I have an abiding interest in national energy policy, and I am old enough to remember the blackouts and brownouts of the 1940s, when strike action in the coalfields showed what could happen to a modern society short of reliable system-wide energy. These days it’s not strike action that is the likely problem, but political action — the notion on the part of governments that it would be a good thing to move from fossil fuels to alternative energy sources. For some of this persuasion the stimulus is energy security (though that’s not the case in Australia). For others it is the lure of a sustainable world. For others still, it is the absence of fossil fuels and a reluctance to go down the nuclear path. There will be mixtures of these and other sentiments.

I’ve written a few times about the bizarre case of the ACT, which has announced that it will be fossil-fuel free, or 100 per cent alternative, in its electricity supply  by 2025, or whatever date you encounter. This is a fantasy and is either disingenuous or dishonest, or both. By and large Australia’s electricity comes from the grid, and 70 per cent of it is generated by burning coal and gas. When you switch on a light in the ACT the fossil fuel component of that light is the same as anywhere else. What the ACT Government has done is to commission solar and wind generators, whose output by the declared date might equate to the actual consumption of electricity in the Nation’s Capital. That is not the same thing at all.

In 2025, the proportion of fossil fields in the generation of the actual light or heat produced by flipping the switch will be the same as now, around 70 per cent. There’s ample evidence for all of this, and I’m not going to go through it all again. Why does the ACT Government go down this very odd path? Well, my guess is that it is a mixture of belief and electoral strategy. The Government recognises that there is a strong belief in AGW (anthropogenic global warming) in the ACT, and if it wants to be re-elected it need to keep those citizens in mind. (They might vote Green, and reduce the number of Labor MLAs…)

But this post is mostly about the world leader in all this: Germany, which has an official policy about the move to alternative energy, which is called die Energiewende — in translation, the energy ‘transition’, turn, or change. The term is 35 years old, and it’s been official policy in Germany for about 15 years, It means a steady move away from fossil fuels and nuclear energy to wind and solar, and has been heavily subsidised and regulated by the German Government under Angela Merkel.

I came across a systematic review of the policy via Judith Curry’s Climate etc website, and thought it worth discussing, because the lessons you can learn from it apply universally, not just in Germany. The author is Davis Swan, who seems to have been a kind of broad-spectrum energy analyst for the past twenty years. You can read his essay here.

His report card is balanced. Germany has produced a really significant base of alternative energy without either going bankrupt or causing real hardship to consumers, and what it has done will have useful flow-on effects to other countries by bringing prices down. On the other hand, the German Government has spent much more money that it needed to to achieve this (i.e.. it was rich and could afford to do it). German consumers have the second-highest retail prices for electricity in Europe (but they are comparatively rich too, so they can afford them). And Germany has only been able to do this (ACT take note) because there is a European grid from which it could purchase power when the alternative systems were not adequate. The purchased power comes from coal and nuclear energy. What’s more, there is no less dependence on fossil fuel power than there was in 2000, and there is no real sign that that dependence will end in the future. The following graph shows what has happened since 2003: alternative energy has increased but mostly at the expense of nuclear power.


Finally, since alternative energy sources are given preference, the base-load coal electricity generators are pushed off line. Since they are designed to run all the time, this preference to sun and wind means the coal plants are only marginally profitable. One should simply shut them all down, or some of them, at any rate. But you can’t because they’re needed for back-up. And there is still no large storage capacity available to reduce the need for back-up, so the system stumbles on.

Davis says that without finding an economical and hugely scaleable energy storage system this approach cannot proceed much further. That’s the summary, and I think I’ve given it fairly. Davis says that the roof-top subsidies for solar panels have been  a $100 billion failure, because there just isn’t enough powerful sunlight to make much difference. Most of Germany lies above the 48th parallel, which means that in terms of the southern hemisphere it’s all well below Tasmania and below the South Island of New Zealand.  And when the sun is at its brightest, the demand for power is at its lowest. Correspondingly, when the demand for power is at its highest, in the depths of winter, sun and wind don’t help much. Solar power then provides about 1 per cent of demand. That is not a great help when everyone wants to be warm, the days are short, the nights are long, and the lights need to be on everywhere.

He finishes his essay, which I recommend, with this sombre statement: I cannot see how Germany can continue to develop significantly more wind and solar resources in the next few years. The imbalances between supply and demand at different times of the day and different months of the year are becoming too extreme. And with so much generating capacity in place it is difficult to imagine utilities building any new plants. What that means when the nuclear plants shut down is anyone’s guess but it does not look like a pretty picture to me.

Why do we do these nutty things? Again, it seems to be a mixture of belief and electoral apprehension, along with a considerable bit of path dependence. Once you’re heavily involved in doing something, and have spent lots of money on it, there is a real reluctance to abandon that path. You can see that in all sorts of things — defence procurements, rail and road plans and hospital building, to offer three examples. But in energy policy the blind belief that alternative energy is the answer, when the evidence is so strongly against it, makes me shake my head in wonder.

Join the discussion 23 Comments

  • Todd Myers says:

    If I might be so rude as to embrace my pedantry, I believe you mean *above the 48th parallel, not below.

    • Don Aitkin says:

      You are right, and I’ve made the correction. Many thanks. I was juggling where Germany would fit in the southern hemisphere, so to speak, and used the wrong preposition.

  • G van Rijswijk says:

    The move to ‘renewables’ cannot be justified on the basis of technology or the environment.

  • Don Amoore says:

    I mentioned the Jonova discussion to friends in Germany and queried if it was indeed true that so many people in Germany could no longer afford electricity. I received the following reply.

    Regarding your question what´s up with the electrical power companies and the households here in this rich country, I would like to tell you that indeed Germany has very high electrical power prices, the highest in Europe, and the people are annoyed about the wrong attitude in this matter by the government. Only two examples: We give free of charge at night-time huge quantities of electrical power to Austria. With that the Austrians are pumping water into the mountain lakes at night time and in the morning via turbines and water power they are able to produce new electrical power and this expensive electrical power is sold at high price to Germany, this is not rational, is it? Secondly, in the Northern Sea there are hundreds of wind power stations. But there weren´t build cables in time from offshore to the mainland. When the wind power stations should stop turning, then they would be damaged, therefore diesel engines had to be be installed in these wind power stations to ensure that the rotor blades on the top keep on rotating. This I call bad planning.

    A rich industrial nation doesn´t necessarily mean that cleverest people are working in highest positions…..be happy that you are living in Australia and not in Europe.

    • bobo says:

      Hi Don Amoore,

      As promised I had a look at David Evans’ notch delay theory. I looked at the notes at

      The main problem with the theory is that it seems to be completely independent of the behaviour of the climate system up to a constant. In other words, it does not describe the climate system at all.

      First I will summarise what Evans’ notch delay model does:

      Essentially the notch delay model takes as its input total solar insolation (TSI), filters, shifts the data, then scales it by a constant to obtain best fit with surface temperature. Observe that in the following diagram


      the peaks in the gold data (TSI) seem to roughly increase in time. So all Evans did is filter out (remove) the 11 year cycles (because there is no 11 year cycle in the blue surface temp data), remove some of the high f noise (wiggles with periods less than or equal to five years are attenuated), and shift all this along by 11 years to obtain the pink plot:


      Since the notch delay model’s pink plot doesn’t fit the blue surface temp plot, Evans subtracts an arbitrary function (the black plot) which he calls “atmospheric nuclear bomb tests”. Since the support of this function above a certain magnitude corresponds to the period of most nuclear testing, he insists that this fudge function has physical significance, but offers no further justification.

      Some more detailed thoughts:

      The input is total solar insolation (TSI). Evans looks at the spectrum (Fourier transform) of TSI and notices there is a peak at 1/(11 years) which is exactly what is expected given the 11 year sunspot cycle. He then observes that there is no discernible response in the surface temp spectrum, so decides there must be a filter in the climate system that is perfectly tuned to filter out the 11 year insolation peaks. To guess at the first iteration of his response function (notch function) he divides surface temp spectrum by TSI spectrum, and because surface temp spectrum is nearly constant (with a bit of high frequency noise and slight drop off at higher frequencies), and because the only feature of interest is the 1/(11 year) peak in TSI, for all intents and purposes the surface temperature spectrum can be replaced with a constant. So his notch function to model this ratio approximates the inverse of the TSI spectrum up to constant – it’s purely a fit to solar data with no non-trivial climate parameter. Essentially the point of this notch function is to delete the 1/(11 year) peak from TSI because no 1/(11 year) cycle is observed in the surface temperature spectrum.

      Next he applies a delay function; essentially this moves the TSI spectrum without the sunspot cycles forward by 11 years, so that his model has an 11 year response time to changes in TSI. Again this time period corresponds to the sunspot cycle which he subsequently claims must be the cause (or more precisely, the cause is apparently the highly speculative, never observed, solar “Force X” which is perfectly in phase with the sun’s magnetic field). So again no climate parameter is used.

      Next a low pass filter is applied which attenuates TSI frequencies higher than 1/(5 years). Essentially this removes a few wiggles in TSI higher in frequency than the sunspot cycle in the TSI spectrum. He tries to justify this by saying (I think) that the radiative output from the virtual radiative surface of the atmosphere lacks high frequency noise; all he has done is remove high frequency noise from the TSI spectrum and there is no parameter introduced in this part of the model that specifically depends on the climate system. He claims the cutoff frequency for the low pass filter is the same as some empirical climate quantity, but really all he’s doing is filtering out TSI noise without any precise physical justification.

      There are a few constants that are also introduced in the construction of the model but these all multiply together to form a scaling constant.

      So as you can see there is no climate data apart from a scaling factor that fits the output of the model with the surface temp data. Up to scaling factor the notch model simply processes solar output without any climate parameters whatsoever.

      From a physical perspective the obvious question is, where does the energy lost in the notch and low pass filtering go? What precisely is the low-pass filtering correspond to (without handwaving)? Why has “Force X” never been observed?

      To see why this theory is wrong, note that the upper stratosphere and mesosphere are cooling down, the rate of cooling seems to increase with height. This is not possible if increasing TSI is causing global warming.

      • JimboR says:

        Thanks bobo for the detailed analysis and the very readable description of his algorithm. I appreciate how much effort went into that.

        Luboš believes the model has been perfectly tuned to the colour of the temperature signal and that any random temperature signal (with the same spectral components as the original signal) could be shown to behave the same way.

        One of his parting shots as he left the review team was to suggest somebody reverse the temperature signal (in the time domain) and re-fit Evan’s model to that signal. He feels that too would demonstrate the same correlation. If it does, then Force X becomes even more impressive. Instead of being an 11 year delay line on an intricate signal, it becomes a 395 year delay line. That nature has a way of storing and replaying a detailed signal 11 years later is impressive, but if it can replay it 395 years later it’s really impressive. That’s got to take a serious amount of “storage” somewhere.

        I generally like to keep an open mind on theories, but I think Force X is so creative as to make me tend to agree with Don’s “no evidence then no theory” rule.

        • David says:

          Yes, kudos to Bobo.

          I am a bit old school, I prefer my explanatory variables to be measurable or at the very least definable.

          • David says:

            Its ironic how reliant climate skeptics are on “unknown unknowns” for their alternate take on AGW. You will find one in every skeptic’s model.

          • JimboR says:

            While despising them in the orthodox view. If you can’t quote me ECS as a universal constant to 3 decimal places, then I’m going to have to assume it doesn’t exist. Meanwhile, can I interest you in some Force X as a viable alternative?

            Mind you, if anyone can identify how nature has implemented an 11 year delay line on a detailed signal, and we can replicate it in silicon, the possible uses are almost endless. Well, maybe not for 11 years, but presumably if we can do it for 11 years we can also do it for much more useful shorter intervals.

        • bobo says:

          “Luboš believes the model has been perfectly tuned to the colour of the temperature signal and that any random temperature signal”

          OK, originally I didn’t think there was any residue of the temp data, but Lubos could well be correct. When Evans divides temp spectrum by TSI spectrum, he gets


          But his modelled ratio looks like


          which doesn’t appear to have any spectral features apart from the notch (which deletes 11 year cycle when convolved with TSI), HOWEVER, the notch may contain some temperature spectral info. (As an aside, note that the weird asymmetry with the higher valued constant at higher frequencies is eventually killed off by the low pass filter).

          This is the surface temp spectrum:

          Around 1/(11 years) there is a small amount of spectral wiggle that may, with closer inspection, influence the shape of the chosen notch function, this may be what Lubos is referring to.

      • Don Amoore says:

        Have you posted this on Dr Evans (jonova) site? If not why not? Constructive criticism is vital. Well done!

  • Mike says:

    Don I agree in essence with what you’re saying here but have some problems with your figures. You see I have the entire database of NEM (National energy market) dispatch data for 2014. This is data taken for every power station on the the eastern power grid of Australia in five minute increments. I have analysed it to see what the actual percentages are.

    In 2014 the fossil fuel output was 171 TW hours. This includes coal, gas, diesel and kerosene. For the same year renewables produced 31 TW hours. So in fact the 2014 figures are that 16% was produced by renewables. If you remove the hydroelectric component you get 7% and if you go further and look at just wind power you get 4%. The 4% is what needs to be focused on because that is the only thing currently that might be expanded. So you see Don it is far worse and you thought.

    Where do these errors come from? I have read that and as yet not confirmed that if you are running a wind farm you can look at the plate capacity let’s say it’s 100 MW and apply that to the government REC formula which will return you a figure that does not represent the actual performance. The figure it uses is 36% of the plate capacity and if you look at websites for wind farms that’s what they claim. So the fossil fuel generators have to purchase REC (renewable energy certificates) that assume you are producing on average 36 MW. Good business if you can get it!

    In finality it is the 4% that would have to be expanded to whatever wild assertion a politician may make. I do not believe in fairies nor the wild assertions that are made about renewables. I do not think it is at all possible no matter what we do.

    • Don Aitkin says:

      I’ve written bout this before, but can’t find it. Yes, I think that it is coal that is 70 per cent and gas an extra fossil-fuel component. Renewables are a small proportion.

    • Mike says:

      Apologies you were writing about Germany I addressed the situation in Australia.

  • Don Aitkin says:

    I find that I am spending more time than I ought to be in responding to comments that call into question something I have written, but don’t seem to me to advance the discussion in any way. I’ve decided to follow Willis Eschenbach, a citizen scientist who writes good things on WUWT. He ends his posts this way:

    ‘If you disagree with me or anyone, please quote the exact words you disagree with. I can defend my own words. I cannot defend someone’s interpretation of my words… If you think that e.g. I’m using the wrong method on the wrong dataset, please educate me and others by demonstrating the proper use of the right method on the right dataset. Simply claiming I’m wrong doesn’t advance the discussion.’

    Of course, anyone is at liberty to write that I am wrong, or misguided, or biased. I’ll respond (again, if I feel I can do so usefully) when the writer shows me why what I have written is wrong, or provides alternative argument or evidence. I like to learn, and have shifted my stance on an number of issues over the years. But I don’t have the time to engage in endless iterations which lead nowhere.

Leave a Reply