Nuclear power is often touted as being an important tool in the array of measures needed to help prevent climate change. Some environmentalists and climate scientists have recently been gaining attention because of their support for nuclear power as a tool for helping reduce our CO2 emissions. However, even though the goal of dramatically reducing CO2 emissions is critically important, there are still many reasons to be skeptical of nuclear power as the solution for creating a long-term sustainable energy system.
We can look at nuclear power from several different points of view. It is true that careful life-cycle analyses show that total CO2 emissions from nuclear power are comparable to those of renewables, and at least 10 to 20 times less than even natural gas electricity for a given amount of electricity. Some argue that uranium mining entails emissions, but a life-cycle analysis is designed precisely to take these upstream factors into account.
Greenhouse gas emissions are only one aspect to consider, although key to the renewed discussion among environmentalists. All currently operating nuclear power plants use uranium, a relatively plentiful element, but still a finite resource. Therefore, issues of scarcity will appear at some point, likely later in this century. Nuclear fuel costs are a relatively small fraction of operating costs and therefore have only a weak impact on the cost of electricity, so scarcity may or may not be a major factor. There are also nuclear power technologies that essentially allow far more efficient use of uranium, or even to breed additional nuclear fuel; thus far, these technologies have not proven successful.
There are far more important economic issues with respect to nuclear power. Nearly all technologies decrease in cost as we gain experience and as the scale of installation increases. We have seen this most dramatically in the case of solar photovoltaics over the past decade, a technology for which costs have dropped by 80-90% from where they were a few years ago. Nuclear power may be the only counterexample to this pattern (see here and here). There has been a steady increase in the cost of building nuclear power plants over time, partly due to the need for redundant and complex safety measures.
Adding nuclear power is now significantly more expensive than most alternatives, including wind power. New plants to be built in England are receiving a 35-year government price guarantee to the power companies that is twice the current cost of electricity there. The European project to build relatively inexpensive reactors with substantially reduced construction times in France and Finland has not worked out that way at all. Likewise, plants being built in Georgia can take advantage of several billion dollars in federal government loan guarantees and work with guaranteed rate-hikes to recover whatever costs are incurred in construction. Finally, in the U.S. there is a federal program that limits costs to the nuclear power industry in case of accidents, a kind of insurance for which we the taxpayers will cover costs for which no private insurer would ever take the risk.
Germany made a decision to exit nuclear power within a decade, by which time the country will be providing 35-40% of its electricity from renewables. In the US we are in an even stronger position to both increase energy efficiency, since we use more energy for the same economic output than other industrialized countries, and at the same time take advantage of our vast renewable energy resources. The National Renewable Energy Laboratory has provided scenarios for reaching 80% renewable energy electricity by the middle of the century, the same goal that Germany has for its power system. However, in a power system with a high penetration of renewables, nuclear power is poorly suited for providing the flexible complementary generation capacity that will be necessary. Furthermore, as renewables gain greater penetration, there will be less and less need for what is conventionally known as baseload power, those sources running 24/7. One secondary consequence of this dynamic is that the economics of nuclear power become even worse if the plants are restricted in the number of hours they are required to operate.
The issue of nuclear waste disposal, which has not yet been satisfactorily resolved, has not even been mentioned yet. In addition, there is no other energy source about which we must continually fear the slightest human error. It is clear that coal power currently kills vastly more people than nuclear power ever has. But the catastrophic events around Chernobyl, Three Mile Island and Fukushima have left long-term scars over large areas and dislocated hundreds of thousands of people.
There is one more issue with nuclear power that is rarely mentioned. If we look at debates right now about Iran's nuclear program, the subtext is that we (Industrialized countries? Western countries? The U.S. alone?) should be allowed to determine who else is admitted to the nuclear club, or at the very least, that we should have the option of tightly controlling access to nuclear technology and fuel. This attitude represents a new kind of energy colonialism and should be unacceptable in the 21st century.
Fundamentally, a sustainable energy system cannot be one that raises fears of societal dislocation, catastrophic accidents or the spread of weapons as a by-product. It is also hard to reconcile sustainability in a global sense with a technology that results in a few countries, the nuclear powers, determining which other countries may or may not have access to the technology. Most importantly, a sustainable energy system cannot be one that leaves its waste to be cleaned up by future generations - especially when other options do exist.