First blog post of the new semester is by our newest environmental economics faculty member, Professor William Pizer of the Sanford School of Public Policy and the Nicholas Institute for Environmental Policy Solutions.
Last week I gave a talk to the Orange/Chatham County Sierra Club on this topic. Since the disaster at the Fukushima Daiichi Nuclear Power Plant last spring, there appears to be a global retreat from nuclear power. Countries planning to expand nuclear power are reconsidering and countries that were already uncomfortable with nuclear power – namely Germany – have made plans to phase it out. What does that mean for climate change? Can we successfully address the threat of climate change without leaning on nuclear power?
To address this question, I first looked at a number of stabilization scenarios produced by various modeling teams over the last few years to see what they said about the potential role of nuclear power and then tried to understand what their analyses implied about both costs and other consequences. Here, stabilization scenarios represent views by various experts about how we might limit atmospheric concentrations of greenhouse gases to certain levels, thereby limiting the impacts of climate change. There are a wide range of views about the appropriate or achievable stabilization level, but they can be loosely grouped in terms of whether they target 450, 550, or 650 ppm (parts per million) of carbon dioxide equivalent concentrations. Carbon dioxide “equivalent” (or CO2e) references the inclusion of other greenhouse gases, expressed in terms that can be added to concentrations of carbon dioxide (CO2). Again, roughly speaking, this amounts to best guess warming of 1.5, 2.5, and 3 °C of average global warming (see discussion here).
Before talking about the specific implications for nuclear power, it is useful to first note that these stabilization scenarios have fairly dramatic implications for fossil fuel use generally. 450 ppm CO2e scenarios imply that global emissions CO2 emissions, which are otherwise forecast to perhaps double by 2050 (see slide 8), need to fall by 50% or even more. 550 ppm CO2e scenarios imply 2050 concentrations that are roughly the same as today, and 650 implies maybe a 50% increase (see slide 6). Because emissions are roughly proportional to fossil fuel use, this implies dramatic consequences for energy use around the world.
What kind of consequences? Generally, four. (1) Less energy use overall; (2) More renewables; (3) More nuclear; and (4) Carbon capture and storage (CCS), where fossil fuels are burned and the CO2 is captured and pumped deep underground. The first two consequences are important, but limited. Energy conservation can help countries that use too much energy per capita, but cannot address the needs of growing countries with large development needs. Currently, 39% of the world’s population does not have access to basic energy services (see slide 9).
Non-biomass, non-hydro renewables – solar and wind – are an important piece of the puzzle, but are limited because of their intermittency. This implies higher costs, because backup power needs to be available, but also penetration limits because of stability issues. Large scale hydroelectric power allows storage, but frequently has adverse environmental impacts (e.g., this report). New developments in biofuels are promising, but raise issues with a potential food-fuel trade-off (see, for example, these essays) as well as land-use and lifecycle emission issues. This leaves two large potential sources of low-carbon emissions: nuclear and fossil with CCS.
We see exactly these two technologies, along with non-biomass renewables, dominating power generation in the scenarios produced by a recent analysis under the U.S. Climate Change Science Program (slide 15). This 550 ppm CO2e global analysis was carried out by 3 of the top modeling teams in the United States. What is interesting is that the two technologies play different roles in the three studies. The MIT team, for example, assumed that nuclear power was constrained to current levels by political and/or proliferation concerns. Meanwhile, the Stanford and Maryland models assumed nuclear power more than doubles by 2050.
A more recent study by the International Energy Agency (IEA), in its World Energy Outlook 2011, had nuclear more than doubling by 2035 in a 450 scenario (slide 10). But their study, which came out after Fukushima, also considered a “low nuclear 450 scenario” where nuclear power actually falls by 50% in 2035 (slide 13).
Finally, in a 2009 analysis of H.R. 2454, the domestic climate change bill passed by the U.S. House of Representatives, the U.S. Energy Information Administration considered a “limited technology” scenario where both capture and storage and nuclear technologies were assumed to be 50% more expensive. While the base case for H.R. 2454 showed nuclear power in the United States doubling by 2030, the “limited technology” case had nuclear power staying at current levels (slide 16).
So at this point, I think it is fair to make two observations. First, stabilization of greenhouse gases at the more ambitious levels envisioned by most scientists will require significant limits if not absolute reductions in carbon dioxide emissions over the next half-century even as energy use rises to meet development needs in emerging economies. Second, significant reductions will lean, to a large extent on nuclear power or capture and storage technologies, along with non-biomass renewables – but the exact amount remains an open question. Generally, less nuclear means more CCS, and vice-versa. Solar and wind are important, but become problematic at high penetration levels due to intermittency.
My final question is what does this mean for costs. The IEA study indicated that their low nuclear 450 scenario only raised costs by 10 percent compared to the ordinary 450 scenario. The EIA study, where both CCS and nuclear were limited, raised allowance prices by almost 100 percent holding other policy variables equal (e.g., comparing the “limited technology/limited international offset” case to the “limited international offset case” in slide 20). While these are just two data points, the lesson is this: limiting nuclear power is not impossible and perhaps not even that much more expensive – but it puts a lot more pressure on other technologies, notably CCS. Without nuclear and CCS, it becomes a lot harder if not impossible to achieve greenhouse gas stabilization.
Ultimately, each of these choices involves troubling risks that are difficult to quantify. Climate change, nuclear power, and CCS – pumping billions of tons of carbon dioxide deep underground – all pose risks that we would prefer to avoid. But we cannot avoid all of them, and putting an absolute premium on avoiding one risk only exacerbates the other risks. While the choice is not really “nuclear power or climate change,” thanks to CCS and other options, prudent policymaking will require a careful balancing of multiple risks.