Is Natural Gas Worse for Climate Change Than Diesel Fuel?

Science published an interesting and useful new paper on methane leaks in natural gas operations yesterday – but the New York Times chose to highlight the one thing in it that’s both unoriginal and shaky. Understanding that flaw reveals some useful pointers for policymakers.

The heart of the Science paper is a review of studies on methane emissions from natural gas. Figure one from the paper (reproduced above) alone justifies the paper’s publication. A quick glance at the figure reveals a couple things. First, most serious estimates of methane emissions are higher than those reported by the EPA. (By “serious” I’m basically referring to estimates that include uncertainty bounds; it’s tough to put much stock in the rest.) Second, though, if you look at the space where those estimates overlap, it’s not much in excess of the EPA estimates. The review points to an incremental, not radical, upward revision of methane leakage estimates.

The Times, however, chooses to focus on a single sentence deep in the paper. The authors write: “Climate benefits from vehicle fuel substitution are uncertain (gasoline, light-duty) or improbable (diesel, heavy-duty)”. This is the only reference to vehicles in the paper. Yet the Times headline is “Study Finds Methane Leaks Negate Benefits of Natural Gas as a Fuel for Vehicles” and the Times story quotes the paper’s lead author as saying: “Switching from diesel to natural gas, that’s not a good policy from a climate perspective”. There are two large problems with this.

The first is that the observation has nothing to do with the analysis in the Science paper. The sole sentence in the paper that addresses diesel-to-gas switching cites a two-year-old Proceedings of the National Academy of Sciences (PNAS) paper – nothing new is added. More problematic is that the 2012 paper, which is mostly excellent, stumbles when it comes to comparing diesel with natural gas.

Why? If you dig up the references that the PNAS paper uses you’ll find that it assumes CNG-fueled vehicles are 20.7 percent less efficient than diesel-fueled ones. There is a citation for this – and indeed many CNG-fueled vehicles suffer a severe efficiency penalty. But this is far from universal. Diving a couple references deep reveals that the figures are not for CNG-fueled trucks in general but for urban buses – one of the worst cases (and perhaps the worst case) for CNG. Moreover, the original reference has pretty big uncertainty bounds, though those are dropped as the paper’s contents are exploited elsewhere. A quick spin through reports unearthed by a Google search about the CNG efficiency penalty reveals a wide range of estimates – from no penalty at all to a bit north of the 20.7 percent that the PNAS authors use – depending on the engine technology chosen and how (and where) the vehicle is driven. If you adopt the more favorable estimates for the efficiency penalty – which tend to correspond to more modern engines (though not universally) and to non-urban applications – switching from diesel to CNG is indeed mildly beneficial for the climate.

This raises the second issue. The PNAS study – which uses the EPA estimates that the Science paper challenges – already claims that diesel-to-CNG switching isn’t beneficial for the climate. The Science paper doesn’t say anything new about that. In particular, contrary to the Times reporting, it isn’t the new estimates of methane leakage that drive the Science conclusion – it’s primarily the implicit assumptions about engine efficiency. In fact, even with zero methane leakage, the methodology in the PNAS paper points to scant climate benefit (only about 10 percent lower emissions) from switching from diesel to CNG.

That suggests that if people are worried about climate dangers associated with switching from diesel to CNG, and they’re focused on methane leakage, then they’re looking in the wrong place – or at least missing the central piece of the puzzle. It might make more sense for policymakers to focus on boosting the efficiency of the natural gas engines that people adopt rather than fixating on methane leakage. Perhaps when CAFE rules give special credit for natural gas vehicles, the rules could impose minimum standards on those engines’ efficiency. This would be analogous to the way different kinds of natural gas fired power plants have been treated by EPA regulations – standards for new plants are far more welcoming to high-efficiency natural gas plants than for lower-efficiency gas technologies.

In the long run, it’s mainly carbon dioxide, not methane, that will determine how much the planet warms. Making sure natural gas powered vehicles are as efficient as possible would cut the carbon dioxide emissions associated with them – and, as a bonus, the amount of methane released too.