The efficiency of wind power

  • The “global” wind power load factor is estimated to be 22.7% for 2012. This is based on an analysis of installed wind capacity and electricity production data for 17 countries published by BP [1]. This estimate is subject to the veracity of the data and methodology applied. The average load factor for the last 9 years is estimated to be 22.5±0.4% (1SD).
  • A number of competing factors are expected to impinge upon wind load factors over time, such as improving turbine efficiency, the move to less windy onshore sites, the move to windier offshore sites, increasing down time of ageing turbines and climate change. Since 1997 average loads have increased (Figure 1) but have changed little in the last 9 years suggesting that positive and negative factors are in balance.

Figure 1 The average load factor in dark blue for the 17 countries listed in Figure 2. Overall load factors have improved with time but in detail have been static for the last 9 years.


BP report installed wind capacity at the end of the year but electricity production for the full year [1]. Since all new capacity built in a given year has not been operational for the full year an adjustment is required, especially when capacity growth is high. The capacity figure used here in any given year is equal to the year end capacity of the previous year plus half of the capacity addition in the given year. This is imperfect but should result in negligible errors where new capacity additions are small.

Capacity additions

Seventeen countries with significant installed wind capacity were selected (Figure 2). Brazil was originally in the mix but the data were rather chaotic so the country was removed. Back in 1997 there was not too much to distinguish the pack (Figure 2). Germany was first to lift off followed by the USA and Spain. There are 5 major players: China, USA, Germany, Spain and India. The UK is in a distant sixth place. Each of these 5 countries are energy importers and 4 of them, China, Germany, Spain and India may be described as enthusiastic to boost indigenous primary energy production although they will all claim their motive is to reduce CO2 emissions.

It is the USA and China that catch the eye. The worlds number one and number two economies with highest CO2 emissions seem Hell bent on winning the wind turbine installation race.

Figure 2 The history of wind capacity expansion in 17 selected countries.

Figure 3 Load factors for the 17 countries shown in Figure 2 calculated using adjusted capacity as descried in Methodology above. As capacity has grown load variability has declined to the point that in most countries loads have been quite uniform for the past 5 years. The 5 year averages are shown in Figure 4. New Zealand stands out as an anomaly. New Zealand has only 603 MW installed capacity (China 68892 MW) and is therefore a very small player along with Egypt (552 MW). It is plausible that a small number of turbines deployed in a windy site may yield a load factor of 40%.

Figure 4 The histogram shows the average loads for the last 5 years. Wind loads vary from 16.3% in China and 18.3% in Germany to 29.8% in the USA and 30.2% in Australia. All of the data are subject to the uncertainty of the input data [1] and the methodology applied. It is surprising to see that the USA has almost double the load factor of China. In the year 2000 American and Chinese loads were similar (Figure 6). Since then, USA loads have grown, which if true points to an industry getting more efficient while in China loads took a strange step down in  2007, the year that exponential capacity growth began. It is possible that in China State edict means that turbines are being installed anywhere and everywhere.

Figure 5 The data plotted in Figure 4. I’d be interested to hear from commenters how these results compare with national statistics. Note that the data show 5 year means, 2008 to 2012. For the UK in 2012, DECC report 35.2% for offshore wind and 26.2% for onshore wind. In 2012, my figure for the whole UK is 30.8% suggesting that the methodology is sound.

Figure 6 Comparison of loads between the USA and China. At face value the US wind industry is growing more efficient while the Chinese industry less so.

Concluding thoughts

The range in wind load factors between countries is roughly 2 fold while for solar it is at least 3 fold [3]. Furthermore, wind likely has a higher ERoEI than solar making it less sensitive to load for embedded energy recovery. This makes wind a more widely deployable renewable energy source. It still makes more sense nevertheless to deploy wind on windy sites in windy countries.

A global average load factor of 22.5% may seem good news for the wind industry but it must be born in mind that nuclear runs with a load factor closer to 90%. The low load factor of wind is a feature of its intermittency and the key arguments against wind power discussed in “Arguments for and against wind power” remain [4].

[1] BP: Statistical Review of World Energy 2013
[2] DECC Regional Renewable Statistics
[3] Energy Matters The efficiency of solar photovoltaics
[4] Energy Matters The Arguments For and Against Wind Power

This entry was posted in Energy and tagged , , , , , , , , , , , , . Bookmark the permalink.

24 Responses to The efficiency of wind power

  1. Roger Andrews says:

    The main reason NZ load factors are higher than everywhere else seems to be that the Kiwis site many of their turbines in areas where topographic funnelling effects amplify wind speeds. Some of these turbines reportedly achieve load factors >50%.

    NZ also generates about 16 times as much of its electricity from load-following plants (53% hydro, 20% gas, 8% coal) as it does from wind (5%). This would make it comparatively easy to balance out fluctuations in wind output and minimize curtailments, but I don’t know how much difference it might make in quantitative terms.

  2. Hi Euan,

    Great topic. In the US, so much of the new wind capacity is brought on line in the 4th quarter that it may be fairer to calculate the capacity factor for a year with the generation in that year and the capacity in the previous year.

    This would pull the US back into the pack.


    • Euan Mearns says:

      Dave, the USA and China are now engaged in the new Hot Air race for global domination. Without the specific data for grid connections it becomes impossible to work out load factors with meaning. But the collapse part when it becomes too expensive to subsidise, rings true.


  3. Glen Mcmillian says:

    It is very hard to separate the pr bovine waste from the real facts but I do not doubt the figures in this article are reasonably accurate.

    It’s often necessary to use less than the full capacity of many kinds of machinery- trucks are dispatched every second of the day with less than a full load of cargo for instance.

    The real question is this one when it comes to determining the economic value of wind power. After allowing for increased expenses involving integrating it into the grid how much is the actual amount dispatched worth in money in a given year from a given wind farm or in the service area of a given utility??

    This value would be primarily dependent on the amount of fuel saved at fossil fuel plants.It must be multiplied by some factor to allow for future fuel savings which will be worth more as fuel prices go up but on the other hand it must be reduced by some amount to reflect the time value of waiting for the future output of the wind farms.

    Since any actual figures are very hard to track down I suspect that they are not as good as the wind industry would like and not as bad as the anti wind fossil fuel advocates imply.Surely the people who actually work in the electricity industry have ready access to such data.

    A whole lot of people are claiming that there are plenty of places where wind farms can achieve forty percent load factors -at least here in the US-but that most of them are too far from large cities and there fore no good unless a lot of new high voltage long distance transmission lines are built.

    This is of course a chicken and egg situation. Who goes first?

    If anybody has links to any serious work on utilizing wind power in existing industries or new industries or old industries that can be relocated I would really appreciate the links.

    There may be a few industries which use a whole lot of electricity that could operate profitably on an intermittent basis forty percent of the time on average- given a cheap kilowatt hour rate..A desalination plant sucks up a hell of a lot of energy for instance. If a desalination plant could be designed to run intermittently on wind power it might be a useful undertaking.Of course the capital costs per unit of output would be about double to triple those a plant operating continuously- but if electricity is a big enough part of the total expense that would not be a deal killler.

    Likewise a desalination plant could run close to twenty percent of the time on solar power in some locations. If solar and wind could be combined at such a plant it might be able to run at as much as fifty to sixty percent or so of its actual capacity.It might be possible to bump this on up to a workable number by using an off line peaking power plant some small portion of the time to flesh out the wind and solar power going to the desalination plant.

    I am not proposing that such a scheme is practical- but rather asking if any such schemes are likely to prove to be practical in the near future.

    • Willem Post says:

      Here is a reality check regarding Germany’s and Vermont’s RE:

      Utility-scale, economically-viable energy storage has not yet been invented. After it has been invented, it will require great cost and many years to deploy it.

      – PV solar energy, without storage, is minimal or zero about 65% of the hours of the year, and,

      – Wind energy in New England is minimal or zero about 30% of the hours of the year, because wind speeds are less than 7.5 mph, too slow to turn the 373-ft diameter rotors; during those hours wind turbines draw energy FROM the grid.

      – There are many hours during the year when the lack of solar energy and lack of wind energy overlap.

      That means almost ALL conventional generating units are required almost ALL hours of the year to provide energy when solar and wind are insufficient.

      It is called having “capacity adequacy” and it is NOT free, because all these generators would need to be:

      – Staffed,
      – Fueled, and
      – Kept in good working order, and
      – Replaced on a scheduled basis with new ones,

      to be ready to serve on a moment’s notice to provide stable energy, 24/7/365, to a modern society.

      At greater annual percentage levels of RE, capacity adequacy, and its operation, including balancing the variable RE energy, adds at least 5 c/kWh to the cost of delivering wind and solar energy to users, a significant “externality”, in addition to the “externality” of an extensive redesign of existing grids.

      Here is a speech to a PV solar stakeholder convention in Germany by Sigmar Gabriel, Vice Chancellor, and Economics and Energy Minister, regarding the Germany’s Renewable Energy project, ENERGIEWENDE, verging on failure. This speech was likely approved by Chancellor Angela Merkel. The audience was stunned to hear the unvarnished truth regarding RE.

      Gabriel: “The truth is that in all fields we under-estimated the complexity of the Energiewende.”

      Vermont RE promoters have also grossly underestimated the RE complexity and costs. The SPEED program for projects 2.2 MW or less produces energy at 3-4 times NE grid prices. Poor Vermont’s 2011 CEP to have 90% of ALL energy (electrical energy is only 1/3 of ALL energy) from RE by 2050, is a much more ambitious goal than rich Germany’s 2050 goal. See URLs.

      Here are the production results for the SPEED Program, 2.2 MW or less:

      2010……..5,980,779 kWh……..0.1387 $/kWh; July – December
      2011……20,172,973 kWh……..0.1644 $/kWh
      2012……29,666,592 kWh……..0.1716 $/kWh
      2013……44,822,813 kWh……..0.1919 $/kWh

      Note the RISING trend, whereas RE promoters were claiming RE rates would decline. NE annual average grid prices are about 5 c/kWh. The more such expensive energy is rolled into electric rate schedules, the higher household and business rates will become. This folly needs to be stopped.

      Gabriel: “The complete exemption from paying feed-in tariffs is a model that is wonderful for you (PV stake holders and PV system owners) as a business model, but is one that is a problem for everyone else.”

      Vermont’s legislature have also shifted the RE benefits to PV stake holders and PV system owners, and shifted the costs, directly or indirectly, to everyone else, a most regressive, undemocratic “taxation” approach.

      • Roger Andrews says:

        Willem: Useful data. Thanks.

        It’s probably worth pointing out that while the US isn’t regarded as a world leader in the fight against global warming, no fewer than 35 states plus the District of Columbia have adopted renewables targets (an eyeball average gives about 20% renewables by 2020). How many, if any, of these targets will be met is of course another question.

  4. Staffell and Green have a paper on wind turbine degradation for the UK fleet, which follows on from Hughes’ earlier report on aging for the Renewable Energy Foundation. They concluded that wind turbines lose 1.6 +/- 0.2% of the their output per year, with average load factors declining from 28.5% at new to 21% at 19 years.

    • Euan Mearns says:

      Graham, this is very interesting stuff at a level of detail I do not have time to follow. The move offshore in the Uk could mask large decline in load of onshore fleet. Andrew Montford on Bishophill I know has pursued this but I just don’t have time to follow up all these leads. David MacKay pitched in with comments on one of The Bishops threads.

      • For those not familiar, Gordon Hughes published a report on age-related degradation of wind farms in the UK. But David MacKay refuted the scale of the degradation, in particular highlighting the statistical problem of identifiability. This led into Staffell and Green’s paper which had input from Hughes, MacKay and others, and was under review for about 5 months, published in “Renewable Energy”. This was also covered on Bishop Hill

        The obvious problem with measuring age-related degradation is the need to correct for wind conditions, but also other confounders include site selection favouring windier sites first and improvements in turbine technology, etc. Since power output is related non-linearly to wind speed, the particular model’s power curve is relevant, there is a difference between monthly average wind and instantaneous wind etc. Hughes study was the first long term fleet-level study into loss of output from wind farms in the open literature, but one wonders whether Vestas and others already have similar data, and wind developers would surely be interested in this. My shorthand conclusion is that the Staffell and Green paper probably provides the best guess at this stage, small but significant age-related degradation and worthy of more research.

  5. Willem Post says:


    Very useful data in this article. The US CF has increased somewhat as a result of recent installations in windy areas.

    US Regions: Here are the official regional 2012 CFs for NEW projects commissioned in 2010 and 2011:

    – Central States………..0.370
    – Great Lakes…………..0.280
    – West Coast……………0.260
    – Northeast………………0.252
    – Southeast………………0.247

    See page 48 of URL.

    Here are some US Northeast data:

    Maine has a CF of 0.25

    Here are the 2013 expected ridgeline wind production results:

    Sheffield…… 40 MW……$120 million…………83,395 MWh………..CF 0.238
    Lowell*………63 MW……$170 million………..113,687 MWh……….CF 0.206
    Georgia……..10 MW……..$28 million………….21,024 MWh……….CF 0.240 Est.

    New York State:
    Here are the actual capacity factors of New York State. They are not anywhere near project-owner claimed values of 0.33, or better, to get permits and subsidies from gullible/complicit government entities.


    This article contains actual production data from various ares of the world:

  6. Yorkshire Miner says:

    Dear Euan
    I used to follow you on The Oil Drum and always used to enjoy your articles immensely. My answer will be subjective as I have very little knowledge on renewable energy but it might help to explain why I think these figure for load factors in the different counties differ so from the theoretical load factors. The difference in the figures for Germany and Denmark, really caught my eye Germany around 18 percent and Denmark around 24 percent. I think the different circumstance in the two countries is on there ability to get rid of the production. The Danes can quiet easily dump there excess production into the the into the Norwegian and the Swedish grid which can easily absorb it by switching off there Hydro production. This is something that the Germans can not easily do. Another factor is the the difference in the size of the Danish and German PV production Germany has about 36 GW of PV production while Denmark has about 200 MW. The Danes don,t have to juggle with fitting PV into the Grid. I live in the Netherlands right on the Germany boarder in fact the end of my Garden is in fact the German Dutch Boarder, About a mile away is a wind farm consisting of 16 turbines of different makes and sizes they are all owned to my knowledge by different local coops. The first two put up in 1997 were just dismantled, they were small 500 KW they have been replace by one of 2.25 MW of the same make. I was talking to the engineer who was doing the dismantling and installation. The two that were dismantled had been sold I think for 100,000 Euros and were off to Sicily for a second lease of life after refurbishing what I used to find interesting is that they always seemed to be running when the other taller ones were stopped. I assume that this is that because they were erected at the beginning of the boom that they had preference on the grid and as the market grew that the connection contracts were more restrictive. When I went out early this morning they were all working, I have just been out and out of the 16 three are not working and they are not the old ones. The problem as I see it basically is that in the fossil fuel system of producing electricity is that we begin with storage and then produce the electricity the piles of coal at the power station being the storage while with renewable unless it is Hydro we start first with the production and have inadequate storage after the production which means that we cannot use the renewable resource to its maximum an earlier comment suggested that desalination could be used as a form of storage an extremely fine idea, a complete waste of time here in Europe and especially Scotland. There are other alternatives though Audi have just opened a methane production plant last year. I could think of a few more but once mass storage starts to be employed then all this talk of load factors will become irrelevant. What is hidden in these figures is the fact of how much renewable energy we are wasting. As I have said before once storage expands then we will have more renewable energy without any real extra costs

    Thanks once again Euan for a fine article.

    • Euan Mearns says:

      Hi Miner, thanks! You may be on to something suggesting that the low load in Germany is down to curtailment. Especially when you look at Poland with a load of 22.4%. But this does suggest that the Germans are wasting a lot of capacity and likely making large curtailment payments.

      Storage is the key to making renewables work. But it is coming up with something that is scalable, energy efficient and cheap. And even then if we were to scale up renewables production to say 50% of energy it would mean covering our countryside with windmills and solar panels and storage facilities.

      • kakatoa says:


        It looks like your government has support for UK’s RE plans- they have a study saying you want it:

        at least the “The European Wind Energy Association (EWEA)” interprets the survey data this way:

        …..”For renewable energy sources in general, 80 percent of the public said they support the use of renewable energy to provide the UK’s electricity, fuel and heat. Moreover, six in ten people said they would be happy to have a large scale renewable energy development in their area….”

        • Euan Mearns says:

          Kakatoa, the people are told that renewable energy is cheap, clean and secure by the government. None of this is true. On the aesthetics front I have no objection to roof mounted solar and little objection to wind turbines most of the time since I live in the city. But I am bothered by the growing number of wilderness turbines for some of the time. A city dweller may not object to a neighbour installing solar PV, especially since most do not know they are paying the subsidy for it, but if you have just retired to a cottage in the country with views of rolling hills and someone comes and plonks 20 turbines a mile from your house you may get rather pissed off.

          Considering that most folks in UK live in towns and cities 6 in 10 will never run the risk of having 50 turbines, a pumped storage scheme or 1000 acres of rape being developed close to them.

          • kakatoa says:

            Morning Euan,

            I am in the second category- We live in the country on some acreage, thank goodness as I can adapt to the ways CA, like the UK, are trying to address CO2 levels. It’s been awhile since I believed in the tooth fairy so I concur with your assessment of the realities of the costs, and say the added complexity of keeping the grid up, to decarbonize society: “the people are told that renewable energy is cheap, clean and secure by the government. None of this is true.”

            A local pumped storage effort is one means of addressing the intermittency of RE here in Northern CA. The problem is you have to admit that there is a problem, which doesn’t fit the narrative…….. Your approach to addressing the realities of the situation by asking rather straightforward questions seems to be the way to address tooth fairy solutions. I wonder if the UK government allows access to the raw data in those surveys- not likely as one could likely come up with rather different conclusions than what the wind power lobby would like to be presented.

          • philsharris says:

            Euan and all
            We live among ‘big farming’ in NE England just short of the Scottish Border. I guess we see more than 1000 acres of oil seed (rape) within 2 mile radius of us. It has been like that for a good while now. Of course it is not all OSR in every field because the crop must be rotated.
            Scotland’s arable area (a small fraction of the total land area) has the best growing conditions for OSR in Britain. I did a calculation a little while ago based on official data that the maximum oil seed crop possible (all of it) in Scotland could provide about 6% of Scotland’s current diesel demand.
            Incidentally some of the OSR varieties round us go as chemical feed stock – contains enough (fairly) toxic acids and glucosinalates as useful starter molecules.

            I agree that renewables in general cannot replace liquid fossil fuel and can only go so far to provide electricity, even with a multi-source pan-European grid – with hydro for peak and balancing. But hey, might be a better legacy than all those tall office buildings dominating our cities and other pieces of nonsense like the swish global HQ constructed by RBS outside Edinburgh before its business model went bust?

            Thanks for good article. ‘Energy payback time’ for a fast-growing industry is a very important metric – as calculated by Michael Dale (Stamford) and presenbted at GES Conference Edinburgh 2013. Wind seemed favourable.


  7. power says:

    Using the same approach as BP we can calculate the load factor for the whole Europe in 2013. The data are available already

    The load factor for wind power is 22.6% and 12.0% for PV power.

    The load factor for nuclear power in the world in 2013 was 75.3%. and even if we exclude Japan, it would not go above 82-83%. It was 75.6% in UK and 77.1% in France in 2013.

  8. jrwakefield says:

    Load factor, capacity averaged for the year, is a completely irrelevant number. Power systems must provide energy on demand in real time. It doesnt matter what wind does over the year, it matters what it can do right now. That means you need to see the Hourly Median Capacity Factor, and the Capacity Value.

    To see the effects of Hourly Median Capacity Factor, you need to break the data up by season. For example, in the summer time in Ontario, the Hourly Median Capacity Factor is a mere 7%. 40% of the time wind produces nothing at all in the summer.

    Capacity Value is what a power plant can produce when demand is within 10% of peak. Again, in Ontario, that number is less than 5%. Typical power plants must be capable of at least 90% Capacity Value.

    Wind power will never be a viable source of electricity because it simply cannot provide power when we need it. This is because wind output spikes up only when a frontal system moves through. Power regulators have no idea how high the power spike will be, or when it will spike, or how long it will fall off the spike. This makes wind completely unreliable, and a serious threat to the stability of the grid.

    This is my analysis of wind in Ontario.

  9. Pingback: Energy articles | Basic Rules of Life

  10. Pingback: Recent Energy And Environmental News – May 27th 2014 | PA Pundits - International

Comments are closed.