September 14th, 2011 by David McCabe, Atmospheric Scientist
The last few months have seen a flurry of academic papers investigating whether using natural gas for power generation creates more global warming than using coal for power generation. A few have reached the startling conclusion that using gas for power is just as bad, or worse, than coal. The most recent of these is by Tom Wigley, a global leader in climate science, and therefore bears special examination. As we’ll argue below, natural gas is no climate panacea, especially over the time scales that Wigley examines. We need zero-carbon energy. But it is also important to consider how we get to that future, and natural gas – coupled with carbon capture and storage and tight controls on methane leaks – will likely have a big role to play there in the next few decades. It is critical that we accurately account for the climate impacts of gas, and we don’t agree with Wigley’s approach in two key areas.
Wigley does not dispute that switching electrical generation from coal to gas will reduce CO2 emissions. But according to Wigley much – or all – of the climate benefit from the reduced carbon would be offset by two other changes in emissions: 1) methane leaks and releases from increased gas development and 2) reductions of sulfates from coal burning.
While nearly all of the press coverage of the paper has focused on the methane, the sulfate is more important than the methane in Wigley’s models for any reasonable natural gas leak rate. The sulfate is more important even if the leak rate for gas is twice what we think it is. Let’s take a look at the sulfate issue first.
It’s well understood that coal-fired power plants emit sulfur compounds, which react in the atmosphere to form a haze of sulfate particles that reflect incoming sunlight and cool the planet. These sulfates are a very serious air pollutant, increasing respiratory illnesses and deaths. They are also the major cause of acid rain. These public health and environmental impacts have driven successful regulations to reduce sulfur. Largely through use of scrubbers, the US has halved sulfur from power plants over the past 20 years.
The drop has been much faster in Europe (pdf, p. 16). During the same period, coal consumption has dropped a bit in Europe, and in the US it has increased.
Sulfur emissions do remain high in other parts of the world, including China, where roughly one new coal plant has come on line every week over the past few years. While there is no chance that these new coal plants will retire any time soon, sulfur emissions from coal plants in China have declined significantly since about 2006, due to installation of scrubbers – even while coal plants have continued to come on-line at a very rapid pace.
Thus we see no support for Wigley’s assumption that over the next 50 years conversion from coal to gas will cause sulfur reductions. Coal to gas conversions that do occur will most likely happen after the sulfur has already been removed, and we will already be experiencing the climate response. That train has left the station, and we are benefiting from the cleaner air.
Now, let’s look at natural gas leaks – but first, we should mention that we consider fixing leaks of methane from natural gas one of the highest climate priorities. EPA has recently proposed rules to reduce those leaks by about a quarter, and we will urge them to tighten those rules considerably. We are also working to make sure that other nations minimize these leaks.
For the US, the higher leak rates that Wigley considers (up to 10%, but much of his analysis uses 5%) are probably too high. Wigley cites Howarth et al‘s 2011 study to justify his values. We’ve noted before that there are large problems with Howarth et al; it is not a credible study. The official US EPA inventory for methane implies a leak rate around 2% for US natural gas systems. This number has increased a lot in the last year, as EPA fixed large holes in the inventory for emissions from drilling and gas wells. There remains uncertainty in those numbers, but there’s not much chance that the leak rate is 5%, let alone 8% (Howarth’s high estimate) or 10%. Some leaks and releases are probably undercounted in EPA’s numbers, but others are quite likely overcounted. If leaks were at the high end of Wigley’s range, scientists would have picked up traces of those leaks in ambient measurements. The measurements we know of point to leak rates roughly around where the inventories currently lie.[1] We badly need better information about methane leaks (and the gas industry needs to stop fighting efforts to improve that information), but there is no real evidence that leaks are in the 5 – 10% range.
If leak rates are around 2-3%, Wigley finds that the methane from natural gas adds little net warming (in part because there is methane from coal mining too). This result is in line with other recent work which finds that leaking gas does add significantly to the climate footprint of electricity from gas-fired power plants, but that gas still has a smaller footprint than coal.
In the bigger picture, Wigley is modeling a scenario that we already knew would be a climate disaster. We need much, much quicker and deeper reductions of CO2 than switching from coal to gas can deliver. Both of the scenarios that Wigley examines increase CO2 emissions from power plants every year until 2100, which will clearly lead to dangerous climate change. The slight difference between the gas and coal CO2 curves is not nearly enough to prevent that, aside from the questions about methane and sulfates.
Our bottom line conclusion on the Wigley paper, then, is as follows:
- Sulfates, which are mainly produced by burning coal, are certainly cooling the planet; removing them will result in warming. However, sulfate removal has been and will be driven by air quality concerns and not by retiring coal plants and replacing them with natural gas.
- The greenhouse gas footprint of coal is still worse than that of natural gas, even when the harmful methane emissions from natural gas wells and pipelines are factored in.
The fact that Wigley’s analysis is flawed doesn’t mean natural gas is a panacea for global warming. Natural gas might be useful for accelerating CO2 reductions over the next decades (especially if the CO2 in the exhaust can be captured and sequestered underground). Burning natural gas – particularly without sequestration – certainly is not a long-term solution to our climate emergency.
To have any hope of stabilizing temperatures in the face of increasing greenhouse gases and decreasing sulfate, we need zero carbon solutions and we need to deploy them quickly. We need to get the leak rate from gas near zero, to rapidly reduce other short-lived climate forcers, to begin sequestering carbon, and to explore geoengineering ideas. Let’s focus on all of that and spend our time carefully analyzing, based on quality data and reasonable assumptions, how to best get to zero carbon – not in a century or two, but in decades.
[1] Before EPA increased their inventories for methane from natural gas, two papers based on ambient observations suggested that the then-current inventories were too low. Very roughly, their results are consistent with the current leak rates, around 2 or 3%. These studies are not accurate enough to rule out a higher leak rate than the current inventories, but they suggest that 5 – 10% is too high.
Tags: air quality, carbon capture and storage, carbon dioxide, climate change, coal-fired power plants, emissions, greenhouse gas emissions, methane, natural gas
This entry was posted on Wednesday, September 14th, 2011 at 7:03 pmand is filed under Climate, Fossil Transition. You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.
