Is nuclear power obsolete?

After a summer hiccough, the new UK government has finally signed the deal with the French nuclear company EDF and its Chinese financial backers to build a new nuclear power station at Hinkley Point. My belief that this is a monumentally bad deal for the UK has not changed since I wrote about it three years ago, here: The UK’s nuclear new build: too expensive, too late.

The way the deal has been structured simultaneously maximises the cost to UK citizens while minimising the benefits that will accrue to UK industry. It’s the fallacy of the private finance initiative exposed by reductio ad absurdum; the government has signed up to a 35 year guarantee of excessively high prices for UK consumers, driven by the political desire to keep borrowing off the government’s balance sheet and maintain the fiction that nuclear power can be efficiently delivered by the private sector.

But there’s another argument against the Hinkley deal that I want to look at more critically – this is the idea that nuclear power is now obsolete, because with new technologies like wind, solar, electric cars and so on, we will, or soon will, be able to supply the 3.2 GW of low-carbon power that Hinkley promises at lower marginal cost. I think this marginal cost argument is profoundly wrong – given the need to make substantial progress decarbonising our energy system over the next thirty years, what’s important isn’t the marginal cost of the next GW of low-carbon power, it’s the total cost (and indeed feasibility) of replacing the 160 GW or so that represents our current fossil fuel based consumption (not to mention replacing the 9.5 GW existing nuclear capacity, fast approaching the end of its working lifetime).

To get a sense of the scale of the task, in 2015 the UK used about 2400 TWh of primary energy inputs. 83% of that was in the form of fossil fuels – roughly 800 TWh each of oil and gas, and a bit less than 300 TWh of coal. The 3.2 GW output of Hinkley would contribute 30 TWh pa at full capacity, while the combined output of all wind (onshore and offshore) and solar generation in 2015 was 48 TWh. So if we increased our solar and wind capacity by a bit more than half, we could replace Hinkley’s contribution; this is indeed probably doable, and given the stupidly expensive nature of the Hinkley deal, we might well indeed be able to do it more cheaply.

But that’s not all we need to do, not by a long way. If we are serious about decarbonising our energy supply (and we should be: for my reasons, please read this earlier post Climate change: what do we know for sure, and what is less certain?) we need to find, not 30 TWh a year, but more like 1500 TWh, of low carbon energy. It’s not one Hinkley Point we need, but 50 of them.

What can’t be stressed too often, in thinking about the UK’s energy supply, is that most of the energy we use (82% in 2015) is not in the form of electricity, but directly burnt oil and gas.

About a quarter of our total energy use – more than 500 TWh – is the gas we burn for heating and hot water, in homes and offices, and in industry. We should ultimately replace this with low carbon electricity, but this is going to be expensive. This is where energy efficiency measures give us the biggest short term gains. Currently overall we’re seeing a long term decrease in energy use of about 2% a year; coincidentally this is roughly on the scale of a Hinkley Point a year. The sort of energy efficiency measure we’re talking about here is unglamorous – it’s about retrofitting houses with better insulation, about tighter building regulations for new builds – but the investments pay for themselves quickly, and it’s difficult to understand why we don’t pursue these more urgently.

Another quarter of our energy consumption comes in the form of directly burning oil for road and air transport. In principle some of this fossil fuel consumption can be displaced by biofuels (which currently account for about 3% of road transport fuel), but I think this is problematic both in terms of a proper accounting of the embedded carbon in the process and for competition for land-use with food crops.

But electric vehicles potentially can help here, as they can displace directly burnt oil by electricity – so if (and only if) that electricity is produced from low carbon sources that represents progress. It’s great that use of electric vehicles increased by 44% in 2015 – but at this level, they are still displacing only about 0.1% of the oil we burn in our cars [1]. This needs to increase by another couple of orders of magnitude before it makes any material impact.

Which brings us back to electricity, which represents about one fifth of the energy we consume. Currently about half the electricity we use comes from gas and coal; it’s a sign of the progress we’ve made that about half of our electricity comes from low-carbon sources. But we should take a careful look at what is meant by low-carbon here. There are four categories here – dubious (industrial biomass and interconnectors), worth having but limited growth potential (hydro, landfill and sewage gas), true renewables with growth potential (wind, solar, and potentially waves and tides) and nuclear.

To begin with the dubious category, 8% is accounted for by burning biomass. A large part of this comes from burning wood-chips, imported from the USA, in converted coal fired power stations. I’d like to be convinced by a proper life-cycle analysis that this genuinely is zero-carbon and sustainable.

Another 6% comes from electricity imported from France and the Netherlands, assumed in the statistics to be low-carbon, on the grounds of the importance of nuclear power in France and wind in the Netherlands. Interconnectors are important for energy security and for better balancing supply and demand, but unless the exporting country has a true surplus of low carbon energy (which would probably be true for Iceland, for example, but is not for either the Netherlands or France) I don’t think this is a viable route for real lowering of the carbon intensity of our energy supply.

Hydroelectricity accounts for 2.5%, a useful contribution but unlikely to get much bigger given the limited supply of floodable valleys.

This leaves 20% of our electricity coming from nuclear, and 13.5% from wind and solar (wave and tide energy’s contribution is currently negligible). Let’s just pause and remind ourselves that most of the energy we use isn’t in the form of electricity, it’s directly burned oil and gas. Nuclear, solar and wind between them produced about 120 TWh of energy in 2015, 5% of the total energy we used, 2400 TWh.

How might this change in the near future? There should be substantial growth in both solar and offshore wind. For solar, the biggest remotely plausible scenario for installed capacity I have seen projects 60 GW of capacity by 2030. It should be stressed that this is possible, not necessarily likely; it would need strong political and regulatory support, which isn’t currently in evidence. Of course, 60 GW of solar capacity doesn’t produce 60 GW of power, averaged over the year, because of the awkward interruptions of night and winter; using the load factor for currently installed base – a little less than 10% – produces an estimate of 50 TWh per year generation. For offshore wind, the potential capacity could be in the region of 30 GW; load factors are higher for offshore wind than solar so this might produce another 100 TWh.

So an achievable, but ambitious, expansion of offshore wind and solar could produce 150 TWh of electricity per year (up from 14 TWh currently) – that’s 6% of our current energy consumption. In fact, it’s better than that – if this is replacing electricity generated by fossil fuels , then the fossil energy consumption it displaces includes not just the final electricity generated, but also the fossil fuel energy lost in the conversion process, which is currently around 40% efficient overall (due largely to the 2nd law of thermodynamics). On this calculation, this much solar and wind displaces 375 TWh of burnt fossil fuel, some 16% of current energy consumption.

What might we expect to happen to nuclear energy by 2030? If we do nothing, all but one of the existing fleet of nuclear power stations will have shut down, reducing nuclear capacity from 8.9 GW currently to 1.2 GW. The output we will lose as a result of this is about 60 TWh – a bit less than a half of the low carbon energy that we would gain from the most ambitious expansion of wind and solar.

Currently planned nuclear new build includes the notorious 3.2 GW of capacity at Hinkley Point which has just got the go ahead, 2.7 GW at Wylfa in Anglesey from the Horizon consortium (led by Hitachi), 3.8 GW at Moorside, Cumbria from the NuGen consortium (led by Toshiba’s nuclear subsidiary Westinghouse). Beyond these relatively firm plans, Horizon plans a second plant at Oldbury, China has expressed interest in building a reactor at Bradwell, Suffolk, and there is the prospect of a consortium to build small modular reactors, with a possible location for an installation at Trawsfynydd in N.Wales.

If the three lead projects did get built by 2030 (and that seems a big if at the moment), this would produce about 90 TWh of low carbon energy. It’s important to realise that this represents only a 20 TWh increase over the current nuclear output – the next decade’s nuclear new build program is largely replacing existing low carbon capacity rather than significantly expanding it.

Energy efficiency is perhaps less glamorous than producing new plant, whether wind, solar or nuclear, but is just as, or even more, important. A continuation of a trend of a 2% p.a. reduction in energy use through efficiency would take out more than 500 TWh of demand by 2030. But measures still need to be taken to capture these gains, whether that’s up-front financial investment, government regulatory actions (such as banning incandescent light bulbs), or government incentives. Not all of these are politically popular in the current environment.

So what’s the best we can hope to do? The current situation, in rough numbers, is that we use 2400 TWh, 5% of which come from wind, solar and nuclear. Let’s say we take out 500 TWh of demand through improved efficiency, that takes us to 1900 TWh. Achieving very high targets for wind and solar, and completing the first wave nuclear new build program, would generate 240 TWh of low carbon energy, which would displace 600 TWh of fossil fuel demand (after taking into account conversion losses in fossil fueled power stations). So we’d end up still using 1300 TWh of fossil fuel based energy – we wouldn’t yet be half-way to decarbonising the energy economy.

Without the initial nuclear new build, we’d lose 60 TWh of nuclear low carbon energy, which would partially offset the 150 TWh of solar and wind; the net 90 TWh increase in low carbon energy would displace 225 TWh of fossil fuel input, leaving us still having to find more than 1600 TWh of fossil fuel energy. (Both this calculation and the last assume that any problems due to the intermittency of solar and wind are solved, either by improvements in energy storage technology or by maintaining sufficient gas generation capacity to cover any gaps.)

The first scenario in particular does depend also on the assumption that there’s an expansion of electricity generation to respond to a much greater uptake of electric vehicles as we currently only use 500 TWh of fossil fuel inputs to generate electricity. Currently, electric vehicles only account for the consumption of 0.1 TWh of electricity, so as I’ve already said this needs to increase by a couple of orders of magnitude. That’s not going to happen without substantial falls in the cost of batteries; the prospects for that needs another discussion.

So, with or without nuclear new build, it’s difficult to see how, by 2030, we can get even halfway to the point of decarbonising our energy economy. What happens after then?

I suspect the 60 GW solar/30 GW offshore wind scenario this is predicated on already brings us to the point of diminishing returns. More offshore wind capacity would take us into deeper waters and increasing expenses, while solar installation on this scale would have used up most rooftops and would be starting seriously to compete for land against agricultural and amenity use.

Much hope has been pinned on carbon capture and storage to fill the gap. Again, this needs a longer discussion; the UK has made very little progress in implementing it at scale. The problem with carbon capture and storage is that its cost is a pure overhead – burning fossil fuels without CCS will always be cheaper than with CCS. For other technologies – solar, wind and nuclear – it is possible that new technology and learning by doing will bring the unsubsidised cost below that of fossil fuels, and we should be working very hard to make that happen. This will never be the case for carbon capture and storage.

So, my conclusion at the end of this very long post is that, if we really are serious about putting our energy economy on a sustainable low carbon footing on a timescale of decades – and we should be – we need to accelerate the implementation of solar and offshore wind. I haven’t talked much about cost and efficiency, or about energy storage, but these are all challenges that need big technological improvements to make effective and affordable. But there’s still a gap – a big one – and I don’t seen how this can be filled without nuclear energy. This makes it all the more a pity that the UK’s new nuclear build programme has turned into such a shambles.

Energy numbers are from the 2016 edition of the UK Government’s Digest of UK Energy Statistics. They have been rounded and simplified for clarity. For the 60 GW solar scenario, see for example Solar powered growth in the UK, Cebr/Solar Trade Organisation, 2014.

[1] Energy used in electric cars: 0.1 TWh. Total energy used in road transport: 40.5 MTOE = 470 TWh, so electric cars represent 0.02% of energy currently used in road transport. Guessing that the direct energy conversion efficiency of electric motors is 5 times greater than internal combustion engines, we can estimate that our current electric car fleet has displaced about 0.1% of the petrol/diesel we use in internal combustion engines.