Future energy availability: the importance of ‘net energy’

Chris Vernon

There is a lot of oil still left in the ground but unless, when it is put to work, it yields perhaps four times the energy it took to extract and refine it, it might as well not be there. As a result, the supply could contract very rapidly indeed and the world may have run out of useful oil by 2050.

All life needs energy. Organisms depend for their survival on their ability to gain energy from their environment with which to weather the elements, survive pathogens, fight or flee from predators and, of course, procreate. The unique genius of Homo sapiens lies in our ability to manipulate our environment’s energy system. Tools are a means of focusing the energy in our muscles, a knife focuses energy onto a fine edge, and a lever multiplies mechanical force. The development of tools elevated us beyond other species and enabled Homo sapiens to colonise the planet successfully.

Focusing internally metabolised energy with tools was just the first step, however. Humanity’s dominance of the Earth today, which has led to the Anthropocene being regarded as new geological era, has come about because we no longer rely solely on the food we eat to energise our way of life but employ secondary and greater sources of energy. A recent book, Catching Fire, by Richard Wrangham, a biological anthropologist at Harvard, claims that a breakthrough in human evolution happened 1.8 million years ago when our forebears tamed fire and began cooking. This use of fire by Homo erectus led to anatomical and physiological changes that adapted us to eating cooked food.

Offshore wind and tidal barrages give good energy returns

Illustration 1: The world does not need just energy – it needs energy that is delivered with very low levels of carbon dioxide emissions (that is, a low-carbon intensity) while still giving a lot more energy back than it took to produce it. This chart, by Evan Robinson, shows the most promising technologies and those to ignore. The half dots indicate where a technology is beyond the limits of the chart. Source: http://evanrobinson.typepad.com/ramblings/science_nature/.

Wrangham argues that hominids’ jaws, teeth and guts were able to shrink, and more calories were available to fuel their expanding brains, because cooking made it easier for our bodies to extract energy from food.

Then, in the Neolithic period, approximately 9,500 BC, the domestication of animals provided a new source of energy, and for the next 10,000 years, Homo sapiens steadily increased its access to energy by burning biomass, using draught animals and, eventually, harnessing water and the wind. The amount of energy that humanity harnessed from transient energy flows provided by the sun increased steadily for many years. The rate of growth in the energy supply accelerated somewhat when the Romans started to employ limited amounts of coal and soared dramatically after the development of James Watts’ condensing steam engine in the 1770s and, more generally, the start of the industrial revolution.

Coal represented something new. For the first time, energy from a different time period was accessible and, more importantly, available on a larger time scale. Before coal, the available energy was limited to the proportion of the transient energy flows the technology of the day could capture. Coal (and the other fossil fuels) enables us to access a stock of energy sequestered over millions of years in the distant past and, to release that energy over a few short centuries.

Just as tools enabled early man to exceed the physical limits of his body by focusing the energy of his muscles, fossil fuel enables us to use more energy than we could obtain from current natural flows by tapping into vast stocks of ancient energy. The rate at which we are drawing down this ancient stock can only lead to its depletion. The characteristics of this depletion are already becoming apparent, years before its total exhaustion. As the stock diminishes, it becomes harder to extract energy from it. In other words, more energy is required by the extraction process, which reduces the net energy available to society.

Net energy

A tree must gather more energy from the sun through its leaves that it expends constructing the foliage. Similarly, a fox must gain more energy consuming the hare than it took to chase it down. Our exploitation of fossil fuels is no different. In order to extract fossil fuels and utilise their embodied chemical energy, the amount of energy expended must be less than the amount we get to use. In the early days of its exploitation, a resource is abundant, easily discovered and takes little energy to extract. The principle of ‘best first’ is adopted automatically, so the large coal seams near the surface and the large onshore oil fields are both the first to be discovered and easiest to exploit. This ease of exploitation results in large amounts of net energy as relatively little energy needs to be expended to extract the fuels.

As the resources become depleted, however, the task becomes harder. In the case of oil, new extraction is increasingly coming from deep-water deposits. The recently announced Keathley Canyon discovery in the Gulf of Mexico is under 1,259 m of water and the well depth is 10,685 m [1] below the sea bed; that’s a greater distance below the surface of the earth than Everest rises above it. Unconventional resources such as shale oil and Canada’s tar sands require the use of a lot of energy to produce a useful product while coal-to-liquids, biofuels and gas-to-liquids require a great deal of post-extraction processing before the fuels can be used [2]. The net energy — the energy return on invested (EROI) — delivered by all these processes is much less than the return from, say, the first oilfields in Texas.

Illustration 2 summaries the concepts of surplus energy and the EROI ratio. Eout represents the magnitude of energy available after the energy extraction costs, Ein, have been accounted for. This is the energy available to society.

Energy has an energy cost

Illustration 2: An energy source can rarely be used directly. An energy extraction process is required to discover, extract and process the resource before its energy is available to society. This process consumes energy itself, a deduction from the energy otherwise available. The energy return on invested is the ratio of surplus energy to energy required to drive the process.

EROI is a dimensionless ratio. If the extraction of 50 barrels of oil takes the energy equivalent of 1 barrel of oil, the ratio is 50:1 and 98% of the embodied energy in the source is net energy available to society. This ratio has dramatically declined over time. Professor Charles Hall at the State University of New York has calculated that for oil extracted in the US:

The EROI for oil… during the heydays of oil development in Texas, Oklahoma and Louisiana in the 1930s was about 100 returned for one invested. During the 1970s it was about 30:1, and from about 2000 it was from 11 to 18 returned per one invested. For the world the estimate was about 35:1 in the late 1990s declining to about 20:1 in the first half decade of the 2000s [3].

This decline has occurred almost invisibly as total extraction has increased. This has been possible as the decline from 100:1 to 30:1 to ~11-18:1 only represents a move from 99% energy availability to 97% to 93%, a trivial change in the face of the magnitude of total production which increased almost four-fold. There has been a large increase in net surplus energy compared with a small decrease in the EROI. However, projecting forward, this is not linear system. Illustration 3 illustrates how the net energy available declines rapidly as the EROI continues to fall.

Impact of declining EROI on energy availability

Illustration 3: When an energy resource’s EROI is high (>10) then most of the gross energy is available as net energy to society. Over time the EROI declines; however, the net energy does not fall significantly until the ‘knee’ of the curve is reached at approximately 5:1. Once the knee is reached, a rapidly increasing amount of gross energy is consumed by the extraction process itself until it is no longer energetically profitable to continue to extract the resource.

Very low-EROI sources (Canadian tar sands, for example, at <5:1 [4]) are already being used; their exploitation is sustained through energy cross subsidy from high EROI sources like natural gas. Large volumes of water (2-4.5 barrels of water for every barrel of synthetic crude) are also required in this case so it is likely that extraction rates will not depend on the tar sand resource at all but rather on other inputs [5]. This works in the short term, for a small volume, and whilst the gas and water is available, but does not guarantee the continued exploitation that some assume going forward.

Calculating EROI

Calculating EROI is not simple, largely because our current system is denominated in monetary terms, not energy terms. Two significant challenges are energy quality and system boundaries.

To a physicist, energy is a simple concept. Measured in joules (after 19th-century physicist James Prescott Joule), it quantifies the amount of work performed on the environment; work against gravity to raise an object, work performed to increase the temperature or velocity of an object, for example. Quality does not come into it. However, for practical applications energy can be considered to vary in quality, complicating direct comparison. The ten megajoules of chemical energy released as heat when 3 kg of coal is burnt cannot power a television for a day because the heat cannot be used directly. An indication of relative energy quality can be obtained from market price. The price for a megajoule of electricity is typically around three times higher than that of a megajoule of natural gas, and represents a willingness to waste as much as two-thirds of the primary energy in the gas when converting it to a higher quality energy, electricity.

System boundaries are particularly troublesome. A simple analysis may look at an oil well and consider the electrical energy used to pump the oil from beneath the ground compared with the energy content of the resulting oil. This is reasonable, and returns the EROI on the day the measurements were taken. However, energy will also have been expended in discovering the oil field, drilling the well (including the three preceding dry holes) and in the manufacturing and transporting of the pumping equipment itself. This will produce a fairer result because to extract oil, one must first discover it. This line of thought can be extended to include the energy costs of the petroleum engineer’s education, food and health care.

Finally, simply producing surplus net energy with an EROI ratio greater than one still is not enough. A barrel of oil at the wellhead cannot be used as it stands. First, it must be refined into products such as petrol or diesel and transported to where it is required. Secondly, the infrastructure with which to use this fuel must be manufactured; the cars, trucks and the very road surface upon which they travel.

The energy used to extract the energy is only one part of the picture. Further energy must be expended in order to use the energy. Too often EROI discussion is centred upon whether a proposal is greater than unity, whether it breaks even and provides a net energy surplus. This break-even point is not nearly enough though. If our energy system were merely to break even, human civilisation would do no other activity apart from energy gathering. There would literally be no energy available for anything else.

EROI must be greater than one, but how much greater? What is the minimum EROI required for civilisation? Three types of energy use can be defined: energy used to harvest energy (this is Ein from Ilustration 1), energy used to build and maintain the infrastructure to use the energy, and energy used for everything else that makes us civilised.

Charles Hall’s research group address the first two in their paper titled “What is the Minimum EROI That a Sustainable Society Must Have?” [3]. They conclude that for oil and corn-based ethanol, the minimum EROI is 3:1 at the wellhead or farm-gate. Below that 3:1 figure, oil and corn-based ethanol cease to be a viable energy source because the energy output would not cover the first two types of energy use listed above: the energy used for extraction or growing and harvesting and for the construction of the roads and vehicles in which the fuel is to be used. There would be no energy left over for all the other activities of society. Civilisation therefore requires energy sources to have an average EROI significantly higher than 3:1. Hall estimates that the overall EROI of the US energy system in 2005 was between 40 and 60 to one. Coal is extracted at high EROI and oil (domestic and imported) lower than this average [6]. Europe achieves similar complexity of society on approximately half the energy per capita, suggesting that significantly lower EROI can support complex society.

Peak oil

Fossil-fuel resources are finite and, on human timescales, non-renewable. It follows that their extraction rate starts at zero and returns to zero once the resource is exhausted. The simplistic representation of this is a bell-shaped curve with extraction rate plotted against time, and the area under the curve being equal to the extractable resource. Graphs showing the output from many oil-bearing provinces have had this bell-shaped form and their extraction rates have steadily declined after a well-defined peak. However, whilst the extraction rate may be approximately symmetrical about the peak, the first half of a province’s life can be characterised by a small number of large, fast flowing fields. The EROI is high. In contrast, the second half of the province’s extraction is made up of many more smaller and more complex fields, requiring secondary or tertiary recovery techniques. The EROI is low. This is only natural since the best first principle leads to the lowest-cost resources being exploited first.

Illustration 4 projects a possible global oil-extraction scenario. It is made up of a peak extraction rate in 2010 followed by a 2% per year decline rate. In the year 2000, the EROI for global oil is taken to be 30:1, which leaves 97% of the energy available to society as surplus. The blue and red curves illustrate how the surplus energy available from oil declines as EROI declines at 2% and 5% per year respectively.

By 2000, 30 years past its peak, US oil extraction had an EROI of 11 to 18:1, down from approximately 100:1 in the 1930s. This represents a rate of decline of a little over 2% per year. Under a 2% decline scenario, the global oil EROI falls to 11:1 by 2050 with 92% of the energy still available to society. However, if EROI decline at the steeper rate of 5% it passes the minimum threshold of 3:1 in 2045.

Energy from global oil

Illustration 4: The continuous black curve projects the annual energy contained in the global oil supply assuming a decline rate in extraction of 2% per year from 2010. The dashed and dotted curves illustrate the net energy available to society. In 2000, the EROI is taken to be 30:1 (97% surplus). The dashed curve assumes EROI declines at 2% per year, dropping to 11:1 by 2050, the dotted curve declines at 5% reaching 2.3:1 by 2050, below the minimum to be considered an energy source.

In other words, although there might be enough oil for global oil-extraction rates to be approximately half today’s level by 2050, how much usable energy humanity will get from it depends on the rate at which EROI declines. If the EROI declines faster than it did in the US during the 20th century, it is possible that the average EROI will be so low that, by then, oil will cease to be the significant net energy source.


Our unique relationship with the energy system has defined our species. Our current reliance is on previously sequestered stocks of energy, which must suffer depletion and, with it, declining energy return on the energy invested in its extraction, processing and distribution. The pre–fossil fuel existence of our ancestors was reliant on the Earth’s energy flows and suffered no such systemic decline. It is imperative that we find a way to move society away from its current reliance on declining, finite energy stocks and back to an energy system based on flows.


  1. BP Press Release, BP Announces Giant Oil Discovery in the Gulf of Mexico, September 2, 2009.http://www.bp.com/genericarticle.do?categoryId=2012968&contentId=7055818 Accessed 14/09/09
  2. Brandt, A.R., Farell, A.E., “Scraping the bottom of the barrel: greenhouse emission consequences of a transition to low-quality and synthetic petroleum resources,”
    Climatic Change, No. 84, Springer Science, 2007
  3. Hall, C.A.S.; Balogh, S.; Murphy, D.J.R. “What is the Minimum EROI That a Sustainable Society Must Have?” Energies 2009, 2, 25–47.
  4. Hall, C.A.S. Unconventional Oil: Tar Sands and Shale Oil – EROI on the Web, Part 3 of 6
    http://www.theoildrum.com/node/3839 Accessed 14/09/09
  5. Canada’s Oil Sands – Opportunities and Challenges to 2015: an update, National Energy Board, June 2006, pp. 38 http://www.neb.gc.ca/clf-nsi/rnrgynfmtn/nrgyrprt/lsnd/lsnd-eng.html Accessed 14/09/09
  6. Hall, C.A.S., Lambert, J.G.L., The balloon diagram and your future http://www.esf.edu/EFB/hall/images/Slide1.jpg Accessed 14/09/09

Featured image: Power. Author: Rodolfo Belloli. Source: http://www.sxc.hu/browse.phtml?f=view&id=709105