NAS: By 2050, it’s possible to cut car petroleum use by 80 percent It would be hard, but easier than cutting CO2 emissions by a similar amount.

A Prius that's been modified to plug into the grid and use its battery for frequency stabilization.

After years of inaction, the US government managed to cut a deal with automobile manufacturers that would prompt the first significant increases in average fuel efficiency in many years. It's part of a plan that would see the average fuel economy clear 50 miles-per-gallon by 2025. Although that may seem like a large leap, it will still leave the US' efficiency standards well below those of other industrialized nations.

A new report by the National Academies of Science looks well beyond these goalposts. It looks at what it would take to get a 50 percent reduction in petroleum use by 2030 and a full eighty percent drop by 2050. To make matters a touch more challenging, it also looks at what it would take to get an 80 percent reduction in greenhouse gas emissions from our light-duty vehicle fleet. The answer is that no single technology is going to be capable of all of this, which means success would require a mix of technologies on the road at the same time. According to the report, none of it will come close to happening without a concerted policy push.

The cars under consideration here—the light duty vehicle category—are the sorts of things that most of us drive for commutes or to run errands. Collectively, they burn through about half of the petroleum consumed in the US and create more than 15 percent of its greenhouse gas emissions. That's enough to make them a significant cost in both financial and environmental terms, and it's a critical factor in the country's strategic decisions and international politics. The report emphasizes repeatedly that it will be expensive to drive the sorts of massive efficiency gains it describes, but it's an investment that will more than pay for itself.

The report also considers the potential for vehicles powered by the fuel that is radically altering US electricity production: natural gas. Compressed natural gas vehicles are a solved problem, but they would require a significant investment of infrastructure to actually deliver the fuel to vehicles. Compressed natural gas also won't help much with carbon emissions, for obvious reasons. The authors note that the fracking boom is expected to last for decades, after which the future of natural gas supplies (and prices) becomes a bit cloudy. The report doesn't note this, but it does say that "several decades" takes us right to the end of the time period it's considering.

That combination of factors—the lack of impact on CO₂ emissions and an uncertain long-term future—may explain why the report places most of its emphasis on other technologies.

One of those is the traditional internal combustion engine. Technology already exists or is in development that can significantly lower vehicle weight and reduce rolling friction. Along with efficiency changes in the engines themselves, better traditional vehicles can contribute to reductions in both fuel use and emissions. A further drop in emissions could be had if drop-in replacement biofuels (either biodiesel or a gasoline analog) development advances continue. The report suggests that efficiencies of 74 MPG are possible by 2050 (and more with hybrids)—but only if the advances go towards fuel efficiency and not, say, acceleration.

Vehicles with batteries are already playing a significant role in transportation, and the report expects hybrid vehicles to only become more common. Battery-only vehicles are expected to increase in range thanks to advances in lithium-ion and lithium air technologies, and these will reach the point where a single charge might cover 100 miles of driving—easily covering most commutes. These will clearly help drop petroleum consumption, but their impact on emissions will depend on further growth of renewable electricity and/or the development of efficient carbon capture and storage.

The one technology that the group felt has been unfairly ignored is fuel cell vehicles. These have the advantages of using well-established technology and handling both emissions and petroleum use (assuming the hydrogen is generated from renewable energy). Additionally, fuel-cell vehicles are as convenient to refuel as current vehicles. The downsides are large, however, including a complete absence of existing infrastructure and a lack of a clear route to hydrogen production. Nevertheless, the report gives them a strong endorsement: "The committee believes that hydrogen/fuel cells are at least as promising as battery electric vehicles in the long-term and should be funded accordingly."

In the end, the committee that prepared the report concluded that it would be very, very difficult to meet the 2030 goals, in part because many of the vehicles that will be in service then would be produced before this decade is out. The 2050 goals are possible, but no single technology would be able to reach them on its own; at least two, and possibly more, might be needed. This would mean having significant numbers of more than one type of vehicle on the road at the same time, something we're just starting to experiment with now thanks to the rise in electric vehicles.

Reaching the goals would also require developments outside the automotive field, such as making the electric grid more heavily powered by renewables and putting a hydrogen production and distribution network in place. All of that would be expensive, as would the vehicles themselves. The authors expect the vehicle of 2050 would cost several thousand dollars more, even when inflation is accounted for (although it would be significantly cheaper to run).

All of this is why the authors expect a concerted policy effort will be required to bring this about. Given that we collectively have a poor track record of predicting which technological changes will prove easiest to develop, they argue that the government should focus on policies that favor any efficiency, such as a tax on carbon emissions from vehicles. They also argue that R&D dollars should be spent on all of the technologies required, at least provided there are no unforeseen breakthroughs that make one a clear favorite.

The effort will also need to be sustained over decades in order to bring what are now nascent technologies into the mainstream. All of which will be hard, but the payoffs of success will be rather large. And as they note, "Even if the nation falls short of the 2050 goals, there are likely to be environmental, economic, and national security benefits resulting from the petroleum use and GHG emissions reductions that are achieved."