Skip to content

How Do They Measure Energy From Wind Power?

June 26, 2020

By Paul Homewood




There has been some discussion around BP’s switch from Mtoe to Exajoules, as their measurement of energy consumption. I have only done a few quick checks, but the change does not seem to have made any real difference to the trends.

However I want to focus now on how they measure energy from wind, solar and nuclear power, as this can make big differences.

Measuring fossil fuel use is relatively simple – a tonne of oil is a tonne of oil, although assumptions may need to be made about its grade. Similarly, gas and coal can be converted to Mtoe/Exajoules, given assumptions about calorific values. (The BP spreadsheet gives conversion factors for anybody interested.

However, wind, solar, nuclear and hydro power are another kettle of fish. Let’s start with a simple example (ignore the actual numbers please, they are just for show):

Suppose it takes 3 Mtoe of oil to produce 1 TWh of electricity in a oil-fired power station. It could therefore be said that producing 1 TWh from a wind farm is equivalent to 3 Mtoe.

However, the energy content of that 1 TWh clearly is not 3 Mtoe, because power stations do not work at 100% efficiency. Typically, it is more likely to be around 33%, meaning that two thirds of the energy has been lost.

In our example, therefore, the energy content of the electricity is only 1 Mtoe.

BP use the first method, called the “input equivalent basis”, so would value 1 TWh as 3 Mtoe. (Conversion to Exajoules this year does not affect the result in any meaningful way, merely serving to fine tune the calculation).

This is how they describe it:

Traditionally, in BP’s Statistical Review of World Energy, the primary energy of non-fossil based electricity (nuclear, hydro, wind, solar, geothermal, biomass in power and other renewables sources)

has been calculated on an ‘input-equivalent’ basis – i.e. based on the equivalent amount of fossil fuel input required to generate that amount of electricity in a standard thermal power plant.

For example, if nuclear power output for a country was 100 TWh, and the efficiency of a standard thermal power plant was 38%, the input-equivalent primary energy would be 100/0.38 = 263 TWh or about 0.95 EJ.

For many years, the efficiency of this standard power plant has been assumed to be 38%. However, in reality, the world average efficiency of fossil fuel-based power changes over time and has risen from around 36% in 2000 to over 40% today.

Moreover, given the much higher efficiency of the most modern power plant (e.g. the thermal efficiency of a modern gas turbine plant is above 55%), the global average is expected to increase in the future.

However, the UK government uses the latter, output-based methodology. (My understanding is that this follows EU and UN rules):

The energy value for hydro-electricity is taken to be the energy content of the electricity produced from the hydro power plant and not the energy available in the water driving the turbines. A similar approach is adopted for electricity from wind generators where the electricity is regarded as the primary energy form because there are currently no other uses of the energy resource “upstream” of the generation.

I suppose both methods could be said to be right. If wind power has replaced, say, oil-fired power, each TWh has displaced 3 Mtoe.

On the other hand, though, if wind power is replacing the direct use of oil for, say, heating or industrial purposes, then it is the energy content that matters.


Given BP’s assumed 38% efficiency of a thermal power plant, they effectively triple the energy value of wind, solar, nuclear and hydro power.

If we look at the UK data for last year, government figures show that wind/solar/hydro only accounted for 3.8% of primary energy consumption:



On the other hand, BP say that wind/solar/hydro provided 0.73 EJ, out of a total of 7.84 EJ, ie 9.3%.

You pays your money, and you takes your choice!

  1. Tony Budd permalink
    June 26, 2020 10:40 am

    Hi Paul: Very informative. However, I’d still like to see a calculation of the total thermal energy released by the processing of fossil fuels and their eventual use for transport, manufacture and heating. All of that thermal energy, plus of course the output from nuclear, hydro, wind and solar, simply ends up eventually in the atmosphere and/or ocean: there is nowhere else for it to go. I suspect that over the last two centuries that has contributed far more to any global warming than the greenhouse gases, but ironically the move away from fossil fuels will increase the total heat output because of the relative inefficiency of “green” sources as mentioned below. Thanks for all your efforts. Yours, Tony Budd

    • MrGrimNasty permalink
      June 27, 2020 8:55 am

      “there is nowhere else for it to go” – just part of the energy the earth is radiating to space.

      There’s several articles/papers out there on the topic easily found via Google (the pay walls can be circumvented – also easily available info. via Google.)

      At one extreme you can find an article saying it is 100% responsible for global warming, but most seem to say it’s a major contribution to UHI but negligible (1%) for global warming, but could be significant by 2300 (or whatever) – according to the same models that predict disastrous greenhouse gas global warming and daftly assuming continued exponential energy production growth.

      It’s probably in the same category as geo-thermal/volcanic heat, obviously there are local affects, but it is insignificant for global warming – although again you will find some who disagree.

      e.g. One published paper:

      “The relative contribution of waste heat from power plants to global warming.”

  2. Rupert Fiennes permalink
    June 26, 2020 11:29 am

    BP assume every fossil fuel plant has a thermal efficiency of 38%? Back 30+ years ago a Rankine cycle coal plant had one of 41%! CCGT is currently around 64%.

    • It doesn't add up... permalink
      June 26, 2020 8:42 pm

      With frequent ramping up and down and startup/shutdown to match variations in demand and wind and solar output, I suspect most CCGT plants in the UK are struggling to do better than about 45%. You can actually see the evidence of this in System Sell/buy prices compared with the levels of CCGT output

  3. A man of no rank permalink
    June 26, 2020 1:28 pm

    Going back to Paul’s post of June 13, about annual renewable subsidies, my sums have to be wrong:
    £10 billion subsidy for 9.3% of UK’s primary energy usage seems rather expensive.
    £10 billion for 3.8% sounds more like a racket!

  4. Ben Vorlich permalink
    June 26, 2020 3:47 pm

    “UK government uses the latter, output-based methodology. (My understanding is that this follows EU and UN rules)”

    Since Brexit has only cut out the EU middleman then this is unlikely to change in the near future

  5. June 26, 2020 4:03 pm

    “although assumptions may need to be made about its grade”

    And the efficiency of its transformation into delivered energy.

  6. Joe Public permalink
    June 26, 2020 5:33 pm

    “Suppose it takes 3 Mtoe of oil to produce 1 TWh of electricity in a oil-fired power station. It could therefore be said that producing 1 TWh from a wind farm is equivalent to 3 Mtoe.”

    ONLY if all oil is used to manufacture electricity.

    Suppose it takes 3 Mtoe of oil to produce 2.5 TWh of process heat. It could therefore be said that producing 2.5 TWh of heat from a wind farm is equivalent to 3 Mtoe.

    The % of oil (or coal, or gas) used to generate electricity is considerably smaller than the %s used for other purposes.

    “You pays your money, and you takes your choice!”

    Lies, damn lies and statistics.

  7. tom0mason permalink
    June 26, 2020 6:42 pm

    “How Do They Measure Energy From Wind Power?”
    By taking the number of hobgoblins they can create in their proposal and financial report and multiplying it by a million.


  8. It doesn't add up... permalink
    June 26, 2020 8:33 pm

    I can confirm that BP used a conversion factor of precisely 10PJ of “input equivalent” per TWh of wind energy generation for years up to 2000, declining thence to exactly 9PJ/TWh in 2017, and to 8.91437935272429 PJ/TWh for 2019. The conversion factors (and the change from the previous year) to 5 d.p. are
    2000…10.00000… 0.00000
    2001… 9.93506… 0.06494
    2002… 9.87097… 0.06410
    2003… 9.80769… 0.06328
    2004… 9.74522… 0.06247
    2005… 9.68354… 0.06168
    2006… 9.62264… 0.06090
    2007… 9.56250… 0.06014
    2008… 9.50311… 0.05939
    2009… 9.44444… 0.05866
    2010… 9.38650… 0.05794
    2011 … 9.32927… 0.05723
    2012… 9.27273… 0.05654
    2013… 9.21687… 0.05586
    2014… 9.16168… 0.05519
    2015… 9.10714… 0.05453
    2016… 9.05325… 0.05389
    2017… 9.00000… 0.05325
    2018… 8.94737… 0.05263
    2019… 8.91438… 0.03299

    They have also revised TWh wind output generated data: the most significant revisions are post 2010, summarised as:


    Odd that a major nation like Canada should see such large revisions

    On the wider changes in electricity generation by source (N.B. – electricity generated, not equivalent primary energy input), they changed the data for 2018 as in this chart (a long scroll through!)

  9. Thomas Carr permalink
    June 26, 2020 9:06 pm

    Anybody done any work to determine the power required to create the components, assemble ,transport, install and commission a wind powered generator. Never mind the fuel needed for access to the wind farms and on-site servicing?
    Plenty of talk about costs and capacity to serve X thousand of consumers on an intermittant basis but I have yet to read about the embedded power of the hardware.

    • Steve permalink
      June 27, 2020 7:15 am

      Plus the maintenance of turbines 500 ft above the sea and replacing these and rotors every 20 or less years. Also the maintenance of the undersea cables and the cost of the additional grid lines and controls. Of course the cost of maintenance of gas and biofuel stations should be in the sums too and an allowance for their inefficiency following wind and solar should be deducted from wind and solar. If only Prof MacKay and his team were still around.

    • John Cullen permalink
      June 27, 2020 10:01 am

      Hello Thomas,

      Professor Kelly (Cambridge Uni.) has done some work on related matters. For example, see page 3 of the following link for the materials requirements to turn the UK vehicle fleet into electric vehicles:-

      Click to access KellyDecarb-1.pdf

      For more detailed work by Kelly please refer to:-

      Click to access lessons_from_technology_development_for_energy_and_sustainability.pdf

      Refer to page 3 in the above link for a discussion on “the energy return on investment” which is what I think you are particularly interested in.

      IIRC professor Kelly used the above paper as the basis of a recent GWPF lecture but, unfortunately, I can’t find the link at the moment. However, I recall he spoke about the number of unicorns that would be required to make this “all electric” world happen. This is the first time I had heard of unicorns being used in such circumstances – but, I’m afraid, I doubt it will be the last.

      John Cullen.

    • John Cullen permalink
      June 27, 2020 10:15 am

      Hello Thomas,

      Sorry for blank space above; clearly the first link I gave was not appreciated by the system. You should find it by following a link made up from the following (but without the spaces!):-
      the gwpf dot org / content / uploads / 2020 / 05 / KellyDecarb-1 dot pdf

      I think you will be able to follow the final link (i.e to unicorns) here:-

      Fingers crossed!
      John Cullen.

      • dave permalink
        June 27, 2020 11:11 am

        There is, for a CONSUMER of electricity, little question of which “measure” of the proportion of electricity produced by non-fossil plant is relevant. Imagine that you hear that all the non-fossil means of making electricity have been destroyed, and that each consumer is to have his supply reduced in proportion – while the fossil production will be exactly unchanged.

        Normally, your bill for a month says 1,000 KWH used. Will your monthly bills in future say (UK Govt ‘method’) 960 KWH used or (BP ‘method’) 900 KWH? Will you have to cut back your use by 4% or by 10%? Which method will give the sensible answer?

        The cost accountants employed by producers of electricity may well be interested in various ‘equivalents’ that do not greatly concern the public.

  10. Nordisch geo-climber permalink
    June 27, 2020 10:27 am

    For comparison of energy sources, search for Euan Mearns’ Energy Externalities days 1 to 13:

    The original article, the comments below, then the daily analyses for each energy source are all very revealing.

  11. It doesn't add up... permalink
    June 27, 2020 10:16 pm

    Just for a bit of fun I took a look at what DUKES reveals about the efficiency factors for different fuels in electricity generation. By cobbling together data from tables 5.1 and 5.3 it is possible to estimate annual averages by fuel. The results are in this chart:

    Coal comes in a little on the low side compared with BP’s assumptions, and shows a very modest decline – doubtless the effects of ageing of existing plants. The apparent decline for oil is more precipitate, but that will be because oil use is now confined largely to diesel generators rather than thermal ones. Gas shows a slightly rising trend, which is perhaps surprising, since there has been very little newbuild. Efficiencies are comfortably above the BP assumptions, although at 46-48% are some way below what good plant can achieve in continuous operation.

Comments are closed.

%d bloggers like this: