The phantom solution: Alan Roberts explains the danger of pursuing nuclear power when it will fail to meet both environmental and energy needs.

It was once hailed as the solution to humanity's energy demands, promising electricity 'too cheap to meter'. But nuclear power has not lived up to its advance publicity. Thousands of stations, it was confidently predicted, would greet the millennium; but they neither exist nor are on order. After more than half a century of development, the nuclear industry can show less than five hundred power stations, yielding in 2002 around 2 per cent of the world's total energy production and only about a sixth of its electrical output. This last fraction had actually slipped back a little from the peak achieved by nuclear power fourteen years earlier in 1988. In its biggest market, the United States, no new nuclear stations have been ordered for over twenty-five years. In the whole world, only three new reactors came on line in the five years from 1998 to 2003.

Resuscitating a Dinosaur?

This is the picture of a stagnant or declining industry, one that many commentators, those opposed to the industry in particular, have labelled a 'dinosaur', with all its future behind it. But in recent years, we have heard voices striking a far different note. A comment in the Guardian (12 August 2004) is typical:

Nuclear power is back on the march. Reviled and rejected for 25 years as man's most dangerous and unsustainable fuel source ... is it possible that public opinion is wrong, and that nuclear should be the fuel of choice of the future?

Other nuclear proponents have stressed how it can save the landscape from unsightly windmills and insulate the economy against future oil shocks. But the preponderant argument by far is environmental: to ward off climate change, we must go nuclear.

It would be wrong to think that this new case for nuclear power is made only by the old and new nuclear club. Listen to quite a different sort of proponent, none other than the scientist and environmentalist responsible for the Gaia hypothesis:

I hope that it is not too late for the world to emulate France and make nuclear power our principal source of energy. There is at present no other safe, practical and economic substitute for the dangerous practice of burning carbon fuels.

Thus James Lovelock. And while his is not the only voice now advocating nuclear power on environmental grounds, he is certainly the one most distinguished and respected in both activist and scientific circles.

Another stream of environmental opinion, while not so emphatic, has unmistakably shifted away from a blanket condemnation of nuclear power. For example: 'Paul Allen, development director at the Centre for Alternative Technology in Machynlleth ... says he is not endorsing nuclear, but is trying to keep an open mind'.

It is evident that an informal body of 'Environmentalists For Nuclear Power' has emerged. Are they right--has the unacceptable now become a lesser evil? How we answer will obviously depend on how seriously we regard the prospect of climate change. To decide, there are two preliminary issues that need to be first clarified: is there indeed a threat of human-induced climate change that would have disastrous effects on our own species and others, and can nuclear power significantly counteract this threat and thus help to avert or at least reduce the damage?

Is Climate Change Really Happening? Is It Human Induced?

In 1988, the World Meteorological Organization and the UN Environment Program set up the Intergovernmental Panel on Climate Change (IPCC). The Panel's Third Assessment Report (2001) drew on the research, authorial and reviewing services of experts from 56 countries, including 16 people from Australia, and left little doubt that significant 'anthropogenic' (human-induced) climate change has already occurred.

The direct indicators are briefly listed in Table 2-1 of its Synthesis Report, (available at <www.ipcc.ch>); one is the global mean surface temperature, which increased by about 0.6 degrees Celcius in the course of the twentieth century. Other indicators include the decreased daily range of land surface temperature in the last half of the twentieth century, and (listed as 'likely') the decrease of cold and frost days and the greater severity and frequency of droughts in some regions (such as parts of Africa and Asia).

Why is this, in all probability, the result of human activity? The clues are in the same table, where the amounts of various gases in the atmosphere are given. Typical is the finding for carbon dioxide: over the last couple of centuries or so--that is, over the industrial era--its concentration in the air increased by around 30 per cent. Now, carbon dioxide acts like the glass roof of a greenhouse: it lets sunlight through easily, but blocks off the waves of milder heat travelling up from the ground. Thus it cuts down this 'upward' heat loss, so there results a higher temperature at ground level than would otherwise be the case. For obvious reasons, it is called a 'greenhouse gas'; so is methane, for example, of which the air now holds some 150 per cent more than in 1750.

The Synthesis Report suggests the most likely climatic sequels of the higher surface temperatures created by this more intense greenhouse effect. They include further ocean warming and glacier melting (with concomitant rise in sea level and risk to populations of small islands and low-lying coastal areas); increased threats to human health; greater risk of extinction for vulnerable species; exacerbated water shortage in dry areas; and more frequent occurrence of 'extreme' weather conditions (hurricanes, heat waves, heavy rainfall) implying increased risk of drought and floods.

Since 1988 when the IPCC was formed, a great deal has been added to our knowledge of the factors involved in climate change and their interaction. Unfortunately, two things can be safely said about this further knowledge--first, that it still leaves great areas of uncertainty; and second, that the likely overall picture it paints is of a world closer than had been thought to a crisis greater than had been thought.

This last disturbing implication of the work was strengthened in the first week of February 2005, when new observations and calculations were presented to a gathering at Exeter of hundreds of scientists from relevant fields. A New Scientist article quoted researchers emphasising 'evidence that the danger is more pressing than was thought. "The sleeping giants are being woken," several said'. This last metaphor referred to the 'runaway' events mentioned above:

Triggering irreversible changes in natural systems, such as the melting of polar ice caps. Once we pass these thresholds there will be no return. And the conference heard that we could be closer to the brink than previously supposed.

If you incline to the view of sceptics such as David Bellamy (botanist) and Michael Crichton (novelist), then of course the question of turning to nuclear power for the solution of the climate-change crisis does not arise, since there is no such crisis needing solution. My own view is that, on the contrary, there is now an overwhelming case that human activities are changing the climate, and that a catastrophic outcome is likely if massive remedial action is not promptly undertaken.

In all this, it is hard to reject one at least of Lovelock's views: 'whatever the objections there are to nuclear, it's nothing like as great a danger as just leaving things as they are and going on burning fossil fuel'. He is right to insist that, if nuclear power is to be rejected, it should not be out of some fundamentalist abhorrence that has it necessarily worse than any other evil. But ironically, the argument that follows is that turning to nuclear power as the answer to climate change would, in practice, carry the danger of essentially the policy Lovelock (rightly) fears: 'just leaving things as they are and going on burning fossil fuel'.

Most Greenhouse Gases Do Not Come from Electrical Power

First let us note an apparently widespread idea--that, if we switched to all-nuclear power stations, the dangerous injection of greenhouse gases into the atmosphere would then be eliminated, or effectively so at least.

This is far from the truth. The bulk of greenhouse gases are produced by human activities quite unrelated to electrical generation. A few figures indicate how subsidiary the climate-changing role of electricity generation is. On 25 February 2004 the European Economic and Social Commission of the European Union (EU) issued a report (called an 'opinion') on nuclear power and electricity generation. It showed inter alia that its fifteen constituent countries discharged over four million tonnes of greenhouse gases into the atmosphere in 1999; out of this amount, the generation of network electricity was responsible for less than a million--in fact, for under 21 per cent of the total.

This tells us the maximum that nuclear power could achieve in the way of reducing EU emissions. Let us suppose that all the non-nuclear stations in the EU countries are replaced by nuclear ones, and that these new stations are constructed, fuelled and operated without any greenhouse gases ever being generated, though we will see below how unreal and indefensibly optimistic this latter point is. Then even so, if this sweeping change in electricity generation were the only step taken, the EU countries would still be pouring into the atmosphere just on 80 per cent of their present load, or over 3,200,000 tonnes of greenhouse gases each year.

This figure for Europe is a pointer to the global role--significant but subsidiary--that power generation plays in inducing climate change. In other regions it can loom larger or, as in the state of California, smaller still. (California's power stations are responsible for only 16 per cent of its greenhouse gas emissions, less than a third of the 58 per cent due to transportation.) On the other hand, the coal-fired stations in Victoria, Australia, reportedly achieve 55 per cent of the state's total emissions. If this figure is accurate it may be an unenviable world record for which Victorians can thank their abundant supplies of inefficiently burning brown coal. Less dominating, and far less than the total weight of emissions, are the figures from the International Energy Agency (IEA) for 1999, which give the world emissions by electrical networks as something less than 39 per cent of the total.

Whichever figures we go by, it is evident that even total removal of the fossil fuel stations from the equation would still see a vast amount of greenhouse gases pouring into the atmosphere every year, continuing the fatal and barely understood process of changing our planet's climate. If catastrophe were waiting for us, this nuclear substitution could achieve no more than to delay its arrival--perhaps, if we were lucky, by a few years.

However, it would be disturbingly novel in both its ethical depth and its social sweep if we actually had to choose between these two fates: risking the destruction of the biosphere, or bequeathing to our descendants the unwelcome effects of one attempted remedy: the poisonous wastes, genetic dangers and virtually permanent responsibilities associated with an expanded nuclear industry.

Is this really the decision we have to make? The Guardian editorial writer (12 December 2004) believes so (after 'more than a week's worth of news on climate change'), and offers an unhesitating verdict:

Since nuclear power, for all its faults, does not contribute to greenhouse gas production, it may be used to make up the gap between today's carbon-bearing energy generators and tomorrow's green power sources.

The editorial does not say anything about the other emissions pouring the bulk of gases into the air--as we have just seen, measured in millions of tonnes each year. And contrary to its bald assertion, nuclear power can give a massive contribution to greenhouse gas production, as will appear below. But first, a few, not uncommon assumptions made in this passage should be examined.

The second question the nuclear camp faces is: where is all this uranium to be found? The first place to look is in all the uranium-bearing ore bodies throughout the world that are at present known or plausibly estimated ('Reasonably Assured Resources plus Estimated Additional Resources'). Taking the figures given by the industry, on the World Nuclear Association (WNA) website (<www.world-nuclear.org>) for example, it is straightforward to work out how long the deposits would last. This calculation has been done by van Leeuwen and Smith, at the Centre for Energy Conservation, Delft, the Netherlands, using the best figures available for 'burnup'--the heat energy actually obtainable from a tonne of uranium ore. Two burnup figures occur in the literature; van Leeuwen and Smith use the higher one, based on extracting energy more efficiently from the uranium fuel. So, if the present global output of electricity were obtained entirely from nuclear reactors, and as efficiently as best practice allowed, for how long could all the uranium of all the known ore bodies in the world keep them going? The answer: just under nine years. Thereafter, the world has no nuclear power stations operating and therefore no power stations at all.

With the operating life of a nuclear station at around thirty years or more, this is hardly acceptable--particularly taking into account that the construction and initial fuelling of a reactor requires significant energy input (see below) and associated greenhouse emissions. Obviously, this suggests we abandon the idea of replacing all the fossil-fuel stations.

Since it is the small amount of uranium in known ore bodies that limits the 'nuclear solution' so severely, it should be asked whether reactor fuel can be obtained elsewhere. Note first that the discovery and opening of new mines previously overlooked, even if equivalent in output to the whole of those known at present--and it is hard to imagine this--would not greatly alter the finding above: the operating period of the 'replacement' reactors would just be stretched to eighteen years, still well short of a normal lifetime.

The dismantling of nuclear weapon warheads under disarmament treaties creates another uranium source, which 'from 2000 ... is displacing about 9000 tonnes of uranium oxide per year from mines, which represents about 14 per cent of the world's reactor requirements'. Recalling that total world electricity output is about six times that of existing reactors, it is easy to see that on total replacement the nuclear stations would, at this rate of supply, obtain from the weapon stocks only about 14 divided by 6 per cent of their fuel requirements each year--that is, a bit over 2 per cent. This would not add even a month to the period over which the 'replacement' reactors could be fuelled--a period still under the nine-year figure quoted above from the van Leeuwen and Smith study.

Thus the known or estimated ore bodies and processed warheads could support between them only a few years of all-nuclear generation. However, the WNA site draws attention to other sources, and these further possibilities should be looked at.

Can Market Forces Produce New Sources of Uranium?

The WNA website states that:

It is now clear that uranium is not scarce and it is known that it averages almost two parts per million of the Earth's crust. There are substantial resources that are not yet fully proven. These so-called speculative resources are likely to be of the order of 10 million tonnes, about three times the known reserves.

It also mentions the enormous stores in granite and sedimentary rock (respectively four and two parts per million) and even in seawater (three parts per billion). These might sound like tiny fractions, but each is part of a colossal whole--the total amount of uranium in seawater alone is estimated at over four billion tonnes. Sketching the market dynamic that makes resource quantity increase with price, the WNA site goes on:

An orebody is, by definition, an occurrence of mineralisation from which the metal is economically recoverable. It is therefore relative to both costs of extraction and market prices. At present neither the oceans nor any granites are orebodies, but conceivably either could become so if prices were to rise sufficiently ... Measured resources of uranium, the amount known to be economically recoverable from orebodies, are thus also relative to costs and prices. They are also dependent on the intensity of exploration effort. Changes in costs or prices, or further exploration, may alter measured resource figures markedly. Thus, any predictions of the future availability of any mineral, including uranium, which are based on current cost and price data and current geological knowledge are likely to be extremely conservative.

The market dynamic acknowledged here by the WNA is important and has not always been allowed for, leading to a figure for 'current stocks' being wrongly treated as an absolute and eternal constraint. Thus, even if the van Leeuwen and Smith study shows that present supplies ('reasonably assured' plus 'estimated additional') are far too small to permit any large-scale replacement of fossil fuel by nuclear power, can we assume that more intense exploration will find uranium elsewhere if demand makes the price rise?

Unfortunately, no. What is needed is not a cost analysis, but an energy analysis. This means calculating the energy input at every step in the industrial process that starts with an untouched deposit and a greenfields site, and ends with uranium fuel rods being loaded as needed into a functioning reactor.

The major part of the van Leeuwen and Smith work cited above is precisely such an energy analysis. Of course, many of the 'energy costs' incurred in order to build a functioning reactor differ little from those met with in building a coal-fired station but, to grasp the major conclusions of the study, it is enough to stress the special character of the fuelling stage, at which the uranium is obtained and processed into form suitable for a reactor.

The van Leeuwen and Smith study calculates the quantity of energy expended in the process of building the reactor and preparing its uranium fuel. The electrical output from a nuclear station--or indeed, from any type of source--is never pure gain, and the study needs to itemise fourteen stages at which energy must be expended to create a functioning reactor.

The Nett Energy Gain from Uranium

The results can be understood in terms of 'nett gain': the amount of energy created from the uranium in the reactor, after subtracting the amount that had to be expended to build and operate it. It turns out that the 'nett gain' from uranium coming out of a given ore body depends crucially on how rich the ore was--that is to say, how much uranium was in each tonne of ore.

The van Leeuwen and Smith study dramatically illustrates this dependence. For a single fuel load, the figure plots the energy used against the richness of the ore. As the ore becomes less rich, the curve soars up to, and then past, the value for the total electrical output from that fuel load. This means, for such poor ores, the uranium is yielding less energy than was used in obtaining it.

This general phenomenon had already been noted years before. The researcher Kistemaker stated in 1976 that, with ore grades less than one part in five thousand, over half the energy in the uranium was lost in the extraction process. The van Leeuwen and Smith value confirms this, and gives the grade of total loss (a nett energy return of zero) as just under one part in ten thousand.

The grades already being worked in the major uranium mines today are mainly in a range well above this critical value, so that the reactors built to date have benefitted from comparatively rich ores and have thus yielded highly positive nett energy returns. But major sources also include values as low as one part in two thousand (for example, Olympic Dam in South Australia) or one part in five thousand (in South Africa). New sources could well be discovered and tapped if the price of uranium rises high enough, but what is crucial is their grade of ore. They would need to be richer than some of the deposits already included when counting today's 'resources'. Take the South African grade of one part in five thousand, for example; about half the energy to be eventually obtained from it would have to be used in making it into fuel--and this from the van Leeuwen and Smith study, which consistently favours the lower number for energy loss from inputs when there is doubt.

Speculative Resources

The WNA quotation above referring to 'substantial resources that are not yet fully proven ... so-called speculative resources', gives their amount as roughly triple known reserves. If indeed these speculative resources proved existent, and their richness was well above the value (about one part in ten thousand) at which the lost input energy becomes significant, they could extend the all-nuclear period from under nine years to nearly thirty years--close to a reactor's lifetime. But in fact we can draw no definite conclusion about their effect, even if they exist, since no indication is given of their likely richness--which determines whether they would be usable. The question is much more clear-cut when it comes to granite (four parts of uranium per million) or sedimentary rock (two!): the van Leeuwen and Smith findings make it obvious that they cannot be 'mined' for nuclear fuel.

Uranium from Seawater: the van Leeuwen and Smith work analyses the mining-milling process, and so is not relevant to the problem of recovering uranium from seawater, on which research still continues. But the Analytical Center for Non-Proliferation Problems, a Russian institute naming the US Department of Energy as a 'major partner', closely follows progress in this field, for obvious reasons. The Center has summarised where these extraction efforts stood--or stagnated--in 2004:

Although research and development for recovery of this low-concentration element by inorganic adsorbents such as titanium oxide compounds, etc. has occurred since the 1960s in the United Kingdom, France, Germany, and Japan, the present status of all such research has been subsequent stoppage due to low recovery efficiency.

The Fast Breeder: In the early days of the industry, great hopes were placed in the 'Fast Breeder' reactor, which could in principle produce more fuel than it used up. While there does not appear to be any firm evidence that this hope has been realised in practice, several fast breeders have run long enough to meet with disturbingly severe vicissitudes. The Fermi fast breeder near Detroit was the first to set this pattern, and the accident it suffered in 1966 was the subject of Fuller's gripping book We Almost Lost Detroit. Since then, the incidents reported include the not atypical examples below.

In Japan:

On January 27 [2003] the Kanazawa branch of the Nagoya High Court ... handed down a decision favoring local residents who had sued the government seeking to nullify its approval of the prototype Monju fast-breeder nuclear reactor in Tsuruga, Fukui Prefecture ... The reactor, which is overseen by the Japan Nuclear Cycle Development Institute (JNC), has been shut down since an accidental leak of sodium in 1995.

In France:

From its entry into operation, into 1994, the [French fast breeder] Superph,nix had operated only the equivalent of 174 days at full power and, in the words of deputy Bataille, in 1996, had 'collected an impressive series of accidents' ... On 3 July 1990, the reactor was shut down because of impurities in the sodium in the core ... in April [1998] Nersa was asked to begin the permanent shutdown process.

Russia operated a large breeder for years, but I have found no actual breeding of new fuel reported. India intends to build a fast breeder at the Indira Gandhi Centre for Atomic Research, Mamallapuram, and hopes to avoid the succession of troubles that have plagued the fast breeder prototypes of nation after nation. 'If our ancestors could build temples that last 13 centuries, it should not be beyond us to build a reactor that lasts 60 years', says Baldev Raj, director of the IGCAR.

Of course, unlike the fast breeder, the Mamallapuram temples were not cooled by liquid sodium, an element that is highly toxic and flames spontaneously if exposed to the air.

As will appear below, the failure to tame the fast breeder into a reliable supplier of fuel has been a crucial barrier to the expansion of nuclear power.

The Thorium Cycle: the WNA website, on its 'Supply of Uranium' page, also points out:

Today uranium is the only fuel supplied for nuclear reactors. However, thorium can also be utilised as a fuel ... Thorium is about three times as abundant in the earth's crust as uranium.

It is worth noting that experience with reactors based on the thorium cycle is minute compared to the history of uranium's usage, that the use of thorium involves complications not encountered with uranium, and that it is hard to imagine less than a few decades as necessary for any move to the thorium cycle. However, for present purposes it is enough to recall that being 'abundant' is not necessarily the same thing as being usable, as we saw above with the massive but useless quantities of poor-grade uranium. How rich then are the thorium ores? From an Oak Ridge National Laboratory paper:

While natural thorium is more abundant than uranium in the very long term, most of the thorium reserves are present as very low grade ores; the amounts of low-cost reserves are about the same for uranium and thorium.

This suggests that resort to the thorium cycle, even if it were satisfactorily achieved, would not allow any significant expansion of nuclear power beyond the limits set by uranium stocks.

What Nuclear Can Do

It has been straightforward, though requiring the detailed analysis of the van Leeuwen and Smith study, to show that ore bodies already under exploitation--that is, mines existing or ready to open--are unable to provide the uranium that would be needed for all-nuclear electrical generation or anything approaching it. Examining the various proposed alternative supplies (poorer deposits, granite, sedimentary rocks, seawater, weapon warheads, breeders) has been more complicated, because these other industry-cited sources needed individual attention, and, to be properly studied, required the van Leeuwen and Smith energy analysis--necessarily much more complex than a cost analysis.

These other sources do not seem able to alter the general picture, unless some surprising developments occur--for example, a scientific breakthrough in the process of extracting uranium from seawater; or the conversion to peaceful purposes of a much greater number of warheads than envisaged up to now; or the discovery of new deposits about as rich as those known now but with about three times the tonnage of uranium. The issue would then have to be re-thought, since alleviating the climate change crisis through nuclear energy could no longer be seen as a simple impossibility.

The nuclear industry has a great advantage over its fossil fuel rivals: its harmful emissions are invisible. The smoky clouds from a coal-burning station manifestly threaten damage (and of course deliver it); but radiation slips by unnoticed. The functionality of a coal-burner thrusts itself into our vision with its grimy chimneys and haphazard brick walls, but a nuclear installation has no huge intake of materials every day to cater for and can be designed to look like a pleasant chemical factory. There is no outward sign of the gigantic energies deep inside, which, from a single human error, can metamorphose into a Chernobyl, making a whole region uninhabitable for generations. Rather, on its green lawns sheep may safely and cleanly graze.

All this helps to shape the nuclear station into an icon of the technological dream. It is a powerful symbol--powerful in its grasp on the imagination, and (supposedly) powerful in what it can accomplish in reality, for who can doubt that power after Hiroshima and Nagasaki? With this genie working for us, surely we need not fear climate change?

In beckoning us into this ideological trance, the nuclear power prescription carries its greatest danger. Of course, it is hardly grounds for celebration when share values in a uranium mining firm soar, or when the beleaguered industry sells a few more reactors. But for all their perils, these are minor matters in comparison. What is of real concern is that pushing the nuclear barrow could substitute for effective action of the required scale against greenhouse emissions.

What Should Be Done--and Is Already Underway

As a 'solution' to climate change, nuclear power fails in every important respect. But it retains an iconic glitter, far better than the plodding, everyday concerns that have nevertheless crippled its development--questions about how much it costs, where to put its waste, what if its operators turn out to be only human and error-prone ... To the dazzled eye, such banalities can be simply irritating. ('Tomorrow is already here, and they want us to worry about cents per kilowatt-hour!') Compared to the breathtaking sweep of the nuclear 'solution' that applies a single, ultra-powerful, hypermodern fix, the actual, effective means of combating climate change could be made to seem banal.

Indeed, the very first policy that recommends itself can be put in a banal enough way: 'We should stop wasting energy!'; or, less brusquely: 'Be more efficient in how we use available energy'. Add to this that it is essential in the long term for sustainable energy sources to be developed, and the fundamentals of an effective policy are laid out. Note that what consumers see as their needs is taken to be essentially unchanged; it is a question only of the processes used in satisfying them. (Many of us, perhaps, would like to see a future in which the pattern of these consumer demands has changed to a greater or less extent, but no such developments are being considered here.)

It then becomes a matter of studying whether the threat of 'inadvertent' climate change can be removed simply by using energy more efficiently. What is first involved here is examining whether the improved methods can reduce the amount of energy used--thus reducing harmful emissions--to keep greenhouse gases below danger levels. Then there is the further question of the economic and political plausibility of these methods.

These are measures to be supported and disseminated, but they should not be overrated. As every climate conference makes increasingly clear, governments are doing too little and may well be too late. Governments were generally irresponsible when the question of nuclear power originally arose, educated only when the public moved to protect generations to come and held those governments to account. To achieve timely action against climate change will probably require a similar public awareness and a similar political pressure.

For a last word, let us turn to the work of Keepin and Kats in a 1988 issue of Energy Policy. A quantitative figure emerges from their analysis when they consider a six-fold expansion of nuclear power suggested by a nuclear advocate, and ask the question:

To what extent does this sixfold nuclear expansion scenario effectively contribute to the greenhouse warming problem, in the sense of diverting funds away from more promising CO2 abatement strategies?

Their analysis allows them to give, in the case of the United States, a startling figure for this 'opportunity cost': Each dollar invested in efficiency displaces nearly seven times more carbon than a dollar invested in nuclear power ... every $100 invested in nuclear power effectively releases an additional tonne of carbon into the atmosphere.

Over the years since 1988, each of the cost figures involved--for nuclear power and for improving efficiency--will have changed, but if anything, these changes would probably tilt the balance today even more heavily against nuclear expenditure. (The price of uranium has gone up, for instance, while the tendency of technological change is to lower the cost of efficiency improvement.)

Reasons have been given above why a large-scale nuclear program is not possible and, when they consider the prohibitively large capital investment required, Keepin and Kats give their own reasons for this. But to do harm, the program does not have to be extensive; even a single nuclear station 'replacement' would have an effect measurable in tonnes--the tonnes of greenhouse gas that need never have been emitted, if the station had not taken funds away from improving energy efficiency. Rather than ineffective placebos like the opening of a few nuclear stations, we need a far-reaching overhaul of our current practices that squander energy while threatening to make the world seriously unliveable.

Alan Roberts is a current member of the (advisory) Nuclear Safety Committee of the Australian Radiation Protection and Nuclear Safety Agency.

This is an edited version of an article that was first published in Arena Journal No. 23, 2005. The original article contains all references. Alan Roberts will be speaking on this issue in September as part of the Public Advocacy Program at Victoria University.