Book Review – Insurmountable Risks

Insurmountable Risks: The dangers of using nuclear power to combat global climate change by Brice Smith. A report for the Institute for Energy and Environmental Research 2006

Books on nuclear power don’t always easily cross the Atlantic, but this one is going to be invaluable. It starts by looking at two representative “nuclear growth scenarios” – one from the Massachusetts Institute of Technology (MIT) and one from the University of Chicago.

MIT suggests a global growth up to 1,000GW by 2050. Since all existing reactors would be closed by then, it means 1,000 new reactors. One might imagine that a tripling of installed capacity would have a significant impact on carbon emissions, but it doesn’t. Due to increases in demand nuclear power accounts for only 19.2% of the world’s electricity in 2050 compared with 16.3% in 2000. In the US, where the largest share of nuclear construction is assumed to take place, carbon emissions from electricity production actually increase.

The failure of nuclear power, so far, according to Smith, has been largely due to its inability to compete economically with coal and natural gas. The very large federal subsidies proposed in the US to make nuclear power appear more attractive to investors, according to Standard and Poor’s: “…may not provide sufficient incentive to pursue new construction.”

MIT says that although the cost improvements suggested for new reactor types are “plausible but unproven” even if they were all somehow achieved, which is highly unlikely, coal would remain cheaper and natural gas would only exceed the price of nuclear power in the highest fuel price scenario. The University of Chicago found that by the time the eighth plant is built – when cost savings due to learning would be complete – the cost of electricity would still remain higher that fossil fuels.

But, of course, cost comparisons with fossil fuels are not the most useful when looking at the role nuclear power might play in mitigating the impacts of climate change. In looking at MIT’s coverage of carbon taxes, Smith points out that the resultant increase in electricity costs would lead to an increased focus on efficiency and reductions in demand growth which would favour more flexible types of generation than large nuclear plants. Smith quotes a UK Government, Policy and Innovation Unit report which suggests that for a cost of between £130 per ton and minus £640 per ton, energy efficiency measures could reduce carbon emissions by 65 million tonnes by 2050. In contrast, nuclear power would cost between £70 and £200 per ton of carbon and could only reduce emissions by 20 million tonnes by 2050. Rather oddly, Smith makes no mention of studies by the Rocky Mountain Institute, and Amory Lovins, which many of us in the UK have found so inspiring.

Combined with energy efficiency, wind power offers the most economically competitive way to provide carbon-free electricity supplies. A joint study by Greenpeace and the European Wind Energy Association shows wind could be providing 10% of global electricity demand by 2020.

One Americanism perhaps is where Smith discusses the advantages of CHP, and says the number of industries large enough to efficiently operate their own generators limits its potential. Is he implying that domestic district heating wouldn’t work in America? If so, then what about University campuses, shopping malls, hospitals, schools and so on. His section on fuel switching suggests using liquefied natural gas as a transition fuel in combined cycle gas turbines – whereas in Europe there is a growing body of opinion that using gas in this way without CHP would be criminally inefficient.

Chapter Three is devoted to discussing nuclear power’s connection to the proliferation of nuclear weapons, which Smith describes as the “largest single vulnerability associated with the expansion of nuclear power”. A large increase in global nuclear capacity implies a proportional expansion of uranium enrichment. The diffusion of knowledge and the increase in trade in specialised materials and equipment would make it progressively more difficult to identify clandestine weapons programmes.

Also worrying is the significant prominence which reprocessing appears to have gained in the US in recent years where up to $580m has been allocated over the next three years to research and Bush has launched the Global Nuclear Energy Partnership to promote, so-called proliferation resistant recycling technologies. Under the MIT scenario the use of MoX would increase. It would take just 0.0025% of the MoX manufactured every year to provide the plutonium for a nuclear bomb. The number of reprocessing facilities required – given the inherent difficulty in attempting to safeguard large plutonium separation plant – would pose a huge proliferation risk.

In a useful section which explains how advanced reprocessing technology might work, Smith says it might offer some non-proliferation benefits compared to current technology, but would still pose significant risks if deployed on a large scale. The MIT study itself concludes that, on proliferation grounds, there are “significant questions about the wisdom of a global growth scenario … if the nonproliferation regime is not strengthened, the option of significant global expansion of nuclear power may be impossible.”

Proposals to restrict access to fuel cycle technologies are as old as the technology itself and are still politically untenable. Dr ElBaradei has suggested they should be under international control. But similar initiatives have failed in the past, and are likely to be just as unacceptable today. Smith summarises the case against using sources such as the Congressional Office of Technology Assessment which says there would be strong objections because it would reinforce “the discriminatory aspects of the NPT”.

A second unique vulnerability of nuclear power is the potential for catastrophic accident, or terrorist attack. The chapter on nuclear safety draws heavily on the work of David Lochbaum and the Union of Concerned Scientists. The accident rate for many complex technical systems follows what Lochbaum calls a “bathtub curve”. In other words, the accident rate is higher during the initial shakedown phase when the plant is new; then things settle down for a while, until ageing problems increase the likelihood of accidents again. Smith gives a detailed discussion of what he calls “the single most important discovery to date regarding degradation of reactor safety due to ageing”, which was the reactor vessel head cracking at the Davis-Besse plant in Ohio. Given the amount of cracking in various different components of UK reactors, this story offers a warning about the dangers of operators and regulators giving too much weight to commercial considerations. On new reactors Smith warns against curtailing public participation in the licensing process – US history shows that interveners have had a positive impact on safety. There is a useful section too on the various studies of the risk of terrorist attacks on spent fuel ponds.

The flawed logic behind Probabilistic Risk Assessment is highlighted by the prediction that a Three Mile Island type accident would occur once every 100,000 years – made only five years before the partial meltdown. But even if these risk levels are correct, the global growth in reactor numbers proposed by MIT would produce an unacceptable accident frequency – with 4 core damage accidents by 2055.

According to MIT, the worldwide deployment of 1000GW of nuclear capacity would require a new Yucca Mountain sized nuclear dump to be opened somewhere every three or four years. But even without new reactors, the US will need a second dump by 2012, if Yucca is to stick to the statutory limit. Smith emphasises the uncertainties regarding geological disposal and highlights a number of examples of errors in previous models used to support Department of Energy decision-making.

An appendix to “Insurmountable Risks” addresses the question of the adequacy of uranium supplies. In this Smith appears to be seeking to shift support away from reprocessing, MoX and fast reactors, which all have weapons proliferation implications, by showing that uranium resources are plentiful. A laudable aim, but unfortunately Smith does not appear to have addressed the argument made by Jan Willem Storm van Leeuwen about the increasing amount of carbon emissions from uranium mining as we are forced to turn to poorer quality ores. Nevertheless this book would be a useful addition to any UK campaigners armoury.


Published: 26 October 2012
Last updated: 23 November 2012