2

Setting the Stage: Regulatory Horizons, Challenges, and Influences

The fuel consumption and greenhouse gas (GHG) emissions of medium- and heavy-duty vehicles (MHDVs) are affected by the intersection of a variety of technological factors, market forces, regulatory programs, and other government and private-sector policies and initiatives. Before examining specific technologies in the main body of this report, this chapter provides the relevant regulatory, policy, and market background and context for the present study. It first describes the future prospects and projections for the regulation of MHDV fuel consumption and GHG emissions in the period from 2030 to 2050 when potential future Phase III standards might be implemented (Section 2.1). It then addresses some key gaps and challenges of the National Highway Traffic Safety Administration (NHTSA)/Environmental Protection Agency (EPA) program (Section 2.2). Other regulatory programs directly or indirectly affecting MHDV fuel consumption and GHG emissions are summarized next (Section 2.3).

2.1 FUTURE REGULATORY PATHWAYS

This report focuses on a potential future Phase III set of NHTSA/EPA MHDV standards (i.e., likely affecting model year (MY) 2028 at the earliest for EPA, and MY 2030 at the earliest for NHTSA). These vehicles will in many cases remain in operation through the middle of the century. Because this report is addressing technologies and strategies for reducing fuel consumption and GHG emissions from MHDVs that will be fully implemented more than a decade into the future, we necessarily take a prospective and horizon-scanning approach in this analysis. Over the time period in which Phase III standards might be adopted and implemented, many important changes and advancements are likely in vehicle technology and costs, transportation systems and planning, fuel availability and price, other measures and policies affecting MHDV fuel economy and GHG emissions, and in the economic and political circumstances of the nation and world. Thus, the forward-looking analysis in this report will obviously necessitate various projections, assumptions, and uncertainties, which will be identified throughout this report.

This section proceeds in three parts. First, projections of future fleet characteristics are discussed, focused on the likely time period for adoption and implementation of the Phase III standards (Section 2.1.1). Next, the potential pathways for reducing fuel consumption and GHG emissions over this time period are assessed (Section 2.1.2). Finally, the timing opportunities and challenges in planning for the Phase III rulemaking based on these analyses are discussed (Section 2.1.3).

2.1.1 Fleet Characteristics During the Phase III Standards Time Horizon

The direction and stringency of a future set of Phase III standards will be dictated by the intersection of technological feasibility, market demands, economic trends, and the policy objectives set by the federal government for reducing fuel consumption and GHG emissions from MHDVs. However, the uncertainties regarding the future characteristics of the MHDV fleet are significant. The extent to which the fleet characteristics are likely to change is difficult to predict. Changes in the fleet may result from changes in the demand for goods and services; the competing alternatives available to provide those goods and services; fuel availability and fuel prices; future standards and regulations; and other policies and exogenous factors such as changes in the economy or change in fuel prices. Even in the absence of policy and regulatory changes, projections on the composition of the fleet fall short of being accurate. In addition, uncertainty in national and international policies creates further difficulties in providing projections.

To illustrate this uncertainty see Figure 2-1, which affects the transportation sector overall. Every year, the Energy Information Administration (EIA) produces estimates of future energy consumption for the next several decades. Seven of these annual estimates, made in calendar years 1985 to 2015, are shown. Also shown is the actual U.S. transportation energy consumption for this period. As the data show, the EIA estimates of future transportation energy consumption often deviate substantially from actual consumption, illustrating the difficulty of making accurate long-term projections. These deviations between projections and actual consumption result from two kinds of uncertainties: (i) uncertainties relating to structural changes in energy markets such as the 1970s OPEC embargo and (ii) fluctuations in supply, demand, and prices within existing structural markets.

The most recent EIA Annual Energy Outlook (AEO; EIA, 2015) projects that the number of vehicle miles traveled by freight trucks (> 10,000 pounds) and commercial light trucks will continue to increase, from about 284 and 72 billion vehicle miles, respectively, in 2015 to about 400 and 105 billion vehicle miles traveled, respectively, in 2040 (see Figure 2-2). AEO 2015 also includes projected fuel efficiency (in miles per gallon) for freight trucks

(> 10,000 pounds), which is expected to increase to 7.7 mpg in the late 2030s and then plateau or even slightly decrease (see Figure 2-3). This projection reflects the implementation of the Phase I MHDV fuel efficiency and GHG standards, but not the Phase II or subsequent standards or any other future standards (EIA, 2015, p. 10).

Fuel consumption from freight trucks (> 10,000 pounds) is also expected to increase according to the EIA AEO’s most recent report, from about 2.8 million barrels per day oil equivalent in 2015 to about 3.4 in 2040, a 21 percent increase (see Figure 2-4). Again, this projection assumes implementation of the Phase I MHDV standards, but not any subsequent potential standards.

Finding: Technological, market, economic, and policy variables make future projection of fuel consumption and GHG emissions by MHDVs highly uncertain. This can result from both structural changes within energy markets and price volatility and fluctuations within market structures.

Recommendation 2-1: Given the uncertainties in projecting future MHDV fuel consumption and GHG emission trends, periodic or ongoing assessments should be made by NHTSA, in cooperation with EPA, to evaluate the implementation of existing fuel efficiency and GHG emission standards for MHDVs, and to adjust planning for future standards.

2.1.2 Assessing the GHG Reductions and Fuel Savings Potential

2.1.2.1 International Studies and Insights

Fuel savings and GHG reductions from transport are each affected and informed by international climate agreements and voluntary goals such as the Kyoto Protocol and the goals adopted in the 21st Conference of Parties in Paris. For example, the European Union adopted aggressive targets under its “20-20-20” policy of reducing greenhouse gas emissions by 20 percent relative to 1990 and achieving this with 20 percent renewable energy and 20 percent improvement in energy efficiency by 2020 (European Commission, 2008) and is now moving to even

more ambitious goals. Large GHG reductions in the transportation sector may include strategies and technologies such as vehicle electrification, fuel switching to biofuels derived from biomass, etc. (IEA, 2015).

Meeting large GHG reduction goals at a national or regional level will require a portfolio of strategies in all sectors, including those outside of transportation such as electricity, buildings, and industry, where mitigation strategies may be the most cost effective. Several studies have considered pathways to stabilize global temperature rise at 2 degrees Celsius (°C) above pre-industrial levels and made varied observations about the abatement of GHGs that might be achievable in different sectors. For example, the Organisation for Economic Co-operation and Development’s (OECD’s) International Energy Agency (IEA; 2010), found that most of the abatement occurs in the electricity sector—a finding shared by Energy Modeling Forum 24 (Clarke et al., 2014, p. 21) and by the Transportation Research Board (2011). In transportation, the IEA’s (2015) scenarios, for example, showed electrification in light-duty vehicles to exceed 40 percent of new cars by 2040. The Global Energy Assessment (GEA, 2012) evaluated a group of energy efficient pathways consistent with 2-degree stabilization that envisaged increased technical efficiency of all transportation coupled with mode shifting to mass transit for personal transportation in developed countries by 2050 and the increase in goods movement is fulfilled mostly by rail transport. The Intergovernmental Panel on Climate Change (IPCC, 2014, p. 24), estimated that, on a global basis, GHG emissions in the transportation sector could be reduced 15 to 40 percent below business-as-usual values by 2050 using a mix of technological and behavioral changes (such as mode shifting), transit-oriented development, and other considerations. Taptich et al. (2015) assessed the potential for GHG intensity reductions for passenger and freight transportation that would be achievable in 2030 and 2050 through the adoption of alternative technologies, mode switching, and electrification across eight regions of the world. The authors argue that non-OECD countries have lower reduction potential than OECD countries for all modes except oceangoing vessels, and that the reduction of GHG intensity occurs more slowly for freight transportation when compared to passenger transportation.

Though the studies do not converge on a particular technology or fuel pathway, all studies find that to achieve deep GHG reductions transformation in the energy system must occur well before 2050, anticipated and driven by large investment in the preceding decades. The Deep Decarbonization Pathways Project (DDPP) of the sustainable Development Solutions Network study finds that incremental investments in transportation of a few hundred billion dollars per year would be needed, peaking in about 2040 before declining as the new technologies became cheaper (DDPP, 2015). The lack of convergence in the results of these models could be due to uncertainty surrounding both the readiness of the various technologies and their potential to abate GHG emissions (Clarke et al., 2014, p. 29).

2.1.2.2 National Studies and Insights

At the national level, several studies have examined emissions reduction potential for the U.S. MHDV sector. The U.S. contribution to the DDPP (2015) examined possible pathways to reduce the aggregate emissions of the U.S. energy system by 80 percent below 1990 levels by 2050. For heavy-duty diesel engines, the study’s modeling assumptions led to scenarios where drop-in replacement biomass-derived fuels dominated, and other scenarios found substantial deployment of heavy-duty vehicles using hydrogen or natural gas. All cases that met the 80 percent economy-wide reduction target showed substantial improvement of 50 percent or greater in heavy-duty vehicle fuel economy. The DDPP scenarios for heavy-duty vehicle (HDV) emissions reduction as part of a national, economy-wide 80 percent GHG emissions reduction strategy are shown in Figure 2-5.

In economy-wide scenarios, MHDV emissions do not generally fall to 80 percent of 1990 levels by 2050, with other sectors taking on larger reductions. If the U.S. government were to set a goal of 80 percent GHG emissions reduction in the MHDV sector specifically,¹ rather than as an economy-wide goal, even more aggressive measures and strategies for MHDV emissions reduction would likely be needed. One study of freight transported by MHDVs in California found that GHG emissions could be reduced in that subsector by 80 percent by 2050 from a 1990 baseline by implementing zero emission vehicle and fuel technologies at an aggressive rate (Nahlik et al., 2015),

___________________

¹ This discussion of an 80 percent emissions reduction goal in the MHDV sector is provided for illustrative purposes only, without implying endorsement of any specific goal, a matter for policy makers and beyond the scope of this report.

but that those reductions would be overwhelmed by increased freight traffic, leading to a net increase in GHG emissions under a scenario involving incremental improvements in existing technologies.

Other scenario analyses have found a similar need for aggressive technology shifts and timelines to meet a goal of 80 percent reduction in MHDV emissions by 2050 (Fulton and Miller, 2015; Sharpe, 2013). The results of the Fulton and Miller scenario analysis for the United States (extrapolated from a model for California) are shown below (Figure 2-6). According to the authors, a substantial growth is expected in MHDV carbon dioxide (CO₂) emissions due to increased goods movement activity, which makes achieving steep reductions in emissions by 2050 very challenging. To achieve an 80 percent reduction in CO₂ emissions from 2010 levels, two scenarios were constructed. One included a near complete turnover of the MHDV fleet to electric and hydrogen fuel cell trucks using very low carbon energy sources. A second scenario included a mix of these technologies (roughly 50 percent of new truck sales by 2050) plus widespread use of very low carbon biodiesel fuel (60 percent blend in conventional diesel).

Yang et al. (2009) assessed whether California could achieve a reduction in transportation GHGs by 80 percent below 1990 levels by 2050. They found that the goal could be achieved with a concerted effort that would include changes in travel behavior, in vehicles and fuels, and argued for a portfolio approach. Morrow et al. (2010) used the National Energy Modeling System (NEMS) from the EIA to assess the implications of several policy scenarios (economy-wide CO₂ prices, fuel taxes, fuel economy standards, purchase tax credits for new vehicle purchases, and combinations of these policies). They concluded that a GHG emissions reduction goal 14 percent below 2005 levels by 2020 cannot be achieved. In Table 2-1 we summarize the studies reviewed, the climate mitigation goals considered, and the suggested alternatives for emissions reductions.

These findings suggest that additional approaches beyond improving the efficiency of conventional combustion engines, trucks, and trailers will be necessary in the 2030-2050 time frame if 80 percent reductions in GHG

TABLE 2-1 Summary of Studies Reviewed That Included Large GHG Emissions Reductions Goals for MHDVs in the United States or a Region Within the United States

emissions for the medium- and heavy-duty sector are to be achieved. At the same time, the above studies noted that meeting these aggressive goals can be done more cost effectively by making the deepest emissions cuts in energy-intensive sectors other than transportation, a strategy consistent with international agreements. In this latter instance, or if a less ambitious GHG reduction goal than 80 percent reduction by 2050 were to be chosen as a national goal (either for the economy overall or the MHDV sector specifically), more flexibility would exist in terms of the types and timing of technologies needed to meet those less stringent goals. As discussed in this report, some of the approaches that may be feasible over the coming decades include electrified drivetrains, low-carbon fuels, and more efficient means of moving goods such as modal shifts. These more radical changes could be described as revolutionary, rather than evolutionary, approaches and will likely require significant lead time to implement. To meet a 2050 milestone that achieves large GHG reductions, fleet turnover can absorb a decade, and another decade may be required for the development and phase-in of advanced propulsion technologies and new fuel and/or freight movement infrastructure. Thus the technologies conceived by 2030 will take until 2050 to completely diffuse into the fleet. Figure 2-7 shows the Phase III standards taking place around 2030 (shown above the blue line), with advanced technology commercialization and alternative fuel infrastructure rollout through approximately 2040, followed by 10 years of fleet turnover to arrive at the full effect of the low-carbon vehicles and fuels at about 2050.

Finding: The MHDV fleets of the next two decades will be comprised of not only traditional technologies such as gasoline- and diesel-powered vehicles, but will likely also include a variety of alternative technologies powered by a variety of energy sources.

2.1.2.3 Public Investment in Energy Research and Development

The public-sector energy research enterprise has relied on a portfolio approach that includes investment in a range of technologies. These portfolios can be explicitly designed to be robust (i.e., to provide technology options) across a variety of possible future states of the world (e.g., one that is carbon constrained or has very high oil prices) as a hedge against the uncertainty of long-range planning (NRC, 2007). Operationalizing a portfolio approach can be challenging, but it requires extending decision points as far into the future as possible without sacrificing the overall objective in order to maximize the number of feasible options that can be considered and developed for as long as possible. Other factors complicate the task of projecting the future incumbent technologies. For

example, attractive technology options can emerge from innovation pathways unforeseen today (NRC, 2009). In addition, present-day estimates may not show a clear winner when, for example, the cost of electricity from several technologies is compared (NRC, 2009, p. 57). Historically, such portfolios yield a mix of market-ready technologies with realized economic and environmental benefits as well as so-called knowledge benefits derived from research results that did not lead to a market-ready technology (NRC, 2001). The portfolio approach obviates the need to choose one technology in which to invest.

The history of technology innovation and policy provides several lessons with respect to the long lead time required to fully implement advanced technologies, fuels, and freight efficiency for the technology pathways that may be considered in a third phase of rulemaking:

First, developing a portfolio of technologies will increase the likelihood that the most effective advanced technologies, fuels, and freight movement approaches—capable of meeting longer-term goals—will be available to support compliance with a Phase III rulemaking. The goal here is not to have the government “pick winners,” which often is unsuccessful (Marchant, 2014; Nelson and Langlois, 1983) given the many challenges and uncertainties in predicting future technology pathways and timelines. Rather, the goal should be for both government and the private sector to play an active role in ensuring a robust portfolio of technology options is available for consideration, which usually does not happen without active nourishment of the most promising options by both the private and public sectors (Jaffe et al., 2005).

The investment needed to develop such a portfolio of technology options and deploy those that are market ready might be less effective were there to be over-investment in technologies that provide only modest energy consumption and GHG reductions beyond what can be achieved by conventional technologies (Nahlik et al., 2015). In addition, the opportunity cost of misdirected investment could slow progress in the more advanced technologies that in some cases might achieve larger GHG reductions and could adversely affect economic competitiveness. Once made, these investments are unlikely to be pushed aside in favor of a more promising advanced technology and/or fuel option that could achieve greater reductions in fuel consumption and GHG emissions, and decision makers would need to deal with a long-lasting and potentially suboptimal infrastructure.

The timing of the investment in the portfolio is also critical: If this investment waits until after a Phase III rulemaking has commenced, the available technology choices may not evolve beyond the mix of diesel and natural gas combustion engines and trucks and tractors that is currently available. The timeline for technological readiness and implementation of additional options—including advanced technologies, fuels, and freight movement approaches—will accordingly be delayed, beyond 2050.

Second, proactive government involvement to provide clearer signals about future environmental performance requirements is essential to drive private innovation in the direction of longer-term solutions and technologies (Norberg-Bohm, 1999).

Third, limiting the scope of the third phase of rulemaking to consider only those technologies and approaches for which NHTSA and EPA have direct authorities (e.g., standards for engines, trucks, and conventional fuels) will limit the opportunities to achieve the maximum feasible and cost-effective GHG and fuel consumption reductions. For example, involvement of agencies with authorities or influence over goods movement (e.g., Federal Highway Administration [FHWA]) or alternative fuels (Department of Energy [DOE]) in the development of the rulemaking could identify credits or incentives to be implemented in parallel with the rulemaking that would provide performance-based rewards to industry for achieving the best-performing advanced technologies, fuels, and goods movement methods.

Finding: Implementation of new propulsion technologies, new low-carbon fuels, and more efficient freight operations and logistics may offer the opportunity to reduce GHG emissions beyond what is achievable from improving the efficiency of combustion engine MHDVs.

Recommendation 2-2: NHTSA in cooperation with EPA should establish what MHDV GHG and fuel consumption reductions need to be achieved in the 2030 to 2050 time frame, consistent with any national goals and international commitments the United States has adopted.

Finding: The commercialization and deployment of new propulsion technologies, low-carbon fuels, and more efficient freight movement methods that are significantly different from those currently in use may need to start as early as 2030 if ambitious national GHG emissions reduction and fuel economy goals are established for the 2050 time frame. Because of the long lead time needed to develop such advanced technologies and strategies, planning and preparation for such measures must begin well before the target dates for commercialization and implementation.

Finding: The current regulatory structure which focuses on engine and truck fuel consumption and GHG emission standards does not appear to have regulatory flexibility to address new fuels and methods of improving the efficiency of goods movement. For example, increases in MHDV size and weight limits could reduce fuel consumption; however, the authority to make such changes lies with another agency not part of this rulemaking process.

Recommendation 2-3: NHTSA should coordinate with EPA to identify a portfolio of technologies and options that could achieve national GHG emission and fuel economy goals for the 2030 to 2050 time frame and to identify the research, planning, and preparation needed to commercialize and implement such technologies and options in the relevant time frame. NHTSA should also coordinate with EPA to engage other agencies in the rulemaking process, such as DOE, the Federal Motor Carrier Safety Administration (FMCSA), and FHWA, who have authorities that can facilitate commercialization of low-carbon fuels and more efficient freight movement methods. NHTSA in cooperation with EPA should evaluate how incentives or other regulatory provisions can be incorporated to facilitate implementation of fuels and freight movement approaches that lie outside of NHTSA and EPA authorities.

2.1.3 Timing: An Opportunity and a Challenge for the Phase III Rulemaking

The long time period until Phase III standards might be proposed, adopted, and implemented leads to two considerations. First, the situation is very different than the NHTSA and EPA’s Phase I and Phase II rulemaking, in which the agencies noted they lacked time and resources to address some fundamental issues that therefore had to be reserved for future rulemakings.² The long interim period provides a window of opportunity to undertake much of the preparation, groundwork, and identification and mitigation of obstacles for some fundamental and broader strategies for effectively controlling GHG emissions and fuel consumption by MHDVs. These issues may include, but are not limited to, development of an effective in-use compliance strategy, a methodology for analyzing and comparing life-cycle GHG emissions of different fuel options, a strategy for integrating fuel economy and emission standards with other policies and methods for reducing GHG emissions and fuel consumption (e.g., intermodal shifts, weight and height regulations, carbon tax, etc.), and a technological roadmap for deployment of vehicles that can potentially have near-zero emissions such as electric and fuel cell vehicles. In the Phase I (EPA and NHTSA, 2011) and Phase II (EPA and NHTSA, 2016) rulemakings, the agencies identified a number of such future issues that they did not have sufficient time and resources to address in those rulemakings, including addressing innovative and integrated designs of tractor-trailer combinations, analyzing the life-cycle consequences of different fuel options, and designing a comprehensive in-use compliance and enforcement program.

Second, the long time period during which NHTSA’s Phase II regulations on fuel consumption will be in effect creates an opportunity and need for a midcourse evaluation of the preparations for any future regulations. This would evaluate the implementation progress of the Phase II regulations and other developments in technology availability and feasibility. The Phase II standards—covering model years 2021 to 2027—and any follow-on regulations, which may then be in the early stages of formulation, both will depend on many assumptions about technology availability, cost, implementation, and effectiveness. An interim evaluation would provide an opportunity to assess those assumptions and projections, and to make a determination whether any adjustments may be

___________________

² For example, in the Phase I rulemaking NHTSA and EPA stated that they would reserve consideration of requirements applicable to trailers to a future rulemaking. In the Phase II rule, the agencies postponed consideration of better integrated tractor-trailer design changes and life-cycle analysis of natural gas and other alternative fuels.

appropriate in the preparations for any future rulemaking. This interim evaluation could also coincide with the next National Academies of Sciences, Engineering, and Medicine report, required every 5 years.

An interim evaluation of its MHDV regulation in the 2021-2022 time period would help improve the latter’s overall effectiveness and value contribution. Its primary focus would be on preparations for any future regulations beyond the Phase II standards.

Finding: NHTSA and EPA have a decade between the promulgation of the second and implementation of the third set of MHDV standards to develop more fundamental approaches and strategies for transforming the MHDV sector, such as planning for the expanded use of low-carbon fuels, development of zero emission technologies, and approaches to increasing the efficiency of moving goods. Such approaches could form a more comprehensive and effective program of reducing fuel consumption and GHG emissions from MHDVs.

Recommendation 2-4: To avoid the problems of having inadequate time and resources to address fundamental issues such as life-cycle analysis and technological, economic, and environmental analysis, in-use compliance and enforcement, and more integrated and comprehensive fuel efficiency and GHG emissions reduction policies, NHTSA in cooperation with EPA should be proactive in fully utilizing the 10-year window before a possible third phase of standards is implemented to sponsor and conduct studies, analyses, pilot projects, workshops, and collaborative efforts, including discussions with other relevant government agencies, to advance, anticipate, and address opportunities and challenges for long-term strategies for achieving more ambitious emissions and fuel consumption reductions in the 2030s and beyond.

Finding: The Phase II regulations, and any post-2027 regulations then being contemplated, will depend on many assumptions about technology availability, cost, implementation, and effectiveness. An interim evaluation would provide an opportunity to assess those assumptions and projections, and to make a determination whether any adjustments may be appropriate in the preparations for future rulemakings. This interim evaluation would also coincide with the next National Academies report, required every 5 years.

Recommendation 2-5: NHTSA, in coordination with EPA, should conduct an interim evaluation of the MHDV program in the 2021-2022 time period, focusing on adjustments to and preparations for future revisions to the regulations.

2.2 KEY REGULATORY GAPS AND CHALLENGES

In addition to the timing issues discussed above, several complexities and challenges in regulating MHDVs result in inherent gaps and limitations for any regulatory program that seeks to regulate such a diverse set of vehicles. Complexities include defining fuel economy for regulatory purposes; the effects of exogenous factors such as volatility in fuel prices; the need to consider vehicle components other than the engine and powertrain since these affect fuel consumption and GHG emissions; the significance of vehicle weight and length for the payload carried (and thus the freight efficiency); the variety and diversity of vehicle types and functions that complicate manufacturing, regulation, and enforcement; the diversity in the size and capabilities of companies in the MHDV sector; differences in the results of life-cycle analyses of different fuel options; challenges in ensuring effective certification and in-use compliance; and miscellaneous other factors affecting fuel consumption and greenhouse gas emissions.

2.2.1 Defining Fuel Economy

The first complexity is in defining fuel economy for MHDVs. The terms fuel economy and fuel consumption are both used to show the efficiency of fuel use in vehicles. Fuel economy is a measure of how far a vehicle will go with a gallon of fuel and is expressed in miles per gallon (mpg). Fuel consumption is the inverse measure—the amount of fuel consumed in driving a given distance—and is measured in units such as gallons per 100 miles.

Fuel consumption is a physical measure and is useful because it can be related directly to the goal of decreasing the amount of fuel required to travel a given distance. A key concept for MHDVs is load-specific fuel consumption (LSFC) (see NRC, 2010, p. 25). The primary purpose of most medium- and heavy-duty vehicles is to deliver freight or passengers (the payload). An extreme example illustrates the problem: A simple way to reduce the fuel consumption of a truck is to leave the cargo on the loading dock. This approach, however, ignores the purpose of these vehicles. Therefore, the way to represent an attribute-based fuel consumption metric is to normalize the fuel consumption to the payload that the vehicle hauls. Adding payload to a vehicle increases fuel consumption, but the higher payload actually improves the efficiency of the vehicle (in terms of LSFC). The Phase I rule employs the U.S. customary system of units, and LSFC is expressed as gallons per 1,000 ton-miles.

2.2.2 Volatile Fuel Prices

A second complexity is the uncertainty and volatility in fuel prices. Fleet owners will make purchasing and operating decisions based on expectations of what fuel prices are likely to be in the future. However, accurate projections of oil prices have proven to be difficult. For example, every year, the EIA of the Department of Energy produces a set of projections of energy quantities using their NEMS, a general equilibrium model. Figure 2-8 shows the projections for oil prices in the dashed grey lines that were produced by the EIA in its AEOs over time, for projections made since year 1985. The actual oil price is shown by the black line. All values are in constant dollars. Two messages come across: (1) oil prices are hard to predict and (2) there is substantial volatility in the prices, and that volatility seems to have increased in recent years.

One potential strategy currently being discussed is the use of natural gas for transportation. In that regard, it is worth pointing out that the same issues mentioned for oil—volatility and difficulty in projecting the prices—apply to natural gas prices, as shown in Figure 2-9.

2.2.3 Factors Other than Engine/Powertrain Affecting Fuel Consumption/GHG Emissions

Improving the efficiency of the engine and drivetrain is an important approach to reducing fuel consumption and CO₂ emissions; however, a third complexity is that many factors impact fuel consumption, aside from engine and powertrain operation. Of the fuel consumed by a Class 8 tractor-trailer moving at constant highway speed, roughly 40 percent provides tractive power to overcome road load—aerodynamic drag and rolling resistance. The National Research Council (NRC, 2014) found that reducing tractor and trailer aerodynamic drag and tire rolling resistance were viable methods of reducing fuel consumption and CO₂ emissions.³ The remainder of the energy is

___________________

³ Note the energy use of a lower-speed vocational (urban) truck will differ, with stop-and-go driving and tire rolling resistance more important than aerodynamics.

consumed by accessories, parasitic loads, and friction and other inefficiencies in the engine and drivetrain (NRC, 2014, p. 68). This subsection briefly identifies potential opportunities to reduce fuel consumption and GHG emissions using strategies other than those focused on the engine and drivetrain. More detailed analysis of some of these strategies is provided later in this report, and a full discussion of vehicle characteristics is presented in Chapter 5.

2.2.3.1 Trailers

Trailers provide an opportunity to reduce CO₂ emissions and fuel consumption. As of 2009, there were about 5.7 million trailers commercially registered in the United States (Roberts, 2016). About half are 53-foot or longer dry van or refrigerated van trailers. The remainder is composed of shorter van trailers, or specialty trailers such as tankers or flat beds. The use of boat tails, fairings, and side skirts has the potential to reduce fuel consumption caused by dry van trailers by up to 10 percent (NRC, 2010, pp. 99-100). Use of low-rolling-resistance tires on a tractor and trailer can reduce fuel consumption by about 3 percent, and tire pressure monitoring and inflation systems can help ensure fuel consumption does not increase due to tire under-inflation (NRC, 2014).

California has adopted GHG standards for trailers, which now are being implemented. This regulation requires tractor operators pulling trailers (53-foot or longer dry or refrigerated vans) to use trailers that improve fuel economy of the tractor-trailer by at least 5 percent compared to a trailer without aerodynamic features. Observations made in California and Arizona showed a greater proportion of trailers using aerodynamic devices than did observations made in Oregon, Texas, Michigan, Pennsylvania, and Maryland, where travel to California is less likely (NRC, 2014, p. 83).⁴

The 2014 NRC report, Reducing the Fuel Consumption and Greenhouse Gas Emissions of Medium- and Heavy-Duty Vehicles, Phase 2: First Report (the first report of National Academy of Sciences “Phase Two”), included a recommendation that the federal government adopt by regulation fuel consumption and GHG standards for new van (box) trailers, and to evaluate if other types of trailers should be included in the regulation. The EPA/NHTSA Phase II rule adopted GHG standards for new trailers beginning in 2018 (fuel consumption standards are mandatory beginning in 2021). The standards increase in stringency through 2027. Manufacturers of new 50-foot and longer box trailers (both dry and refrigerated) must achieve the greatest reduction in GHG emissions and fuel consumption, which may be achieved with a combination of aerodynamic devices, low-rolling-resistance tires, and tire monitoring or inflation systems. Other types of trailers have less stringent standards based on difficulties in using some technologies, and others may comply with design standards based on use of low-rolling-resistance tires and tire pressure monitoring or inflation systems. CO₂ emissions reductions range from 2 to 9 percent compared to what would be expected in the absence of the regulation. Although the Phase II regulations respond to the recommendations of the NRC Phase Two First Report to adopt performance standards for most new trailers, opportunities to further reduce the energy use caused by trailers, such as trailer redesign, aerodynamic improvements to non-box trailers, weight reduction, and better integration with tractors, may be possible in the time frame of a third phase of rulemaking.

2.2.3.2 Tires

Tires represent another vehicle component that provides potential for reducing the vehicle’s fuel consumption and GHG emissions. Tires are addressed in the Phase I and Phase II standards, but additional progress may be possible. Most manufacturers of truck, tractor, and trailer tires produce low-rolling-resistance (LRR) models for all wheel positions. Many of these are verified by the EPA SmartWay program to reduce tire rolling resistance by at least 15 percent, which equates to about a 3 percent improvement in fuel economy for a tractor-trailer. The best-in-class tires reduce rolling resistance by 30 percent.

Manufacturers reported in 2013 that the majority of new tractors and trailers are equipped with LRR tires. Greater use of LRR tires can be expected as new truck manufacturers rely on them to help comply with the federal

___________________

⁴ The Oregon survey was taken on I-84, which is an east-west highway that connects Portland to Idaho and points east, not toward California.

fuel economy standards. Currently, about three of every four new tires sold are for replacement, representing one of the few technologies available for improving fuel efficiency of existing vehicles. However, most of these replacement tires are not LRR, and less than half of the tire models available are SmartWay verified (NRC, 2014).

To ensure the fuel economy benefits of both new and replacement tires are maximized, the committee in its 2014 report (NRC, 2014) recommended that NHTSA should further evaluate and quantify the rolling resistance of new tires, especially those sold as replacements. If additional cost-effective fuel savings can be achieved, the report recommended that NHTSA and EPA adopt a LRR performance standard for all new tires sold, including those sold as replacements. In the Phase II rule, the agencies did not adopt a LRR performance standard for all new tires. The agencies expect that LRR tires will likely be used as a means of complying with the GHG emission and fuel consumption standards for new trucks and trailers. However, this does not address the key fact that more replacement tires are sold each year than tires sold for installation on new trucks and trailers, and given the incremental higher costs of LRR tires there remains a lack of incentive for those who only own trailers to invest in fuel-saving technologies such as replacement tires. Thus the committee continues to believe in the appropriateness of its previous recommendation that NHTSA and EPA evaluate the cost effectiveness of low-rolling-resistance tires, especially replacement tires, and consider adopting a LRR performance standard if indicated (NRC, 2014, p. 80). More detailed analysis of the potential fuel consumption savings from tires is provided in Section 5.3 in Chapter 5.

2.2.3.3 Cabin Air Conditioner Refrigerant

Most vehicle air conditioners leak refrigerant during their life, either by slow leakage through hoses or seals, or from the failure of the various components.⁵ Refrigerants can also be released from collisions and during servicing and scrappage if not properly reclaimed.

The refrigerant currently in service for truck cabin conditioning is HFC-134a, which has a 100-year global warming potential (GWP) of 1,430.⁶ Truck and trailer refrigerant units (reefers) also use R-404A which has a still higher GWP of 3,922. The Phase II rule included a minimum leakage standard for Class 2b to Class 8 vocational trucks, beginning in 2021.

Other factors may favor a move away from HFC-134a to lower-GWP refrigerants in MHDVs. EPA has finalized a rule delisting approval of HFC-134a in light-duty vehicles by MY 2021 (EPA, n.d.).⁷ The European Commission has eliminated the use of refrigerants having GWP above 150 effective January 1, 2017, in new cars and vans. The European Commission reviewed and reaffirmed the directive⁸ after one manufacturer and Germany’s Federal Motor Transport Authority (KBA) identified a risk of flammability of HFO-1234y, a substitute for HFC-134a, in head-on collisions.⁹

Finding: NHTSA and EPA have adopted requirements in the Phase II rule that will reduce fuel consumption and/or GHG emissions caused by nonengine components such as trailers, tires, and air conditioning. However, fuel consumption and GHG emission reductions beyond those adopted in Phase II appear possible.

Finding: The use of refrigerants in light-duty vehicles for cabin air conditioning has moved to refrigerants having lower GWP than HFC-134a. The use of such refrigerants may be feasible in medium- and heavy-duty vehicles.

Recommendation 2-6: The committee recommends that each of these nonengine components be further evaluated for feasibility and cost effectiveness during the period prior to issuing a Phase III proposal. Specifically, to achieve

___________________

⁵ This discussion of hydrofluorocarbons (HFCs) affects the GHG emissions of a MHDV but not its fuel consumption.

⁶ GWP is defined as the ratio of the time-integrated radiative forcing of a gas to that of an equal mass of CO₂. The period of time integration is user defined.

⁷ Protection of Stratospheric ozone: Change of listing status for certain substitutes under the significant new alternatives program. Federal Register 80:42870.

⁸ See http://europa.eu/rapid/press-release_MEMO-14-50_en.htm (accessed December 1, 2019).

⁹ See http://europe.autonews.com/article/20130808/ANE/130809861/german-officials-provide-mixed-ruling-on-honeywell-refrigerant (accessed December 1, 2019).

greater reductions in fuel consumption and GHG emissions, the design of trailers, and better integration with tractors, should be evaluated; and the rolling resistance of new tires, especially those sold as replacement tires, should be quantified and evaluated. If additional cost-effective fuels savings can be achieved, NHTSA should adopt a regulation establishing a low-rolling-resistance performance standard for all new tires designed for tractor and trailer use.

2.2.3.4 Length and Weight

Truck size and weight federal policy was enacted to preserve the infrastructure and ensure that heavy vehicles could operate on the road system within the load-carrying capacity of roads and bridges. It also ensured that vehicle size was controlled to allow operation within geometrical constraints dictated by bridge height, lane widths, and curve radius constraints, which limit vehicle maneuverability. Since the policy limits vehicle length and weight, it de facto limits options to improve vehicle productivity. U.S. federal truck size and weight policy has been frozen since 1982. Meanwhile, other countries have reformed their policies, yielding improved freight efficiency and resulting in substantial fuel emissions reductions for a given freight task. It is estimated that the reduction in vehicle miles traveled (VMT) and CO₂ emissions attributable to potential size and weight policy reform is in the range of 10 to 20 percent (Woodrooffe et al., 2009). A 10 percent reduction in VMT translates to an annual reduction of approximately 11 billion liters of diesel fuel and 28 million metric tons CO₂ and has an estimated safety benefit of about 330 fewer fatalities and over 4,200,000 fewer injuries due to reduced vehicle exposure. The annual value of these savings is approximately $16 billion (Woodrooffe, 2016). Heavier and longer vehicles could affect and be affected by road infrastructure, although the Federal Highway Administration recently completed a truck size and weight study (FHWA, 2015) that examined how selected heavier and longer vehicles would influence infrastructure life. The study found that the infrastructure costs for most vehicles examined were minimal and there was no significant infrastructure risk identified. This issue of size and weight and its relationship to reducing fuel consumption and GHG emissions is discussed further in Chapter 9.

2.2.4 Complexity and Variety of Vehicle Types

A fourth complexity for MHDV regulations is the variety of vehicle types and duty cycles. Passenger vehicles are relatively uniform, varying mainly by weight, footprint, and frontal area (aerodynamic drag), and primarily perform movement of people. Most passenger vehicle models are produced in relatively high annual volumes. By contrast, medium- and heavy-duty (MHD) trucks vary greatly in weight, size, power, and drag. A local delivery truck and an 18-wheel tractor-trailer share much less in common than do a subcompact car and a large sport utility vehicle. The MHDV on-road vehicle industry is composed of a variety of vehicles with missions spanning small business and utility service to delivery and commercial cargo trucks (see Figures 2-10 and 2-11). The complexity and variety of vehicle types and tasks makes all aspects of the industry more complicated, from manufacturing to regulation to enforcement.

Most MHDVs have historically been powered by diesel engines; however, the impact of environmental regulation compounded by high fuel costs has caused the evolution of a more complex set of technologies designed for these vehicles in recent years. At the lighter end of the weight class for these vehicles, gasoline- and diesel-fueled heavy pickups compete based on fuel economy and vehicle price. At the heavier end of the weight class, favorable pricing of natural gas versus diesel fuel has brought about an increase of natural gas–powered vehicles in the MHDV markets. Compounding this is the need for a myriad of engine, transmission axle, tire, and programmable electronic features that satisfy the specific work each vehicle is designed to achieve. With the rapid development of electronics, many MHDVs in the late 2020s and 2030s will also be equipped with features that selectively power vehicle functions. Finally, the role of the operator cannot be discounted in how effective reductions in fuel consumption can be. Whether benefits and/or incentives are available to the driver will play a significant role in whether the vehicle is driven in the most efficient manner. All this will produce a complex mix of technology features and behaviors which will play a role in the control of GHG emissions and improve fuel economy in the future. Chapter 9 explores these considerations further.

2.2.5 Large Versus Small Companies

A fifth complexity the regulation of medium- and heavy-duty vehicles must address is the range and variation of the size of companies potentially subject to or otherwise affected by the regulation. Especially in the fleet ownership sector of the industry, companies will range from large commercial fleet owners such as Walmart, FedEx, and UPS, with substantial regulatory resources and expertise, to small (one- to three-truck) family businesses with much more limited resources and expertise for regulatory compliance. The Phase I standards exempted small businesses from compliance with the standards, but the Phase II standards will no longer exempt small businesses. However, NHTSA and EPA have provided additional flexibility to small businesses subject to the Phase II standards, notably small business trailer manufacturers, such as providing a 1-year delay for compliance and relaxing certification and reporting requirements. Going forward, continued flexibility and accommodation is important for small businesses subject to the standards, as well as small businesses that may be affected by the regulations (e.g., small business fleets that purchase new trucks).

2.2.6 Life-Cycle Analysis of Vehicle/Fuel Systems

The importance of life-cycle analysis, particularly for GHG emissions, is another complexity MHDV efficiency regulations must address. The current regulatory design only includes GHG emissions from the tailpipe and to

some extent the air conditioning system. There are some important limitations with undertaking this approach, which may lead to negative unintended outcomes from the policy. For example, the direct tailpipe GHG emissions associated with electric MHDVs are zero; however, if the electricity used to power such vehicles is produced using a substantial share of coal, the life-cycle GHG emissions could be higher than for other technologies that directly emit GHGs (Tamayao et al., 2015; Tong et al., 2015a,b; Yuksel et al., 2016).

Similarly, direct GHG emissions from natural gas MHDVs could lead to the conclusion that these are net emissions savers. However, leakage of methane during production and distribution of methane (natural gas) has been identified as a potentially significant source of GHGs (Brandt et al., 2014; Tong et al., 2015a,b) and is not addressed by the current regulations. A high leakage rate in methane production (i.e., in the “upstream” emissions) may more than offset the GHG reductions of natural gas MHDVs, resulting in an overall increase in GHGs. These are discussed in greater depth in Chapter 6.

Quantifying life-cycle emissions of alternative fuels and propulsion systems is one approach to addressing this regulatory gap. Life-cycle emissions can include upstream and wheel-to-wheel emissions associated with different vehicle pathways. It should also include the energy consumption of auxiliary and other power sources used on a vehicle. The advantage of this type of system analysis is that—in its full scope—it accounts for the emissions (or energy) associated with manufacturing, use, and disposal of MHDVs. The disadvantage is that these processes and their emissions change over time in response to changing economic and regulatory influences, and differ

regionally, which may make policy development and implementation more difficult. The effect of accounting for life-cycle GHG emissions and fuel use should be quantified, including uncertainties to determine if accounting for life-cycle emissions is necessary to create a level playing field for emerging fuels and technologies or for incorporation into future regulatory design is necessary.

Figure 2-12, reproduced from the committee’s 2014 First Report, provides a framework to assess potential fuel pathways to serve different MHDVs.

Finding: New fuel and power source options (including hybrid-electric systems and auxiliary power sources) may be needed to achieve significant further progress in reducing fuel consumption and GHG emissions from MHDVs over the next couple of decades. In addition to direct fuel consumption and emissions from the vehicle, these fuel options will likely differ significantly in upstream and nontailpipe emissions. These nontailpipe and upstream emissions may have an impact on program benefits achieved.

Recommendation 2-7: NHTSA, in coordination with EPA, should evaluate and quantify the life-cycle GHG emissions and fuel consumption of all fuels and technologies whose use could contribute to meeting a third phase of standards, and take them into consideration in developing a third phase of regulation. It will be critically important to incorporate a life-cycle perspective in those instances where some fuel-technology pathways’ life-cycle emissions may lead to an increase, rather than a decrease, in emissions. However, given the resources needed for detailed life-cycle analysis and uncertainty characterization, and given the resource constraints from the agencies, we would recommend assessing carefully which fuel-technology pathways may deserve specific attention in a life-cycle perspective.

2.2.7 Certification and Real-World Compliance

For passenger vehicles, EPA requires the vehicle manufacturer to perform pre-sale emission and fuel economy testing of the completed vehicle to validate that it is capable of meeting emission and fuel economy standards. The test cycles are designed to mimic real-world operation. During the vehicle life following sale, EPA performs testing of many models using the same test cycles to determine if the vehicles continue to comply in-use with standards. Models that do not may be recalled. EPA also performs extensive testing using a broader range of test cycles, including use of on-vehicle portable emission measurement equipment, to determine the emission performance of the vehicle fleet as a whole, which provides data that allow an assessment of regulatory program effectiveness. Together these testing requirements help ensure (1) new vehicles, before they are sold, are capable of meeting standards; (2) vehicles meet standards during most of their operational life, and if they do not, remedial actions are taken; and (3) the overall regulatory program is effectively reducing emissions and fuel consumption (NRC, 2015). The test results may also suggest ways the effectiveness of the regulations can be increased, and the cost of compliance reduced. These compliance programs have identified many in-use light-duty vehicles with high emissions, for which design changes were made through recalls, and also identified opportunities to improve the effectiveness of the regulations.

For MHDV Class 4 and above, the current regulation for GHG emissions and fuel economy, when compared to the program for passenger vehicles, falls considerably short of providing an assurance of compliance and program effectiveness. Pre-sale certification emission testing applies to the engine, but not the vehicle. Certification of the vehicle is accomplished using a simulation model with no direct vehicle GHG emission or fuel consumption testing. Once vehicles are sold, in-use testing is sparse and limited to on-vehicle emission measurements made during random truck operation. There are generally no state and emissions inspections of the sort used for light-duty vehicles (LDVs). This approach does not lend itself to determining in-use compliance since the on-road testing differs from the certification engine test and the cycles used in the simulation model, and it is the on-road testing rather than the certification testing that provides the necessary real-world measurement of emissions. Removing engines and testing them in the same manner as pre-sale certification can be done, but it is far too costly and time consuming to be a viable and comprehensive method of determining in-use compliance. It also remains unclear if an on-road vehicle test that suggests the vehicle is exceeding standards can be used as a basis for enforcement given the vehicle was originally certified using a simulation model rather than an emission test.

This regulatory gap is discussed in more detail in Chapter 3, Certification, Compliance, and Enforcement, and some possible alternative approaches to ensuring compliance and measuring program effectiveness are identified that should be evaluated by NHTSA and EPA prior to proposing a third phase of regulations.

2.2.8 Regulatory Baselines and Metrics

The overall objective of the MHDV fuel efficiency and GHG emissions reduction regulations is to achieve real-world benefits in reducing fuel consumption and GHG emissions in the MHDV sector. As explained in the committee’s 2014 report (NRC, 2014, Chapter 4), a reliable baseline of pre-regulatory fleet characteristics, miles driven, fuel consumption, and emission data is critical for measuring real-world impacts. As explained in the 2014 report, NHTSA and EPA currently lack a reliable source of data on real-world vehicles that can be used to establish a regulatory baseline. Phase II is based on an assumption of compliance with Phase I standards as the baseline for the Phase II regulations, but this modeled baseline is not connected to or validated by real-world measurements and thus does not provide a valid baseline that can be used to measure real-world performance. The nearly decade-long gap between the Phase II and a possible “Phase III” rulemaking provides an opportunity for NHTSA and EPA to consult with other governmental and private entities to coordinate efforts to obtain real-world vehicle data that can be used to establish a reliable baseline.

Finding: NHTSA and EPA currently lack reliable data on real-world vehicles that can be used to establish a regulatory baseline. While engine certification data are available and useful, they do not address the need for on-road real-world truck data. A reliable regulatory baseline based on real-world data is essential for evaluating the effectiveness and success of the regulatory program, and identifying future regulatory priorities and directions.

Recommendation 2-8: NHTSA and EPA will have almost a decade between the Phase II and a possible “Phase III” rulemaking in which to coordinate with other governmental and private entities to collect real-world vehicle data that can be used to establish a reliable baseline. The agencies should commit resources to collecting real-world fuel consumption and GHG emissions data from a robust and representative sample of pre-control trucks and for each model year subject to Phase I and Phase II standards. These data can be used to establish a regulatory baseline that can be used to evaluate program effectiveness and future regulatory priorities.

2.2.9 Other Factors Affecting Fuel Efficiency

There are a number of other factors that can affect the fuel efficiency and GHG emissions performance of a MHDV. For example, vehicle speed is an important factor, with higher vehicle speeds resulting in greater fuel consumption needed to transport a given load a specified distance. As a result, many fleet owners are now installing speed limiters on their vehicles to require lower speeds and save money through fuel conservation. Another important factor is driver behavior, as inefficient driver behavior (e.g., nonoptimal gear shifting or excessive idling) can result in significant fuel wastage. Many vehicle owners are now using telematics and surveillance technologies to monitor driver behavior, creating sensitive tensions between optimal performance and driver privacy. Yet another important factor is road condition and traffic management, all of which can also dramatically affect fuel efficiency. Because these and other factors have a significant effect on vehicle fuel efficiency and GHG emissions, it is important to consider how fuel efficiency and GHG emission regulations will interact with or be affected by these factors.

Some additional market and technology factors, some of which are discussed in more detail in the following chapters, include (i) economic trends such as construction and manufacturing levels; (ii) the availability of industry incentive programs such as tax credits for electric vehicles and other energy-conserving technologies (Diamond, 2009); (iii) the development and deployment of connected and automated vehicle systems; (iv) modal shifts between trucks, ships, rail, and water vehicles; (v) changes in transportation planning and traffic management; (vi) emerging technologies such as 3D printers and unmanned aerial vehicles that may affect transportation and delivery systems (Basulto, 2014; Harford, 2013; Vance, 2010); and (vii) the expansion of the Panama Canal and other infrastructure projects.

2.3 OTHER REGULATORY PROGRAMS

A variety of other regulatory programs at the state, federal, and international levels seek to control directly the fuel consumption and GHG emissions of MHDVs, as does the NHTSA/EPA program. These other programs, summarized below, may provide useful lessons and synergies for the NHTSA/EPA program. In addition to these regulatory programs that directly regulate the fuel conservation or GHG emissions of MHDVs, there are a number of other regulatory programs, also discussed below, that indirectly affect fuel consumption and GHG emissions associated with the MHDV sector.

2.3.1 Other Programs Directly Regulating MHDV Fuel Efficiency and GHG Emissions

2.3.1.1 States

Under federal clean air law, only California can adopt its own emission limits for new vehicles, which must be at least as stringent as the EPA standards. For new heavy truck engines, the California Air Resources Board (CARB) required stricter emission standards for criteria air pollutants than EPA until the 1990s, when CARB and EPA agreed to jointly adopt uniform standards that meet both federal and California’s urban air pollution control needs. This continued with the adoption of the Phase I MHDV GHG standards, and the agencies have worked cooperatively in developing the Phase II GHG standards, which were adopted in 2016. In addition, California is the only state to regulate certain truck operators by requiring large trailers to have improved aerodynamic performance, which reduces tractor fuel use, and requiring use of low-rolling-resistance tires.

Other states are preempted from adopting their own vehicle emission standards under the Clean Air Act, but section 177 of the statute allows other states under certain circumstances to adopt and enforce the California emission standards for LDVs or MHDVs. Nine states (mostly in the Northeast) have used this statutory provision to require California certification of new MHDVs sold in their states. Because the California and federal criteria emission standards for MHDVs have been essentially the same since the 1990s, and EPA was first to adopt MHDV GHG standards, most of these states have not updated their regulations to require California-certified heavy trucks, but the authority to require sale of California vehicles remains should it be needed.

2.3.1.2 Other National Programs

A number of other nations have adopted their own MHDV fuel efficiency or GHG regulations. The European Union (EU) has been an international leader in regulating overall GHG emissions but has not yet adopted standards for fuel efficiency or GHG emissions specifically for MHDVs. The Commission of the European Union has developed a computer simulation tool called VECTO to calculate CO₂ emissions from new MHDVs. The Commission issued a strategy for low-emission mobility which indicates the EU will need to introduce measures to actively curb CO₂ emissions from lorries (trucks), buses, and coaches, as other parts of the world have. The Commission will speed up analytical work and public consultation to prepare proposed standards (European Commission, 2016).

Japan in 2005 was the first country to establish fuel economy standards for new heavy-duty trucks and buses over 3.5 tonnes. The standards apply only to diesel-fueled trucks and buses, and must be met on average for a vehicle class (typically weight based). The standards are set by the best-performing vehicle per class in a base year (2002), and are enforceable beginning in 2015, providing a 12 percent reduction in fuel consumption from new vehicles. Compliance is based on engine dynamometer testing and a simulation model with actual engine and drivetrain inputs and standard assumptions for typical driving cycle, drag, rolling resistance, and weight. The concept, called Top Runner, is that all trucks and buses of a given class should be able to perform equal to or better than the best in class given time for redesign. The 10-year lead time acknowledged that standards for NO_x and PM_2.5 were also strengthened in 2009. No in-use compliance program is specified (Olivares and Wagner, 2014a).

China has adopted fuel economy standards for heavy trucks and buses over 3.5 tonnes, fueled by diesel or gasoline. The second-phase standards went into effect in July 2015 and vary by weight, and five truck or bus types. Each truck must meet its appropriate standard. The standard is expected to reduce fuel consumption by about 11

percent compared to the first phase of standards. Compliance is based on a chassis emission test for a base model representing the worst-case truck in a class, with a simulation model used to determine fuel consumption for closely related variants to the base model. The standards are less stringent for gasoline trucks. No in-use compliance program is specified (ICCT, 2014).

Canada has adopted GHG standards nearly identical to those of the United States, beginning with 2014 models. Vehicles certified to U.S. standards are deemed in compliance with Canadian standards (Olivares and Wagner, 2014b).

India in February of 2016 proposed to adopt the EuroVI standards and procedures for criteria pollutants, with implementation beginning with 2020 models (Olivares et al., 2016). India has been evaluating the GHG and fuel consumption of heavy-duty trucks and is currently developing fuel efficiency standards to address the unique vehicles and driving conditions there (ICCT, 2014).

2.3.1.3 Voluntary Programs

The U.S. EPA initiated a public-private partnership in 2004 called SmartWay®, the objective of which is reducing GHG and pollutant emissions resulting from movement of freight across a firm’s supply chain (EPA, 2017).¹⁰ The program works with corporate partners and recognizes those fleet operators and logistics firms that make voluntary reductions in their trucking emissions (Berth, 2014). SmartWay partners agree to provide data on their operations, which are input into standardized tools to produce a measurement of environmental efficiency, such as grams pollutant per ton-mile. These benchmark results are compared to similar categories of freight movement, for example, dry vans, and ranked in quintiles. Fleet operators and shippers use the results to improve their efficiency, identify green options, and achieve recognition. SmartWay also includes a Technology Program that develops test procedures and verifies the performance of technologies, equipment, and strategies that save fuel and reduce emissions.

With the implementation of mandatory GHG emission and fuel economy standards, beginning with the 2014 models, many of the fuel-saving features for new tractors encouraged by the SmartWay program will be used on new tractors in order to meet government standards. This includes low-rolling-resistance tires and idle-reduction technologies. The government regulation effectively supplants this portion of the (voluntary) SmartWay program for new trucks. However, the trailer standards in the Phase II rule would not be fully implemented until 2027, and more importantly would affect only new trailers, and not those already in use. Thus SmartWay can continue to encourage improvements to existing tractor-trailer combinations.

2.3.2 Other Regulatory Programs That Indirectly Affect Fuel Consumption and GHG Emissions

Many other regulatory programs indirectly affect fuel consumption and GHG emissions of MHDVs and may intersect with or otherwise affect the NHTSA and EPA MHDV standards. Such programs include the regulations of tailpipe emissions from vehicles of both so-called criteria pollutants and of carbon dioxide, vehicle safety regulations, and fuel programs.

2.3.2.1 NO_x Emissions and Regulation

Efforts to improve fuel efficiency of diesel engines can be impacted by requirements for reducing emissions of the ozone precursor NO_x. The committee received testimony from CARB that the need for significant additional NO_x reductions in California requires an integrated approach in future GHG and NO_x emissions reduction regulation (Sax et al., 2014). Analyses by CARB and the South Coast Air Quality Management District indicate that the greater Los Angeles area needs additional NO_x emissions reduction by 2023 and 2031 to meet federal ozone ambient air quality standards. Figure 2-13 indicates that compared to 2023 NO_x emissions that reflect only current regulatory requirements, NO_x emissions must be reduced by another 50 percent in 2023 and by 65 percent before

___________________

¹⁰ This program is discussed in more detail in the committee’s Phase Two First Report (NRC, 2014).

2032. Also shown in the figure is that on-road heavy-duty diesel trucks are the largest source of NO_x emissions. Thus it will be necessary to further reduce their emissions. Eleven state and regional air pollution agencies, most from outside of California, have petitioned EPA for tighter NO_x standards for MHD engines.¹¹ In December 2016, EPA responded to the petition indicating it would initiate a rulemaking to propose revisions to the on-road heavy-duty control program to reduce NO_x emissions, starting with the 2024 model year. Many regions across the country in addition to California are also facing challenges in complying with the national ambient air quality standard for ozone (see Figure 2-14). Additional reductions in NO_x emissions from MHDVs would likely be welcomed by many of these states to help them meet the ozone standard in their region.

CARB and the South Coast Air Quality Management District have begun the process of determining how to achieve the additional NO_x reductions needed from diesel trucks. The agencies are sponsoring studies of approaches to optimize selective catalytic reduction control efficiency, and to achieve lower emissions from natural gas vehicles. In the California Sustainable Freight Action Plan issued in July 2016, CARB laid out its time frame for adopting a more stringent NO_x standard for new MHDVs. CARB states it will work with and/or petition EPA to adopt a nationwide low-NO_x engine standard by 2019 to be implemented with the 2024 models. CARB has stated that, were EPA not to act, it would adopt a low-NO_x standard for new engines sold in California, beginning with 2023 models. While EPA did not propose or adopt a more stringent NO_x standard for new MHD engines, in the preamble to the 2016 Phase II rule EPA states it plans to engage with stakeholders to discuss opportunities to develop more stringent standards to reduce NO_x from heavy-duty on-highway engines through a coordinated effort with CARB. In addition, it should be noted that the need or desire of some states to achieve greater reductions in ground-level

___________________

¹¹ Petitioners represent four state agencies (CT, DE, NH, and WA) and seven local or regional agencies (South Coast [CA], Bay Area [CA], Akron [OH], New York City [NY], Pima County [AZ], Puget Sound [WA], and Washoe County [NV]).

ozone precursors or fine particulates may be the driver for those states to opt into the California program, which would also include the California GHG emission standards.

Finding: CARB has made a finding that additional reduction of NO_x emissions is needed to attain the federal ambient air quality standard for ozone in California. Other state and local air quality agencies from several parts of the country have also indicated a desire for additional NO_x reductions. In California, emissions of NO_x are primarily from mobile sources, and on-road heavy-duty diesels are one of the largest sources. These factors suggest that the current NO_x standard for MHD engines may be tightened either by California or on a national level.

Recommendation 2-9: Because reductions of NO_x may increase fuel consumption and emissions of CO₂, it is important that EPA and NHTSA thoroughly evaluate the effect of current and emerging technologies on this relationship, and determine an optimum approach for addressing this issue in time for consideration as part of a third phase of regulatory development.

2.3.2.2 Climate Change Regulation

President Barack Obama issued a Climate Action Plan in June 2013 (White House, 2013). In addition to the adoption of GHG emission standards for new passenger vehicles and MHDVs, the plan directed EPA to issue GHG emission limits for new and existing power plants, which were finalized in August 2015. The Climate Action Plan sets a goal of reducing annual U.S. GHG emissions in the range of 17 percent below 2005 levels by 2020. In a subsequent agreement with China, a goal of reducing by 2025 U.S. GHG emissions by 26 to 28 percent below 2005 levels was announced. This goal was submitted as the United States’ intended Nationally Determined Contribution as part of the 21st Conference of Parties under the United Nations Framework Convention on Climate

Change held in Paris in December 2015 (White House, 2015). On June 1, 2017, President Trump announced his intention to withdraw the United States from the Paris Agreement, and the new administration has not yet set any specific goals for GHG emission reductions.

At the state level, California requires GHG emissions to be reduced to 1990 levels by 2020, and 40 percent below this level by 2030. This is being accomplished by source-specific regulatory action, such as the GHG standards for new cars and trucks, renewable fuel requirements for in-state power plants, the Low Carbon Fuel Standard for transportation fuels, GHG reduction targets to be achieved by local governments through improved planning, recovery of hydrofluorocarbons, and a cap-and-trade program which generates funds that are used to achieve reductions in GHG emissions. Relevant to this report, cap-and-trade funds of about $150 million yearly are being allocated to support commercialization of near-zero-emission MHDVs. The results from these pilot programs will demonstrate if new technologies can be commercially viable under normal use and, if they are, can provide data valuable to the development of a third phase of MHDV regulations.

Many city and county governments have also taken actions to reduce GHG emissions. For example, the C40 Cities Climate Leadership Group, with over 70 megacity members, has committed to over 8,000 actions with the goal of reducing more than one gigaton of CO₂ emissions by 2020.

Finding: A growing number of actions, commitments, and goals of national, regional, and local governments are supporting the reduction of GHG emissions from energy use including MHDVs.

2.3.2.3 Natural Gas and Hydraulic Fracturing Regulation

The rapid increase in hydraulic fracturing in the United States over the past decade has greatly expanded the availability, and reduced the cost, of natural gas produced from shale rock, potentially increasing the use of natural gas as a fuel in the MHD trucking sector. The economic feasibility of natural gas production through hydraulic fracturing is sensitive to fuel prices, and the recent decline in oil prices has suppressed fracturing activity, but this form of energy production is capable of rapidly expanding again as fuel prices increase (Mills, 2016). Federal and state regulation of hydraulic fracturing can also limit activity in this area. Regulation can focus on a variety of different elements, including limits on discharges to groundwater, surface waters, and air, restrictions on the chemicals used in the fracturing process, requirements to disclose information, or outright prohibitions (Richardson et al., 2013). In addition to environmental concerns, some jurisdictions are also considering restrictions to protect against the possibility that hydraulic fracturing may contribute to seismic activity. Finally, a number of lawsuits have been filed against companies engaged in hydraulic fracturing seeking monetary damages for alleged personal injury and property damages as well as injunctive relief (Hall, 2012). Notwithstanding such legal risks and economic costs, hydraulic fracturing will continue to play an important role in U.S. energy production, and hence MHDV fuel options and costs, for many decades to come.

2.3.2.4 Infrastructure and Road Regulation

Infrastructure and road regulation pertaining to trucks is governed by the Federal Highway Administration, by individual state governments, and by municipalities in the form of size and weight regulations, seasonal axle weight limits, road classification, lane limits, high occupancy vehicle and truck lanes, traffic calming, weigh in motion, and road pricing. In addition, in August 2016, NHTSA proposed to require all newly manufactured U.S. trucks, buses, and other vehicles with a gross vehicle weight rating more than 26,000 pounds to come equipped with speed limiting devices. These regulatory programs and their potential for reducing MHDV fuel consumption and GHG emissions are discussed in greater detail in Chapter 9.

2.3.2.5 Safety Regulations

Commercial vehicle safety regulations at the federal level are governed by two separate agencies. NHTSA controls regulations governing “new vehicles.” These regulations are generally performance based and deal with

occupant safety, vehicle component performance such as brakes and rear underride systems, and conspicuity. The FMCSA safety regulations focus on vehicle operations. They cover, develop, and enforce data-driven regulations that balance motor carrier (truck and bus companies) safety with efficiency; harness safety information systems to focus on higher-risk carriers in enforcing the safety regulations; target educational messages to carriers, commercial drivers, and the public; and partner with stakeholders including federal, state, and local enforcement agencies, the motor carrier industry, safety groups, and organized labor on efforts to reduce bus- and truck-related crashes.

As discussed in greater detail in Chapter 9, some fuel efficiency technologies may create safety trade-offs, although these trade-offs are not as obvious and controversial as have been the case for LDV fuel economy standards (NRC, 2002). Some fuel efficiency measures, such as lower speed levels, may have beneficial safety implications for MHDVs. Other technologies, such as alternative fuels and low-rolling-resistance tires, could potentially have negative safety trade-offs. NHTSA has comprehensively examined these positive and negative safety trade-offs with fuel efficiency regulations in a report produced as part of the Phase II regulatory proceeding (NHTSA, 2015).

2.3.2.6 Fuel Regulations and Government Programs

EPA has adopted a renewable fuel standard for transportation which is being phased in through 2022. The standard requires specified volumes of renewable transportation fuel, principally ethanol and biodiesel, in four categories that have varying lower life-cycle GHG emission footprints compared to petroleum fuels. The 2013 requirement for biomass-based diesel, one of the four categories, is 1.13 percent of total diesel fuel volume (1.28 billion gallons), and this fuel must have a carbon footprint at least 50 percent lower than petroleum-derived diesel. In 2014, 1.63 billion gallons of biomass-derived diesel were produced. On November 30, 2015, EPA adopted a requirement that 2.0 billion gallons of biomass-based diesel be produced in 2017 (EPA, 2015).

California adopted in 2009 a Low Carbon Fuel Standard for transportation fuels. The regulation requires that each producer or importer of gasoline or diesel fuel for use in California gradually reduce the life-cycle GHG emissions of its products by 10 percent by 2020. Compliance can be achieved by directly reducing the fuel’s carbon intensity, or by buying credits generated by third-party producers or importers of lower-carbon fuels. Credits generated to date are 60 percent from ethanol, 28 percent from biodiesel, 10 percent from natural gas (28 percent less CO₂ per unit enthalpy than diesel), and 2 percent from electricity, and the available credits exceed the current year regulatory requirement by about 65 percent. Almost 170 biofuel facilities have registered to provide low-carbon fuels to California (CARB, 2014). The state has invested $60 million in supporting the development of advanced biofuel production facilities. Beginning in January 2015, transportation fuels have been added to California’s cap-and-trade program.

These government programs are encouraging, incentivizing, and in some cases requiring the use of cleaner substitutes for petroleum diesel fuels, and a nascent market for their use is developing. Alternative fuels that can be used in diesel-like engines, but require engine modification and new fueling infrastructure, are also being evaluated for commercial viability.

Another approach to reducing GHG emissions from MHDVs is to produce a fuel from renewable and other sources such that the fuel has a much smaller well-to-tank carbon footprint. For example, biomethane can be produced from waste products. Similarly, dimethyl ether (DME) can be efficiently produced from waste products and combusted in a diesel cycle engine with a very low net production of CO₂ relative to standard diesel fuels (Semelsberger et al., 2006).

Synthetic diesel fuel can also be made from hydrocarbon sources such as natural gas, or coal, which would reduce conventional diesel consumption while avoiding the need for engine modifications and new fuel infrastructure. This approach can reduce our dependence on petroleum, but even with carbon capture and storage, well-to-wheels GHG emissions from indirect liquefaction of coal are roughly the same as that of petroleum fuels from conventional reservoirs due to high energy use and resulting GHG emissions during fuel production (NRC, 2013). Conversion of natural gas to petroleum-like liquid fuel, without carbon capture and sequestration, also results in GHG emissions similar to conventional petroleum fuels. Vocational trucks and buses can be powered by electricity or hydrogen fuel cells with much lower CO₂ emissions, depending on how the electricity or hydrogen is produced.

Finding: Improvements in the efficiency of diesel and gasoline engines, trucks, and tractor-trailers have been the focus of regulatory efforts to reduce fuel use and GHG emissions. However, if the practical application of these approaches is inadequate to achieve desired GHG reductions, other alternative fuels such as natural gas, propane, DME, biomethane, and renewable low-carbon diesel may offer opportunities for further reductions in GHG emissions. Battery electric and fuel cell electric powertrains could also provide additional GHG reduction in urban trucks.

2.4 REFERENCES

IEA (International Energy Agency). 2010. World Energy Outlook 2010. Paris: Organisation for Economic Cooperation and Development.

IEA. 2015. World Energy Outlook Special Report on Energy and Climate. Paris: Organisation for Economic Co-operation and Development.

IPCC (Intergovernmental Panel on Climate Change). 2014. Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, Core Writing Team, R.K. Pachauri and L.A. Meyer (eds.). Geneva: IPCC, 151 pp.

Jaffe, A.B., R.G. Newell, and R.N. Stavins. 2005. A tale of two market failures: Technology and environmental policy. Ecological Economics, 54(2-3):164-174.

Marchant, G.E. 2014. Complexity and anticipatory socio-behavioral assessment of government attempts to induce clean technologies. UCLA Law Review 61:1858-1894.

Mills, M.P. 2016. After the carnage, shale will arise again. Wall Street Journal, January 18.

Morrow, W.R., K.S. Gallagher, G. Collantes, and H. Lee. 2010. Analysis of policies to reduce oil consumption and greenhouse-gas emissions from the US transportation sector. Energy Policy 38(3):1305-1320.

Nahlik, M.J., A.T. Kaehr, M.V. Chester, A. Horvath, and M.N. Taptich. 2015. Goods movement life cycle assessment for greenhouse gas reduction goals. Journal of Industrial Ecology 20(2):317-328.

Nelson, R.R., and R.N. Langlois. 1983. Industrial innovation policy: Lessons from American history. Science 219(4586):814-818.

NHTSA. 2015. NHTSA Study: Review and Analysis of Potential Safety Impacts and Regulatory Barriers to Fuel Efficiency Technologies and Alternative Fuels in Medium- and Heavy-Duty Vehicles. Washington, DC: NHTSA.

Norberg-Bohm, V. 1999. Stimulating “green” technological innovation: An analysis of alternative policy mechanisms. Policy Sciences 32(1):13-38.

NPC (National Petroleum Council). 2012. Advancing Technologies for America’s Transportation Future. Washington, DC: U.S. Department of Energy.

NRC (National Research Council). 2001. Energy Research at DOE: Was It Worth It? Energy Efficiency and Fossil Energy Research 1978 to 2000. Washington, DC: National Academy Press.

NRC. 2002. Effectiveness and Impact of Corporate Average Fuel Economy Standards. Washington, DC: National Academy Press.

NRC. 2007. Prospective Evaluation of Applied Energy Research and Development at DOE (Phase Two). Washington, DC: The National Academies Press.

NRC. 2009. America’s Energy Future: Technology and Transformation. Washington, DC: The National Academies Press.

NRC. 2010. Technologies and Approaches to Reducing the Fuel Consumption of Medium- and Heavy-Duty Vehicles. Washington, DC: The National Academy Press.

NRC. 2013. Transitions to Alternative Vehicles and Fuels. Washington DC: The National Academies Press.

NRC. 2014. Reducing the Fuel Consumption and Greenhouse Gas Emissions of Medium- and Heavy-Duty Vehicles, Phase 2: First Report. Washington DC: The National Academies Press.

NRC. 2015. Cost, Effectiveness and Deployment of Fuel Economy Technologies for Light-Duty Vehicles. Washington DC: The National Academies Press.

Olivares, E., and V. Wagner. 2014a. Japan—Heavy-Duty: Emissions (last updated 2014). Available at http://www.transportpolicy.net/index.php?title=Japan:_Heavy-duty:_Emissions (accessed September 23, 2019).

Olivares, E., and V. Wagner. 2014b. Canada: Heavy-Duty: GHG (last updated 2014). Available at http://transportpolicy.net/index.php?title=Canada:_Heavy-duty:_GHG (accessed September 23, 2019).

Olivares, E., V. Wagner, M. Morgan, and E. Foukes. 2016. India: Heavy-Duty: Emissions (last updated 2016). Available at http://transportpolicy.net/index.php?title=India:_Heavy-duty:_Emissions (accessed September 23, 2019).

Richardson, N., M. Gottlieb, A. Krupnick, and H. Wiseman. 2013. The State of State Shale Gas Regulation. Resources for the Future. May. Available at http://www.rff.org/files/sharepoint/WorkImages/Download/RFF-Rpt-StateofStateRegs_Report.pdf (accessed December 16, 2019).

Roberts, D. 2016. 1.8 million American truck drivers could lose their jobs to robots. What then? Vox. August 3.

Sax, T., S. Yoon, C. Misra, J. Collins, J. Herner, D. Hawelti, S. Lemieux, et al., California Environmental Protection Agency Air Resources Board. 2014. An Integrated Approach to Heavy Duty GHG and NO_x Control. Testimony presentation to the National Research Council’s Committee on Committee on Assessment of Technologies and Approaches for Reducing the Fuel Consumption of Medium- and Heavy-Duty Vehicles, Phase Two, Ann Arbor, MI, July 29.

SCAQMD (South Coast Air Quality Management District). 2016. Petition to EPA for Rulemaking to Adopt Ultra-Low NO_x Exhaust Emission Standards for On-Road Heavy-Duty Trucks and Engines. Available at http://www.aqmd.gov/docs/default-source/default-document-library/news-docs/nox-petition-to-epa-june-2016.pdf?Status=Temp&sfvrsn=2 (accessed September 23, 2019).

Semelsberger, T.A., R.L. Borup, and H.L. Greene. 2006. Dimethyl ether (DME) as an alternative fuel. Journal of Power Sources 156(2):497-511.

Sharpe, B. 2013. Examining the Costs and Benefits of Technology Pathways for Reducing Fuel Use and Emissions from On-Road Heavy-Duty Vehicles in California. Report No. UCD-ITS-RR—3-17. Davis, CA: University of California, Davis.

Tamayao, M., J. Michalek, C. Hendrickson, and I.L. Azevedo. 2015. Regional variability and uncertainty of electric vehicle life cycle CO₂ emissions across the United States. Environmental Science & Technology 49(14):8844–8855.

Taptich, M.N., A. Horvath, and M.V. Chester. 2015. Worldwide greenhouse gas reduction potentials in transportation by 2050. Journal of Industrial Ecology 20:329-340.

Tong, F., I.L. Azevedo, and P. Jaramillo. 2015a. A comparison of life cycle greenhouse gases from natural gas pathways for light-duty vehicles. Energy and Fuels 29(9):6008-6018.

Tong, F., I.L. Azevedo, and P. Jaramillo. 2015b. A comparison of life cycle greenhouse gases from natural gas pathways for medium and heavy-duty vehicles. Environmental Science & Technology 49:7123-7133.

Transportation Research Board. 2011. Policy Options for Reducing Energy Use and Greenhouse Gas Emissions from U.S. Transportation: Special Report 307. Washington, DC: The National Academies Press. https://doi.org/10.17226/13194.

Vance, A. 2010. 3-D printing spurs a manufacturing revolution. The New York Times, September 13.

White House. 2013. The President’s Climate Action (June). Available at https://obamawhitehouse.archives.gov/sites/default/files/image/president27sclimateactionplan.pdf (accessed December 16, 2019).

White House. 2015. U.S. Cover Note, INDC and Accompanying Information, available at https://www4.unfccc.int/sites/submissions/INDC/Published%20Documents/United%20States%20of%20America/1/U.S.%20Cover%20Note%20INDC%20and%20Accompanying%20Information.pdf (accessed March 31, 2015).

Williams, J.H., B. Haley, F. Kahrl, J. Moore, A.D. Jones, M.S. Torn, and H. McJeon. 2014. Pathways to deep decarbonization in the United States: The U.S. report of the Deep Decarbonization Pathways Project of the Sustainable Development Solutions Network and the Institute for Sustainable Development and International Relations. San Francisco, CA: Energy and Environmental Economics, Inc.

Woodrooffe, J. 2016. Opportunity cost for society related to US truck size and weight regulation: Freight efficiency. Transportation Research Record: Journal of the Transportation Research Board 2547:25-31.

Woodrooffe, J., B.M. Belzowski, J. Reece, and P.F. Sweatman. 2009. Analysis of the Potential Benefits of Larger Trucks for U.S. Businesses Operating Private Fleets. Prepared for the National Private Truck Council. University of Michigan Transportation Research Institute. May.

Yang, C., D. McCollum, R. McCarthy, and W. Leighty. 2009. Meeting an 80% reduction in greenhouse gas emissions from transportation by 2050: A case study in California. Transportation Research Part D: Transport and Environment 14(3):147-156.

Yuksel, T., M-A. Tamayao, C. Hendrickson, I.L. Azevedo, and J. Michalek. 2016. Effect of regional grid mix, driving patterns and climate on the comparative carbon footprint of gasoline and plug-in electric vehicles in the United States. Environmental Research Letters 11(4):044007.