Friday, May 8, 2009

Is David MacKay "trying to make wind sound useless"? Let's look at more data

I'm delighted to see that the response to Sustainable Energy - without the hot air so far has been remarkably positive. There's just one or two folks who have become convinced that I am anti-wind, that I am deceitfully making wind sound worse than it really is; and they have been running round leaving comments on blogs (for example, you can find one lurking in the comments on this excellent article about the financial cost of wind power (the oil drum), who asserts "MacKay has made a serious error in his calculations of on-shore wind energy resources. ... Some of the wind farms initially built were in poorer locations but close to electric transmission lines, so his calculations are not good examples of what is possible in UK.")
I've written three blog posts about this topic already, encouraging people to provide real data rather than just spreading poisonous rumours. I've now worked through the ordnance survey maps and ROC register entries for about 15 windfarms around the UK, and included the data and maps in a presentation I made at a wind energy conference in St Andrews this week. I am still working on this; what I have focussed on so far is mainly the newest windfarms for which data is available, with the largest numbers of turbines, with biggest diameters, and mainly on scottish hilltops or welsh hilltops or near to the coast. The new data starts at slide 30 and is summarised on slide 41. These onshore wind farms have powers per unit area between 2 and 4.6 watts per square metre. To indicate the rough scale of windfarms required to deliver large amounts of power, I assumed in the book a power per unit area of 2 watts per square metre. So yes, there are windfarms that have powers bigger than 2 watts per square metre. Was I deliberately "making wind power seem worse than it is"? No. I chose 2 watts per sq metre as an estimate of what we could get if we put up lots of wind farms (with the area of Wales), which is obviously going to be less than the power per unit area of the very best spots. Yes, I willingly agree that if we want wind to make only a small contribution (for example, less than 1 kWh per day per person), then it would be appropriate to assume a higher power per unit area - perhaps 3 or 3.5 W/m2 instead of 2 W/m2, if we keep building in the best spots.

As evidence that I am not deliberately biased against wind, take a look at the data for offshore wind farms.

In my book I assumed a power per unit area of roughly 3 watts per square metre for offshore wind. But the two offshore windfarms in my data have powers per unit area below 2.5 watts per square metre.
There are several other scientists who have used a power per unit area similar to mine when estimating wind resources. For example, Socolow from Princeton uses 2 watts per square metre when discussing his "wedges". On page 234 of my book I cite a study by Elliott et al. (1991) in which windfarms in the best locations in America, covering an area equal to that of California, were estimated to have an average power density of 1.2 W/m2.
While my book is technology-neutral, the truth is that personally I am pro-wind! I think wind farms are brilliant, and I'd be very happy be within eye-shot of one almost anywhere in the ordinary countryside.
Please could the commentors call off the dogs?
Thanks! David


Andrew Smith-Gibbs said...

I welcome your statement that you believe that we should have more onshore wind, though I note that you haven't said how much, here. Have you shifted from the position you express in SEWTHA about how much onshore wind is possible, or reasonable?

David MacKay FRS said...

I haven't "shifted position" because in SEWTHA I didn't really take a position, apart from this: "For me, the bottom line is that the plan must add up; I'm happy with any plan that adds up". I'm reluctant to specify any particular amount of wind, since the really important message of the book is not "DJCM advocates x wind + y tidal +..." but "we the people have to choose where to get our energy from". If the British people asked me to decide for them on a zero-carbon plan, I would be inclined to say "let's put up enough onshore wind to roughly cover all electric cars and vans (with the aim being that all cars and vans would be electric as soon as is reasonable) and to cover roughly all building-heating using air-source or ground-source heat pumps (with the aim being that most buildings would switch to good quality heat pumps as soon as is reasonable). Wind fluctuations could then be largely absorbed by smart demand management of these two big loads." This would correspond to 20-30 kWh per day per person of electricity on average from wind, and would obviously involve a radical change in British attitude to wind farms. The onshore area would be 10-15% of the country (or perhaps a bit less, depending how inaccurate my 2W/sq m estimate turns out to be!). Stronger winds in winter would align reasonably nicely with bigger heating demand in winter. If we can get tidal stream power going, I'd get lots from that too. A sort of honorary wind-power. The other 30 kWh per day per person would come from other sources, some of them firm (eg waste to energy), as in the "five plans" chapter. I would recommend this vast amount of wind (which would currently be viewed as potty by many) because it is clear that it would work, and it would work for ever, and the cost would be reasonable (i.e., a bit more expensive than today's 'free' fossil fuel power, but not terribly much more; and wind power's price would probably drop the more of it we got). Having looked at the data, I think I would recommend onshore rather than offshore (as Dale Vince from Ecotricity has argued) because it is so much cheaper to put up and maintain onshore wind farms. The key thing is to build up electric vehicles and electric heat pumps demand at the same rate as wind supply. Roughly 1000 electric vehicles per (2MW) wind turbine.
I think people living in warm homes and driving electric cars could rapidly come to love the sight of powerful wind turbines in the countryside, powering their lifestyle.
But I'm very easy-going. If the people want another plan that adds up, with less wind, that would be fine with me too.

Andrew Smith-Gibbs said...

On variability, which you touch upon, please do bear in mind that managing variability is a grid-wide issue: it's not specific to wind. It's all about balancing total demand and total supply - there won't be a parallel grid for wind+EVs, they'll all participate in the national grid, and, within a few years, the unified trans-EU grid. The balancing happens at system-level, not at the level of individual technologies.

Given our current fossil-fuel fleet, we could in the next 5 years, if we chose (though I don't suppose that either this or the next government will choose it), extend our onshore wind to 40GW total nameplate capacity, giving us - what - about 4-6kWh/d of cheap zero-carbon energy, requiring only reinforcements to the transmission network to make it happen, and nicely filling the forthcoming UK energy deficit, as well as contributing to our carbon budgets and reducing our dependence on Russian gas. That, happily, would give us 5 years headstart on working towards creating a more dynamic, manageable demand, and putting in place some serious energy efficiency measures.

Looking beyond that, then yes, making more demand time-shiftable, enables us to increase wind capacity. Roll-out of both heat pumps and electric vehicles is likely to be slow, owing to supply-chain issues, so they're probably the place to look for large-scale developments in the 2020s; but there are plenty of other ways to make demand time-shiftable in the next few years: the obvious one being immersion heaters in hot-water tanks, coupled with smart meters.

I like the idea of tidal stream power as "honorary wind power" but to be honest, I'm not really sure what that you mean by it. Sounds nice, though.

I'm sure we'll be supplementing all that onshore wind with at least as much again from offshore wind: it's not either-or, and our offshore wind resource is immense, of course, and carries with it benefits off geographic diversity of supply, and large interconnectors to continental European grids, with big consequential advantages for load balancing.

It's hard to see waste-to-energy as sustainable - though it may be marginally better than letting it rot to methane in landfill. Is it any more than a useful stop-gap? Oil->plastics->incinerator is as sure a way to increase CO2 emissions as Oil->petrol->car.

Unknown said...

Here's my own little contribution to commercial and dynamic load "shiftability" in the UK...

I know have this (or rather the automatable flags alongside it) control one of my co-hosted Web servers in London, and my dishwasher coming on at night; both to roar into action when kgCO2/kWh intensity drops below mean and median for the last 24 hours, which in future may of course be as the wind picks up rather than just as evening demand drops away.



Andrew Smith-Gibbs said...

Interesting, Damon. But why zero carbon for nuclear? Mining and processing uranium releases a lot of carbon, and those quantities will grow as we run short on high-grade uranium ore. Some nuclear processes also release chlorine and fluorine compounds, including CFCs which have unit greenhouse forcings 4 orders of magnitude higher than CO2. So there is a contribution to anthropogenic global warming from marginal nuclear production, and nuclear has no place in a decarbonised grid.

David MacKay FRS said...

@Andrew... you said 'nuclear has no place in a decarbonized grid'. This is going a bit off-topic, but may I suggest talking in numbers, rather than dogma? I do love numbers, and good old Lightbucket has supplied numbers for the carbon intensities of several forms of power generation, from nine independent life cycle analyses. These numbers can be found in the book's Wiki.
Nuclear and wind actually come out pretty similar.
[Oh, and if you didn't understand what I meant by tidal stream power, please read the book.]

David MacKay FRS said...

I have updated my data on windfarms.
I've gone through roughly 75 farms, finding the power per unit area. You can see the results here:

Andrew Smith-Gibbs said...

I specifically wondered why you consider tidal stream power to be "an honorary wind-power", not "what tidal stream power is". "honorary wind power" sounds sweet, but I don't know what you mean by it - it seemed an unusual turn of phrase. Just curiosity, no significance.

On GHG emissions from nuclear, the figures from the ISA, University of Sydney, in Life-Cycle Energy Balance and Greenhouse Gas Emissions of Nuclear Energy in Australia, are: around 60g CO2-e/kWh(e) for light water reactors [range 10-130g], and around 65 g CO2-e/kWh(e) [range 10-120g] for heavy water reactors. The figures from Smith & Storm van Leeuwen are in the range 84-122g CO2-e/kWh(e)

Now, how do we decarbonise wind? Well, moving the British, Danish and German grids to 100% renewables (achievable within 20 years), would do it, if we use for steel and turbines manufactured in those countries - and as that is happening now, the GHG output from wind is decreasing. According to the Vestas LCA for their 3MW turbine, using figures from 2006, 4.6g CO2/kWh(e) for onshore, and 5.2g CO2/kWh(e) for shallow offshore.

How do we decarbonise nuclear? You need mining and processing operations in Australia, Russia and Niger to decarbonise. And that's much harder, for technical and political reasons. As the availability of high-grade uranium ores decrease, then energy inputs increase, and the GHG output from nuclear increases. (Secure Energy - Civil Nuclear Power, Security and global warming - Oxford Research Group): even without a dash to nuclear, it becomes as GHG intensive as CCGT by 2050. If the IEA forecasts come true and the world does dash to GenIII and GenIII+ nuclear technology, then we'll reach that point quite soon - possibly before any new British nukes have become net exergy producers (given that both construction time and energy payback time are 5-10 years each).

Alex Terrell said...

When I started looking at wind power about 6 years ago, I deduced from the rules of thum for turbine separation, that a capacity of 10MW/km2 is possible. This figure was broadly confirmed by industry insiders but relates to smallish wind farms, not the 8000km2 Dogger Bank wind farm. 10MW/km2 would be possible, but you'd start to get efficiency losses, so, given space is not yet a problem, you might go for 5MW/km2 of capacity, which should come to 2MW/km2 average power. Going higher is possible, but may not be economical.

I was revisiting the piece in your book on storage to cater for lulls in wind output.

I concluded that if we are to allow a little carbon output, then these lulls are not an issue for Liquified Natural Gas.

The gas storage tanks at Milford Haven each hold 160,000m3 of LNG. At 50% efficiency, 1 tank holds 22 GW days of power.

And gas plants are cheap. A 2GW plant is about to be built for €1 billion in Pembrokeshire.

The figures you give above imply 60GW average. So a 5 day lull, or 300GW days, would need 15 storage tanks, and €30 billion of CCGT plant. That's cheap compared to the cost of 150 GW of wind capacity.

Kelly Cowan said...

Although wind farms range over a large area, their actual footprint is only about 1-2% of the land they sit on (i.e. for towers, roads, and transmission). The land continues to serve other functions, most commonly agricultural. So, it seems misleading to talk of "covering" large areas with wind parks. Turbines are often sited in some of the least populated areas and allowances can be made for scenic areas and potential impacts on wildlife. According to this, wind power's overall footprint may be the lowest on several measures among all leading energy sources.

Gary Moran said...

Can you elaborate on the concept of using electric vehicles to mitigate against a lull in wind farm capacity? 30 million vehicles may provide that capacity, but surely they are required at that time for transport not for powering the national grid? Are people expected not to use their vehicles at this time to help the grid out?

I have to say this seems far too expensive, elaborate and fragile to be practical. You are relying on: a smart network; vehicles designed specifically for this dual role (who pays the extra expense, or is it mandated, and therefore an extra tax?); behavior of drivers to not use vehicles when they are required for power (perhaps the vehicles would lock down so owners couldn't use them?); plus a support infrastructure of battery storage and maintenance (again designed as much around delivering backup power to the grid as with transportation). It is likely to be a big enough challenge to make practical electric vehicles just for transportation, let alone adding the requirement that they back-up the national grid as well!

Your work highlighting the fact that our first step is to get the basic numbers to add up has been essential, and many thanks for that (loved the book). However, a lot more needs to be done to put working solutions in place. Solutions must be reliable, robust and economic. If politicians want to get the public on-side they need to demonstrate that solutions meet all these criteria - and no one is doing that.

Andrew Smith-Gibbs said...

a few quick bullet points

* balancing supply and demand is a bit too complicated for back-of-the-envelope calculations. If you want to learn more about variability, I recommend this book

* All forms of generation need backup - it's not just about wind's intermittency. Backup is provided at the system-wide level, so it wouldn't just be electric vehicles [EVs] backing up wind; it would be EVs adding resilience to the whole grid.

* Most cars are unused most of the time. So they would be available to the grid, as long as there's an incentive to keep them plugged in

* There are some very good papers out there - do a google scholar search for the abbreviation V2G (vehicle to grid), which is used as shorthand for using electric vehicles as backup for the grid.

* The financial motivation for V2G is that one can earn a lot of money by being able to provide power at times when supply is struggling to meet demand - wholesale electricity prices just increase until they balance. So, for owners for EVs, owners of backup-generators, and biomass plant-builders, there are incentives to make sure that they can serve power at those times.


Gary Moran said...

* balancing supply and demand is a bit too complicated for back-of-the-envelope calculations. If you want to learn more about variability, I recommend this book

Quoted from this: “Although difficulties and constraints are highlighted, it is taken for granted that renewable energy forms an important component in future energy supplies for the electricity supply industry, the more so in the UK with increasing dependence upon imported gas and the future retirement of coal and nuclear stations. Problems raised, therefore, should be seen as problems to be solved – in some cases by more research, and in others by the development of technology.”

If this is how the whole book proceeds, then: It is based on an assumption that we must use substantial renewable, an assumption many do not share; solutions that require future technology, or societal behavior changes, or have many dependencies, are unlikely to prove practical or economic.

* All forms of generation need backup - it's not just about wind's intermittency.

This is somewhat disingenuous. Yes we need to balance base load and peak continuously, regardless of whether there are significant non-dispatchable plants on the grid. As an example the pumped storage plant at Dinorwig was built (as I understand it) to balance baseload and peak power needs around Britain’s then upcoming nuclear plants. However the scale which David MacKay is talking about is of a completely different magnitude!!! You imply that its really not much different to what we do at the moment, well it is!

* There are some very good papers out there - do a google scholar search for the abbreviation V2G (vehicle to grid), which is used as shorthand for using electric vehicles as backup for the grid.

Not a paper, but one of the first pages I found googling on V2G, pretty well sums up quite a few of my objections:

I’m highly skeptical of wind plants, but I’m trying to understand whether they really will ever provide more than supplemental power in any realistic time frame. I enjoyed David’s book, but didn’t take the chapter on mitigating wind’s lulls seriously - I took it merely as a numbers exercise (which is what the whole book is, of course). In other words if magic was real and we could just wave a wand and solutions would appear ready made, then yes perhaps it will work – but as a real world project its too fragile. I work with IT projects, and we see lots of these sort: dependent on immature technology and lots of dependencies – which are obviously never going to work in the real world, we call them Hogwarts projects. So I was surprised to see David say that if it was up to him he’d build 30GW of wind farms and back it up with EVs and heat pumps., because to me it is an obvious Hogwarts project.

Like David I don’t care where our power comes from as long as the numbers add up – but this applies equally to the money, and the timescales.

Andrew Smith-Gibbs said...

Sorry you feel its disingenuous. Nevertheless, backup is planned at the level of the grid as the whole, not at the level of an individual technology.

No plant has 100% availability. One key difference is that renewables availability is forecastable. Tidal power, you can predict most of the variations decades in advance. Wind, you can predict 3 days in advance with reasonably reliability. That's not true of all technologies.

If you have nuclear, you have to plan for there being years where you'll have less than 50% availability on average across all your nukes, over the whole year; and during the year there will often be little advance warning of unavailability. And if you have no viable native uranium, you also have to manage the risks of having imports cut off, just as we are vulnerable to having our gas or oil imports cut off.

When we have tidal barrages, we'll have to schedule around 12-hour and 28-day cycles. When we have lots of wind, we'll have to plan around wind forecasts of 1 hour, 12 hour, 3 day, 1 month ahead. If we have a lot of wave and tidal stream, we'll also have near-term forecasts to manage around. Biomass and geothermal plants will be load-following.

There's been a lot of systems modelling done to see if Britain and Europe can power themselves from 100% renewables, and manage the question of variability. And the answer is that they can. See work by Mark Barrett and Gregor Czisch, for example.

On electric vehicles: there are different V2G models - the swappable battery is one option; the alternative is sufficient connection points that a car can be on the grid at most parking places.

V2G is only one component of managing variability. We shall use a panoply of measures. We shall become part of a grid that spanned a much larger geographic area, which smooths variability considerably. We shall build interconnectors to those parts of UCTE and NORDEL where there are tens of gigawatts of hydro, and sufficient hydro storage for a month's supply of electricity to the EU15+Norway. That's a hell of a lot of backup. Together with pumped hydro in Britain, thermal storage, the smart grid, backup biomass plants designed for a 10% load factor, and so on. Maybe even vanadium flow batteries. But the point is, that technological innovation isn't necessary - we have all the tools we require now. Innovation is useful, because it may well offer us lower-cost and/or faster pathways to decarbonise and reach sustainability, so there's plenty of scope for research too.

SEWTHA suggests a hell of a lot more than 30GW of wind. There's about 50GW of onshore wind in there, which would be around 200-300GW of peak capacity. And really, there is no sane alternative for most of our energy supplies. Wind is the only thing that can be rolled at the scale of tens of gigawatts quickly. Onshore wind is also very cheap. And, if we committed to it properly, turbine factories would be built in Britain, so that buying tens of gigawatts of wind turbines would be inward investment.

Given that our growing awareness of dangerous climate change means that near-term carbon targets must be tightened considerably, we probably need to decarbonise the grid by 2030. And wind is going to have to be a very large part of that.

Gary Moran said...

Andrew Smith-Gibbs
Our differences are ideological? I suspect you desire that all of our energy production is provided by renewable, as this aligns with your ideology of sustainability. I suspect you regard nuclear as unsustainable? If this is the case then there is probably little point in trading comments about the viability of renewables, you will always minimize the problems with renewable, maximize the problems with nuclear and over state the danger of climate change.
A few points:
SEWTHA does not agree that Britain can power itself on its own renewables.
Because nuclear is available from many sources and because the volume of fuel is small there is little fear of being held hostage to producers.
There is little point discussing the merits of V2G because we have practically no EVs out there – when we start to buy them in bulk, then I’m sure some city somewhere will trial V2G; before that happens it really is pie in the sky.
The problems around the intermittency of wind are well known, there are real problems there that for the moment we are struggling to solve (Denmark does best at this – but there are still issues). This report by Poyry illustrates where we are likely to be if we continue to push and subsidize wind.
When we consider our response to scientific evidence of anthropogenic climate change, it should be rational, and have due regard and understanding of the nature of the evidence. See this article here by Mike Hulme a climate scientist who believes in AGW . You may believe we need to decarbonise the grid by 2030, but this is an ideological or political view not a scientific one.

I really don’t care where we get our electricity from. I’d like some security of supply, but I’m not paranoid. I’d like to decarbonise where its practical but I won’t pay a premium for the privilege. And I don’t want the lights to start going out. I suspect most of the British public have a similar outlook, and our views are rational and should be considered.
My fear is that the push for wind is an ideological one, and that we will pay the price down the line when we have insufficient dependable generating plant and very expensive electricity.

Gary Moran said...

Very quick, stumbled across this paper today:

I think this is brilliant. Its intent is to reduce CO2 by 80% by 2100, but it fully recognizes the enormity of the challenge. It is above all pragmatic and free of wishful thinking. Highly recommended.

It also name-checks David MacKay.

Andrew Smith-Gibbs said...

Desire? Nah. Just interested in getting clean, sustainable energy in the best way for Britain & the world.

You're right, SEWTHA does claim that Britain can't power itself on renewables. The book is wrong on that. On the book's own numbers, you can see that Britain can power itself on renewables, even without any demand reduction: current demand (98kWh/d) < renewable potential (180 kWh/d). Furthermore, that 180 kWh/d excludes most of Britain's offshore wind resource (actual resource exceeds 130kWh/d), and most of the wave resource too (actual resource could be 700TWh/y, which is 32kWh/d).

Climate change isn't a matter of "belief", it's science. Do you consider it to be belief? If so, do you believe?

I haven't had chance to read the full Poyry report, only the summary. Do you have a link to the full version, please?

I do consider 3G/+ fission reactors to be unsustainable, given what's happened in the nuclear industry in the last 50 years. I do think it's conceivable that in the future, we may get next-generation fast breeders or fusion working properly, in a safe, sustainable way, and that research should continue on both. And I think it's conceivable that we'll find systems of management and ownership that can actually deliver a proper safety regime. I also recognise that the last 50 years show that we haven't yet worked those systems out, and that people still do things like ignoring automated safety warning systems, forging safety certificates, and triggering large leaks of toxic materials. I think it's quite a good idea to fix those systemic problems before we build any more nukes.

In the mean time, we need to decarbonise quickly. So I look around at what can deliver low-carbon energy fast. Energy efficiency is a big one - fast, cheap, effective. Lots more wind, similarly - these two give us the best chance of meeting our interim carbon budgets at lowest cost. In the medium term, the fastest low-carbon energy we can deploy at scale is wind, tidal barrage, biomass, solar thermal. To 2030, we can get additional significant contributions from other technologies, with wave, tidal stream, geothermal and nuclear fission becoming available in quantity. However, we can only develop the first three natively, in time. Developing a British nuclear industry would take a generation or more, as it did last time, so we'll be dependent on importing both technology and skilled, experienced labour. Whereas we could build our own wind, wave, tide, geothermal, biomass industries within a few years, if we chose.

National Grid are quite happy about managing the variability issues, at least to a 35% penetration of renewables by 2020.
There is a lot of very good material on addressing variability out there. There's the UKERC study of intermittency; Godfrey Boyle's book (as previously mentioned); and Poyry's report is just one of three on the subject this year.

And finally, the 2030 target for decarbonising the grid comes from the first report of the Committee on Climate Change.

Gary Moran said...

Andrew Smith-Gibbs

Your objections to nuclear appear ideological: you are highlighting what are, no doubt, real instances, yet it is likely that these are exaggerated and out of context; and that the incidents themselves are comparable to other industries. I think we need to understand that Chernobyl was a different case, and excepting that, the nuclear industry has a safety record comparable to any large scale industrial.

Climate change is science – a belief that climate change will be dangerous or catastrophic is not, please read the link I posted by Mike Hulme.

The full Poyry report should be there under the link study, its 30 pages.

Your point about the national grid is an interesting one. I wouldn’t know whether they are “happy” (the mix of plant has essentially been mandated), but this does appear to be regarded as manageable. They now refer to BERR Publication URN 08/1021 June 2008 available from here: . I think we can pretty well regard this as the definitive work on the practicalities of implementing renewable in the UK for the coming decade. There are some interesting points in here:

You can get renewables (predominantly wind, onshore and offshore) up to 41% of nameplate capacity and generation (44GW and 152 TWh) before diminishing returns really start to kick in, based on using conventional plant to mitigate intermittency. This will give you a 48% reduction in CO2 emissions (from 46MtC to 24MtC). However this will increase costs by an eye watering 31% (if we go for 35% renewable it goes down to a 25% increase). This comes from additional renewable investment and network costs of some £63 billion. Lifetime levelised cost (3/MWh) is higher for all renewables (exluding ROC subsidies), but onshore wind is only slightly more expensive at 60.4 than CCGT at 59.8 (this is on the basis that CCGT is now providing standby for 41% renewable, in a conventional scenario it would be 65.7 to 56.5). Nuclear is considerably cheaper at 37.9. Interestingly another issue of wind power mentioned frequently here, but which I’ve not heard before is curtailment, where for reasons of practical supply and demand management they simply can’t take what the wind can potentially deliver.

My conclusions from this are:
Real reductions in CO2, and intermittency at these levels are manageable (real good news, I wasn’t convinced of this before reading).
Beyond 40% (and this would be less for onshore wind only), we are running into diminishing returns trying to add non-dispatchable renewables to the grid using off the shelf technology. At that point we would need to utilize nuclear to decarbonize further, and additional nuclear capacity would need to be adjustable like French PWR plant (due to wind curtailment issues).
This is a mandated solution, and may not be the most economical available to us! The issues and solutions, IMO, have not been openly debated. The public may have been railroaded into a solution that is more expensive but is perceived to be greener (wind vs nukes). I’d be interested to see an alternative 2020 solution based on nukes, but concede we are probably a bit late for this now.
The level of investment required in renewable is very high, presumably this has the real possibility of being significantly underestimated, resulting in even greater costs to the tax payer. (this is an inherent risk in all large capital projects, and so should be considered)
There is little reduction in conventional plant (we gain 38 GW on renewables, but reduce conventional only by 15 GW). On the plus side if some of the renewable projects go awry we won’t be so exposed.
From this we can see that the Committee on Climate Change 2030 target to completely decarbonize the grid is nothing more than “pie in the sky” wishful thinking.

Gary Moran said...

In my previous post 3/MWh should be £/MWh

Andrew Smith-Gibbs said...

No, your Poyry link really does just give the 30-page summary report, not the full study, I'm afraid.

Nuclear doesn't help much at all with variability: adding nuclear onto a grid with 30%+ renewables makes things worse, not better. Even the major nuclear operators acknowledge that, which is why they objected to the UK's renewable targets. What's needed are more interconnectors (which allows export of surplus power rather than curtailment), and dispatchable plant. Conventional fossil-fuelled plant is one part of it, but by no means the only part: some renewables are dispatchable: geothermal, biomass, energy from waste, some hydro. Tidal barrages offer some dispatchability: you can choose whether to over-pump, and choose when to start generation on ebb, as long as you build for one-way generation rather than two. Which you would do, because you get more and cheaper energy that way (despite what SEWTHA says).

And whatever we do, will cost hundreds of billions of pounds - that's one of many areas where our host gets it right: we need to build tens of gigawatts of carbon-free plant.

On variability, there is no "definitive" answer - just a process of learning for the entire industry. A wider reading will reveal many of the possibilities. We are indeed probably 10 years from needing to tackle the issue of what happens beyond 40%, but there are plenty of developments in the pipeline for those 10 years that will help. Such as introducing the smart grid - an internet for energy - which brings whole new areas of possibility that we've only just begun to explore. But getting us to that 40% does give us a no-regrets decarbonisation pathway, as it reduces our fuel imports, and renewable generators generally have a net positive end-of-life value - the materials can be reused and recycled.

The CCC target of decarbonisation of the grid by 2030 is perfectly deliverable, technically. There really are no technical barriers: the problems are political. That's not to belittle those problems in any way: political hurdles can be far more serious than technical ones.

Gary Moran said...

Just read the "committee on climate change" paper section on decarbonizing electricity supply. I didn't see the 2030 date as a target, more a statement about what might be possible if certain assumptions prove to be true, I took these to be: ramping up of nuclear post 2017, and applying CCS to existing fossil fuel plant. In its context in the report I had no problems with the statement.

Your interpretation of completely decarbonised by 2030, particularly your statement that there are no technical hurdles, and that it is perfectly deliverable, however is plain wrong. If a proven scalable technology does not currently exist to remedy a problem, in the absence of supernatural abilities to predict the future, then as far as we can know the problem is not solvable in any given time frame. That is not to say it will not ever be solvable.

They plainly disagree with your intepretation that past 30% renewable penetration cannot be assisted by nuclear, as does David Mackay who often talks of 30% renewable and 30% nuclear. French PWR reactors are adjustable, and the BERR report (the basis on which the national grid is planning its implementation) states that this technology could be considered and would help prevent curtailment. It has the advantage of being off the shelf technology.

Andrew Smith-Gibbs said...

The reactors proposed for Britain can be adjusted twice a day, or 200 times in a year, at most. That's really not much of a contribution to load-following. Wind has zero short-run marginal cost, so should be the first to get used; however, always-on nukes are going to conflict with that.

On 2030: from the CCC report:
p xvii "Our modelling suggests that the least-cost path is likely to entail ... the radical decarbonisation of power
generation by 2030,"

p xxiv "It is important that this opportunity is grasped given the almost full decarbonisation of the power sector required by 2030"

p78 "decarbonisation of electricity generation, which needs to be close to complete by 2030, with g/kWh below 70 by 2030 in the 80% scenario and below 40 in the 90% scenario"

Nuclear is around 65g/kWh (see report from Bilek et al referenced elsewhere on this site); coal+CCS is over 160g/kWh, so neither can play much of role in the 90% scenario, by 2030. And CCS can't have much of a role in the 80% scenario - the CCC report assumed much greater potential for CO2 reductions from CCS than the technology looks likely to deliver.

I agree with you that there is no single scalable technology solution. There is, however, a portfolio of scalable technologies that do provide the solution. Gregor Czisch has demonstrated just such scenarios, using an HVDC inter-continental grid, lots of wind, some solar, and pumped hydro storage. So, it's already been established that it's technically feasible. He's modelled the variability and the economics, and the long-run average cost of electricity is no higher than it is today, with zero carbon. If you've found errors in Czisch's work, then please do let us know what they are.

Soylent said...

"How do we decarbonise nuclear? You need mining and processing operations in Australia, Russia and Niger to decarbonise."

That's not were significant carbon emissions occur; you've been drinking the Storm-smith kool-aid. Using the formula he extracted from where the sun does not shine he predicted the Rössing mine in Namibia would consume more energy than the entire country of Namibia. He's off by about 2 orders of magnitude.

The reason Storm van Leeuwen would perpetrate such an obvious smear job is that he's a member of the Club of Rome. This is the same gaggle of genocidal morons who've called mankind a cancer on the Earth, want to deindustrialize the world and reduce the population to 500 million to 1 billion, condemns the use of DDT because it saves lives and adds to "overpopulation". The Club of Rome doesn't actually have much political power to do anything; so they have to resort to advocating energy policies that lead to energy strangulation, organic agriculture, banning of DDT against malaria mosquitos and the like.

The reason the emissions vary so widely for nuclear is because of enrichment. Gaseous diffusion uses ~50 times more energy than a gas centrifuge. If you then assume that this gaseous diffusion is powered by coal power you get a fairly large CO2 intensity.

If you use gas centrifuges or you power the remaining cold-war era gaseous diffusion plants only with nuclear energy, you get to the lower end of that range.

Typical mining machinery consist of gargantuan mining trucks(diesel), draglines(electric or diesel), pumps and slurry pipelines(usually electric), crushers(usually electric), conveyors(usually electric), dynamite(most of the energy comes from natural gas or coal) and various chemicals and heat involved in processing ore. Most of the energy consumption is already electric or could be electrified; the only thing that's reasonably hard to electrify is the mining trucks(huge, glorified dump trucks that move material out of an open pit mine).

Andrew Smith-Gibbs said...

Hi Soylent - my reference was M. Bilek, C. Hardy, M. Lenzen, and C. Dey, "Life-Cycle Energy Balance and Greenhouse Gas Emissions of Nuclear Energy in Australia"; Integrated Sustainability Analysis, University of Sydney; 2006.

They explicitly discuss and then discount the Stom-Smith van Leeuwen figures in their analysis.


Unknown said...

In the fact page( )to the BBC Countryfile programme on Sun 4 Oct 2009 there is a quote attributed to the "The government's new energy advisor" (presumably Professor MacKay ?) that by 2016 the country could be facing power blackouts because renewable energy is not coming online quickly enough

This of course is not correct. To avoid power blackouts, the UK needs to have thermal power plant with sufficient capacity to meet peak demand for when the wind is not blowing. It is the existence of that back-up plant that secures supply, not the wind power itself!

Please could Prof Mackay confirm or deny this apparent misquote

Gary Moran said...

Agreed. We are not having an informed rational debate about energy supply in this country. The populace in general are being led to believe that adding more wind plant is our only option (because of global warming) and will guarantee supply, this is somewhat disengenuous.

You can't accuse the population of not making the hard choices (as you do here:, when the choices available are not being explained.

I suspect that if we had an informed debate the consensus would be towards a more economical and conservative approach: medium term ramp up of nuclear plant - with wind only being implemented where it was economic to do so.

Unknown said...

Thank you Gazza Mraz. The link you provided is to the BBC news page that contains Professor MacKay’s (presumably original and approved) version in which he says that the UK could face blackouts by 2016 because green energy is not coming on stream fast enough.

It is clear that the error is on the CountryFile webpage where in order to score a point for the programme clip, which was about windfarms, the word green was changed to renewable

Sloppy reporting like this does the BBC no credit at all and does not help the debate

Of course professor MacKay is still not entirely precise. To avoid the danger of blackouts we need sufficient thermal plant (fossil and nuclear) plus hydro & pumped storage to match peak electrical demand. Any additional capacity from wind (which may or may not reduce CO2 emissions – that is another issue) is not relevant to this particular argument. Is he adding just a touch of hot air here?

One hopes that Professor MacKay will find no conflict of interest between serving his new employer and his intellectual integrity that we value so much.

DanH said...

Further to KC's comment, this report makes some interesting observations about attempts to measure the time-average power output per unit land area used of wind and nuclear energy. The essential point is that, for each of these two energy sources, there is enough wriggle room in the definition of "land area used" to introduce an uncertainty of nearly two orders of magnitude in the measured time-average power output per unit land area used. The implication is that the uncertainty ranges for wind and nuclear overlap by rather a lot.

The report appears not to be peer-reviewed, but its bibliography roots it fairly firmly in the peer-reviewed literature, and I've looked over this part of its argument pretty carefully without detecting any obvious fallacies.

Gary Moran said...

The paper is referenced here: and here:

Bitcoin said...

Wow those wind turbines are beautiful what ugly brick moldy castle though!