Article pubs.acs.org/est
Role of the Freight Sector in Future Climate Change Mitigation Scenarios Matteo Muratori,*,†,§ Steven J. Smith,† Page Kyle,† Robert Link,† Bryan K. Mignone,‡ and Haroon S. Kheshgi‡ †
Pacific Northwest National Laboratory, Joint Global Change Research Institute, 5825 University Research Court, Suite 3500, College Park, Maryland 20740, United States § National Renewable Energy Laboratory, 15013 Denver West Parkway, Golden, Colorado 80401, United States ‡ ExxonMobil Research and Engineering Company, Corporate Strategic Research, 1545 U.S. 22, Annandale, New Jersey 08801, United States S Supporting Information *
ABSTRACT: The freight sector’s role is examined using the Global Change Assessment Model (GCAM) for a range of climate change mitigation scenarios and future freight demand assumptions. Energy usage and CO2 emissions from freight have historically grown with a correlation to GDP, and there is limited evidence of near-term global decoupling of freight demand from GDP. Over the 21st century, greenhouse gas (GHG) emissions from freight are projected to grow faster than passenger transportation or other major end-use sectors, with the magnitude of growth dependent on the assumed extent of long-term decoupling. In climate change mitigation scenarios that apply a price to GHG emissions, mitigation of freight emissions (including the effects of demand elasticity, mode and technology shifting, and fuel substitution) is more limited than for other demand sectors. In such scenarios, shifting to less-emitting transportation modes and technologies is projected to play a relatively small role in reducing freight emissions in GCAM. By contrast, changes in the supply chain of liquid fuels that reduce the fuel carbon intensity, especially deriving from large-scale use of biofuels coupled to carbon capture and storage technologies, are responsible for the majority of freight emissions mitigation, followed by price-induced reduction in freight demand services.
1. INTRODUCTION
with implications for energy, climate change and related policies, economic development, and infrastructure.3−5 Historically, the growth of freight demand has been closely related to GDP. A study conducted by The World Bank that examined 33 countries at different development stages suggested that GDP could explain 89% of the variation in road freight volumes.6 More recent studies suggest that if economies shift increasingly to services, then freight demand might increase at a rate slower than overall economic growth, which is often referred to as “decoupling” (defined by Tapio as elasticity values under 1 for the relationship between transport volume growth and economic growth7). While it is possible that freight demand may partially decouple from GDP over the 21st century (as an increasing share of economic growth comes from nonmaterial sectors and freight loading and procurement is further optimized), it is also possible that growing demands for consumer goods and continued global trade may result in a
Movement of goods and commodities across, within, and between countries has been a foundational activity of economies throughout history. Freight is transported today by trucks, railroads, ships, and planes, predominately fueled by petroleum-derived fuels. Global energy consumption for freight transport currently accounts for about 40% of total transportation sector consumption and a similar share of greenhouse gas (GHG) emissions.1 Trucking dominates the overall energy use of the freight sector,2 and generally has a higher energy intensity than shipping or rail. While the global passenger vehicle fleet is expected to more than double by 2040, increasing from about 800 million to about 1.7 billion vehicles,3 the increase in energy consumption (and related CO2 emissions) from the growth of light-duty vehicles is expected to be nearly offset by improvements in fuel efficiency and use of alternative fuels.4 As a result, global energy demand for personal transportation is expected to be relatively flat over the next few decades, and even decrease in the U.S.3,5 However, global freight demand for energy is expected to rise, driven by a projected increase in economic activity and an associated increase in movement of goods and commodities, © 2017 American Chemical Society
Received: Revised: Accepted: Published: 3526
September 6, 2016 January 21, 2017 February 27, 2017 February 27, 2017 DOI: 10.1021/acs.est.6b04515 Environ. Sci. Technol. 2017, 51, 3526−3533
Article
Environmental Science & Technology
less energy-intensive modes (e.g., switching from trucking to rail), and reduction of the carbon content of the fuel used (e.g., switching from petroleum-derived fuels to biofuels). SI Section S2 proposes a decomposition analysis to compute the role of each of these terms. Changes in the energy intensity of freight transport due to improvements in fuel economy or load factors could further reduce emissions, but the trajectory of energy intensity improvement is exogenous in GCAM and generally assumed to be driven by factors other than the prescribed carbon price. The implications of such alternative energy intensity assumptions are explored in more detail in the sensitivity cases reported in SI Section S4.3. 2.1. Freight Transport Demand Scenarios. A key set of assumptions that will determine the future role of freight transport in overall emissions growth is how demand for freight services will depend on large-scale driving forces such as overall increases in income and population, industrialization, and trends in global trade. Freight transport has grown substantially over the past decadesas shown in the detailed discussion provided in SI Section S3.1. Global international shipping volume, for example, grew approximately linearly as GDP increased, over the period 1970−2010 as shown in Figure 1, and grew 30% more than GDP from 1990 to 2010.20
strong long-term relationship between freight demand and economic growth. Emissions from the global freight sector will depend not only on trends in demand for freight service, but also on the energy and carbon intensities of that service. The trend in modeshifting in the freight sector is toward trucking (a comparatively energy-intensive mode), which is favored by “just-in-time” practices, further increasing energy consumption.8,9 Moreover, historical patterns of aggregate carbon intensity from freight are different than those from other sectors. While aggregate carbon intensity declined for other sectors, aggregate carbon intensity for freight increased, primarily due to shifts in modal structure toward more carbon-intensive modes, such as trucking.10 In contrast, technology (e.g., fuel economy) and logistics (e.g., load factor) improvements could reduce freight energy intensity, but historical trends have been inconsistent across regions.2 Finally, transport CO2 per capita is higher today in OECD than in non-OECD countries, but in the future nearly 90% of all transport CO2 emissions growth is expected to come from non-OECD countries.1 In the EIA International Energy Outlook 2014 Reference Case, world liquid fuels consumption increases by more than one-third from 2010 to 2040,11 and non-OECD regions account for virtually all of the increase in demand for petroleum and other liquid fuels. Despite the expected growth in freight emissions, trends in freight emissions and the role of the freight sector in long-term global climate change mitigation have often been overlooked altogether, or examined only at a regional scale,12 and has received limited attention in climate mitigation assessments.13−15 Previous work has examined the role of passenger transportation in climate mitigation, showing that technological advances in lower-carbon options such as fuel economy improvements in conventional vehicles, vehicle hybridization, or electrification, and use of alternative fuels can provide significant emissions reductions and can reduce the cost of mitigating carbon emissions.16−19 By contrast, the Fifth Assessment Report of the IPCC acknowledges that reducing global greenhouse gas emissions from freight will be challenging, since the continuing growth in freight demand, which parallels the growth in GDP, could outweigh mitigation measures unless transport emissions become strongly decoupled from economic growth.15 To fill in the gaps left by previous studies, this paper explores the potential role of the global freight sector in reducing GHG emissions under different climate change mitigation scenarios simulated by the Global Change Assessment Model (GCAM). This study uses a decomposition analysis to evaluate and compare the contributions of different factors to CO2 emissions reduction in the freight sector. To explore future demand uncertainty, in part driven by uncertainty in the relationship between transport demand and economic growth, two different scenarios for future freight demand are considered in this analysis.
Figure 1. Global international shipping volume (total goods loaded) vs global GDP (MER), 1970−2013. Data from United Nations20 and World Bank GDP statistics.
If sectoral composition were to stay constant, then demand for freight services would be expected to continue to increase in line with GDP growth. Alternatively, if economies shift increasingly to services, then freight services might be expected to increase at a rate slower than overall economic growth, which is often referred to as decoupling. The literature on the topic, analyzed in SI Section S3.1, is not consistent. While many long-term scenarios assume an eventual decoupling between freight demand and income,21,22 there is limited evidence for any such decoupling on a global scale to date. Long-term projections absent any decoupling, however, appear unrealistic, especially considering that a significant portion of the economic growth will be associated with services, thus reducing the freight intensity of GDP. We will, therefore, examine the uncertainty in the GDP− freight demand relationship by considering two scenarios for future freight demand, described in detail in SI Section S3.2. The first is a “high demand” scenario in which global freight demand grows at a rate similar to GDP until midcentury, after which economic growth and freight demand partially decouples. The second is a “low demand” scenario in which global freight demand decouples from GDP before 2050. These scenarios are in line with current literature results and allow us
2. METHODS A state-of-the-art integrated assessment model, the Global Change Assessment Model (GCAM), is used in this paper. A description of GCAM is provided in the Supporting Information, SI Section S1. In climate change mitigation scenarios, the freight sectoras modeled in GCAMhas three alternatives to reduce GHG emissions: reduction in demand for freight services induced by increased service price, shifting to 3527
DOI: 10.1021/acs.est.6b04515 Environ. Sci. Technol. 2017, 51, 3526−3533
Article
Environmental Science & Technology
Figure 2. Global CO2 emissions by sector in 2010 and under the “No Mitigation” scenario in 2100 for the two freight demand scenarios considered.
destroyed in the atmosphere but redistributed among atmospheric, terrestrial and oceanic reservoirs (on a century to millennial time scale). The stricter the forcing target, the sooner emissions decline, as shown in Figure S4. Our results show that mitigation of emissions in the transport sector is less rapid and more limited than other demand sectors, as illustrated in Figure 3. Other demand
to examine the impact of a range of assumptions for future freight demand.21,22 2.2. Climate Change Mitigation Scenarios. We explore the role of freight transport in climate mitigation by comparing a scenario without a global price on GHG emissions (“No Mitigation” scenario), to four climate change mitigation scenarios stabilizing radiative forcing at 6.0, 4.5, 3.7, and 2.6 W/m2 by 2100. These radiative forcing targets mirror those used by the IPCC to generate Representative Concentration Pathways or “RCPs”.23 The scenarios produced by GCAM approximate least-cost pathways to these radiative forcing targets and will be called “LCPs”. In the “No Mitigation” scenario, while GHG emissions are not priced, other existing policies, such as air quality mandates and energy efficiency improvements, are represented. More details are reported in SI Section S3.3.
3. RESULTS 3.1. Emissions Results. Over the 21st century, in a “No Mitigation” scenario, the share of CO2 emissions from freight is projected to grow faster than other sectors across the different demand scenarios considered, with the magnitude of growth dependent on the assumed extent of long-term decoupling of freight demand, as shown in Figure 2. CO2 emissions from the transportation sector include direct, or tailpipe, emissionsalso referred to as tank-to-wheels emissionsas well as indirect emissionsalso referred to as well-to-tank emissionswhich include any CO2 emissions related to the production of transportation fuels, such as liquid fuels or electricity. Note that indirect emissions reported here only include fossil and industrial emissions related to fuel production, including any sink of CO2 from CCS, but not including land use change emissions from the production of bioenergy. Unless otherwise noted, CO2 emissions reported in this paper include both direct and indirect contributions. In a “No Mitigation” scenario, global fossil carbon dioxide emissions in the high demand scenario are 16% higher in 2100 compared to the case with faster decoupling of freight demand from GDP (i.e., the low demand scenario). This increase is primarily due to the increased fuel use by the freight sector (an increase of about 70%). The increase in CO2 emissions, relative to the increase in freight demand, varies significantly across regions. The difference between emissions in cases with alternative assumptions about energy intensity improvement (explored in SI Section S4.3) is qualitatively similar to the difference in emissions between the demand cases discussed here, although the magnitude of the latter impact is greater than the former. Under scenarios that stabilize radiative forcing to mitigate climate change, carbon dioxide emissions must eventually be decreased toward zero. This is because carbon dioxide is not
Figure 3. Global CO2 emissions by sector over time for different leastcost climate change mitigation pathways (LCPs) for the low freight demand scenario (strong freight demand decoupling from GDP). Results for the high demand scenarios are shown in Figure S12. Table S3 reports results in tabular form for both figures.
sectors (i.e., buildings and industry) rely more heavily on electricity and more easily electrify, and since the electric sector decarbonizes more rapidly in climate change mitigation scenarios due to more cost-effective abatement opportunities, these demand sectors also decarbonize more rapidly. While it is possible to further electrify all of the demand sectors, including transportation, such actions tend to be less cost-effective than other abatement options in the transportation sector and may be impractical in freight due to the specific requirements of heavy-duty vehicles and long-distance transportation modes (while the same vehicle technology options are in principle available for both passenger and freight transport, the higher energy density and range requirements in freight applications limit the actual practicality). Total CO2 emissions from the transportation sector in 2100 fall below emissions in 2010 only in the most stringent climate change mitigation scenario (LCP 2.6). In all the scenarios, mitigation of emissions in the freight sector is more limited than mitigation in other demand sectors, including passenger transport. In scenarios that stabilize radiative forcing, the share of remaining direct cumulative carbon emissions that comes from freight (reported in Figure S9 and Table S2) increases with increasing climate change mitigation stringency (decreas3528
DOI: 10.1021/acs.est.6b04515 Environ. Sci. Technol. 2017, 51, 3526−3533
Article
Environmental Science & Technology
Figure 4. Global regional freight and international shipping final energy use by service and fuel for the “No Mitigation” and LCP2.6 scenarios assuming low freight demand growth. Note that the contribution of coal, natural gas, and biofuels to the total freight final energy use is small in historical time periods, making it hard to see. However, the contribution of nonpetroleum-derived fuels in the transportation sector has been growing in the recent past. For future time periods, GCAM assumes a competition among fuels based on their relative projected costs.
levels under LCP 2.6, depending on the projected growth of freight demand (i.e., strength of the decoupling), compared to 86−128% in the “No Mitigation” scenarios. In the most stringent climate change mitigation pathway (LCP 2.6), about one-third less energy is consumed in 2100, compared to the “No Mitigation” scenarios (bottom panels), due to the priceinduced effects on the freight sector. In addition, in LCP 2.6, GCAM projects a significant transformation of the freight energy mix over the 21st century. Coal is no longer used to produce liquid fuels, given its carbon intensity, and the share of petroleum is reduced. The share of natural gas used in the freight sector in 2100 remains constant across scenarios at about 40%, and electricity accounts for about 3% of total freight energy use in 2100, compared to about 1% in the “No Mitigation” scenarios. The decline of coal and petroleum-derived fuels is partially offset by the use of biomass, which increases by 30% in 2100. Moreover, over 60% of the biofuels production is coupled to carbon capture and storage technologies by 2050, with nearly all biofuels production coupled to CCS in 2100. The negative emissions deriving from large-scale use of bioenergy with CCS technologies offset emissions from continued use of petroleum and natural gas as transportation fuels.24 The energy results in Figure 4 can be further understood in terms of changes in the underlying economics that result from the introduction of a carbon price (carbon price pathways are shown in Figure S5). The cost of freight services is more sensitive to a carbon price compared to passenger services,12 since fuel cost is a greater portion of total service cost in the freight sector. Thus, under climate change mitigation scenarios, the total cost of transportation services increases more for freight services than for passenger services. For example, under LCP 2.6, the cost of transportation services, reported in Figure S6, increases by 15% for regional freight and doubles for international shipping in 2100. For comparison, the increase in the total cost of passenger transport is 7%. The higher service cost increase in the freight sector compared to the passenger sector would tend to make freight more responsive to a carbon
ing stabilization level), since in GCAM it is more costly to abate emissions in the freight sector than in other sectors. In the “No Mitigation” scenario, the contribution of indirect emissions (related to fuel production) increases as coal-toliquid technologies, whose production is almost three times more carbon-intensive than the production of liquid fuels from petroleum, are adopted in the liquid fuels sector. In addition, in climate change mitigation scenarios, the contributions of indirect emissions decreases due to negative emissions resulting from biofuel production with CCS. These two effects cause the difference in emissions or “wedge” between the “No Mitigation” scenarios and climate mitigation scenarios (shown in Figure S10) to increase when indirect emissions are included. Freight direct (tailpipe) emissions by 2100 are about 40% lower in the low demand, “No Mitigation” scenario compared to the corresponding high demand scenario (driven almost entirely by lower freight demand), and the difference between the two demand scenarios is almost halved by 2100 under LCP 2.6. When considering indirect emissions as well as direct emissions, this difference between the high and low demand scenarios is magnified. Nonetheless, while global CO 2 emissions are negative in the most stringent mitigation scenario (LCP 2.6), total (direct + indirect) CO2 emissions from freight are still positive in 2100, due to the large fraction of petroleumderived liquid fuels that are still used in the freight sector. 3.2. Energy Results. The emissions results discussed above can be explained by examining the ways in which the energy system evolves over time across scenarios. Figure 4 shows final freight energy use by service and fuel for the “No Mitigation” and LCP2.6 scenarios assuming low freight demand growth (Figure S11 reports results for the corresponding high demand scenarios). In the “No Mitigation” scenario (top panels), freight energy use increases by a factor 4−7 by the end of the century (the range indicates the variation across the two freight demand scenarios), with fossil fuels continuing to power the freight sector, and natural gas playing an increasing role. Energy use in the freight sector in 2040 increases by 70−100% from 2010 3529
DOI: 10.1021/acs.est.6b04515 Environ. Sci. Technol. 2017, 51, 3526−3533
Article
Environmental Science & Technology
Figure 5. Decomposition of factors associated with freight CO2 emissions reductions in LCP 2.6. Note that the energy intensity of the freight sector increases in this scenario, causing CO2 emissions to increase (purple wedge). This is due to the more widespread use of natural gas-fueled trucks and ships (driven by the climate change mitigation policy), which emit less CO2, but have higher energy intensities.
evaluate the effect of different assumptions on the cost of climate change mitigation. Our results show that, in GCAM, carbon prices (reported in Figure S5) are higher throughout the century in scenarios with higher freight demand, and in scenarios with lower stabilization levels. In the LCP 4.5 scenario, for example, the freight demand increases in the second half of the century in the high demand scenario requiring a 68% higher carbon price (by 2100) to keep emissions along the stabilization trajectory compared to the low demand scenario (note that the two LCP 4.5 scenarios have similar CO2 emission paths, but significantly different carbon prices). This increase in late-century carbon price is driven by limited options for mitigating emissions from freight transport. Emissions reductions then occur throughout the economy, including in the freight sector, in order to maintain the specified radiative forcing targets. The carbon abatement curve at this point, where emissions are relatively low (abatement is relatively high), is relatively steep, resulting in the substantial increase in carbon price. The impact of different freight transport assumptions on the carbon price is almost entirely due to changes in assumptions about regional freight. This is because of the relatively high efficiency of international shipping and its smaller share of total freight energy consumption (∼25%): the majority of freight fuel consumption is due to regional trucks. The modeled overall mitigation cost to society, measured in GCAM by the discounted sum of total costs−or the integral under the marginal abatement supply curve−in every year between 2010 and 2100, is sensitive to the growth in freight demand. Relative to the high demand case, the low demand case is 49%, 24%, 17%, and 20% less costly for the LCP 6.0, 4.5, 3.7, and 2.6 scenarios, respectively. The smaller relative impact in more stringent mitigation scenarios reflects the larger increase in the total mitigation cost in those sectors. More specifically, there is about an order of magnitude increase in cost between LCP 6.0 and LCP 4.5, and an additional order of magnitude increase between LCP 4.5 and LCP 2.6. We also examine, in SI Section S4.3, the impact of more rapid improvement of freight energy intensity, as a result of technology (e.g., fuel economy) and logistics (e.g., load factor) improvements. While these results highlight the role that freight energy intensity improvements can play in reducing freight CO2 emissions and limiting the carbon price required to meet climate change mitigation targets, future freight demand is a larger driver of mitigation cost than the rate of energy intensity improvement in our scenarios.
price and induce technology and mode switching more easily, but the suite of cost-effective alternative technologies and modes is limited, reducing the extent of the carbon priceinduced transition in the freight sector (see the sensitivity test in the SI Section S4.1). Fuel-shifting plays a significant role in reducing emissions from regional freight and international shipping. On top of a significant adoption of biofuels, the use of natural gas as a fuel for regional freight increases from negligible levels in 2010 to 28% in 2100 in the “No Mitigation” scenario and to 37% in the LCP 2.6 scenario. This reduces CO2 emissions from regional freight by about 10%. For comparison, in the passenger sector, natural gas use increases from near negligible levels in 2010 to 24% in the “No Mitigation” scenario and to 22% in the LCP 2.6 scenario. Natural gas is also used in international shipping. If LNG is removed as a fuel option for marine vessels, then CO2 emissions from international shipping increase by about 20%, while energy consumption remains essentially the same. Another option to reduce emissions from freight transport is to shift from truck to rail or ship transport, which have lower energy intensities. The extent to which this shift could occur will depend on infrastructure and the location of freight destination. Consumer goods would ultimately need to be delivered to stores and homes by truck, while some destinations for bulk freight goods are accessible by rail or ship, allowing for a choice of freight modes. We examine the sensitivity of results to mode shifting in the SI Section S4.2, finding that the ability to shift freight transport to lower carbon intensity modes has only a small impact on the energy and economic outcomes in GCAM. This is consistent with the mode-shifting response in passenger transport, which evolves over time, but not significantly between the mitigation and “No Mitigation” cases. Energy intensity of road freight can also be lowered by assuming that fuel economy or load factors increase, as discussed in SI Section S4.3. Generally, lowering the energy intensity would tend to have a similar effect on the energy system response as lowering freight demand growth. 3.3. Economic Implications. One measure of the cost of climate change mitigation in any given year is the marginal cost of abatement, which is indicated by the carbon price necessary to achieve emissions reductions consistent with the stabilization objective. The carbon price associated with a mitigation stringency (stabilization level) differs significantly between integrated assessment models as is evident in a recent assessment by the IPCC.25 However, comparing the carbon prices among scenarios in the same model can be used to 3530
DOI: 10.1021/acs.est.6b04515 Environ. Sci. Technol. 2017, 51, 3526−3533
Article
Environmental Science & Technology 3.4. Decomposition of Freight Emissions. Having discussed the overall emissions results and described the underlying mechanisms in the energy system that give rise to these emissions results, we next consider a formal decomposition of freight emissions reductions in climate change mitigation scenarios. Carbon dioxide emissions reductions from the freight sector can be broken down, as shown in SI Section S2, into three main factors: demand changes, energy intensity changes, and fuel carbon intensity changes. The former indicates how changes in demand for freight services affect CO2 emissions. The second captures effects related to mode shifting and modal energy intensity. The third conveys information about changes in the carbon intensity of the fuels used in the freight sector, such as the introduction of biofuels. These three factors can also be seen as levers to reduce freight emissions under climate change mitigation scenarios. In such scenarios the energy intensity trajectories of different freight modes are exogenous and do not respond to the carbon price. This assumption is consistent with the fact that a carbon price does not change the overall service cost sufficiently to affect vehicle choice, when the additional capital cost of more efficient vehicles is taken into account (see Figure S6). That said, the energy intensity of freight may change for a number of reasons, including policy measures other than a price on GHG emissions, and for this reason we have conducted a sensitivity analysis (see SI Section S4.3). A key finding from this sensitivity is that, from the perspective of understanding the decomposition of abatement, altering energy intensity is qualitatively similar to altering freight demand, although the effect of the latter is larger in magnitude in our scenarios. Figure 5 shows a decomposition of global freight CO2 emissions reductions under LCP2.6 for both the high and low freight demand scenarios. The magnitude of the effect of decoupling freight demand and economic growth can be measured by the difference between the solid black line in the left-hand panel and the solid black line in the right-hand panel. A slower growth of freight demand (faster decoupling of freight demand from economic growth) reduces global cumulative CO2 emissions over the 21st century by about one-third, and the difference in freight emissions between the high and low freight demand scenarios reaches 45% in 2100. Under LCP 2.6, global cumulative CO2 emissions from freight are reduced by 68 or 73% (for low or high demand scenarios, respectively) compared to the corresponding “No Mitigation” scenarios. This reduction is achieved by (a) price-induced freight demand reduction, driven by the carbon price; (b) mode and technology shifting in the freight sector; and (c) reduction in the carbon intensity of fuels used in the freight sector. Among these emissions reduction factors, the reduction in the carbon intensity of liquid fuels is the largest. Specifically, the production of biofuels coupled to CCS has the greatest contribution, indicating the greater potential of the fuels production sector to mitigate freight emissions in these scenarios, relative to other mitigation options in GCAM. The preference for fuel switching follows from the fact that the adoption of liquid fuels derived from biomass coupled to CCS becomes an economically attractive option at sufficiently high carbon prices in GCAM.26 Other changes in the freight sector that reduce CO2 emissions by reducing carbon intensity include technologyswitching, such as greater use of natural gas trucks and ships, and to a lesser extent, mode-shifting to electric rail transport. However, the use of natural gas leads to higher energy
intensities since trucks and ships fueled with natural gas emit less CO2, but are less energy efficient (see Table S1, Section S1). This is why the emissions contribution from energy intensity changes is positive rather than negative (purple wedge above the solid black line in Figure 5). A decomposition of freight CO2 emissions by the different factors over time is reported in Figure S13 and in the TOC. In the low freight demand scenario under LCP 2.6, the growth in freight service demand during the first half of the century is roughly offset by reductions in fuel carbon intensity deriving from use of natural gas and biofuels, leading to a net CO2 emissions increase of 12% from 2010 to 2050. In the second half of the century, the reduction in fuel carbon intensity is more pronounced and outweighs increases in freight service demand, leading to an overall reduction in freight CO2 emissions of 55% by 2100, relative to 2050.
4. DISCUSSION Currently, the transportation sector is virtually entirely fueled by petroleum. In addition, transportation emissions are highly distributed and are difficult to capture. Thus, mitigation of emissions from the transportation sector would require changes in transportation modes and technologies or in the carbon intensity of the fuels consumed (e.g., use of biofuels or lowcarbon electricity). However, there has been limited assessment of the global potential for GHG emissions reduction from the transportation sector. Among the studies that do consider the role of transport, fewer focus on freight than on passenger transport, which makes the potential for reductions and the underlying transformation pathways much less certain for the freight sector than for the passenger sector.15 The increase in energy consumption and CO2 emissions from the growing number of light-duty vehicles (anticipated to more than double by 2040) is expected to be nearly offset by improvements in fuel efficiency and use of alternative fuels.3−5 In contrast, global freight demand growthand associated growth in energy consumption and CO2 emissionsmay be greater but is uncertain. While some future scenarios posit a decoupling, in which freight demand grows more slowly than overall economic activity, recent history indicates that demand for freight services is continuing to grow in proportion to global GDP. We examined the implications of this uncertainty for climate change and climate change mitigation using two different freight demand growth scenarios. The scenarios explored in this study suggest six main insights: • Future emissions from freight will be driven by the extent of demand growth (reflecting in part the extent of decoupling of freight demand from GDP) as much as climate change mitigation, and demand growth is projected to be predominately in non-OECD regions. The extent of energy intensity improvement will also affect freight emissions in a way that is qualitatively similar to effect of freight demand growth. • Cost of freight services is more sensitive to a carbon price compared to passenger services, due to the higher importance of the fuel cost in the total service cost. Thus, under climate change mitigation scenarios the total cost of transportation service increases more for freight service than for passenger services. While this cost increase may have broader implications for other sectors or the economy as a whole, the suite of cost-effective 3531
DOI: 10.1021/acs.est.6b04515 Environ. Sci. Technol. 2017, 51, 3526−3533
Article
Environmental Science & Technology technologies to mitigate emissions is smaller for freight than for passenger transport, limiting the extent of the carbon price-induced transition in the freight sector. • Mitigation of freight emissions (including the effects of demand elasticity, mode and technology shifting, and fuel substitution) is more limited than mitigation in other demand sectors, including passenger transport, under both demand assumptions considered here. Therefore, the share of remaining carbon emissions from freight increases with decreasing stabilization level. This result follows again from the limited set of mitigation options in the freight sector in GCAM. • Decarbonization of the freight sector in GCAM is largely due to changes in the supply chain of liquid fuels that reduce the fuel carbon intensity, especially deriving from large-scale use of biofuels coupled to carbon capture and storage technologies. The introduction of lower-carbon fuels are responsible for the majority of freight emissions mitigation, followed by the price-induced demand reduction. The reliance on bioenergy with CCS raises important questions about the practical implications of this technology, including the pace of technological advance, the scale of land use change required, along with uncertainty in associated land use change emissions, and various other barriers to deployment, including the acceptability of CCS.27,28 Technologies and practices that improve energy intensity (e.g. increased fuel economy, higher load factors) can further reduce emissions.29 Such outcomes do not follow from the imposed carbon price in GCAM, but could be driven by specific policies, such as increased fuel economy standards. • Assumptions about freight transport demand have a large impact on the carbon price necessary to meet a climate change mitigation target in GCAM because of the lack of options to mitigate emissions from the freight sector. That is, under increased freight demand scenarios, the relative carbon price increase is larger than the relative emissions abatement increase because of the curvature of the overall marginal abatement cost curve. These results point to the need to better understand potential long-term pathways for freight transport growth. One way to improve understanding about the future is to increase understanding of past trends. While global energy consumption statistics for fuel consumption are readily available, these reports do not typically disaggregate freight and passenger transport. National statistics are available for many countries, but in many cases only for a few decades. In addition, more explicit modeling of freight transport in longterm models would be helpful to improve future freight transportation projections and scenarios of transformation pathways. Better understanding and representation of all the drivers of freight emissionseconomic growth, the link between economic growth and demand, mode-shifting, technology adoption, and fuel substitutionwould be key elements of such an effort.
■
■
choice model used to allocate production among different alternatives. A decomposition methodology to breakdown carbon dioxide emissions among different drivers. A study design section that describes historical freight transport growth, including an extensive literature review, and detailed assumptions on the freight demand and climate change mitigation scenarios used. Three sensitivity analyses to better explore how the freight sector compares to the passenger sector in GCAM; the sensitivity of the results reported in this paper to mode shifting in the freight sector; and the impact of increased technology improvement (reduced future energy intensity) for road freight technologies. Several additional figures and tables are also available (PDF)
AUTHOR INFORMATION
Corresponding Author
*Phone: +1 303-275-2927. Fax: +1 303-275-3765. E-mail:
[email protected] (M.M.). ORCID
Matteo Muratori: 0000-0003-1688-6742 Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS The PNNL authors are grateful for research support provided by ExxonMobil Research and Engineering Company. The NREL author is grateful for research support provided by the U.S. Department of Energy Vehicle Technologies Office. The Pacific Northwest National Laboratory and the National Renewable Energy Laboratory are operated for the U.S. Department of Energy under contract DE-AC05-76RL01830 and DE-AC36-08GO28308, respectively. The views and opinions expressed in this paper are those of the authors alone.
■
REFERENCES
(1) International Energy Agency (IEA), Transport Energy and CO2: Moving Towards Sustainability; OECD Publishing: Paris, 2009. (2) Kamakaté, F.; Schipper, L. Trends in truck freight energy use and carbon emissions in selected OECD countries from 1973 to 2005. Energy Policy 2009, 37, 3743−3751. (3) ExxonMobil. The Outlook for Energy: A View to 2040; Irving, TX, 2014. (4) International Energy Agency (IEA), World Energy Outlook 2014; 2014. (5) U.S. Energy Information Administration (EIA), Annual Energy Outlook 2015; 2015. (6) Bennathan, E.; Fraser, J.; Thompson, L. S. What Determines Demand for Freight Transport?; World Bank Publications, 1992; Vol. 998. (7) Tapio, P. Towards a theory of decoupling: degrees of decoupling in the EU and the case of road traffic in Finland between 1970 and 2001. Transport Policy 2005, 12, 137−151. (8) Schipper, L.; Saenger, C.; Sudardshan, A. Transport and carbon emissions in the United States: the long view. Energies 2011, 4, 563− 581. (9) Eom, J.; Schipper, L.; Thompson, L. We keep on truckin’: Trends in freight energy use and carbon emissions in 11 IEA countries. Energy Policy 2012, 45, 327−341. (10) Greening, L. A.; Ting, M.; Davis, W. B. Decomposition of aggregate carbon intensity for freight: trends from 10 OECD countries for the period 1971−1993. Energy Economics 1999, 21, 331−361. (11) U.S. Energy Information Administration (EIA), International Outlook 2014; 2014.
ASSOCIATED CONTENT
* Supporting Information S
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.6b04515. A detailed description of the GCAM model, with particular focus on the freight sector and on the logit 3532
DOI: 10.1021/acs.est.6b04515 Environ. Sci. Technol. 2017, 51, 3526−3533
Article
Environmental Science & Technology (12) Yin, X.; Chen, W.; Eom, J.; Clarke, L. E.; Kim, S. H.; Patel, P. L.; Yu, S.; Kyle, G. P. China’s transportation energy consumption and CO2 emissions from a global perspective. Energy Policy 2015, 82, 233− 248. (13) IPCC, Contribution of Working Group III to the Third Assessment Report of the Intergovernmental Panel on Climate Change; 2001. (14) IPCC, Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change; Metz, B., Davidson, O. R., Bosch, P. R., Dave, R., Meyer, L. A., Eds.; 2007. (15) IPCC, Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change Edenhofer, O., PichsMadruga, R., Sokona, Y., Farahani, E., Kadner, S., Seyboth, K., Adler, A., Baum, I., Brunner, S., Eickemeier, P., Kriemann, B., Savolainen, J., Schlömer, S., von Stechow, C., Zwickel, T., Minx, J. C., Eds.; 2014. (16) Orsi, F.; Muratori, M.; Rocco, M.; Colombo, E.; Rizzoni, G. A multi-dimensional well-to-wheels analysis of passenger vehicles in different regions: Primary energy consumption, CO2 emissions, and economic cost. Appl. Energy 2016, 169, 197−209. (17) Johansson, B.; Åhman, M. A comparison of technologies for carbon-neutral passenger transport. Transport Res. D: Tr E 2002, 7, 175−196. (18) Kim, S. H.; Edmonds, J.; Lurz, J.; Smith, S. J.; Wise, M. The ObjECTS Framework for Integrated Assessment: Hybrid Modeling of Transportation. Energy Journal 2006, SI2006, 63−91. (19) Kyle, P.; Kim, S. H. Long-term implications of alternative lightduty vehicle technologies for global greenhouse gas emissions and primary energy demands. Energy Policy 2011, 39, 3012−3024. (20) United Nations, Conference on Trade and Development (UNCTAD 2015); 2015. (21) Harvey, L. Global climate-oriented transportation scenarios. Energy Policy 2013, 54, 87−103. (22) Girod, B.; van Vuuren, D. P.; Grahn, M.; Kitous, A.; Kim, S. H.; Kyle, P. Climate impact of transportation: A model comparison. Clim. Change 2013, 118, 595−608. (23) Moss, R. H.; et al. The next generation of scenarios for climate change research and assessment. Nature 2010, 463, 747−756. (24) Muratori, M.; Kheshgi, H.; Mignone, B.; Clarke, L.; McJeon, H.; Edmonds, J. Carbon capture and storage across fuels and sectors in energy system transformation pathways. Int. J. Greenhouse Gas Control 2017, 57, 34−41. (25) Clarke, L. et al. Assessing Transformation Pathways. In Climate Change 2014: Mitigation of Climate Change. Contribution of Working Group III to the Fifth Assessment Report of the IPCC. 2014. (26) Muratori, M.; Kheshgi, H.; Mignone, B.; McJeon, H.; Clarke, L. The future role of CCS in electricity and liquid fuel supply. Energy Procedia 2017. (27) Fuss, S.; Canadell, J. G.; Peters, G. P.; Tavoni, M.; Andrew, R. M.; Ciais, P.; Jackson, R. B.; Jones, C. D.; Kraxner, F.; Nakicenovic, N.; Le Quere, C.; Raupach, M.; Sharifi, A.; Smith, P.; Yamagata, Y. Betting on negative emissions. Nat. Clim. Change 2014, 4, 850−853. (28) Muratori, M.; Calvin, K.; Wise, M.; Kyle, P.; Edmonds, J. Global economic consequences of deploying bioenergy with carbon capture and storage (BECCS). Environ. Res. Lett. 2016, 11, 095004. (29) Nealer, R.; Matthews, H. S.; Hendrickson, C. Assessing the energy and greenhouse gas emissions mitigation effectiveness of potential US modal freight policies. Transport Res. A: Pol 2012, 46, 588−601.
3533
DOI: 10.1021/acs.est.6b04515 Environ. Sci. Technol. 2017, 51, 3526−3533
