Bad and good news from the IEA

The International Energy Agency reported today that global CO2 emissions hit a new record in 2010, and are well above where they should be for a path to stabilise CO2(+equivalent) concentrations at 450 ppm. The Global Financial Crisis has had a significant impact in the US and Europe but (not surprisingly) hardly any in China, where the impact of the crisis was short-lived, and rapidly offset by a strong fiscal stimulus. With the failure of policy in the US, things are not looking good. On the other hand, after playing the wrecker’s role at Copenhagen, China now seems to have embraced the idea of becoming the world leader in renewable energy.

The real good news is a new study undertaken by the IEA that refutes negative views about the variability of supply from PV and wind power (expressed by quite a few commenters here over the years, and the subject of numerous amateur analyses at blogs like Brave New Climate) and concludes that “the challenges of integrating large shares of variable renewables in power systems are far from insurmountable“. The analysis suggests that starting with existing grid characteristics, and employing balancing technologies now available, it would be possible to supply between 20 per cent (Japan) and 60 per cent (Denmark) of electricity generation using variable renewables, with an average of around 30 per cent. No specific date is given, but the discussion implies a time horizon around 2030.

Unfortunately, the PDF containing the detailed analysis is on sale at a price of 80 euros, which I don’t intend to pay, but the executive summary is online and gives a general idea of the argument.

An important point is that the most natural partners for variable renewables are sources that can be turned on and off easily. Hydro-electricity is the best example, but scope for expansion is limited. The next best case is a mixture of gas and variable renewables, and that seems like the sensible path to take over the next decade or two.

In the absence of any equally authoritative critique of the IEA analysis, I intend to treat this question as settled from now on, as with the prospects for nuclear power. Anyone seeking to make unsupported counter-claims based on their own intuition, BNC-style amateur analysis and so on should take them to the nuclear sandpit.

Summing up the news so far, if the world’s governments are willing to act to stabilise the global climate, they can do so at very low cost. It remains to be seen whether or not they will.

40 thoughts on “Bad and good news from the IEA

  1. OK, John, read the executive summary. To claim that that report shows that the problem of integrating renewables into the grid is “solved” is a misreading.

    Firstly, replacing one-third of the grid with renewables isn’t close to where we need to get to.

    Secondly, it says nothing about the effect on generation mix of such a change; and the consequent change in emissions profile. If the net result is that more MWh come from open-cycle gas turbines, and fewer from combined-cycle gas, it’s not going to be nearly as big an emissions win as you’d otherwise assume.

  2. An old work mate of mine made the suggestion that roofing material should incorporate solar panels. We would thus essentially solar-panel our houses as we roofed. It’s an interesting idea which might work in some format.

    Will solar panels give us the power-return (EROEI) we need? How long does it take a solar panel to return the energy put into making it, transporting it, installing it, maintaining it, and retiring/recycling it? How much longer than this will it keep giving us energy? How will we make solar panals without the “subsidy” of petroleum/gas based energy? Does anyone in this blog know the answers to these questions?

    The key issue in the long run will be to discover if the EROEI of renewables is great enough (and soon enough in the oprational life) to power our civilization and to power ongoing re-production of solar capacity without energy subsidies from any other source.

    We’d better hope it works as everything else will run out; like oil, coal, gas and uranium. I think it may work but not to our current level of wasteful energy use and not with an earth of 7, 8 or 9 billion people.

  3. An important point is that the most natural partners for variable renewables are sources that can be turned on and off easily. Hydro-electricity is the best example, but scope for expansion is limited. The next best case is a mixture of gas and renewables, and that seems like the sensible path to take.

    Erm, are there any examples of renewables that can be turned on and off easily, yet are sufficiently scaleable and affordable, apart from hydro? No. But we can overbuild I guess

    In the absence of any equally authoritative critique of the IEA analysis, I intend to treat this question as settled from now on, as with the prospects for nuclear power. Anyone seeking to make unsupported counter-claims based on their own intuition, BNC-style amateur analysis and so on should take them to the nuclear sandpit.

    Your blog, your rules, but a pretty unsupportable position apart from a personal preference/interpretation of netiquette. I have no idea what it says about the prospects of nuclear power, you need to elucidate that.

  4. How much longer than this* will it keep giving us energy?

    * By “this” I meant “how much longer will it keep giving us energy after it has broken even in energy terms with respect to its lifecycle energy costs.

  5. An important point is that the most natural partners for variable renewables are sources that can be turned on and off easily. Hydro-electricity is the best example, but scope for expansion is limited. The next best case is a mixture of gas and renewables, and that seems like the sensible path to take.

    Much more sane and cheaper to build gas plants only and ignore renewables entirely. At least for the forseable future. Renewables such as wind and solar remain a token to green fantasies. Without some technological advance they should be mothballed. Far better to spend the money on additional gas plants.

  6. p.s. If we scrapped MRET but had a significant carbon tax and kept the ban on nuclear then gas fired plants is what would get built. MRET distorts the market more than a carbon tax would.

  7. Wilful, you’ve missed the point. The question being asked is, what kind of existing system is most complementary to variable renewables. The answer, as I’ve said is hydro then gas.

    On nuclear, my point is that the nuclear option is clearly off the table for the foreseeable future, and I don’t intend to debate it, or have discussion of relevant issues derailed by this pointless debate. So I’ve quarantined it to the sandpits, with beneficial effects on discussion, in my view. I see the “variable renewables supply vs stable baseline power” talking point, central to the claims of nuclear fans, as belonging in the same space.

    Responding to Robert M, a system of one-third renewables, and two-thirds gas would have about one third of the emissions of one based on coal (I’m ignoring hydro, nuclear etc on both sides for simplicity). Double energy efficiency (entirely practicable with existing technology for most applications) and you have an 85 per cent reduction in the emissions for a given supply of energy services. That’s pretty close to where we need to get.

    As you say, it matters significantly whether gas is open-cycle or closed-cycle, but that question appears (to me) unrelated to the interaction between gas and renewables.

  8. Well in Australia it isn’t hydro, there are no more dams. So it’s gas, as the only possibly decently sized renewables available in reasonable timelines, solar thermal and wind, are the intermittents that we’re trying to deal with.

    We’ll run out of that particular fossil fuel before too long. 30 years with current rates of extraction and export, I’ve read. What then?

    Without belabouring the point of nuclear issues (and completely supporting the quarantining effort), it’s a gross overstatement to say that baseload is a first order argument for nuclear proponents. It’s simply not. Anyway, I will take further comment on that point to the nuclear sandpit.

  9. @John Quiggin

    As you say, it matters significantly whether gas is open-cycle or closed-cycle, but that question appears (to me) unrelated to the interaction between gas and renewables.

    Intuitively, if you’re running your gas-fired generators less often because of greater variability in demand for gas-fired electricity, that’s going to increase the incentive to minimize your capital costs for the required amount of gas capacity.

    Intermittent renewables will increase the variability of demand for gas-fired electricity.

    Hence, more OCGT and less CCGT.

  10. I take your point Robert, although of course that choice will also be affected by carbon prices and gas prices.

  11. I’ve updated to make the point that the implied frame of analysis is the period to 2030, over which it seems reasonable to work on the basis of incremental improvements to known technologies.

    While it is impossible to forecast technology past 2030, it is safe to predict, given the kind of carbon price we need to induce large-scale adoption of renewables, that innovations permitting further substantial reductions in emissions will be available by then.

  12. @John Quiggin

    Also point taken.

    But you’ve touched on the other great unknown here: what happens to gas prices should the world’s coal-fired power stations be replaced by some mix of renewables and gas.

  13. #3 “…are there any examples of renewables that can be turned on and off easily, yet are sufficiently scaleable and affordable, apart from hydro?”

    Geothermal, anybody? There was a segment on it on The Science Show a while ago – I guess it needs a few years yet, but comes with an on/off switch, I think.

  14. From Wikipedia:
    EGS / HDR technologies, like hydrothermal geothermal, are expected to be baseload resources which produce power 24 hours a day like a fossil plant.
    The largest EGS project in the world is a 25 megawatt demonstration plant currently being developed in the Cooper Basin, Australia.
    The Cooper Basin project has the potential to develop 5–10 GW. Australia now has 33 firms either exploring for, drilling, or developing EGS projects. Australia’s industry has been greatly aided by a national Renewable Portfolio Standard of 25% renewables by 2025, a vibrant Green Energy Credit market, and supportive R&D collaboration between government, academia, and industry.

  15. I’m not an engineer, but I wonder about this idea that hydro cannot be expanded to provide more variable power. Without damming any more rivers couldn’t you just put more turbines in existing dams? Certainly this wouldn’t allow the creation of more hydro electricity over the course of a whole year, but it should allow an expansion of peak power. You could have most of the turbines shut most of the time, but open them all when the wind stops and the sun isn’t shining.

  16. @AndrewD

    HFR geothermal is a promising technology – so many promises, so little actual power generated…

    In any case, HFR geothermal makes wind and solar kind of moot. It’s capital intensive but the “fuel” is free, so there’s no reason not to run it at 100% capacity all the time.

  17. The other part of the grim prognosis from IEA was that 2 degrees rise is now the minimum we face, not the maximum we can aim for, given the success of the denial industry. The Guardian ran the report with a quote from Fatih Birol, chief economist at the IEA saying the IEA supported nuclear power – “People may not like nuclear, but it is one of the major technologies for generating electricity without carbon dioxide.” Astonishing the nuclear push continuing even after Japan.

    The other worrying news, which complements this report, is that methane release from the warming Arctic is about to take off. And then it’s goodnight Irene.

  18. @Robert Merkel
    That’s not my impression. As I understand it, most geothermal generators operating at their peak draw heat out of the ground faster than it can be “replaced” from underneath. This means eventually they have to be stopped for a time while the heat reservoirs are recharged. It makes sense then, to only run them at peak when electricity prices are high.

  19. @Ikonoclast
    An old (2004) publication by National Renewable Energy Lab shows the payback time for PV is around three years. I read somewhere that has significantly reduced, so assuming a 30 year operating period on your module you get around 28 years of free electricity. There is a similar calc for CO2 in there I think.

    Click to access 35489.pdf

  20. “Free” meaning energy and CO2 free. The financial payback is highly variable.

  21. If you wan’t to get a good idea of the variability of wind power in current power systems and the associated technical problems (and solutions) there is some good information on the IEA Wind website ( In particular the Task 25 report which looks at the performance of power systems with significant wind integration.

    In the Task 25 report the systems/wind installations analogous to the Australian situation look at Ireland and Texas. They have analogous grids that is they do not have synchronous connections to other large grids. They also have a similar connection history with wind power: large wind farms connected to the high voltage network rather than small wind farms connected to the low voltage distribution network.

  22. @Robert Merkel There are a bunch of technologies in this class: geothermal, CCS, various storage methods, and a couple I’ve consigned to the sandpit. In the last decade, only wind has really delivered, but it’s already clear that solar PV is going to be competitive in the immediate future. Combined with gas, that provides the path to 2030 if we choose to take it. By then, standard probabilistic arguments say that one or more of the “promising” technologies will have stopped promising and started delivering.

  23. This discussion has yet to touch on the perspective that although some renewable supplies are highly variable, many loads are highly variable too. One of the authoratitive voices in the sphere of energy efficiency and demand management is Alan Pears who argues that we would almost do away with off peak power if we hadn’t created the incentives to run equipment in off peak periods in the first place. For me, the key to unlocking the true value of variable renewables and of embedded generation, is smart grids. Denmark has run trials to show that they can micromanage parts of their grid with 100% renewables – they just needed the right smarts in place to balance the supplies and the loads. While I couldn’t find a good Alan Pears article on the subject, I thought the following article on “smart grids demand smart policy” might trigger more ideas and commentary:

  24. Australia already has plenty of existing coal plants, some of which have decades worth of life left in them. There is no particular reason why they cannot be run as load following or peak plants. Just how easy or economical it is to do this depends on their type, with newer plants generally being more flexible.

    Besides running them less often, CO2 emissions from coal plants can also be reduced by burning biomass in them. This can be done by co-firing where biomass is mixed with coal, or plants can be modified to operate solely off biomass.

    Using biomass to simply replace coal at its current level of consumption is an option, but depending on how it’s produced, replacing the 130 million or so tons of coal we annually burn with biomass might require several percent of Australia. If it costs $200 to replace a ton of black coal with biomass the wholesale price of biomass electricity might be around 8 cents a kilowatt-hour, which is more expensive than current wind power. It would also be more expensive than a lot of point of use PV if we could install it for the same cost they do in Germany. As the cost of wind and solar should both continue to fall, I think biomass will mostly be used to meet peak demand. Of course, Australia might still want to produce large quantities of biomass. It could be a valuable export.

  25. I’m prompted to wonder whether it’s time to take another look at pumped storage — especially seaboard pumped storage. While it will be expensive to set up per unit of potential energy stored, the effective life of such plants is likely to exceed the life of any existing stationary facility by quite a large margin, their footprint likely to be quite small, and these would complement virtually energy technology that we now use or are currently contemplating. They could underpin the utility of CCGT, wind or solar thermal very well. They could also operate as dual personality — providing low-cost desal when power supply was in lesser demand than potable water.

    Can any government paying lipservice to replacing the Collins Class subs say this technology is too expensive for what it offers? It’s hard to see how.

  26. @optimist
    Exactly Optimist. I totally agree. We are talking about using new technologies as if we are constrained to the old delivery infrastructure. Who says we are if we really want something clean to take into the future. Its a whole picture not just a costing per kw.hour using existing grid. This is far too narrow way of looking at it. Its not all about price price price. Its not all about the market we can get “now”. It may be that the price we see for solar now is not competitive but with greater us and better infrastructure – then hey the price falls in the future. Short term pain for long term gain. I agree with Blanchett when she says she wants to be able to look her children in the eye and this is not about “todays price”.

    Its about connecting the dots. We already know in Sydney that a few extra trains are capable of blowing the grid, as are surges of power from existing Solar installations. The next logical step is to consider adapating the grid so that it can handle cleaner technologies and trains are a lot cleaner than cars and solar is cleaner than coal and far less dangerous and difficult to manage long term, than nuclear. The picture needs to be integrated. If we want to clean up how we use energy, power and transport are related, household energy use and transport are therefore also related and the essential infrastructure linkages between the two need to be developed in an integrated way.

  27. It is amusing that Rupert’s flunkies attacked Cate Blanchett for her relatively modest wealth but ignore the wealth of their master or the wealth of a succession of business ‘leaders’ who have made delusionist attacks and demands for business exemptions and compensation that can at best be interpreted as self-serving.

    The best they can say is that Blanchett is better placed than most to bear the additional cost of a carbon tax; they cannot say that she will benefit personally from the policy, except to the extent that action on climate change is good policy that will benefit us all.

  28. @Optimist I’ve written at length on the blog about pricing, making many of the points you do. That’s one reason I’m getting tetchy about the repetition of long-refuted claims about the impossibility of dealing with variable supplies.

  29. Advocates of biomass power (straw, bagasse, woodchips etc) consistently overlook the fact that a lot of diesel is needed to harvest the material then preferably return the residues to the field. Most countries like Australia that have limited hydro and biomass therefore need gas turbine generators to complement the variability of wind power. What happens when both diesel and gas are expensive? This must happen well before mid century.

    Apart from future expense another problem with gas backup for renewables is CO2. In both the UK and Texas it has been found that a megawatt-hour of wind power saves less than half a tonne of CO2 whereas if it fully displaced coal we would expect a whole tonne of CO2 to be saved. This is because the gas turbines need to be kept on standby mode and not operating at their optimum efficiency.

    Therefore I agree with Fran that instead of blowing billions on perhaps unnecessary military toys we should invest in energy storage trials. Pumping seawater to holding tanks on the top of cliffs may work and this has already been done in Okinawa, Japan. Battery storage of wind power has been tried on King Island in Bass Strait and found wanting. If the alternative low carbon base load option is unpalatable we should try the energy storage path. Again I think we’re talking billions for a very modest outcome. Power guzzling industries like aluminium smelting might have to fall by the wayside.

  30. Something that hasn’t been mentioned is this: energy use per capita. Australia, USA, Canada use 2 X as much energy per capita per annum as comparable European countries (and Japan) with comparable standards of living. I guess McMansions and 4WDs (SUVs) go some way toward that. We could therefore 1/2 our energy use without really being any worse off. OK you may not like giving up your 300 kW V8 but I think it is becoming unsustainable. The Gov needs to make policy to get us toward a lower emissions per capita. This would (roughly speaking) 1/2 our emissions without costing a cracker.

  31. I found this list in Wikipedia. It is of energy consumption per capita. You can click on the consumption “twistie” in column 2 to put it in consumption ranking. It looks like 2003 data.

    The first point of interest is the first 8 countries which are small but have very high energy consumption per capita. The reasons for this (I guess) are any or all of;

    1. cheap oil used for desalination (Qatar, UAE, Bahrain, Kuwait)
    2. cheap geothermal used for heating (Iceland)
    3. transport/refueling hub (Trinidad and Tobago? Luxemborg)?

    The first reasonably populous country with use not skewed by special considerations is (arguably) Canada. Canada is cold so no doubt it has high heating costs. But Sweden is also cold yet gets by with a per capita use rate barely more than Australia.

    Germany might be a good model to emulate in that it uses half the energy per head as Canada and considerably less than US and Australia. Yet, Germany is still very productive industrially and has a better solar energy program than most countries around the world.

  32. Hermit, a megawatt-hour of wind power displaces about half the CO2 that a coal plant generates because wind power displaces gas, not coal and gas produces roughly half the CO2 that coal does per megawatt-hour. There is no reason why wind cannot displace coal, it just displaces gas first because gas has a higher fuel cost than coal does. The difference is not because gas turbines need to be kept on standby mode. If it did, it would mean gas turbines would have to spend about 50% of the energy in gas just to overcome friction, which is something that would only be expected if they had a major mechanical failure, or someone stuffed a dead body inside them or something.

    An important thing to keep in mind is that, here in Australia at least, no one is paid to keep a gas turbine on standby. There’s no gas turbine standby pension that power companies can apply for. If a power company says to a distributer that they want $10,000 for keeping a gas turbine idling all day, the distributer will tell them to go jump. They might be more polite about it than that, but there is still a good chance they will be much, much ruder. Power companies may keep a gas turbine idling if they believe it will help them make more money the next paid time period, and they may idle a gas turbine if it will help them avoid getting in trouble for not meeting their obligations, but no one pays anyone to put a gas turbine on standby just because a wind farm is connected to the grid.

    Now intermittent sources of electricity can increase the variability in a grid and result in fossil fuel plants being used less efficiently. But it’s a mistake to think they were being used efficiently in the first place. In Australia demand is a bit of a dog’s breakfast. The old rule of thumb was that 20% wind penetration was required before its intermittency started causing problems, but the figure is very dependant upon grid characteristics and, as more experience has been gained with wind power systems and predicting their output, the amount of penetration that can be achieved before problems arise has increased. Here in Australia wind farms are very good at predicting their output over the short term. So while it is possible for intermittent energy sources to reduce the efficiency of fossil fuel power plants, in reality, such as in South Australia, they don’t make much difference.

  33. The idea that the problems at King Island prove that battery technology has been “found wanting” seems to me a huge exaggeration. As I understand it there were some teething problems that were fixable, but by the time they emerged the company that had built the battery had been taken over and effectively shut down. The new owners were not willing to take responsibility for fixing it, and there as a trial technology there wasn’t an industry with people available who knew what needed to be done.

    Since then same technology with small improvements has been tried in several places around the world and seems to be going well.

  34. Ikonoclast, with regards to your first comment, new thin solar cells can apparently pay back their energy costs within one year in a sunny location. A wind turbine requires about 6 months to make back its embodied energy. (I recently did a back of the envelope calculation for a 1.5 megawatt wind turbine and got 4.5 months, but I won’t pretend I was accurate.) Solar cells are typically guaranteed for 25 years and 40 year old solar cells still produce electricity. Solar cells don’t require oil or gas to make. There are solar cell factories powered by solar cells and there is plenty of electrically powered mining equipment available to extract raw materials.

  35. Even the drill people see that we can not drill our way to independence no matter what, the evidence is overwhelming. So they use their slogans and stay clear of serious questions that they can not answer truthfully.

  36. @wilful
    Good – take it to to the nuclear sandpit Wilful where no-one wants to talk about it with you, so on the nose the pro nuclear argument is (and where you only get me banned by using language more offensive than mine).

  37. The Executive Summary makes this claim:

    Solar (PV) plants can produce electricity even under cloud cover, so output is never less than around 20% of rated capacity (during daylight).

    And then references a footnote with a bit of hand waving.

    I’d like to see some proper justification of this claim as it seems to be just plain wrong. The “real time” PV output for Germany can be seen here:

    Unfortunately you have to click back through months to see the winter performance, but I did keep an eye on it at the time and there were quite a few days where maximum output was well under 1GW of the nominal 18GW. If anybody knows of a decent source of historical data for German PV production, then we could test the claim made in the Executive Summary. I’d bet it’s wrong.

    The claim may well be true for more suitable climates, but that is not what it says and when I read this sort of thing it sets my nerves on edge and it leads one to ask – “Why did they not get this correct?”

    Complaint number 2. Why charge 80 euros for the full report? It is high time that all such studies are made available for free. And that goes for all IEA reports.

    On the topic of accommodating variability, the UK Climate Change Committee in “The Renewables Report” estimates the cost at 1 p/kWh which is tolerable. No doubt this will be somewhat dependent on the local demand profile and grid. It also doesn’t see much of a role for solar in the UK and the deployment scenarios are basically nuclear+wind. If the UK does go down this road (which looks more likely than not) they will decarbonize their electricity supply sooner than Germany.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s