What is Peak oil?
"The term Peak Oil refers to the maximum rate of the production of oil in any area under consideration, recognising that it is a finite natural resource, subject to depletion."
An analysis of World Energy Outlook 2012 as preparation for an interview with Science
Submitted by Kjell Aleklett on Thu, 2012-11-29 12:26.
On 12 November the IEA’s World Energy Outlook report for 2012 (WEO-2012) was presented by the chief economist of that organisation, Dr Fatih Birol. When he did so there was one idea that the journalists in the audience latched on to – that by 2020 the USA would become the world’s largest oil producer. The USA would even produce more oil than Saudi Arabia! The New York Times led with the news, “U.S. to Be World’s Top Oil Producer in 5 Years, Report Says” and in their article they wrote that, “The United States will overtake Saudi Arabia as the world’s leading oil producer by about 2017 and will become a net oil exporter by 2030, the International Energy Agency said Monday.” Of course, I wondered if these were Fatih Birol’s actual words.
On 23 November Dr Birol was invited to Stockholm by Sweden’s Minister for Energy Anna-Karin Hatt to present WEO-2012. Representatives from the diplomatic corps in Stockholm, from industry, from academia and others were invited to attend the presentation. An invitation even arrived in my mailbox. Dr Birol gave a very polished presentation, the main message of which was that, in future, we would use our energy resources as efficiently as possible. The presentation was given at KTH, the Royal Institute of Technology, and is now accessible via their “Video Library”. To do so, go to http://www.kth.se/ece/stabsfunktioner/kth-medieproduktion/showroom/kth-live-1.318500 and select “Live Show [livestream] Fri”. The presentation is 1 hour, 24 minutes long and begins with 10 minutes where there is no talking. Dr Birol’s presentation begins at the 18th minute. Those magical words on the USA’s future can be heard at 31 minutes into the video:
”As the result of a new technology developed there [the USA] we have seen a growth coming from unconventional oil replacing the decline in the conventional oil and we expect as a result of this development, the United States in five years of time, will overtake Saudi Arabia as the largest oil producer of the world, a development that was difficult to imagine a few years ago.”
That the USA would become self-sufficient and able to export oil in the future was not stated. Instead, Dr Birol gave the USA’s import needs as approximately 30% of consumption in 2035.
After Dr Birol’s presentation there was time for three questions and the third was put by me with a question on the USA’s future oil production. I tried to explain my reason for asking my question but was not allowed to finish. (My question comes at 1 hour and 20 minutes into the video.) Dr Birol began his answer by saying, “One of the things that you say is completely right, that we know each other since long time, that is completely right. Since long time you kindly say that we are going to run out of oil, we are not”.
This was a rather clever rhetorical trick to make it sound as though everything else I was saying was incorrect. To assert that I have said that oil will run out shows once again that he does not understand the concept of Peak Oil. Peak Oil means that we cannot increase production. During the coming 50 years the production of oil will fall by half but we will not run out of oil.
But let us return to our discussion of the USA’s future oil production. It was not only the New York Times that wrote about the future’s new world order of oil production. The news spread rapidly around the entire world. On 16 November the leading scientific journal Science decided it would cover this story of the USA’s new energy future and booked an interview with Dr Birol for 19 November. They wanted to balance Dr Birol’s perspective with another viewpoint and so (I am happy to say) they contacted me (Kjell Aleklett). I received an email from Richard Kerr, who has been a journalist with Science for 35 years and we discussed the interview that he was to have on Monday, one week after the release of WEO-2012. He would first ring Dr Birol and then me. I had not yet had time to read WEO-2012 so I had a challenging weekend ahead. I focussed on WEO-2012’s discussion of oil since Dr Kerr wrote to me that, “The possible subjects are varied, but they would all pertain to how WEO 2012 has world oil production rising to 2035”. Below, I will discuss Chapter 3 in WEO-2012 from those notes that I prepared before the interview with Richard Kerr and also from the presentation made by Dr Birol in Stockholm.
First, let us conclude the discussion of the USA’s future oil production. In London and Stockholm, Dr Birol showed the graph below, which is a simplified version of a more complicated graph in WEO-2012:
We see that conventional oil production continues to decline but that unconventional oil production, estimated at around 2 million barrel of oil equivalents per day (Mboe/d) in 2010, will continue to increase. Together they give an estimated total oil production of over 10 Mboe/d in 2020. It is then that the USA will reportedly be producing more oil than Saudi Arabia. Dr Birol mentions nothing about 2035.
A summary of the scenario calculations for Saudi Arabia given in various WEO reports:
World Energy Outlook ....... 2004 2006 2008 2010 2012
Something dramatic must have happened with the IEA’s views on Saudi Arabia. Since 2004, Saudi Arabia has been the saviour of the IEA’s WEO-scenarios and has enabled these to project continuing growth in oil production. Eight years ago, when I began to criticise the WEO projections, WEO-2004 predicted oil production in Saudi Arabia of 22.5 Mb/d in 2025. Now WEO-2012 paints a completely different picture. In 2011 Saudi oil production was 11.1 Mb/d and the IEA now predicts this will decline to 10.6 Mb/d by 2020 before growing to 10.8 Mb/d in 2025! Then production will continue to grow to reach 12.3 Mb/d by 2035. In the criticism that I advanced in 2004 I said that 22.5 Mb/d for Saudi Arabia in 2025 was completely unrealistic. The IEA’s current prediction of 10.8 Mb/d in 2025 shows that I was correct. In the graph that Dr Birol presented on the USA’s future oil production (see above and note that it gives production quantities in barrels of oil equivalent, not simple volumetric barrels) it is difficult to determine the exact production but with help of a computer program that can determine quantities from graphical figures I obtained a value of 10.4 Mboe/d in 2025 which is not more than they give for Saudi Arabia in the same year (10.8 Mboe/d). In 2035 Saudi Arabia is definitely the world’s largest oil producer with 12.3 Mb/d of production.Now, referring to WEO-2012 and Figure 3.18 of that report (below) we get a different picture of past and future oil production in the USA. Here, oil production is 11 Mb/d in 2020 and thus greater than in Saudi Arabia (predicted by the IEA to be 10.6 Mb/d in 2020). Note that the slow decline in (conventional) crude oil production shown requires that new oilfields are found that can provide 40% of crude oil production in 2035. (The production is now given as Mb/d and not as Mboe/d as in the other figure. 1 barrel of NGL is 0.7 barrel of oil equivalents, this and the exclusion of “other unconventional oil” could explain the difference in numbers between the two graphs.)
Natural Gas Liquids (NGL) is another important component that is reported as part of the USA’s oil production in WEO-2012: “In addition to light tight oil, shale gas plays contribute to the rise in liquid hydrocarbons production in the United States, with the output of NGLs projected to grow by 1 mb/d between now and 2020, reaching 3.2 mb/d, before slowly decreasing to 2.3 mb/d in 2035, as the known liquids-rich gas plays are exhausted and gas production refocuses on dry gas.”
NGL is produced as a byproduct of natural gas. Historically NGL constitutes about 15% of the production from a natural gas field. If the production of natural gas in the USA is lower than predicted the NGL production will be lower than reported.
Note that the USA’s crude oil production continues to decline and new discoveries of conventional oil are needed to support today’s declining trend to 3035. Around 50% of the USA’s crude oil production in 2035 is to come from fields yet-to-be found. New regions must be opened up for exploration.
The USA’s Energy Information Administration (EIA) has also made predictions of future US oil production. In its report, “Annual Energy Outlook 2012 with Projections to 2035” released in June 2012 the EIA also included biofuels as part of oil production with volumes of 1.0 Mb/d in 2010, 1.1 Mb/d in 2020 and 1.7 Mb/d in 2035. These should be subtracted if the EIA figures for oil production are to be compared with those of the IEA and, if we do so, we see that the EIA predicts oil production at 10.0 Mb/d for 2020 (compared with the 11 Mb/d of the IEA). If we also subtract “xTL” (second generation biodiesel) and coal-to-liquids (CTL) production from the EIA’s 2035 production figure we get 10.2 Mb/d in that year (compared with 9 Mb/d from the IEA). Thus we can conclude that the IEA believes that shale oil will attain a higher peak of production than predicted by the EIA but that this production will also decline faster than predicted by the EIA. The ultimate result is that the EIA and IEA predict similar cumulative volumes of shale oil produced by 2035.
In Figure 112 above from the EIA report we see the crude oil production predicted to be contributed by various regions in the USA. We see that production in Alaska is declining to a trickle after its heyday in the 1980s – a typical example of Peak Oil. The fairly constant level of future production shown for “Lower 48 offshore” requires substantial new discoveries to be realised. For the “Lower 48 offshore” it is worthwhile citing a comment by the EIA, “Growth in lower 48 onshore crude oil production comes primarily from the continued development of tight oil resources, mostly from the Bakken and Eagle Ford formations. Tight oil production surpasses 1.3 million barrels per day in 2027 and then declines to about 1.2 million barrels per day in 2035 as “sweet spots” are depleted”. It is also remarkable that the IEA has a figure for “light tight” oil production that is twice as large as that given by the EIA.
In the quote above it is interesting that the EIA mentions “sweet spots” and points out that oil production from the sweet spot area of the Bakken shale in Montana reached a maximal production of 300,000 barrels per day in the third quarter of 2006 (see the figure below).
Our conclusion is that the pronouncement that the USA will be a greater oil producer than Saudi Arabia can be regarded as a clever sales trick to generate interest in the WEO-2012 report.
According to the e-mail from Science I should focus on issues that make it possible for me to comment on the possibility of increasing oil production to the levels discussed in World Energy Outlook 2012. I now would like to point to the note “Spotlight” on page 38. In that we can read about “the biggest source of uncertainty for energy prospects”. In my seminars, (normally around 50 per year), I frequently say that we have entered an era when physics is more important than economics. I’m a professor in physics (and global energy systems) and it is the laws of physics that are limiting to oil production in the future. It made me smile when the IEA cited a Nobel Prize laureate in physics, Niels Bohr, as they started to discuss uncertainties. I think it is important to read the whole note:
<em>As Niels Bohr said, “it is tough to make predictions, especially about the future”. In common with all attempts to describe future market trends, the projections in WEO-2012 are subject to a wide range of uncertainties. Indeed, it is unlikely that the future will follow any of the precise paths described in our scenarios. But that is not the aim of the WEO: none of the scenarios is a forecast. Each is intended to demonstrate how markets could evolve under certain conditions. How close those scenarios are to reality not only hinges on how well the model represents the way energy markets work and the validity of the assumptions that underpin that model, but also on the occurrence of “game changer” events in the economy at large. Many past forecasts and outlooks have been confounded by events such as the 1997-1998 Asian financial crisis, the US subprime mortgage crisis and subsequent recession and the more recent “Arab Spring”. Key drivers of energy markets are hard to predict, in part because they interact with each other. Yet some are inherently easier to predict than others, especially in the near term. Population, even by region, is unlikely to deviate much from the assumptions used for this Outlook. And we can be reasonably confident about how technology is likely to evolve in the short to medium term, even if there are surprises, such as the improved technologies that have recently unlocked huge unconventional gas and oil resources in the United States and elsewhere.
Economic prospects in the near term are much more uncertain and, given the strong link between economic growth and energy demand in most countries, are perhaps the biggest source of uncertainty in the medium term – even more so than usual, given the fragile state of the global economy. But most forecasts of economic growth at the world and regional levels over the long term fall within a relatively narrow range, even if there may be significant divergence between countries.
History has shown that energy prices are notoriously difficult to predict. But even if prices follow a very different path than what we have assumed, the overall impact on demand may not be very pronounced: demand for some forms of energy is less sensitive to price changes than demand for many other goods. Nevertheless, the dramatic fivefold increase in the oil price that we have witnessed over the last decade could still have an unexpectedly large impact on long-term demand, especially if (as our assumptions suggest) these price levels persist.
In the very near term, even government policy is reasonably predictable, as changes tend to be incremental: short of a breakthrough agreement on climate change or a major geopolitical event, it is unlikely that any policy moves in the next twelve months would have a large impact on key global energy trends in the medium term. But in the longer term, policy making is the area in which the greatest uncertainty exists, particularly when it comes to issues such as the extent to which action is taken to mitigate climate change, developments in energy subsidies, decisions on nuclear power, and the pricing and production strategies of the major oil and gas exporters.</em>
When reading articles about the findings in WEO-2012, these scenarios, although not forecasts, are treated as being the truth. We can conclude that it is unlikely that the future will follow any of the precise paths described in the IEA’s scenarios. The IEA also understands that the <em>“validity of the assumptions”</em> is important for the accuracy of its scenarios. However, when the IEA discusses uncertainties none of them are related to the physical conditions of the resources to be produced. We at Uppsala University have specifically studied these physical conditions and we can point on a number of important factors.
At a recent industry conference on reserves and resources representatives of the oil industry pointed out that they like to discuss oil production according to the figure below. To understand future oil production it is important to follow this scheme. When I make comments about WEO-2012 I will refer back to this figure as Figure 1.
When an oil field is discovered the resources in the field are evaluated. Oil companies listed on the New York Stock exchange must follow strict rules when it comes to evaluating reserves and in principal they are only allowed to report developed resources as reserves. Other oil companies can follow other rules. Reported increases in reserves are just reclassification of resources and this is also what WEO-2012 tells us. The critical sentence from WEO 2012 is the following: <em>“In 2011, 12 billion barrels were discovered, equal to 40% of the oil produced during the year.”</em> Since we are using 30 billion barrels per year we see as a fact that the available resources are shrinking.
In Figure 3.12 the IEA accepts the reserve numbers that the Oil & Gas Journal collects every year. These numbers have been questioned by many. The next doubtful analysis that is reported in WEO-2012 is the R/P ratio. <em>“The global reserves-to-production ratio, which is sometimes used as an indicator of future production potential, has increased steadily in recent years, to around 55 years at end-2011.”</em> The signal to the reader is that you don’t need to worry for the next 50 years. Everyone in the oil industry knows that the reported reserves according to the Oil & Gas Journal can never be produced during the next 50 years. All oilfields in the world show increasing production to a maximum before they begin to decline. The most dramatic decline in production that any oilfield has shown is that seen for the super-giant Cantarell. Once Cantarell was the oilfield with the world’s second-highest rate of production but its production has been declining extremely rapidly in recent years; <em>“it is currently producing only about 400 kb/d, down from a peak above 2.1 mb/d in 2003-2004”. </em> A statistical analysis made at Uppsala University shows that, generally, the peak production of giant oilfields is reached before 50% of their oil is produced. My recommendation is that IEA stops reporting the very unscientific R/P ratio number.
The world oil demands that the different scenarios generate are supported by a supply scenario. This scenario is regarded by many as a supply forecast. The supply that the WEO-2012 generates as a scenario output is reported in Figure 3.15. It is worth repeating the IEA’s statement that: <em>“none of the scenarios is a forecast. Each is intended to demonstrate how markets could evolve under certain conditions”</em>
The majority of the oil we consume is crude oil and this statement from WEO-2012 should be noted by all: <em>“Crude oil output from those fields that were in production in 2011 falls by close to two-thirds, to only 26 mb/d by 2035 (Figure 3.15). Thus, the projected production of 65 mb/d in 2035 requires almost 40 mb/d of new capacity to be added over the projection period. Of this capacity, 26 mb/d, or 66%, comes from discovered fields yet to be developed, most of which are in OPEC countries, and the remaining 13 mb/d from fields that have yet to be found, mainly in non-OPEC countries.” </em>
In WEO-2010 the production from fields yet to be found was set to 19 Mb/d in 2030 and in our publication “The Peak of the Oil Age” we showed that this was not attainable with the discovery volumes predicted by the IEA. Our reanalysis of the IEA data produced a number of 9 Mb/d. Adding 5 more years beyond 2030 should increase that number and the fact that the IEA now has a lower number than in WEO-2010 shows that their calculations are becoming more realistic in that respect. The rate of production decline stated by the IEA for the fields that were in production in 2011 is also what we have reported in a publication but the rate of oil production from fields yet to be developed is still too high, the IEA’s 26 Mb/d should probably be lower by 10 Mb/d. In 2004 when I for the first time said that the WEO-2004 was wrong they were stating that crude oil production in 2030 would be 108 Mb/d. The IEA has lowered the scenario output for 2030 by 42 Mb/d down to 66 Mb/d. That is a change of 40% in 8 years!
The scenario results for unconventional oil, the lower right quadrant of Figure 1, are as follows: “Output of oil sands in Canada grows rapidly, from 1.6 mb/d in 2011 to 4.3 mb/d in 2035, on the assumption that public concerns about the environmental impact of their development can be addressed. Venezuela is the other contributor to increased production of extraheavy oil, with production from the Orinoco Belt expected to reach 2.1 mb/d in 2035 – a rise of 1.5 mb/d.” We will discuss this later.
In the upper left quadrant of Figure 1 we have the new shale oil production in the USA and the scenario numbers for supply from the USA in the future have, as said before, dominated the media picture of WEO-2012. Some more words from WEO 2012:
“The resurgence of oil production in North America in the last few years is projected to continue. Production in the United States hit a low of 6.9 mb/d in 2008 and has since been rising strongly, reaching 8.1 mb/d in 2011. We project this upward trend to continue, with production climbing to a plateau of 11 mb/d before 2020 and then declining gradually from the late 2020s to around 9 mb/d by 2035 (Figure 3.18). This is a significant upward revision from last year’s Outlook. Light tight oil continues to rise into the 2020s (Box 3.3), supplemented by deepwater output and by supplies of unconventional CTL, GTL and kerogen oil, particularly towards the end of the projection period.”Box 3.3
“The rise in light tight oil production in the United States in the last few years has been nothing short of spectacular. Production from the Bakken formation had grown to more than 600 kb/d by mid-2012 in North Dakota alone (the formation also extends into the state of Montana in the United States and provinces of Manitoba and Saskatchewan in Canada). Production from the Eagle Ford shale in south Texas, adjacent to the Mexican border, is also expanding rapidly, from almost nothing three years ago to more than 300 kb/d by mid-2012. Combined US production from the Bakken, the Eagle Ford and other emerging plays, such as the Niobrara (which straddles Colorado, Kansas, Nebraska and Wyoming), the Texas Permian basin and the Californian plays, is expected to reach over 3.2 mb/d by 2025 in the New Policies Scenario. Resources are believed to be large enough to sustain production at that level until about 2030, before a decline in production sets in.”
The whole of Chapter 3 in WEO-2012 is about oil and the chapter begins with “highlights” . We should now look at these highlights in more detail:
“The projections for each of the scenarios are derived from the IEA’s World Energy Model (WEM) – a large-scale partial equilibrium model that is designed to replicate how energy markets function over the medium to long term. The WEM comprises six modules, covering final demand (by sector and sub-sector), power and heat generation, refining and other transformation activities, supply (by primary energy source), carbon-dioxide (CO2) emissions and investment in energy supply infrastructure.”
In the WEM documentation the IEA team (comprising over 30 people) gives the general structure of the model for the scenarios. We can see that the economy provides “Drivers” and that it is the output from the “economic analysis” that generates the “demand for useful energy”. (This is important to remember when we later discuss unconventional oil production.) The next step is “Technology/fuel allocation” and by applying efficiency levels the model delivers the final demand.
Figure 2.8 is a very interesting figure showing the global energy system in 2010 in terms of Mtoe (1 toe can be 7.11, 7.33, or 7.4 boe, barrel of oil equivalents, depending on type of oil. I will use 7.3 boe). From Figure 2.8 we have that 4113 Mtoe from oil went into the system in 2010 (that is 30.0 Gb or 82.2 Mb/d). [As a comparison, BP reports oil production for 2010 to be 82.48 Mb/d (excluding liquid fuels from other sources such as biomass and coal derivatives). For 2011 BP gives the number 83.6 Mb/d.] This leads us to a very important question and that is what it is that should actually be reported as oil production. From Figure 2.8 it is logical that “processing gains” is not part of oil production, but normally the IEA and US Energy Information Administration (EIA) do report processing gains as oil production. For a realistic discussion about future oil production we should limit ourselves to the number without processing gains. From Figure 2.8 we can also calculate that 4% of the produced oil is consumed in oil production and that 6.7% is directly used for electricity and heat production. As an example around 50% of the electricity produced in Saudi Arabia comes from oil. That leaves us with 76.4 Mb/d of oil as possible input to industry including transport. BP reports 75.2 Mb/d as “Oil - Refinery throughputs” and since that oil ends up in refineries that leaves us with 1.2 Mb/d of NGL as directly used as transport fuel or used in industry. The conclusion to be drawn from the above discussion is that the numbers for the starting point, year 2010, are well understood and are very reasonable.
The demand net growth (including processing gains) is 12.3 Mb/d (99.7 Mb/d in 2035 minus 87.4 Mb/d in 2011) and as all of this growth comes from the transport sector it requires an increase of “refinery throughputs”. From the 2011 level of 75.6 Mb/d this is an increase of 16%. If this increase is real we should expect increased investments in the refinery industry. There is today nothing to indicate this is happening or is likely to happen. Another important factor is that the IEA tells us that crude oil production will not increase which means that these hypothetical new refineries must handle unconventional oil. It would be interesting to see some proposed real projects. For this to happen, the industry must first believe in the new demand.
It appears that the IEA finally understands that processing gains are not the same as oil production. Once again the IEA accepts that we have had Peak Oil for crude oil, (or conventional oil), and they see a decline by 3 Mb/d to 65 Mb/d between 2011 and 2035. Note that the huge increase of oil production in Iraq is crude oil. Applying the statement “the increase coming entirely from natural gas liquids and unconventional sources” we get that the increase in unconventional oil must be (97-84-3 Mb/d =) 16 Mb/d. The light, tight oil production such as shale oil is part of this. The statement “light tight oil production grows above 4 Mb/d in the 2020s, mainly from the United States and Canada”, tells us that today’s production of about 1 Mb/d will increase by 3 Mb/d to 4 Mb/d, but in Box 3 they state 3.2 Mb/d.
First I would like to list the 12 member states of OPEC; Algeria, Angola, Ecuador, Iran, Iraq, Kuwait, Libya, Nigeria, Qatar, Saudi Arabia, United Arab Emirates. As the deep water production in Angola and Nigeria is crude oil we see that Venezuela is the only OPEC member that will produce unconventional oil during the period 2011 – 2035. The call on OPEC for increased oil production is 12 Mb/d in total and it is realistic to expect than only 2 Mb/d of that can be provided by expanded production from Venezuela (WEO-2012 tells us 1.5 Mb/d). The remaining 10 Mb/d, must be increased crude oil (conventional oil) production by the OECD nations. In the WEO-2012 text it is stated that oil production from OPEC will increase from 36 Mb/d to 46 Mb/d which is consistent if one is discussing crude (conventional) oil. This is a 28% increase in crude oil production. Total world crude oil production is currently 68 Mb/d and will decline by 3 Mb/d to 65 Mb/d by 2035. If OPEC is to produce 46 Mb/d then the volume required from the rest of the world (non-OPEC) will be 19 Mb/d. This is a reduction by 13 Mb/d (32 – 19 Mb/b) or 40%. Since OPEC’s giant fields were put into production, there have been more newly-discovered fields put in production by non-OPEC nations than by the OPEC nations. To expect that OPEC’s giant fields will not be affected by large declines in production during the coming 25 years is completely unrealistic. (For example see Cantarell in Mexico as discussed above.) That we will see declining production outside OPEC and increasing production inside OPEC during the next 25 years is not logical. Once again we see the IEA telling a fairy tale about OPEC that lacks any foundation in reality.
And now the last point:
Earlier I mentioned that a large increase in refinery capacity was required to meet the demand for transport fuels. That would be very expensive to put in place. Regarding the required level of investment stated above, I am not able to judge whether it would be sufficient.
Figure 3.19 can be used as background for some final remarks.
Very few oil producing nations show an increase in crude oil production in the period to 2035. Instead most nations show a decline. The future production in three nations, Iraq, Brazil and Kazakhstan will be very important. If this production increase, about 10 Mb/d, does not eventuate the future for the world economy can be very difficult. The IEA sees the largest future decline in crude oil production as occurring for Russia and that is also what I have described in my book “Peeking at Peak Oil”. In a collaborative publication with China Petroleum University in Beijing we have also described the decline in production in China.
To summarise my analysis of the IEA’s WEO-2012 report, I find it very hard to understand how IEA can smile and say “don’t worry, be happy”.