Editing IPCC:AR6/WGIII/Chapter-10 (section)

== 10.4 Decarbonisation of Land-based Transport ==

<div id="10.4.1" class="h2-container"></div>

<span id="light-duty-vehicles-for-passenger-transport"></span>
=== 10.4.1 Light-duty Vehicles for Passenger Transport ===

<div id="h2-13-siblings" class="h2-siblings"></div>

LDVs represent the main mode of transport for private citizens ( [[#ITF--2019|ITF 2019]] ) and currently represent the largest share of transport emissions globally ( [[#IEA--2019d|IEA 2019d]] ). Currently, powertrains depending on gasoline and diesel fuels remain the dominant technology in the LDV segment ( [[#IEA--2019d|IEA 2019d]] ). HEVs, and fully battery electric vehicles (BEVs), however, have become increasingly popular in recent years ( [[#IEA--2021a|IEA 2021a]] ). Correspondingly, the number of lifecycle assessment (LCA) studies investigating HEVs, BEVs, and fuel cell vehicles have increased. While historically the focus has been on the tailpipe emissions of LDVs, LCA studies demonstrate the importance of including emissions from the entire vehicle value chain, particularly for alternative powertrain technologies.

Figure 10.4 presents the cumulative lifecycle emissions for selected powertrain technologies and fuel chain combinations for compact and mid-sized LDVs. This figure summarises the harmonised findings from the academic literature reviewed and the data submitted through an IPCC data collection effort, as described in Appendix 10.1 ( [[#Hawkins--2013|Hawkins et al. 2013]] ; [[#Messagie--2014|Messagie et al. 2014]] ; [[#Bauer--2015|Bauer et al. 2015]] ; [[#Tong--2015b|Tong et al. 2015b]] ; [[#Ellingsen--2016|Ellingsen et al. 2016]] ; [[#Gao--2016|Gao et al. 2016]] ; [[#Kim--2016|Kim and Wallington 2016]] ; [[#Cai--2017|Cai et al. 2017]] ; [[#Evangelisti--2017|Evangelisti et al. 2017]] ; [[#Ke--2017|Ke et al. 2017]] ; [[#Lombardi--2017|Lombardi et al. 2017]] ; [[#Miotti--2017|Miotti et al. 2017]] ; [[#Valente--2017|Valente et al. 2017]] ; [[#Cox--2018|Cox et al. 2018]] ; [[#de%20Souza--2018|de Souza et al. 2018]] ; [[#Elgowainy--2018|Elgowainy et al. 2018]] ; [[#Luk--2018|Luk et al. 2018]] ; [[#Bekel--2019|Bekel and Pauliuk 2019]] ; [[#Cusenza--2019|Cusenza et al. 2019]] ; [[#Hoque--2019|Hoque et al. 2019]] ; [[#IEA--2019a|IEA 2019a]] ; [[#Rosenfeld--2019|Rosenfeld et al. 2019]] ; [[#Shen--2019|Shen et al. 2019]] ; [[#Wang--2019|Wang et al. 2019]] ; [[#Wu--2019|Wu et al. 2019]] ; [[#Ambrose--2020|Ambrose et al. 2020]] ; [[#Benajes--2020|Benajes et al. 2020]] ; [[#Hill--2020|Hill et al. 2020]] ; [[#Knobloch--2020|Knobloch et al. 2020]] ; [[#Prussi--2020|Prussi et al. 2020]] ; [[#Qiao--2020|Qiao et al. 2020]] ; Wolfram et al. 2020; [[#Zheng--2020|Zheng et al. 2020]] ; [[#Sacchi--2021|Sacchi 2021]] ; [[#Valente--2021|Valente et al. 2021]] ). The values in the figure (and the remaining figures in this section) depend on the 100-year global warming potential (GWP) used in each study, which may differ from the recent GWP updates from WGI. However, it is unlikely that the qualitative insights gained from the figures in this section would change using the update 100-year GWP values.

<div id="_idContainer026" class="Basic-Text-Frame"></div>

[[File:0ff3ecc17f2c895321a4c28d0092cef0 IPCC_AR6_WGIII_Figure_10_4.png]]

'''Figure 10.4 | Life cycle greenhouse gas emissions intensities for mid-sized light-duty vehicle and fuel technologies from the literature.''' The primary x-axis reports units in gCO 2 -eq vkm –1 , assuming a vehicle life of 180,000 km. The secondary x-axis uses units of gCO 2 -eq pkm –1 , assuming a 1.5 occupancy rate. The values in the figure rely on the 100-year GWP value embedded in the source data, which may differ slightly from the updated 100-year GWP values from WGI. The shaded area represents the interquartile range for combined vehicle manufacturing and end-of-life phases. The length of the box and whiskers represent the interquartile range of the operation phase for different fuel chains, while their placement on the x-axis represents the absolute lifecycle climate intensity, that is, includes manufacturing and end-of-life phases. Each individual marker indicates a data point. ‘Advanced biofuels’ refers to the use of second-generation biofuels and their respective conversion and cultivation emission factors. ‘IAM EMF33’ refers to emissions factors for advanced biofuels derived from simulation results from the integrated assessment models EMF33 scenarios. ‘PM’ refers to partial models, where ‘CLC’ is with constant land cover and ‘NRG’ is with natural regrowth. ‘Hydrogen, low-carbon electricity’ is produced via electrolysis using low-carbon electricity. ‘Hydrogen, natural gas SMR’ refers to fuels produced via steam methane reforming of natural gas.

Furthermore, note that the carbon footprint of biofuels used in Figure 10.4 are aggregate numbers not specific to any individual value chain or fuel type. They are derived by combining land use-related carbon emissions from [[IPCC:Wg3:Chapter:Chapter-7|Chapter 7]] with conversion efficiencies and emissions as described in [[#10.3|Section 10.3]] . Specifically, land-use footprints derived from the three modelling approaches employed here are: i) Integrated Assessment Models – Energy Modelling Forum 33 (IAM EMF33); ii) Partial models assuming constant land cover (CLC), and, iii) Partial models using natural regrowth (NRG). The emissions factors used here correspond to scenarios where global production of biomass for energy purposes are 100 EJ/year, with lower emissions factors expected at lower levels of consumption and vice versa. Further details are available in Box 10.2 and Chapter 7.

The tailpipe emissions and fuel consumption reported in the literature generally do not use empirical emissions data. Rather, they tend to report fuel efficiency using driving cycles such as New European Driving Cycle or the US Environmental Protection Agency Federal Test Procedure. As a result, depending on the driving cycle used, operating emissions reported in literature are possibly underestimated by as much as 15–38%, in comparison to real driving emissions ( [[#Fontaras--2017|Fontaras et al. 2017]] ; [[#Tsiakmakis--2017|Tsiakmakis et al. 2017]] ; [[#Triantafyllopoulos--2019|Triantafyllopoulos et al. 2019]] ). The extent of these underestimations, however, varies between powertrain types, engine sizes, driving behaviour and environment.

Current average lifecycle impacts of mid-size ICEVs span from approximately 65 gCO 2 -eq pkm –1 to 210 gCO 2 -eq pkm –1 , with both values stemming from ICEVs running on biofuels. Between this range of values, the current reference technologies are found, with diesel-powered ICEVs having total median lifecycle impacts of 130 gCO 2 -eq pkm –1 and gasoline-fuelled vehicle 160 gCO 2 -eq pkm –1 . Fuel consumption dominates the lifecycle emissions of ICEVs, with approximately 75% of these emissions arising from the tailpipe and fuel chain.

HEVs and plug-in HEVs (PHEVs) vary in terms of degree of powertrain electrification. HEVs mainly rely on regenerative braking for charging the battery. PHEVs combine regenerative braking with external power sources for charging the battery. Operating emissions intensity is highly dependent on the degree to which electrified driving is performed, which in turn is user- and route-dependent. For PHEVs, emissions intensity is also dependent on the source of the electricity for charging. HEV and PHEV production impacts are comparable to the emissions generated for producing ICEVs as the batteries are generally small compared to those of BEVs. Current HEVs may reduce emissions compared to ICEVs by up to 30%, depending on the fuel, yielding median lifecycle intensities varying between 60 gCO 2 -eq pkm –1 (biofuels, EMF33) and 165–170 gCO 2 -eq pkm –1 (biofuels, partial models NRG). Within this wide range, all the combinations of electric and fossil-fuelled driving can be found, as well as the lifecycle intensity for driving 100% on fossil fuel. Because HEVs rely on combustion as the main energy conversion process, they offer limited mitigation opportunities. However, HEVs represent a suitable temporary solution, yielding a moderate mitigation potential, in areas where the electricity mix is currently so carbon intensive that the use of PHEVs and BEVs is not an effective mitigation solution ( [[#Wolfram--2017|Wolfram and Wiedmann 2017]] ; [[#Wu--2019|Wu et al. 2019]] ).

In contrast to HEVs, PHEVs may provide greater opportunities for use-phase emissions reductions for LDVs. These increased potential benefits are due to the ability to charge the battery with low-carbon electricity and the longer full-electric range in comparison to HEVs ( [[#Laberteaux--2019|Laberteaux et al. 2019]] ). Consumer behaviour (e.g., utility factor (UF) and charging patterns), manufacturer settings, and access to renewable electricity for charging strongly influence the total operational impacts ( [[#Wu--2019|Wu et al. 2019]] ). The UF is a weighting of the percentage of distance covered using the electric charge (charge depleting (CD) stage) versus the distance covered using the internal combustion engine (charge sustaining (CS) stage) ( [[#Paffumi--2018|Paffumi et al. 2018]] ). When the PHEV operates in CS mode, the internal combustion engine is used for propulsion and to maintain the state of charge of the battery within a certain range, together with regenerative braking ( [[#Plötz--2018|Plötz et al. 2018]] ; [[#Raghavan--2020|Raghavan and Tal 2020]] ). When running in CS mode, PHEVs have a reduced mitigation potential and have impacts comparable to those of HEVs. On the other hand, when the PHEV operates in CD mode, the battery alone provides the required propulsion energy ( [[#Plötz--2018|Plötz et al. 2018]] ; [[#Raghavan--2020|Raghavan and Tal 2020]] ). Thus, in CD mode, PHEVs hold potential for higher mitigation potential, due to the possibility of charging the battery with low-carbon electricity sources. Consequently, the UF greatly influences the lifecycle emissions of PHEVs. The current peer-reviewed literature presents a wide range of UFs mainly due to varying testing protocols applied for estimating the fuel efficiency and user behaviour ( [[#Pavlovic--2017|Pavlovic et al. 2017]] ; [[#Paffumi--2018|Paffumi et al. 2018]] ; [[#Plötz--2018|Plötz et al. 2018]] ; [[#Plötz--2020|Plötz et al. 2020]] ; [[#Raghavan--2020|Raghavan and Tal 2020]] ; [[#Hao--2021|Hao et al. 2021]] ). These factors make it difficult to harmonise and compare impacts across PHEV studies. Due to the low number of appropriate PHEV studies relative to the other LDV technologies and the complications in harmonising available PHEV results, this technology is omitted from Figure 10.4. However, due to the dual operating nature of PHEV vehicles, one can expect that the lifecycle GHG emissions intensities for these vehicles will lie between those of their ICEV and BEV counterparts of similar size and performance.

Currently, BEVs have higher manufacturing emissions than equivalently-sized ICEVs, with median emissions of 14 tCO 2 -eq per vehicle against approximately 10 tCO 2 -eq per vehicle of their mid-sized fossil-fuelled counterparts. These higher production emissions of BEVs are largely attributed to the battery pack manufacturing and to the additional power electronics required. As manufacturing technology and capacity utilisation improve and globalise to regions with low-carbon electricity, battery manufacturing emissions will likely decrease. Due to the higher energy efficiency of the electric powertrain, BEVs may compensate for these higher production emissions in the driving phase. However, the mitigation ability of this technology relative to ICEVs is highly dependent on the electricity mix used to charge the vehicle. As a consequence of the variety of energy sources available today, current BEVs have a wide range of potential average lifecycle impacts, ranging between 60 and 180 gCO 2 -eq pkm –1 with electricity generated from wind and coal, respectively. The ability to achieve large carbon reductions via vehicle electrification is thus highly dependent on the generation of low-carbon electricity, with the greatest mitigation effects achieved when charging the battery with low-carbon electricity. The literature suggests that current BEVs, if manufactured on low-carbon electricity as well as operated on low-carbon electricity would have footprints as low 22 gCO 2 -eq pkm –1 for a compact-sized car ( [[#Ellingsen--2014|Ellingsen et al. 2014]] ; [[#Ellingsen--2016|Ellingsen et al. 2016]] ). This value suggests a reduction potential of around 85% compared to similarly-sized fossil fuel vehicles (median values). Furthermore, BEVs have a co-benefit of reducing local air pollutants that are responsible for human health complications, particularly in densely-populated areas ( [[#Hawkins--2013|Hawkins et al. 2013]] ; [[#Ke--2017|Ke et al. 2017]] ).

As with BEVs, current HFCVs have higher production emissions than similarly-sized ICEVs and BEVs, generating on average approximately 15 tCO 2 -eq per vehicle. As with BEVs, the lifecycle impacts of FCVs are highly dependent on the fuel chain. To date, the most common method of hydrogen production is steam methane reforming of natural gas ( [[#Khojasteh%20Salkuyeh--2017|Khojasteh Salkuyeh et al. 2017]] ), which is relatively carbon intensive, resulting in lifecycle emissions of approximately 88 gCO 2 -eq pkm –1 . Current literature covering lifecycle impacts of FCVs shows that vehicles fuelled with hydrogen produced from steam methane reforming of natural gas offer little or no mitigation potential over ICEVs. Other available hydrogen fuel chains vary widely in carbon intensity, depending on the synthesis method and the energy source used (electrolysis or steam methane reforming; fossil fuels or renewables). The least carbon-intensive hydrogen pathways rely on electrolysis powered by low-carbon electricity. Compared to ICEVs and BEVs, FCVs for LDVs are at a lower technology readiness level, as discussed in section 10.3.

Two-wheelers, consisting mainly of lower-powered mopeds and higher-powered motorcycles, are popular for personal transport in densely populated cities, especially in developing countries. LCA studies for this class of vehicle are relatively uncommon compared to four-wheeled LDVs. In the available results, however, two-wheelers exhibit similar trends for the different powertrain technologies as the LDVs, with electric powertrains having higher production emissions, but usually lower operating emissions. The lifecycle emissions intensity for two-wheelers is also generally lower than four-wheeled LDVs on a vehicle-kilometre basis. However, two-wheelers generally cannot carry as many passengers as four-wheeled LDVs. Thus, on a passenger-kilometre basis, a fully occupied passenger vehicle may still have lower emissions than a fully occupied two-wheeler. However, today, most passenger vehicles have relatively low occupancy and thus have a correspondingly high emissions intensity on a pkm basis. This points to the importance of utilisation of passenger vehicles at higher occupancies to reduce the lifecycle intensity of LDVs on a pkm basis. For example, the median emissions intensity of a gasoline passenger vehicle is 222 gCO 2 -eq vkm –1 , and 160 gCO 2 -eq vkm –1 for a gasoline two-wheeler ( [[#Cox--2018|Cox and Mutel 2018]] ). At a maximum occupancy factor of four and two passengers, respectively, the transport emissions intensity for these vehicles is 55 and 80 gCO 2 -eq pkm –1 . Under the same occupancy rates assumption, BEV two-wheelers recharged on the average European electricity mix, achieve lower lifecycle GHG intensities than BEV four-wheeled LDVs. On the other hand, FCV two-wheelers with hydrogen produced via steam methane reforming present higher GHG intensity than their four-wheeled counterparts, when compared on a pkm basis at high occupancy rates.

ICEV, HEV, and PHEV technologies, which are powered using combustion engines, have limited potential for deep reduction of GHG emissions. Biofuels offer good mitigation potential if low land-use change emissions are incurred (e.g., the IAM EMF33 and partial models, CLC biofuels pathways shown in Figure 10.4). The literature shows large variability, depending on the method of calculating associated land-use changes. Resolving these apparent methodological differences is important to consolidating the role biofuels may play in mitigation, as well as the issues raised in [[IPCC:Wg3:Chapter:Chapter-7|Chapter 7]] about the conflicts over land use. The mitigation potential of battery and fuel cell vehicles is strongly dependent on the carbon intensity of their production and the energy carriers used in operation. However, these technologies likely offer the highest potential for reducing emissions from LDVs. Prior work on the diffusion dynamics of transport technologies suggests that ‘the diffusion of infrastructure precedes the adoption of vehicles, which precedes the expansion of travel’ ( [[#Leibowicz--2018|Leibowicz 2018]] ). These dynamics reinforce the argument for strong investments in both the energy infrastructure and the vehicle technologies.

To successfully transition towards LDVs utilising low-carbon fuels or energy sources, the technologies need to be accessible to as many people as possible, which requires competitive costs compared to conventional diesel and gasoline vehicles. The lifecycle costs (LCCs) of LDVs depend on the purchasing costs of the vehicles, their efficiency, the fuel costs, and the discount rate. Figure 10.5 shows the results of a parametric analysis of LCC for diesel LDVs, BEVs, and FCVs. The range of vehicle efficiencies captured in Figure 10.5 is the same as the range used for Figure 10.4, while the ranges for fuel costs and vehicle purchase prices come from the literature. The assumed discount rate for this parametric analysis is 3%. Appendix 10.2 includes the details about the method and underlying data used to create this figure.

<div id="_idContainer028" class="Basic-Text-Frame"></div>

[[File:8a8351f7a84f60bc73f12af895f7f5ce IPCC_AR6_WGIII_Figure_10_5.png]]

'''Figure 10.5 | LCC for light-duty internal combustion engine vehicles, battery electric vehicles, and hydrogen fuel cell vehicles.''' The results for ICEVs represent the LCC of a vehicle running on gasoline. However, these values are also representative for ICEVs running on diesel as the costs ranges in the literature for these two solutions are similar. The secondary y-axis depicts the cost of the different energy carriers normalised in USD per gigajoule for easier cross-comparability.

Figure 10.5 shows the range of LCC, in USD per passenger-kilometre, for different powertrain technologies, and the influence of vehicle efficiency (low or high), vehicle purchase price, and fuel/electricity cost on the overall LCC. For consistency with Figure 10.4, an occupancy rate of 1.5 is assumed. Mid-sized ICEVs have a purchase price of USD20,000–40,000, and average fuel costs are in the range of USD1–1.5 per litre. With these conditions, the LCC of fossil-fuelled LDVs span between USD0.22–0.35 pkm –1 or between USD0.17–0.28 pkm –1 , for low- and high-efficiency ICEVs respectively (Figure 10.5).

BEVs have higher purchase prices than ICEVs, though a sharp decline has been observed since AR5. Due to the rapid development of the lithium-ion battery technology over the years ( [[#Schmidt--2017|Schmidt et al. 2017]] ) and the introduction of subsidies in several countries, BEVs are quickly reaching cost parity with ICEVs. Mid-sized BEVs’ average purchase prices are in the range of USD30,000–50,000 but the levelised cost of electricity shows a larger spread (USD65–200/MWh) depending on the geographical location and the technology (Chapter 6). Therefore, assuming purchase price parity between ICEVs and BEVs, BEVs show lower LCC (Figure 10.5) due to higher efficiency and the lower cost of electricity compared to fossil fuels on a per-gigajoule (GJ) basis (secondary y-axis on Figure 10.5).

FCVs represent the most expensive solution for LDV, mainly due to the currently higher purchase price of the vehicle itself. However, given the lower technology readiness level of FCVs and the current efforts in the research and development of this technology, FCVs could become a viable technology for LDVs in the coming years. The issues regarding the extra energy involved in creating the hydrogen and its delivery to refuelling sites remain, however. The levelised cost of hydrogen on a per GJ basis is lower than conventional fossil fuels but higher than electricity. In addition, within the levelised cost of hydrogen, there are significant cost differences between the hydrogen-producing technologies. Conventional technologies such as coal gasification and steam methane reforming of natural gas, both with and without carbon capture and storage, represent the cheapest options ( [[#Bekel--2019|Bekel and Pauliuk 2019]] ; [[#Parkinson--2019|Parkinson et al. 2019]] ; [[#Khzouz--2020|Khzouz et al. 2020]] ; [[#Al-Qahtani--2021|Al-Qahtani et al. 2021]] ). Hydrogen produced via electrolysis is currently the most expensive technology, but with significant potential cost reductions due to the current technology readiness level.

<div id="box-10.3" class="h2-container box-container"></div>

<span id="box-10.3-vehicle-size-trends-and-implications-on-the-fuel-efficiency-of-ldvs"></span>
=== Box 10.3 | Vehicle Size Trends and Implications on the Fuel Efficiency of LDVs ===

<div id="h2-1-siblings" class="h2-siblings"></div>

'''Vehicle size trends.''' On a global scale, SUV sales have been constantly growing in the last decade, with 39% of the vehicles sold in 2018 being SUVs ( [[#IEA--2019d|IEA 2019d]] ). If the trend towards increasing vehicle size and engine power continues, it may result in higher overall emissions from the LDV fleet (relative to smaller vehicles with the same powertrain technology). The magnitude of the influence vehicle mass has on fuel efficiency varies with the powertrain, which have different efficiencies. Box 10.3 Figure 1 highlights this relationship using data from the same literature used to create Figure 10.4. Higher powertrain efficiency results in lower energy losses in operation, and thus requires less energy input to move a given mass than a powertrain of lower efficiency. This pattern is illustrated by the more gradual slope of BEVs in Box 10.3 Figure 1. The trend towards bigger and heavier vehicles, with consequently higher use phase emissions, can be somewhat offset by improvements in powertrain design, fuel efficiency, lightweighting, and aerodynamics ( [[#Gargoloff--2018|Gargoloff et al. 2018]] ; Wolfram et al. 2020). The potential improvements provided by these strategies are case specific and not thoroughly evaluated in the literature, either individually or as a combination of multiple strategies.

'''Lightweighting.''' There is an increasing use of advanced materials such as high-strength steel, aluminium, carbon fibre, and polymer composites for vehicle lightweighting ( [[#Hottle--2017|Hottle et al. 2017]] ). These materials reduce the mass of the vehicle and thereby also reduce the fuel or energy required to drive. Lightweighted components often have higher production emissions than the components they replace due to the advanced materials used ( [[#Kim--2016|Kim and Wallington 2016]] ). Despite these higher production emissions, some studies suggest that the reduced fuel consumption over the lifetime of the lightweighted vehicle may provide a net mitigation effect in comparison to a non-lightweighted vehicle ( [[#Kim--2013|Kim and Wallington 2013]] ; [[#Hottle--2017|Hottle et al. 2017]] ; [[#Milovanoff--2019|Milovanoff et al. 2019]] ; [[#Upadhyayula--2019|Upadhyayula et al. 2019]] ; Wolfram et al. 2020). However, multiple recent publications have found that in some cases, depending on, for example, vehicle size and carbon intensity of the lightweighting materials employed, the GHG emissions avoided due to improved fuel efficiency do not

offset the higher manufacturing emissions of the vehicle ( [[#Luk--2018|Luk et al. 2018]] ; [[#Wu--2019|Wu et al. 2019]] ). In addition, these advanced materials may be challenging to recycle in a way that retains their high technical performance ( [[#Meng--2017|Meng et al. 2017]] ).

'''Co-effects on particulate matter.''' Lightweighting may also alleviate the particulate matter (PM) emissions arising from road and brake wear. BEVs are generally heavier than their ICEV counterparts, which may potentially cause higher stress on road surfaces and tyres, with consequently higher PM emissions per kilometre driven ( [[#Timmers--2016|Timmers and Achten 2016]] ). Regenerative braking in HEVs, BEVs and FCVs, however, reduces the mechanical braking required, and therefore may compensate for the higher brake wear emissions from these heavier vehicle types. In addition, BEVs have no tailpipe emissions, which further offsets the increased PM emissions from road and tyre wear. Therefore, lightweighting strategies may offer a carbon and particulates mitigation effect; however, in some cases, other technological options may reduce CO 2 emissions even further.

[[File:aba244cce80340c9fefc467e765ca3a1 IPCC_AR6_WGIII_Box_10_3_Figure_1.png]]

'''Box 10.3, Figure 1''' '''| Illustrationof energy consumption as a function of vehicle size (using mass as a proxy) and powertrain technology.''' FCVs omitted due to lacking data.

<div id="10.4.2" class="h2-container"></div>

<span id="transit-technologies-for-passenger-transport"></span>
=== 10.4.2 Transit Technologies for Passenger Transport ===

<div id="h2-14-siblings" class="h2-siblings"></div>

Buses provide urban and peri-urban transport services to millions of people around the world and a growing number of transport agencies are exploring alternative-fuelled buses. Alternative technologies to conventional diesel-powered buses include buses powered with CNG, LNG, synthetic fuels, and biofuels (e.g., biodiesel, renewable diesel, dimethyl ether); diesel hybrid-electric buses; battery electric buses; electric catenary buses; and hydrogen fuel cell buses. Rail is an alternative mode of transit that could support decarbonisation of land-based passenger mobility. Electric rail systems can provide urban services (light rail and metro systems), as well as longer-distance transport. Indeed, many cities of the world already have extensive metro systems, and regions like China, Japan and Europe have a robust high-speed intercity railway network. Intercity rail transport can be powered with electricity, however, fossil fuels are still prevalent for long-distance rail passenger transport in some regions. Battery electric long-distance trains may be a future option for these areas.

Figure 10.6 shows the lifecycle GHG emissions from different powertrain and fuel technologies for buses and passenger rail. The data in each panel came from a number of relevant scientific studies ( [[#Cai--2015|Cai et al. 2015]] ; [[#Tong--2015a|Tong et al. 2015a]] ; [[#Dimoula--2016|Dimoula et al. 2016]] ; [[#de%20Bortoli--2017|de Bortoli et al. 2017]] ; [[#Valente--2017|Valente et al. 2017]] ; [[#Meynerts--2018|Meynerts et al. 2018]] ; [[#IEA--2019e|IEA 2019e]] ; [[#de%20Bortoli--2020|de Bortoli and Christoforou 2020]] ; [[#Hill--2020|Hill et al. 2020]] ; [[#Liu--2020a|Liu et al. 2020a]] ; [[#Valente--2021|Valente et al. 2021]] ). The width of the bar represents the variabilityin available estimates, which is primarily driven by variability in reported vehicle efficiency, size, or drive cycle. While some bars overlap, the Figure may not fully capture correlations between results. For example, low efficiency associated with aggressive drive cycles may drive the upper end of the emission ranges for multiple technologies; thus, an overlap does not necessarily suggest uncertainty regarding which vehicle type would have lower emissions for a comparable trip. Additionally, reported lifecycle emissions do not include embodied GHG emissions associated with infrastructure construction and maintenance. These embodied emissions are potentially a larger fraction of lifecycle emissions for rail than for other transport modes ( [[#Chester--2012|Chester and Horvath 2012]] ; [[#Chester--2013|Chester et al. 2013]] ). One study reported values ranging from 10–25 gCO 2 per passenger-kilometre ( [[#International%20Union%20of%20Railways--2016|International Union of Railways 2016]] ), although embodied emissions from rail are known to vary widely across case studies ( [[#Olugbenga--2019|Olugbenga et al. 2019]] ). These caveats are also applicable to the other figures in this section.

<div id="_idContainer030" class="Basic-Text-Frame"></div>

[[File:63378fe3140074f8a2a80581eec4c6f9 IPCC_AR6_WGIII_Figure_10_6.png]]

'''Figure 10.6 | Lifecycle greenhouse gas intensity of land-based bus and rail technologies.''' Each bar represents the range of the lifecycle estimates, bounded by minimum and maximum energy use per passenger-kilometre, as reported for each fuel/powertrain combination. The ranges are driven by differences in vehicle characteristics and operating efficiency. For energy sources with highly variable upstream emissions, low, medium and/or high representative values are shown as separate rows. The primary x-axis shows lifecycle GHG emissions, in gCO 2 -eq pkm –1 , assuming 80% occupancy; the secondary x-axis assumes 20% occupancy. The values in the figure rely on the 100-year GWP value embedded in the source data, which may differ slightly from the updated 100-year GWP values from WGI. For buses, the main bars show full lifecycle, with vertical bars disaggregating the vehicle cycle. ‘Diesel, high’ references emissions factors for diesel from oil sands. ‘advanced biofuels’, refers to the use of second-generation biofuels and their respective conversion and cultivation emissions factors. ‘IAM EMF33’ refers to emissions factors for advanced biofuels derived from simulation results from the integrated assessment models EMF33 scenarios. ‘PM’ refers to partial models, where ‘CLC’ is with constant land cover and ‘NRG’ is with natural regrowth. ‘DAC FT-Diesel, wind electricity’ refers to Fischer- Tropsch diesel produced via a CO 2 direct air capture process that uses wind electricity. ‘Hydrogen, low-carbon renewable’ refers to fuels produced via electrolysis using low-carbon electricity. ‘Hydrogen, natural gas SMR’ refers to fuels produced via steam methane reforming of natural gas. Results for ICEVs with ‘high emissions DAC FT-Diesel from natural gas’ are not included here since the lifecycle emissions are estimated to be substantially higher than petroleum diesel ICEVs.

Figure 10.6 highlights that BEV and FCV buses and passenger rail powered with low-carbon electricity or low-carbon hydrogen, could offer reductions in GHG emissions compared to diesel-powered buses or diesel-powered passenger rail. However, and not surprisingly, these technologies would offer only little emissions reductions if power generation and hydrogen production rely on fossil fuels. While buses powered with CNG and LNG could offer some reductions compared to diesel-powered buses, these reductions are unlikely to be sufficient to contribute to deep decarbonisation of the transport sector and they may slow down conversion to low- or zero-carbon options already commercially available. Biodiesel and renewable diesel fuels (from sources with low upstream emissions and low risk of induced land-use change) could offer important near-term reductions for buses and passenger rail, as these fuels can often be used with existing vehicle infrastructure. They could also be used for long haul trucks and trains, shipping and aviation as discussed below and in later sections.

There has been growing interest in the production of synthetic fuels from CO 2 produced by direct air capture (DAC) processes. Figure 10.6 includes the lifecycle GHG emissions from buses and passenger rail powered with synthetic diesel produced through a DAC system paired with a Fischer-Tropsch (FT) process, based on [[#Liu--2020a|Liu et al. (2020a)]] . This process requires the use of hydrogen (as shown in Figure 10.2), so the emissions factors of the resulting fuel depend on the emissions intensity of hydrogen production. An electricity emissions factor less than 140 gCO 2 -eq kWh –1 would be required for this pathway to achieve lower emissions than petroleum diesel ( [[#Liu--2020a|Liu et al. 2020a]] ); for example, this would be equivalent to a 75% wind and 25% natural gas electricity mix (Appendix 10.1). If the process relied on steam methane reforming for hydrogen production or fossil-based power generation, synthetic diesel from the DAC-FT process would not provide GHG emissions reductions compared to conventional diesel. DAC-FT from low-carbon energy sources appears to be promising from an emissions standpoint and could warrant the R&amp;D and demonstration attention outlined in the rest of the chapter, but it cannot be contemplated as a decarbonisation strategy without the availability of low-carbon hydrogen.

At high occupancy, both bus and rail transport offer substantial GHG reduction potential per pkm, even compared with the lowest-emitting private vehicle options. Even at 20% occupancy, bus and rail may still offer emission reductions compared to passenger cars, especially notable when comparing BEVs with low-carbon electricity (the lowest-emission option for all technologies) across the three modes. Only when comparing a fossil fuel-powered bus at low occupancy with a low-carbon powered car at high occupancy is this conclusion reversed. Use of public transit systems, especially those that rely on buses and passenger rail fuelled with the low-carbon fuels previously described, would thus support efforts to decarbonise the transport sector. Use of these public transit systems will depend on urban design and consumer preferences ( [[#10.2|Section 10.2]] , Chapters 5 and 8), which in turn depend on time, costs, and behavioural choices.

Figure 10.7 shows the results of a parametric analysis of the LCCs of transit technologies with the highest potential for GHG emissions reductions. As with Figure 10.5, the vehicle efficiency ranges are the same as those from the LCA estimates (80% occupancy). Vehicle, fuel, and maintenance costs represent ranges in the literature ( [[#Eudy--2018b|Eudy and Post 2018b]] ; [[#IEA--2019e|IEA 2019e]] ; [[#Argonne%20National%20Laboratory--2020|Argonne National Laboratory 2020]] ; [[#BNEF--2020|BNEF 2020]] ; [[#Eudy--2020|Eudy and Post 2020]] ; [[#Hydrogen%20Council--2020|Hydrogen Council 2020]] ; [[#IEA--2020b|IEA 2020b]] ; [[#IEA--2020c|IEA 2020c]] ; [[#IRENA--2020|IRENA 2020]] ; [[#Johnson--2020|Johnson et al. 2020]] ; [[#Burnham--2021|Burnham et al. 2021]] ; [[#IEA--2021c|IEA 2021c]] ; [[#IEA--2021d|IEA 2021d]] ; [[#US%20Energy%20Information%20Administration--2021|US Energy Information Administration 2021]] ), and the discount rate is 3% where applicable. Appendix 10.2 provides the details behind these estimates. The panels for the ICEV can represent buses and passenger trains powered with any form of diesel, whether derived from petroleum, synthetic hydrocarbons, or biofuels. For reference, global average automotive diesel prices from 2015–2020 fluctuated around USD1 per litre, and the 2019 world average industrial electricity price was approximately USD100 per MWh ( [[#IEA--2021d|IEA 2021d]] ). Retail hydrogen prices in excess of USD13 per kilogram have been observed( [[#Eudy--2018a|Eudy and Post 2018a]] ; [[#Argonne%20National%20Laboratory--2020|Argonne National Laboratory 2020]] ; [[#Burnham--2021|Burnham et al. 2021]] ) though current production cost estimates for hydrogen produced from electrolysis are far lower ( [[#IRENA--2020|IRENA 2020]] ) (and as reported in Chapter 6), at around USD5–7 per kg with future forecasts as low as USD1 per kg ( [[#BNEF--2020|BNEF 2020]] ; [[#Hydrogen%20Council--2020|Hydrogen Council 2020]] ; [[#IRENA--2020|IRENA 2020]] ) (and as reported in Chapter 6).

<div id="_idContainer032" class="Basic-Text-Frame"></div>

[[File:e0b77af9b26643bd31af8caba33cce7e IPCC_AR6_WGIII_Figure_10_7.png]]

'''Figure 10.7 | Lifecycle costs for internal combustion engine vehicles, battery electric vehicles, and hydrogen fuel cell vehicles for buses and passenger rail.''' The range of efficiencies for each vehicle type are consistent with the range of efficiencies in Figure 10.6 (80% occupancy). The results for the ICEV can be used to evaluate the lifecycle costs of ICE buses and passenger rail operated with any form of diesel, whether from petroleum, synthetic hydrocarbons, or biofuel, as the range of efficiencies of vehicles operating with all these fuels is similar. The secondary y-axis depicts the cost of the different energy carriers normalised in USD/GJ for easier cross-comparability.

Under most parameter combinations, rail is the most cost-effective option, followed by buses, both of which are an order of magnitude cheaper than passenger vehicles. Note that costs per pkm are strongly influenced by occupancy assumptions; at low occupancy (e.g., &lt;20% for buses and &lt;10% for rail), the cost of transit approaches the LCC for passenger cars. For diesel rail and buses, cost ranges are driven by fuel costs, whereas vehicles are both important drivers for electric or hydrogen modes due to high costs (but also large projected improvements) associated with batteries and fuel cell stacks. Whereas the current state of ICEV technologies is best represented by cheap vehicles and low fuel costs for diesel (top left of each panel), these costs are likely to rise in future due to stronger emission/efficiency regulations and rising crude oil prices. On the contrary, the current status of alternative fuels is better represented by high capital costs and mid-to-high fuel costs (right side of each panel; mid-to-bottom rows), but technology costs are anticipated to fall with increasing experience, research, and development. Thus, while electric rail is already competitive with diesel rail, and electric buses are competitive with diesel buses in the low efficiency case, improvements are still required in battery costs to compete against modern diesel buses on high efficiency routes, at current diesel costs. Similarly, improvements to both vehicle cost and fuel costs are required for hydrogen vehicles to become cost effective compared to their diesel or electric counterparts. At either the upper end of the diesel cost range (bottom row of ICEV panels), or within the 2030–2050 projections for battery costs, fuel cell costs and hydrogen costs (top left of BEV and FCV panels), both battery- and hydrogen-powered vehicles become financially attractive.

<div id="10.4.3" class="h2-container"></div>

<span id="land-based-freight-transport"></span>
=== 10.4.3 Land-based Freight Transport ===

<div id="h2-15-siblings" class="h2-siblings"></div>

As is the case with passenger transport, thereis growing interest in alternative fuels that could reduce GHG emissions from freight transport. Natural gas-based fuels (e.g., CNG, LNG) are an example, however these may not lead to drastic reductions in GHG emissions compared to diesel. Natural gas-powered vehicles have been discussed as a means to mitigate air quality impacts ( [[#Khan--2015|Khan et al. 2015]] ; [[#Cai--2017|Cai et al. 2017]] ; [[#Pan--2020|Pan et al. 2020]] ), but those impacts are not the focus of this review. Decarbonisation of medium- and heavy-duty trucks would likely require the use of low-carbon electricity in battery electric trucks, low-carbon hydrogen or ammonia in fuel-cell trucks, or bio-based fuels (from sources with low upstream emissions and low risk of induced land-use change) used in ICE trucks.

Freight rail is also a major mode for the inland movement of goods. Trains are more energy efficient (per tkm) than trucks, so expanded use of rail systems (particularly in developing countries where demand for goods could grow exponentially) could provide carbon abatement opportunities. While diesel-based locomotives are still a major mode of propulsion used in freight rail, interest in low-carbon propulsion technologies is growing. Electricity already powers freight rail in many European countries using overhead catenaries. Other low-carbon technologies for rail may include advanced storage technologies, biofuels, synthetic fuels, ammonia, or hydrogen.

Figure 10.8 presents a review of lifecycle GHG emissions from land-based freight technologies (heavy- and medium-duty trucks, and rail). Each panel within the figure represents data in GHG emissions per tonne-kilometre of freight transported by different technology and/or fuel types, as indicated by the labels to the left. The data in each panel came from a number of relevant scientific studies ( [[#Tong--2015a|Tong et al. 2015a]] ; [[#Frattini--2016|Frattini et al. 2016]] ; [[#Nahlik--2016|Nahlik et al. 2016]] ; [[#Zhao--2016|Zhao et al. 2016]] ; CE Delft 2017; [[#Isaac--2017|Isaac and Fulton 2017]] ; [[#Song--2017|Song et al. 2017]] ; [[#Valente--2017|Valente et al. 2017]] ; [[#Cooper--2019|Cooper and Balcombe 2019]] ; [[#Lajevardi--2019|Lajevardi et al. 2019]] ; [[#Hill--2020|Hill et al. 2020]] ; [[#Liu--2020a|Liu et al. 2020a]] ; [[#Merchan--2020|Merchan et al. 2020]] ; [[#Prussi--2020|Prussi et al. 2020]] ; [[#Gray--2021|Gray et al. 2021]] ; [[#Valente--2021|Valente et al. 2021]] ). Similar to the results for buses, technologies that offer substantial emissions reductions for freight include: ICEV trucks powered with the low-carbon variants for biofuels, ammonia or synthetic diesel; BEVs charged with low-carbon electricity; and FCVs powered with renewable-based electrolytic hydrogen, or ammonia. Since ammonia and Fischer-Tropsch diesel are produced from hydrogen, their emissions are higher than the source hydrogen, but their logistical advantages over hydrogen are also a consideration ( [[#10.3|Section 10.3]] ).

<div id="_idContainer034" class="Basic-Text-Frame"></div>

[[File:5815ee5fb13140b09cd642ab60fbb83b IPCC_AR6_WGIII_Figure_10_8.png]]

'''Figure 10.8''' | '''Lifecycle greenhouse gas intensity of land-based freight technologies and fuel types.''' Each bar represents the range of the lifecycle estimates, bounded by minimum and maximum energy use per tkm, as reported for each fuel/powertrain combination. The ranges are driven by differences in vehicle characteristics and operating efficiency. For energy sources with highly variable upstream emissions, low, medium and/or high representative values are shown as separate rows. For trucks, the primary x-axis shows lifecycle GHG emissions, in gCO 2 -eq tkm –1 , assuming 100% payload; the secondary x-axis assumes 50% payload. The values in the figure rely on the 100-year GWP value embedded in the source data, which may differ slightly from the updated 100-year GWP values from WGI. For rail, values represent average payloads. For trucks, main bars show full lifecycle, with vertical bars disaggregating the vehicle cycle. ‘Diesel, high’ references emissions factors for diesel from oil sands. ‘Advanced biofuels’ refers to the use of second-generation biofuels and their respective conversion and cultivation emission factors. ‘IAM EMF33’ refers to emissions factors for advanced biofuels derived from simulation results from the EMF33 scenarios. ‘PM’ refers to partial models, where ‘CLC’ is with constant land cover and ‘NRG’ is with natural regrowth. DAC FT-Diesel, wind electricity refers to Fischer- Tropsch diesel produced via a CO 2 direct air capture process that uses wind electricity. ‘Ammonia and Hydrogen, low-carbon renewable’ refers to fuels produced via electrolysis using low-carbon electricity. ‘Ammonia and Hydrogen, natural gas SMR’ refers to fuels produced via steam methane reforming of natural gas.

Trucks exhibit economies of scale in fuel consumption, with heavy-duty trucks generally showing lower emissions per tkm than medium-duty trucks. Comparing the lifecycle GHG emissions from trucks and rail, it is clear that rail using internal combustion engines is more carbon efficient than using internal combustion trucks. Note that the rail emissions are reported for an average representative payload, while the trucks are presented at 50% and 100% payload, based on available data. The comparison between trucks and rail powered with electricity or hydrogen is less clear – especially considering that these values omit embodied GHG from infrastructure construction. One study reported embodied rail infrastructure emissions of 15 gCO 2 per tonne-kilometre for rail ( [[#International%20Union%20of%20Railways--2016|International Union of Railways 2016]] ), although such embodied emissions from rail are known to vary widely across case studies ( [[#Olugbenga--2019|Olugbenga et al. 2019]] ). Regardless, trucks and rail with low-carbon electricity or low-carbon hydrogen have substantially lower emissions than incumbent technologies.

For trucks, Figure 10.8 includes two x-axes representing two different assumptions about their payload, which substantially influence emissions per tonne-kilometre. These results highlight the importance of truckload planning as an emissions reduction mechanism, for example, as also shown in [[#Kaack--2018|Kaack et al. (2018)]] . Several studies also point to improvements in vehicle efficiency as an important mechanism to reduce emissions from freight transport ( [[#Taptich--2016|Taptich et al. 2016]] ; [[#Kaack--2018|Kaack et al. 2018]] ). However, projections for diesel vehicles using such efficiencies beyond 2030 are promising, but still far higher emitting than vehicles powered with low-carbon sources.

Figure 10.9 shows the results of a parametric analysis of the LCC of trucks and freight rail technologies with the highest potential for deep GHG reductions. As with Figure 10.8, the vehicle efficiency ranges are the same as those from the LCA estimates (80% payload for trucks; effective payload as reported by original studies for rail). Vehicle, fuel and maintenance costs represent ranges in the literature ( [[#Moultak--2017|Moultak et al. 2017]] ; [[#Eudy--2018b|Eudy and Post 2018b]] ; [[#IEA--2019e|IEA 2019e]] ; [[#Argonne%20National%20Laboratory--2020|Argonne National Laboratory 2020]] ; [[#BNEF--2020|BNEF 2020]] ; [[#IRENA--2020|IRENA 2020]] ; [[#Burnham--2021|Burnham et al. 2021]] ; [[#IEA--2021c|IEA 2021c]] ), and the discount rate is 3% where applicable (Appendix 10.2). The panels for the ICEV can represent trucks and freight trains powered with any form of diesel, whether derived from petroleum, synthetic hydrocarbons, or biofuels. See discussion preceding Figure 10.7 for additional details about current global fuel costs. Under most parameter combinations, rail is the more cost-effective option, but the high efficiency case for trucks (representing fuel-efficient vehicles, favourable drive cycles and high payload) can be more cost-effective than the low efficiency case for rail (representing systems with higher fuel consumption and lower payload). For BEV trucks, cost ranges are driven by vehicle purchase price due to the large batteries required and the associated wide range between their current high costs and anticipated future cost reductions. For all other truck and rail technologies, fuel cost ranges play a larger role. Similar to transit technologies, the current state of freight ICEV technologies is best represented by cheap vehicles and low fuel costs for diesel (top left of each panel), and the current status of alternative fuels is better represented by high capital costs and mid-to-high fuel costs (right side of each panel; mid-to-bottom rows), with expected future increases in ICEV LCC and decreases in alternative fuel vehicle LCC. Electric and hydrogen freight rail are potentially already competitive with diesel rail (especially electric catenary ( [[#IEA--2019e|IEA 2019e]] )), but low data availability (especially for hydrogen efficiency ranges) and wide ranges for reported diesel rail efficiency (likely encompassing low capacity utilisation) makes this comparison challenging. Alternative fuel trucks are currently more expensive than diesel trucks, but future increases in diesel costs or a respective decrease in hydrogen costs or in BEV capital costs (especially the battery) would enable either alternative fuel technology to become financially attractive. These results are largely consistent with raw results reported in existing literature, which suggest ambiguity over whether BEV trucks are already competitive, but more consistency that hydrogen is not yet competitive, but could be in future ( [[#Zhao--2016|Zhao et al. 2016]] ; [[#Moultak--2017|Moultak et al. 2017]] ; [[#Sen--2017|Sen et al. 2017]] ; [[#White--2017|White and Sintov 2017]] ; [[#Zhou--2017|Zhou et al. 2017]] ; [[#Mareev--2018|Mareev et al. 2018]] ; [[#Yang--2018a|Yang et al. 2018a]] ; [[#El%20Hannach--2019|El Hannach et al. 2019]] ; [[#Lajevardi--2019|Lajevardi et al. 2019]] ; [[#Tanco--2019|Tanco et al. 2019]] ; [[#Burke--2020|Burke and Sinha 2020]] ; [[#Jones--2020|Jones et al. 2020]] ). There is limited data available on the LCC for freight rail, but at least one study IEA (2019g) suggests that electric catenary rail is likely to have similar costs to diesel rail, while battery electric trains remain more expensive and hydrogen rail could become cheaper under forward-looking cost reduction scenarios.

<div id="_idContainer037" class="Basic-Text-Frame"></div>

[[File:cdb50a6cd47a32b076eaf064e5ca4533 IPCC_AR6_WGIII_Figure_10_9.png]]

'''Figure 10.9 | Life cycle costs for internal combustion engine vehicles, battery electric vehicles, and hydrogen fuel cell vehicles for heavy-duty trucks and freight rail.''' The range of efficiencies for each vehicle type are consistent with the range of efficiencies in Figure 10.8. The results for ICEV can be used to evaluate the lifecycle costs of ICE trucks and freight rail operated with any form of diesel, whether from petroleum, synthetic hydrocarbons, or biofuels, as the range of efficiencies of vehicles operating with all these fuels is similar. The secondary y-axis depicts the cost of the different energy carriers normalised in USD per GJ for easier cross-comparability.

<div id="10.4.4" class="h2-container"></div>

<span id="abatement-costs"></span>
=== 10.4.4 Abatement Costs ===

<div id="h2-16-siblings" class="h2-siblings"></div>

Taken together, the results in this section suggest a range of cost-effective opportunities to reduce GHG emissions from land-based transport. Mode shift from cars to passenger transit (bus or rail) can reduce GHG emissions while also reducing LCCs, resulting in a negative abatement cost. Likewise, increasing the utilisation of vehicles (i.e., % occupancy for passenger vehicles or % payload for freight vehicles) simultaneously decreases emissions and costs per pkm or per tkm, respectively. Within a given mode, alternative fuel sources also show strong potential to reduce emissions at minimal added costs. For LDVs, BEVs can offer emissions reductions with LCCs that are already approaching that for conventional ICEVs. For transit and freight, near-term abatement costs for the low-carbon BEV and FCV options relative to their diesel counterparts range from near USD0/tonne CO 2 -eq (e.g., BEV buses and BEV passenger rail) into the hundreds or even low thousands of dollars per tonne CO 2 -eq (e.g., for heavy-duty BEV and FCV trucks at current vehicle and fuel costs). With projected future declines in storage, fuel cell, and low-carbon hydrogen fuel costs, however, both BEV and FCV technologies can likewise offer GHG reductions at negative abatement costs across all land-transport modes in 2030 and beyond. Further information about costs and potentials is available in Chapter 12.

<div id="10.5" class="h1-container"></div>

<span id="decarbonisation-of-aviation"></span>