Fossil Fuels and Human Civilisation

Return: Main Page / Energy Considerations

The Industrial Revolution which began in late 18th century England was only possible because of the abundant coal resources available there and to its relative ease of extraction. The subsequent industrialisation of Europe, Russia, North America and Japan was all fuelled by cheap and plentiful coal. It is interesting to note that it is mainly those countries with abundant coal that got industrialised first. Coal was the fuel of the 19th century as it powered virtually all of the machines of the time.

Though oil has been known since antiquity, it is only from the mid 19th century that it became extracted in non-negligible quantities in Pennsylvannia (USA), Romania and Baku in the Caucasus. Initially it was mainly burned for lighting purposes. The invention of the internal combustion engine at the very end of the 19th century had a tremendous impact on the oil industry of that time, for it created a vast market for oil. Before the First World War, motor vehicles and diesel powered ships were already becoming widespread. The second phase of the Industrial Revolution could begin.

By the fifties, oil had surpassed coal as the main source of fuel for the whole world. In 2004, 40% of energy consumed is oil, 20% gas, 20% coal and the remaining 20% shared between hydro-electric power, biomass and a little renewable energy. Furthermore 95% of all transport requires oil. Indeed, globalisation is possible only because the use of oil allows cheap intercontinental air travel and transport on a very large scale. Industrial food manufacture and its transport to markets require large amounts of oil, a fact that is often forgotten. Oil is needed to run agricultural machinery and natural gas is used as feedstock for the manufacture of fertilisers. Of course oil is the pre-requisite for the myriad of plastic used by an average human being in a modern society. As Yergin wrote in his excellent book "The Prize", published in 1991, modern man is really hydrocarbon man due to the over-whelming importance of hydrocarbons in everyday life.

From the above it follows that the present form of the Human Civilisation (and of course its past and current impacts on the environment) and its future trajectory (and future impacts) are very much dependent upon the widespread availability and abundance of cheap hydrocarbon energy. Thus for anyone remotely interested in the interaction between human societies and the environment, it is very important to examine closely its sources of energy. Hydrocarbons being so important for the reasons depicted above and in view of its share of the energy mix, we shall first examine hydrocarbon and its future.

The Future of Oil and Other Hydrocarbons

Hydrocarbons are basically chains of carbon atoms of varying lengths and to which are attached hydrogen atoms. Oil and natural gas are the two main types. The overwhelming consensus amongst geologists is that hydrocarbons are of fossil origin, formed in the distant past by geologic processes that acted upon the biological remains of algae and other marine organisms. There is an alternate theory of the formation of hydrocarbon that basically defends the idea that hydrocarbons are formed continuously in the deep hot interior of the Earth, its main proponent is astro-physicist Dr Gold. However there is no evidence for it. Our starting point will therefore be that all hydrocarbons are of fossil origin.

Once the fossil origin of hydrocarbons is accepted, it follows that the vast amount of hydrocarbons created in the past is both finite and non-increasing with a renewal rate of zero for all practical purposes. Furthermore, once the hydrocarbon is burnt, the energy contained within it is dispersed as heat and the fuel cannot be reused under any circumstances. Hence fossil fuels are finite, non-renewable and non-recyclable.

In view of the importance of hydrocarbons to the economies of nearly all countries and of its finite nature, some essential questions must be asked about its future availability and affordability. Out of all hydrocarbons, oil holds a special place due to the fact that 40% of traded energy world-wide is oil and 95% of transport world-wide depends upon it. Oil has become so important because of the very special properties it has, which are:

  1. It is an abundant and concentrated source of energy with a high-energy profit ratio.
  2. It is a liquid at room temperature that can be readily stored and transformed in a number of different fuels usable by cars, aircraft, lorries and ships. Indeed nearly anything that moves can be made to use some form of fuel derived from oil.
  3. It is readily transformable into a large range of plastics.

To examine the future availability and affordability of oil is a huge undertaking, ridden with uncertainties and conflicting information. To have an informed opinion on oil matters we have drawn heavily on the work of petroleum geologists and engineers, physicists, energy analysts and economists. What follows must not be taken as final and therefore as unchanging, but more akin to an opinion based on the best scientific facts and figures the IELS has been able to uncover and above all understand.

Oil Has To Be Found First

To extract any oil, it is evident that it has to be found first. There can be no oil extraction without prior discoveries, hence the pattern of discoveries is a vital piece of information. This is a fundamental fact that requires careful examination. We need not delve into the technical details of how oil is discovered or how each individual reservoir is assessed for its oil potential because it is a far too complex subject to tackle here. Furthermore it is not vital to have a precise understanding of oil exploration science and technology in order to have an intelligent grasp of oil supply and demand dynamics. This is because we shall use data from the oil industry itself (British Petroleum, ExxonMobil), from the United States Geological Service (USGS), the Energy Information Administration (EIA) of the US Government and from the Association for the Study of Peak Oil (ASPO).

The first piece of evidence is the historical record of annual discoveries of conventional oil from 1930 to 2000 (Figure 1, source: ExxonMobil and ASPO).

It is seen that the general shape of the discovery curve starts at a very low level of discoveries in the thirties, rises quickly to a maximum in the early sixties and since then the discovery trend has been declining steadily. In the early forties, discoveries amounted to 30 Billion of Barrels (Gb) per year when some very large finds were made in the US and Middle East. Then discoveries dropped sharply during the Second World War but rose again to over 50 Gb by the end of the decade. During the fifties, discoveries steadily increased to reach a maximum of over 50 Gb by 1962. Thereafter discoveries fell steadily with occasional peaks of 40 Gb in the late seventies. In the nineties, discoveries fell to about 10 Gb per year. In 2000 and onwards average yearly discoveries fell to below 10 Gb. The overall profile is thus one of rising discoveries from 1950 onwards reaching a maximum in 1964 approximately and a gradual fall ever since.

The obvious question is: can this decline in oil discoveries be reversed? Nobody knows for sure, but there are indications that it is unlikely that it will be reversed. The first indication is that historically, large oil deposits tend to be found early on in any area under exploration, for an obvious reason: they are too large to be missed. The second indication is that overall, planet Earth has been searched for oil pretty thoroughly. Few areas are left for exploration. Indeed exploration has now shifted to very inhospitable places like the Arctic or the deep-sea, off continental shelves.

This fact, in itself, is indicative that the easy and large finds have already been made and that only the difficult and smaller potential oil fields remain to be discovered. Oil will definitively be found but in smaller quantities than ever before. The era of gigantic finds like those in the fifties and sixties is most probably over. Economists such as Lynch (2003) often argue that the decline in oil discoveries is due to a decline in drilling. Alas the record shows that whilst oil discoveries fell in the seventies and eighties, the number of wildcats (exploratory drills) increased dramatically in the seventies but then fell sharply by 1985 wielding few large discoveries (Figure 2). It is no longer safe to assume that any increases in exploration will yield large finds.

Figure 2 (source: Campbell, 2002)

Whilst discoveries have declined drastically over the last 40 years, consumption has not. On the contrary, world consumption increased from 21 million of barrels per day (bbd) in 1960 to 79 million bbd in 2003 (EIA, US government). That increase in the rate of oil consumption coupled with the declining rate of oil discoveries mean that as from 1980, the world has been consuming more oil than it has been discovering. This trend is continuing as oil consumption for 2001 has been 27 Gb whilst discoveries barely 8 Gb (ASPO Newsletter January 2002). This is a very significant fact. As production can only come from already discovered reserves, it follows that world-wide reserves are declining at the rate of 19 to 20 Gb per year. Until the rate of discovery matches or even exceeds the rate of consumption, reserves are in decline.


Conventional and Unconventional Oil

At this point, it becomes important to discuss the difference between conventional and unconventional oil. There is no exact and precise distinction between the two categories. Broadly speaking, conventional oil refers to oil that flows readily as a liquid at ambient air temperature and pressure. It includes oil from onshore fields and offshore fields. However note that oil found in very deep water (in excess of 500 metres) or found in Polar areas are classified as unconventional. Oil sands and oil shale, from which a form of crude oil can be extracted, are also classified as unconventional. Polar and Deep Sea oil are termed unconventional for the simple reason that to extract oil in such places requires highly advanced technologies that pushes men and machines to their operational limits. Heavy and very heavy oils hardly flow at ambient air temperature tend to be classified as conventional oil. The pattern of oil discoveries referred to includes conventional (on shore and off shore), but excludes deep water and polar oil, oil sands and oil shale.

Oil sands, also known as tar sands, can be found in Canada, Russia, Venezuela, the US and a few other countries. The endowment of oil sands is vast (approximately 5 trillion barrels) and much larger than the endowment in conventional oil (approximately 2 trillion barrels). However, for technical and geological reasons, only a fraction of it is deemed usable. Presently, only Canada and Venezuela produce crude oil from tar sands with a production of about 2 million barrels per day.

The vast majority of world oil production, 85% actually, comes from conventional oil. The remaining 13% to 14% comes from deep sea regions off Angola, the Gulf of Mexico and Brazil for instance or from Polar regions in Russia whilst oil sands contribute perhaps 3% of world oil production. What this distribution means is that World oil production is dominated and will continue to be dominated by conventional oil dynamics in the foreseeable future. A vital fact to be borne in mind.

The Typical Production Profile of a Conventional Oil Field

Once an oil field is discovered, production can commence. It is trivial, but important, to stress that production commences at zero. Very quickly, production can be increased as new wells are drilled and oil springs forth under the natural pressure of the oil reservoir. As the natural pressure begins to decline, secondary methods of extraction are used whereby either water, gas or chemicals are injected into the field to prop up the pressure or increase oil fluidity thereby maintaining high production flows. A number of constraints, technical, economic and political amongst others, mean that oil production from a given oil field will not be allowed to increase to maximum capacity but production will be capped so as to achieve a relatively long plateau. This production plateau can maintain itself for a number of years. In the absence of additional discoveries a decline period will begin whereby gradually the oil production of a given field decreases year by year. The decline rate can be of a few percent per year or more sometimes. Once this decline has set in, it cannot be reversed on the long term, though it might be arrested for a few years with better technology. The tail end production profile can be far extended in time, for instance there are oil wells in Pennsylvania (USA) that are still in production after a hundred years of operation but are now producing only a few barrels of oil per day.

This typical production profile can be seen in a number of large fields, such as Prudhoe Bay in Alaska, Samotlor in Russia, The Brent and Forties field in the North Sea, just to name a few. The point to bear is that production profiles for different oil fields will not all be exactly the same. But they will have common important characteristics, such as production begins at zero, rises quickly to a plateau and then declines gradually till abandonment of the field at some future point in time. It follows that the area under the curve of the production profile is the total oil extracted in the time period considered and that will be less to or equal to the amount discovered in a given oil field.

The United States’ Oil Production Profile

The US has a very long history of oil extraction and still is a very important producer of oil. Its 2000 production was 7.7 million bbd, compared to Saudi Arabia’s 8.4 million bbd. World production for 2000 was 77 million bbd. What makes the US oil production so important to study is not only its magnitude and very long oil history but also that it is in decline, and has been so since the early seventies when its oil production peaked.

For discussion purposes, US oil production can be segregated into crude oil production from the so-called Lower 48 States, from Alaska and form the Gulf of Mexico. Considering crude oil from the Lower 48 States, it is seen that production in 1949 was 5 million bbd. It gradually increased to reach a maximum of 9.3 million bbd in 1970 and then gradually declined to reach 4.8 million bbd in 2002 (EIA), figure 3. It is interesting to note that the peak in oil production came about 40 years after the peak in oil discoveries for the lower 48 states. Indeed it is in the early thirties that oil discoveries reached a historic maximum with the discovery of the giant East Texas Field. Ever since the thirties, yearly oil discoveries in the lower 48 states declined steadily and never rose again, in spite of obvious major advances in the geological sciences and related technologies. The reason for this drop in discoveries is simply that the Continental US has been thoroughly explored by scores of geologists for decades and that the best and large finds have already been made. Furthermore, the large production volumes from Alaska and deep-water offshore projects from the Gulf of Mexico were not enough to offset declines elsewhere. Indeed, total US production of crude oil (inclusive of production from the Lower 48 States, Alaska and Gulf of Mexico) was 6.3 million bbd in 1954 and reached a maximum of 9.6 million bbd in 1970 and had fallen to 5.8 million bbd by 2002 (EIA). One should note that US production is declining in spite of the use of the best available technology world-wide by highly experienced oil companies. Furthermore the US is the largest oil consumer in the world, it has the necessary infrastructure already in place and the US market can afford to buy all the oil it needs. If under the above near ideal conditions, the US cannot reverse decline once established, then it is safe to conclude that elsewhere in the world, declining production is most probably irreversible once it has taken hold. This is a very important conclusion.

There is another aspect of US oil production worth pointing out. Oil production in the US has largely been unconstrained, meaning that through out the 20th century, oil production was never interrupted by wars, sabotage, civil unrest, revolution or embargoes. Oil being produced as per demand which never really slacked. Hence it shows what the typical unconstrained conventional oil production profile of a large producer can be.


Countries with declining oil production

We have seen how the US oil production has increased, peaked then declined over the years. The US is far from being the only country having done so. An analysis BP’s statistical review of world energy 2004 (which covers oil production from 1965 to 2003), it can be seen that in 2003, 21 countries had oil production below their respective historical maximum oil production. In many cases, the decline in production from their respective historical maximum is very large as shown below.

Table 1


Historic Maximum Production (thousand of bbl)

Year of Maximum Production

2003 Production

% decline





























































United Kingdom













































Source: BP






In the seventies, only four countries (Venezuela, USA, Romania, Indonesia) had peaked, two in the eighties (Tunisia, Peru), nine in the nineties (Egypt, India, Gabon, Cameroon, Argentina, Colombia, United Kingdom, Congo, Uzbekistan) and two (Australia, Oman) in the period 2000-2003. It is too early to be sure for Norway, Yemen, Angola and Denmark, they might still be able to reverse the downward trend as they might not. The main point here is that the rate at which countries are peaking in terms of oil production has definitively accelerated compared to the seventies and eighties. As yet there is no indication that those countries where oil production is in decline will be able to reverse the trend. The case of continual decline in US production since the seventies does indicate that once the decline is well established (a decline of more that 15 to 20% for instance), it is irreversible.

A few countries to watch

In addition to the above four countries, (Norway, Yemen, Angola and Denmark) that might peak soon, it is important to bear in mind that another very important oil producer, China, might also peak quite soon. The reason being that whilst its consumption increased by 20% over the last four years, domestic production has increased by a paltry 4.4% as shown below. It does indicate that domestic Chinese oil production cannot increase fast enough to satisfy its domestic consumption. Indeed the rate of growth of Chinese oil production has slowed down considerably since the seventies. It is a strong indication that a maximum in oil production might be reached in the next few years.

Table 2

China, Oil Production and Consumption (thousand of bbls per day)


Oil Production

Oil Consumption













Source: BP


The Meaning of Reserves

Reserves are definitively the most controversial aspect of the discussion for several reasons. The first reason is that there is no universally accepted definition of reserves, nearly everybody having their own version. Hence reserves are not necessarily a scientific measure of the amount of oil or gas that can be extracted from already discovered oil fields. Secondly, most reserve estimates are unverifiable as they tend to be state secrets. Thirdly, reported reserves in open publications tend to increase with time in spite of substantial production. For instance, in 1985 reserves were reported to be 700 Gb and by 2000 this had jumped to 1100 Gb, (source: Annual Statistical Review 2002 BP). Hence considerable caution must be exercised in analysing the meaning of reserves.


Figure 4


The vast majority of world oil production tends to be carried out by either State owned companies or by Western companies which are quoted on the New York Stock Exchange for instance. The latter companies have a legal obligation to report to investors what is termed "proved reserves". Under this term, it is understood that only those reserves that can be profitably exploited given the current level of technology and price structure can be termed as "proved" and hence booked as assets by the company. Other reserves must be classified as either "probable" or "possible". Western companies such as Shell, British Petroleum, ExxonMobil and so on have therefore reported "proven reserves". However, State companies, which are obviously not quoted under any stock exchanges, are under no obligation to follow any regulation, and therefore State companies report whatever level of reserves they want to or are told to by their political masters. Hence, world oil reserves are a combination of "proved reserves" and reserves reported by individual countries.

This combination is in itself troublesome given the uncertain nature of reserves reported by countries who control their own national oil production. For example, during the eighties OPEC countries raised their reserves by 80% altogether and their reserve level remained static ever since. A very unusual situation, to say the least, given the substantial oil production since and the absence or low level of new discoveries in OPEC countries. Hence it is highly possible that the reported reserves of OPEC countries might have been overestimated by some unknown margin. Western companies can also be prone to overestimating their reserves. For instance, in January 2004 Shell downgraded 23% of their reserves from the "proven" category to "probable". By June 2004, British Petroleum also had downgraded 1.2% of their reserves from "proven" to "probable". NordskHydro, a State Norwegian company did the same to some of its reserves.

"Proved reserves", as reported by Western companies, can also be misleading for over the years proved reserves have grown in spite of large volumes of production. Why is that? It appears that whenever a new field is discovered and subsequently assessed by geologists, they make a range of estimates of the recoverable amounts of oil called the P90, P50 and P10 estimates. P90 refers to the 90% probability that so much of the resource will be recovered. P50 and P10 refer to the 50% and 10% probability that so much of the resource will be recovered, respectively. Obviously, the lower the probability, the larger the volume of oil thought to be recoverable.

Once discovered, exploitation can begin and it will be the P90 estimate that initially will be used to assess the profitability of the venture and it will be the P90 estimate that will be booked as "proved reserve". As production becomes on going, the initial P90 estimate is revised upwards and slowly drifts to the P50 estimate of recoverable oil. This revised estimate, a legitimate engineering exercise, gives rise to the phenomenon of "reserve growth" which gives the impression that the more oil is extracted, the more is left to extract.

The main sources of reserve growth are due to upward revisions of the amount recoverable as mentioned earlier or due to improved recovery techniques. The presence of shallower or deeper satellite pools but overall smaller pools next to the main one will also cause reserves to grow. It is clear that "reserve growth" is not a physical phenomenon at all, it is an artefact that arises due to a better understanding of the physics of the reservoir. The amount of the resource initially there is fixed and unchanging, it is called the "original oil in place". This essential statistic is rarely published, if ever. Out of the "original oil in place" volume, only a fraction will be extracted that varies from 20% to over 80%. This fraction is really a function of the viscosity of the oil, the porosity and permeability of the reservoir rock, which is the rock that yields the oil. The fraction that can be recovered is called the "Estimate of Ultimately Recoverable Resource" or EUR for short. It appears that the P50 estimate and the EUR are very close to each other.

Considering that whether we are dealing with the P90 or P50 or the EUR, all of these numbers are estimates and nothing more. It necessarily follows that the term "proved reserve" is an oxymoron. A reserve is above all a statistical estimate subject to revisions and errors. Hence, it is to be expected that these estimates will change with time. This must be borne in mind. It is also important to realise that if the P90 estimate of field A is a 100 barrels and the P90 estimate of field B is 150 barrels, the P90 estimate of field A and B is NOT 250 barrels. P90 estimates are not simply additive. In fact given that the P90 estimate means that there is a 90% probability of extracting so much barrels from a given field, the probability that 100 barrels AND a 150 barrels will be extracted from field A AND field B is

P(A and B): 0.9 x 0.9 = 0.81

The more fields are added to the total the more the probabilities must be multiplied throughout. Very quickly the probability of extracting the cumulative totals of a large number of fields falls to very low values and converges to zero, a very useless exercise. Hence whenever figures of "proven reserves" are mentioned it is good to remind oneself that this figure may mean very little, it is not a scientific measure of what is left in the ground to be extracted. The only way out is to use the EUR or the P50 estimates that tend to be statistically neutral. This means that overall the probability that estimates go up or down is the same for all fields and these estimates already account for potential large reserve growth during the life time of oil fields.


The EUR (Estimate of Ultimately Recoverable Resource) of Conventional Oil for the World

As we have argued earlier, the EUR or the P50 estimates are most probably the safest numbers to rely on. For the world as a whole it is the EUR that is used. Throughout the years there has been many estimates for the recoverable resource. They have varied from barely 800 Gb to 4000 Gb. However, most estimates have averaged out around 2000 Gb of conventional oil. Indeed, this figure seems to be relatively uncontroversial. Cumulative production has been estimated to be around 920 Gb (ASPO newsletter July 2004). If the EUR of 2000 Gb is correct, then the remaining oil left to be produced would be 1080 Gb. A figure that is very close to the "proved reserves" figure of 1100 Gb as reported from the BP Statistical Review. This would imply that very little oil (comparatively) is left to be discovered world-wide, a conclusion that is borne out by the declining trend in new discoveries as shown in figure 1. Furthermore, it appears that cumulative discoveries as at 2002 was of the order of 1800 Gb (figure 1). The declining trend in discoveries (< 10Gb per year) does indicate that by 2020 cumulative discoveries will have converged to 2000 Gb (1800 + 10 x 20 years) by assuming that 10 Gb of new oil will be discovered yearly from 2000 to 2020. Hence, the inescapable conclusion is that in absence of very large discoveries, the world EUR of conventional oil will hover around 2000 Gb. This conclusion is of the greatest importance. Larger estimates of the EUR of the order of 3000 Gb or even 4000 Gb can only be taken as being excessively optimistic and not reflecting the reality of hard figures.


Mathematical Modeling of US and Conventional World Oil Production

In 1956, the chief research geologist of Shell in the US, King Hubbert, a very well known figure of the time, published several estimates of the EUR for the Lower 48 States. Using those estimates, he calculated that crude oil production from the Lower 48 States would peak in the late sixties or early seventies. Although ridiculed at the time, history proved that he was correct. Crude oil production did peak in 1970. Although he never made clear which type of curves he used to model future oil production in the US, the characteristics of the curves are that production starts at zero, rises exponentially to a maximum, then declines gradually till extinction into the far future. More importantly, the area under the curve is the cumulative production for the period and that should be equal to or be less than the EUR. Hubbert, subsequently tried to model future world oil production by using the same methods and obtained a world production that would peak in 1995. Fortunately for us he was a bit off the mark. Bartlett (2000), used the same approach as Hubbert but this time used a Gaussian curve to model conventional US and world oil production. What is very important in this paper is that Bartlett used a range of figures for the world EUR and for each figure he calculated the expected peak date and peak oil production. The results of this analysis are shown in Table 1.





Peak Date

Peak Production

2000 Gb


26.5 Gb

3000 Gb


33.0 Gb

4000 Gb


39.5 Gb

Table 1

From the above, it is obvious that if the EUR is only 2000 Gb, the peak production is imminent (2004). The analysis shows that a doubling of the EUR to 4000 Gb would only push the peak date by a mere 26 years! Furthermore, the mathematical analysis shows that to push the peak date by only one year, additional discoveries of 60 Gb of oil are needed. This must be contrasted with actual annual discoveries that amount to barely 10 Gb at best.

Roper (1976, 2002) outlined a depletion theory of non-renewable resources and applied it to US and world crude oil production. His model based upon the differential of a logistic equation shows that a maximum in world crude oil production of around 23 Gb should be reached during the first decade of the 21st century.


The World Oil Peak or Plateau and its Significance

If the above reasoning is correct, then a maximum in world oil production will be (or is being) reached within the first decade of the 21st century. Thereafter oil production will decline steadily never to rise again. At this point it is important to understand that the mathematical models assume unconstrained production. In our world this is not necessarily the case, there are production cappings, embargoes, wars, strikes and acts of sabotage that contrive to constrain world oil production. Hence it is probable that instead of a peak in production, there will be a plateau during which production hovers around a given level for a number of years before dropping steadily. In effect the production curve is flattened a bit. It is not possible to predict accurately when the production plateau will begin and where it will end marking the steady fall in production. The best that can be said is that a plateau in production will be reached during the period 2000-2010, or even 2015 at the latest if major finds of new oil are made, but that thereafter a decline in production is inevitable.

It is often argued that should there be a fall in the production of conventional oil production, unconventional oil resources will make up the short fall. Hence it is expected that production from deep-sea oil, tar sands or very heavy oil will increase dramatically in due time so that the world will make an easy transition from one form of oil to the other (David Deming, 2000). Alas, things may not be that straight forward. For instance, although deep sea oil is thought to be a vast and barely tapped resource, over the last 20 years, barely 30 Gb of deep sea oil has been found (Werner Zittel, Jörg Schindler, January 2003) and currently deep sea oil production is about 5 mbd. Unless, truly gigantic discoveries are made, deep-sea oil is unlikely to make much difference. Tar sands and very heavy oil are also put forward as substitutes to conventional oil. Although the resource is very large, oil extraction from tar sands is very energy intensive (with an energy profit ratio of 2:1), highly polluting and production still very low, about 2 mbd. Furthermore, it appears that production cannot be increased rapidly. For example, Canadian tar sands oil production is about 1 million barrels per day, the Canadian oil industry itself estimates that they will be able to double production, at best, only by 2020. Furthermore, the production of crude oil from tar sands consumes vast amount of natural gas, and that Canadian natural gas production has been flat over the last few years or so. Any increase in crude oil production from tar sands will require much more natural gas that may not be there. This will be a major constraint for any future production of non-conventional oil from Canada, a country that already produces half of the non-conventional crude oil world-wide. Venezuela is the other country that produces large amounts of non-conventional oil, its current production is 1 million barrels per day and there are no plans to increase that production in the near future. In view of the above, it is reasonable to conclude that non-conventional oil sources will become more important in the future. However, it is doubtful whether they will be able to make up for the drop in conventional oil production, and even less to provide growth in the total supply of hydrocarbons so dearly needed for economic growth.

In view of the considerable importance oil has become for modern economies, the peak or plateau in oil production will be a pivotal event for the 21st century. Once supply will repeatedly fail to satisfy demand, prices will rise and become unstable. The economic impact of this event will be profound and long lasting. Just as cheap and abundant oil transformed the 20th century, the lack of this essential resource will transform the 21st century.

To Sum Up

Oil is a finite, non-renewable resource that must be discovered before production can begin. Conventional Oil discoveries world-wide peaked in 1962 and have declined ever since. The planet has been thoroughly surveyed since and large finds are unlikely. The EUR for conventional oil for the planet is reckoned to be around 2000 Gb out of which 900 Gb has been produced. If this figure for the EUR is correct, mathematical modelling implies that a peak in conventional oil production is imminent (2004). To delay the peak by one year requires the discovery of 60 Gb of oil when the current annual discovery rate is less than 10 Gb. Non-conventional sources of oil may soften or even delay momentarily the onset of decline in world oil production. However it is unlikely that such sources can make up any short fall in conventional oil production. Once supply fails to meet demand repeatedly, prices will rise and become unstable given the over reliance of modern economies on oil. This will be a pivotal event of the 21st century. Just as cheap and abundant oil transformed the 20th century, the lack of it will profoundly transform the 21st century.



BP statistical review of world energy June 2004,

Bartlett Albert A., An analysis of US and World Oil Production Patterns Using Hubbert-Style Curves, Mathematical Geology, Vol 32, No 1, 2000.

C.J.Campbell, Peak Oil: an Outlook on Crude Oil Depletion - Revised February 2002,

David Deming, Oil: are we running out of oil? Second Wallace E. Pratt Memorial conference, "Petroleum Provinces of the 21st century", January 12-15, 2000, San Diego, California

Department of Energy, US government, , date of access 16th January 2004.

Michael C. Lynch, The new Pessimism about Petroleum Resources: Debunking the Hubbert Model (and Hubbert Modelers), 2003

Roper L. David, Crude Oil Depletion, 2003,, date of access 12th April 2004

Roper L. David, Depletion Theory, Department of Physics, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24060-0435, 1976

Werner Zittel, Jörg Schindler, Future World Oil Supply, L-B-Systemtechnik GmbH, Daimlerstrasse 15, D-85521 Ottobrunn, January 2003


Web sites: (Official Site of the Association of Peak Oil Study, ASPO) (Many papers on the subject matter) (Many papers on energy and sustainability) (Daily news articles on energy) (Good introduction on oil depletion) (Web site of Simmons & Co, Energy Bankers, Very good free presentations on energy issues) (Latest prices on crude oil) (very good site on energy issues) (Many interviews on oil depletion) (Oil industry news) (Energy Information Adminstration US Government, excellent source of reliable oil statistics)



Date on the web: 31st of July 2004

Last Update: 5th of February 2005

URL of this document: