Gareth Aubrey has posted a technical defence of the safety and effectiveness of nuclear power, adding "In any case, if I were to post an extended pro-nuclear polemic, the grief received would not be worth it."

Let me take up this masochistic baton and see where I end up. I have blogged on nuclear power before here criticising the weakness of the arguments used against nuclear by the Sustainable Development Commission, but not coming to a firm conclusion.

The latest round of arguments seems to have been spurred by an article Greens aid destruction of planet in the Times. Doubtless that sort of headline sells papers, but at the cost of polarising a debate where clear heads are needed.

The environmentalist arguments against nuclear are typically a) that it is dangerous and that the problem of waste has not been solved, and b) that it is less effective than alternative ways of reducing carbon emissions.

Gareth Aubrey tackles the safety argument head-on, arguing in a very worst case scenario, the failure of deep storage would cause a "chance of contracting cancer from radiation is increased by the same amount as that of a person who moves house from London to Truro". (Is that due to the Radon in Truro?) It would be nice to see the figures for this calculation. And it places a burden on opponents of deep storage to produce their own calculations showing a greater risk.

The effectiveness argument also rather depends on figures. It is often said that if all the money spent on nuclear had been spent on windmills, we would have fantastic windmills by now. And it is probably true. But where are the figures indicating likely returns on future expenditure? This argument is asking us whether we want renewables or nuclear. But that is a loaded question. If we are serious about global warming, the question is whether we want fossil fuels or nuclear, in addition to whatever renewables we can develop.

Different generation technologies have different cost and performance characteristics which mean that it is good to have a mix. Nuclear is good at baseload generation because not a lot is saved by switching it off. Coal is also good at baseload because it is slow to start up and shut down. Gas is good at responding quickly to peaks in demand.

Renewables will reduce the amount of fuel needed in other plant, but don't replace capacity because they will not always be operating. They will increase uncertainty in the supply-demand balance and therefore the demands on responsive gas plant. Incidentally, this uncertainty can be mitigated by linking up grids supplied by renewables over large areas - because the variations tend to cancel out, and because the wind is always blowing somewhere. But this the opposite of decentralised distribution that we hear about from Greenpeace. Smart metering can also reduce this uncertainty and is a good idea however much or little renewables we have.

In addition a mix of generation technologies reduces dependence on a single type of fuel (or weather). With Russian gas likely to be turned off any time Putin wants to throw his weight around, relying on gas for baseload is an unfortunate risk.

So while an energy mix can be maintained without nuclear, a better mix would include it. In particular nuclear's role as a provider of reliable baseload is more suited to gas and coal than to renewables, so it is gas and coal that it is in competition with, more than renewables.

So I don't think we can question the usefulness of nuclear power, in reducing carbon emissions, and playing a different role in electricity generation to that played by renewables. Are there other environmental issues to consider? Gareth again:

If you lived one mile from both a nuclear power plant and a coal-fired power plant, which would give you the highest radiation dose? Coal contains uranium, thorium, radon and carbon-14, all of which is ejected directly to the environment when it is burnt. The answer, counter-intuitive though it may be, is that the coal station gives you a higher dose.

This is an extremely compelling argument. Can we see the figures, Gareth? It is up to the environmental movement, or the coal lobby to admit this or to provide alternative calculations.

But if this is right, we are left with the conclusion that nuclear is environmentally better all round than coal. It is not as good as renewables, but it doesn't have to be that good to qualify on environmental grounds for membership of the energy mix.

So it boils down to the question of whether nuclear is a cost-effective way of reducing carbon emissions and improving energy security, not a question of whether it is wrong in principle. How does Greenpeace argue its corner now?

Greenpeace campaigns for 'distributed generation', their new name for Combined Heat and Power, and contrasts it with nuclear. CHP sells the waste hot water produced in generation as well as the electricity, giving an overall thermal efficiency around 80% rather than 35-40%. It is a good idea, but is limited by the infrastructure for distributing hot water, and by demand. It has its place, as part of the mix but is not an alternative to any other single type of generation. So what Greenpeace doesn't seem to be doing is tackling head-on the arguments for nuclear.

The Lib Dems as a pro-environment party has taken the environmentalist line on nuclear. Environmentalism is sincere and full of dedicated people, and it is right to defer to that movement in the first instance on environmental questions. But sincerity doesn't guarantee accuracy. And arguments that we don't like still have to be addressed on their merits, it is no good dismissing anything said by the nuclear industry as tainted. Windmills have a lobby too.

The question at the end of the day, is whether, in addition to everything else that we do, the cost of nuclear is worth paying for further reductions in carbon emissions. Global warming skeptics say no. Environmentalists say no. What an alliance! I am not nearly so complacent about global warming.