inputs such as coal

The life-cycle non-operational components are sometimesresponsible for the majority of CAP emissions so reductiongoals should consider non-operational processes SO2

emissions for all modes are heavily influenced by director indirect electricity use Similarly significant NOX

emission reductions can be achieved through cleaner electricitygeneration but also the reduction of diesel equipmentemissions in transport and material extraction operationsThe reductions could be achieved by decreased or cleanerelectricity consumption using equipment with cleaner fuelinputs or through the implementation of improved emissionscontrols While automobile CO emissions are mainly fromactive operation (with a large portion attributed to the cold startphase) rail emission reductions are best achieved by reducingthe use of concrete in stations A switch away from dieselor gasoline equipment or stronger emission controls can havestrong implications for aircraft total CO emissions in trucktransport and GSE operations

This study focuses on conventional gasoline automobilesand it is important to consider the effects of biofuels andother non-conventional energy inputs on life-cycle resultsLCAs of biofuels are starting to be developed and willprovide the environmental assessments necessary for adjustingprimarily the lsquofuel productionrsquo component of this LCAInputs such as electricity for plugin hybrid electric vehiclescould also significantly change several components in thisstudy Batteries in vehicle manufacturing differing operationalcharacteristics and electricity production (especially wind andsolar) are just some of the components that would affect theresults presented here This study creates a framework forcomprehensive environmental inventorying of several modesand future assessment of non-conventional fuels and vehiclescan follow this methodology in creating technology-specificresults

Future work should also focus on environmental effectsnot quantified herein such as the use of water [30] generationof waste water and toxic emissions [31] Detailed assessments

of the end-of-life fate of vehicles [32] motor oil [33] andinfrastructure [34] should also be factored into decisions

Through the use of life-cycle environmental assessmentsenergy and emission reduction decision-making can benefitfrom the identified interdependencies among processesservices and products The use of comprehensive strategiesthat acknowledge these connections are likely to have a greaterimpact than strategies that target individual components

Acknowledgments

This material is based upon work supported by the UCBerkeley Center for Future Urban Transport and theUniversity of California Transportation Center (by a 2005grant)

References

[1] Davis S C and Diegel S W 2007 Transportation Energy DataBook US Department of Energy Office of Energy Efficiencyand Renewable Energy Oak Ridge National Laboratory

[2] National Research Council 2002 Effectiveness and Impact ofCorporate Average Fuel Economy (CAFE) Standards(Washington DC National Academies Press)

[3] US Environmental Protection Agency 2005 Emission Standardsand Test Procedures for Aircraft and Aircraft EnginesEPA420-R-05-004 (Washington DC EPA Publication)available at httpwwwepagovotaqregsnonroadaviation420r05004pdf

[4] Energy Independence and Security Act of 2007 Public Law110-140 Washington DC

[5] State of California 2007 Low Carbon Fuel Standard ExecutiveOrder S-01-07 Sacramento CA

[6] MacLean H L and Lave L B 1998 Environ Sci Technol 32322Andash30A

[7] MacLean H L and Lave L B 2003 Environ Sci Technol37 5445ndash52

[8] US Department of Energy 2007 The Greenhouse GasesRegulated Emissions and Energy Use in Transportation(GREET) Model (Argonne IL Office of Energy Efficiencyand Renewable Energy Argonne National Laboratory)

[9] Delucchi M A 2003 Lifecycle Emissions Model (LEM)Lifecycle Emissions from Transportation Fuels MotorVehicles Transportation Modes Electricity Use Heatingand Cooking Fuels and Materials (Davis CA University ofCalifornia Davis) available at httppubsitsucdavisedupublication detailphpid=273

[10] US Environmental Protection Agency 2008 Fuel EconomyReporting available at httpwwwfueleconomygov

[11] National Highway Traffic Safety Administration 2008 VehicleSafety Information available at httpwwwsafercargov

[12] Wardrsquos Communications 2006 Wardrsquos Motor Vehicle Facts andFigures Southfield MI

[13] US Department of Transportation 2005 Air Carrier Statistics2005 Form 41 Table T-100 (Washington DC Bureau ofTransportation Statistics) available online at httpwwwtranstatsbtsgov

[14] US Environmental Protection Agency 2003 Mobile 62 MobileSource Emission Factor Model (Washington DC Office ofTransportation and Air Quality)

[15] US Department of Transportation 2005 National TransitDatabase (Washington DC Federal Transit Administration)available at httpwwwntdprogramgovntdprogram

[16] Fels M 1978 Energy 3 507ndash22[17] Bay Area Rapid Transit 2007 Energy Efficiency Assessment of

Bay Area Rapid Transit (BART) Train Cars San FranciscoCA available at httpwwwbartgovdocsBARTenergyreportpdf

[18] Fritz S G 1994 J Eng Gas Turbine Power 116 774ndash83

7

Environ Res Lett 4 (2009) 024008 M V Chester and A Horvath

[19] Facanha C and Horvath A 2006 Int J Life Cycle Assess11 229ndash39

[20] Chester M V 2008 Life-cycle environmental inventory ofpassenger transportation modes in the United States PhDDissertation University of California Berkeley CAavailable online at httprepositoriescdliborgitsdsUCB-ITS-DS-2008-1

[21] Horvath A 2003 Life-cycle Environmental and EconomicAssessment of Using Recycled Materials for AsphaltPavements (Berkeley CA University of CaliforniaTransportation Center) available at httpwwwuctcnetpaperspapersuctcshtml

[22] Horvath A and Hendrickson C 1998 Transp Res Rec1626 105ndash13

[23] Hendrickson C and Horvath A 2000 ASCE J Constr EngManag 126 38ndash44

[24] European Environment Agency 2006 EMEPCORINAIREmission Inventory Guidebook Copenhagen Denmarkavailable at httpreportseeaeuropaeuEMEPCORINAIR4

[25] US Department of Transportation 2007 EDMS 502 Emissionand Dispersion Modeling System Software (WashingtonDC Federal Aviation Administration)

[26] Deru M and Torcellini P 2007 Source Energy and EmissionFactors for Energy Use in Buildings (Golden CO USDepartment of Energy National Renewable Energy

Laboratory Golden CO) available at httpwwwnrelgovdocsfy06osti38617pdf

[27] Pacific Gas and Electric 2008 PGampErsquos Electric Power MixDelivered to Retail Customers (available at httpwwwpgecommyhomeedusafetysystemworkselectricenergymix)

[28] Carnegie Mellon University 2008 Economic InputndashOutputAnalysis-Based Life-Cycle Assessment Software (PittsburghPA Green Design Institute) available at httpwwweiolcanet

[29] US Environmental Protection Agency 1999 Technical Supportfor Development of Airport Ground Support EquipmentEmission Reductions EPA420-R-99-007 (Washington DCEPA Publication) available at httpearth1epagovotaqstateresourcespolicytranspairportsr99007pdf

[30] Stokes J and Horvath A 2006 Int J Life Cycle Assess11 335ndash43

[31] Horvath A Hendrickson C T Lave L B McMichael F C andWu T S 1995 Environ Sci Technol 29 86ndash90

[32] Boughton B and Horvath A 2006 Resources Conserv Recycl47 1ndash25

[33] Boughton B and Horvath A 2004 Environ Sci Technol38 353ndash8

[34] Vieira P S and Horvath A 2008 Environ Sci Technol42 4663ndash9

8