Peak Watch

Arctic Dreams

By Dave Cohen

Back in October of 2005, the New York Times announced that the Arctic will be the next big oil & gas province to be exploited by humankind in As Polar Ice Turns to Water, Dreams of Treasure Abound.  The Times gave this news a mixed review, and deservedly so. Exploration & production in the Arctic represents an unusual climate change positive feedback loop in which humans participate. With respect to the northern polar regions, the longer term process works like this —

1. Humans burn fossil fuels to release the buried solar energy stored therein
2. Atmospheric carbon dioxide levels rise and the natural "greenhouse effect" is amplified
3. The oceans warm and ice melts
4. Melting ice opens up new prospective oil & gas provinces for exploration & production
5. Humans burn more fossil fuels ... and back to (2)

Such a chain of events illustrates why the atmospheric chemist Paul Crutzen, who won the Nobel Prize for his work on the ozone layer, introduced the term Anthropocene, a "Geology of Mankind" in which the natural world is evermore dominated by human activities. Will the fossil fuels in the Arctic be developed over time? The answer is almost certainly 'yes', so there are two pertinent questions of interest — 1) What resources does the Arctic contain? and 2) Does this affect the peaking of world oil production?

The size of the resources in the Arctic is unknown. However, the answer is becoming better delineated due to the high stakes involved. An often quoted (Houston Chronicle) estimate from the USGS Assessment 2000 report places 25% of the world's remaining oil in the Arctic. It is not clear whether that 25% is in addition to, or, is part of, some estimate for the world's ultimately recoverable conventional oil reserves. However, a quick look at a summary of the study reveals that the USGS made no "official" estimate whatsoever for resources in the Arctic as a whole, although some polar regions, such as the petroleum systems in the Timan-Pechora basin and the North & South Barents Sea basins (both Former Soviet Union) were evaluated. All cited USGS numbers for the entire Arctic refer to undiscovered, or "yet-to-find," oil.

A Wood Mackenzie/Fugro Robertson study (press release) from November 1, 2006, provides some additional information about the size of Arctic resources.

The U.S. can no longer consider the Arctic as a long-term strategic energy supply source, according to a new joint study by Wood Mackenzie and Fugro Robertson, "Future of the Arctic." The study found the Arctic potential is significantly less than previous estimations had suggested, and the mix of resources have been found to contain much less oil and more gas.

"These findings are disappointing from a world oil resource base perspective," said lead study author, Andrew Latham, Vice President, Energy Consulting at Wood Mackenzie.

Unlike the U.S. geological survey's wild-ass guess (WAG), the new assessment included a "detailed geoscience analysis of individual basins" which was independently verified by discovery, drilling and production data from operators working in the Arctic. "Our approach mirrors what those in the industry do when creating their reserve estimates," Latham said. "It's not nearly as optimistic." Surely, this can be called a step forward. The Arctic basins assessed, which include areas adjacent to, or offshore in, the Barents, Kara, Laptov, Chukchi, Beaufort and East Siberian seas, are shown in the graphic (left, click to enlarge).

Here is a summary of the main Wood Mackenzie/Fugro Robertson conclusions about the resource volumes —

• The Arctic is gas prone. The study estimates that "85 percent of the discovered resource and 74 percent of the exploration potential" is natural gas. Of the gas resources, 69% belongs to Russia. The rich get richer.

• In the North American Arctic and Greenland, reserves were determined to be only about 25% of the volumes found in previous studies. Alaska's North Slope, including the ANWR 1002 area, was estimated to contain approximately 6 billion barrels (oil equivalent) in undiscovered resources. Previously, the USGS assessment had estimated that total resources could amount to approximately 50 billion barrels on the North Slope. Greenland's northeast coast may have great potential, with resources exceeding 10 billion barrels.

• The Kara and Barents Sea basins are the most prospective. Russia's South Kara Yamal basin is thought to have nearly 90 billion barrels of oil in undiscovered reserves, while the East Barents Sea may contain more than 10 billion barrels.

Research on Arctic resources continues.The AAPG hosted a special breakout session on Arctic Basins and Hydrocarbon Systems at their recent April annual meeting. Wood Mackenzie's Latham gave a talk, as did the USGS's Donald Gautier, a principal investigator of petroleum systems in the Arctic. Gautier presented a new map of sedimentary successions, a key step toward ranking various polar basins. There is still a lot of uncertainty about resource volumes, as Gautier's abstract indicates.

The Arctic has globally significant upside petroleum potential, coupled with high geologic, technologic, and economic uncertainty. While a few basins, such as the North Slope of Alaska, Timan-Pechora, and the West Siberian Basin, are known to contain large volumes of hydrocarbons, most of the Arctic remains unexplored, untested, and unevaluated.

As for the second question — the relationship between the Arctic and peak oil — significant oil production from undeveloped Arctic resources of uncertain size is many years away. The Wood Mackenzie/Fugro Robertson study forecast that production flows from all the northern polar regions will eventually peak at 3 million b/d at least two decades from now. Natural gas flows may reach 5 million barrels of oil equivalent in the same timeframe.

Enormous aboveground hurdles must be overcome to produce oil in the Arctic. Another Wood Mackenzie press release, Exploration in the Arctic has a Long Term Future, issued prior to the the Oil & Gas of the Arctic Shelf Conference in Murmansk, held in November, 2006, describes some of the difficulties in developing the prospective Russian basins.

... a key challenge to maximizing the value of the Russian offshore will be deploying development solutions which are currently at the cutting edge of technology, and are currently untested in Russian waters.

"Subsea technology, which can operate underneath pack ice, is the key to the economic development of much of the gas in the Arctic, explained [Alan] Murray [Senior Petroleum Economist at Wood Mackenzie]. The limits of this technology are still being tested, and as these solutions are deployed more often, the costs of this should reduce, he said."

Transportation of gas from such remote parts of the globe presents a great challenge to the energy industry. The Yamal peninsula is 2,000 km from Moscow and 4,000 km from Western Europe, however, the existence of future pipeline capacity in Western Siberia is key to the economic development of gas resources in the Russian Arctic. This contrasts sharply with similarly remote parts of Canada and Greenland.

Given the ballpark estimate of future Arctic oil flows, the technological and logistical challenges, and the timeframe, it is safe to say that E&P in the Arctic has no bearing whatsoever on the question of a near-term peak in world oil production. Even if the peak turned out to be a few years later than many anticipate — for example, in 2020, following the low case scenario suggested by Richard Nehring at the AAPG April meeting — oil production from the Arctic could not possibly affect the outcome.

In light of these considerations, it is baffling that the USGS's Gautier made this statement in Richard Kerr's The looming oil crisis could arrive uncomfortably soon (Science, 20 April 2007, Vol. 316. no. 5823, p. 351) —

Nehring is getting some attention but not many converts. "Richard did a good service in holding this Hedberg Conference," says oil assessment specialist Donald Gautier of the U.S. Geological Survey in Menlo Park, California. But there's so much uncertainty, Gautier says, from when Arctic ice might melt out of the way to when new technology can be developed, that predicting the peak may not be worthwhile. A decade or so could tell.

At this moment, there is probably far less uncertainty about sea ice declines, if current trends continue, than there is about the development of Arctic oil & gas resources. Gautier seems to imply that the world is simply waiting for the "ice [to] melt out the way", a view that strangely understates the problems attending the production of fossil fuels in the Arctic, and does not support his intended point that predicting the peak of world oil production may not be a worthwhile pursuit.

Is it appropriate for a representative of the USGS to suggest that organizations like ASPO-USA are wasting their time worrying about peak oil, just when the GAO is issuing reports on the subject? There is no convincing evidence that the production of oil & gas resources in the Arctic basins is germane to a peak in world oil production in any timeframe one might care about. The peak flow rate for the entire region is estimated at 3 million b/d, a bit less than 60% of what Ghawar, the world's largest oil field, has produced on a daily basis for many decades. The rate of warming in the high northern latitudes and uncertainty about the size of the resource base in these polar areas serve only as distractions that may prevent the public from arriving at an honest assessment of the future oil suppy.

Deconstructing the New York Times

by Dave Cohen

Robert McChesney gives a reasonable definition of the two indispensable functions that journalism should provide in self-governing societies. The first is the "watchdog" role over those in power. The second, of more interest here, is explained this way —

... the media system must (1) provide reliable information and (2) a wide range of informed opinions on the important social and political issues of the day. No single medium can or should be expected to provide all of this; but the media system as a whole should provide easy access to this for all citizens [numbering, italics added]

There is little doubt that the current and future state of the world's oil production is "an important social and political issue." Given the warning signs of a looming crisis and the large associated uncertainties, it is hard to think of an issue more critical to industrial economies than maintaining a reliable oil supply in the coming decade and beyond.

In its recent coverage of the world's oil supply, the New York Times has failed to meet either of McChesney's criteria for covering this paramount social issue. Although no one media outlet is responsible for examining all aspects of an issue, as McChesney points out, the NY Times considers itself the "paper of record" and, therefore, it should be held to a higher standard than other media sources. The Times' failure to provide balanced, accurate coverage is apparent in recent articles by energy reporter Jad Mouawad. A close reading of his work reveals that the Times consistently covers the oil supply from the point of those who promote a narrowly optimistic view of it. Examples include selected representatives of the International Oil Companies (IOCs), Nansen Saleri of Saudi Aramco, and Cambridge Energy Research Associates (CERA).

The most egregious example is Oil Innovations Pump New Life Into Old Wells, published on March 5, 2007. Mouawad's article purports to demonstrate that new oil technologies are raising reservoir recovery factors. The clear implication is that the application of such technologies makes it unnecessary to worry about the future oil supply —

Within the last decade, technology advances have made it possible to unlock more oil from old fields, and, at the same time, higher oil prices have made it economical for companies to go after reserves that are harder to reach. With plenty of oil still left in familiar locations, forecasts that the world’s reserves are drying out have given way to predictions that more oil can be found than ever before.

The technology advances in question consist of steam injection for the recovery of viscous heavy oil at the Kern River field in California and CO₂ injection for stranded recovery at the Means field in Texas. Neither of these methods is new. Gas injection — whether it is H₂O or CO₂ — has been around for decades. As the Times' own graphic (below, left) shows, steam injection began at Kern River in the mid-1960's. The Times' graphic also clearly shows that a production rate of only 140 thousand b/d was achieved at Kern River after using this technique for almost 20 years there. This mainstay U.S. Lower 48 field is now producing only about 90 thousand b/d. Contrast this with another large field, Dalia in the deepwater offshore Angola, where a field estimated to hold 1 billion recoverable barrels of light oil will yield a peak flow of about 240 thousand b/d. The Kern River field is past its prime and is now in irreversible decline. Similar remarks apply to Indonesia's Duri heavy oil field, also mentioned by the Times, where steam injection began in 1984.

Using McChesney's criteria, is the Times' reporting reliable? The story's slant is how technology increases the recovery factors. While it is often true that enhanced oil recovery (EOR) techniques allows operators to pump more of the original oil in place, such success depends on a number of factors. For example, using carbon dioxide injection to recover stranded oil depends on the suitability of the reservoir for miscible CO₂ flooding, the availability of the gas and the oil price, which must be high to make such EOR economic. Finally, even where carbon dioxide is successfully applied, recovery rates are slow and thus production must take place over a long period of time. For many of these reasons, the Norwegian Petroleum Directorate (NPD) rejected using CO2 flooding on Norway's continental shelf.

What about steam injection for recovering heavy oil at the Kern River? An underlying concern, unstated by the Times, is that much of the new oil being produced globally is becoming heavier over time, as in offshore Brazil. The Times neglected to mention that low API gravity oil (between 10° and 22.3°) is harder and more costly to produce. Heavier crudes have less commercial value because they yield fewer high value liquids when they are refined, as even Time Magazine understands. Consequently, the nature of the oil supply is changing — and not for the better. 

An uninformed reader of the Times article could easily come away with the misleading message that there is no reason to worry about the oil supply. Mouawad glosses over the complex issues involved in increasing reservoir recovery factors. Indeed, the entire science of petroleum engineering is reduced to a reassuring lesson that "all is well."

What about McChesney's criteria for presenting a wide range of opinions? Mouawad's Oil Innovations article provides the usual quote from CERA chairman Daniel Yergin, who reminds us yet again that “it’s the fifth time to my count that we’ve gone through a period when it seemed the end of oil was near and people were talking about the exhaustion of resources.” Yergin's sound bite echoes the article's theme that as prices rise, technology and the magic of the markets will make enough oil available to meet future demand. By contrast, Mouawad provides "equal time" to the peak oil hypothesis.

There is still a minority view, held largely by a small band of retired petroleum geologists and some members of Congress, that oil production has peaked, but the theory has been fading...

“I am very, very seriously worried about the future we are facing,” said Kjell Aleklett, the president of the Association for the Study of Peak Oil and Gas. “It is clear that oil is in limited supplies.”

... Many oil executives say that these so-called peak-oil theorists fail to take into account the way that sophisticated technology, combined with higher prices that make searches for new oil more affordable, are opening up opportunities to develop supplies.

Leaving aside the question of whether the peak oil hypothesis is fading, Mouawad falsely characterizes the "so-called" theorists studying pending oil supply problems — a diverse group of people worldwide, many of whom are veterans of the oil & gas industry. So, it is reasonable to ask who he is getting his information from. In the past, CERA has also narrowly identified "peakists" as a fringe group of geologists. Moreover, Aleklett's remarks are not presented in any context in which they might make sense to the reader. Mouawad sets up Aleklett as an easy target to be knocked down by Yergin, Saleri and other selected experts.

Information omitted by the Times might have provided a partial context for Aleklett's remarks. Digging a bit deeper, Mouawad fails to note that real declines are happening in many large supplier nations and there are only 15 countries left that have the ability to increase production in the future, as Yergin told the US House of Representatives Committee on Foreign Affairs on March 22, 2007. Obviously, there are worrisome geopolitical concerns associated with many of them, including Iraq, Iran, Venezuela and Nigeria.

Another article by Mouawad, Nowadays, Angola Is Oil’s Topic A, published on March 20, 2007, raises an obvious question — why is Angola, the best success story the IOCs have to offer lately, topic A? The Times quotes George L. Kirkland, a senior executive at Chevron, who comes right to the point. Kirkland is referring to the advantages of importing oil from a geographically diverse set of countries.

It’s a good story for consumers. The greater the diversity, the less your risk — not just political, but also technical and commercial. [emphasis added] "

One proper function of journalism regarding important social issues is to alert the public to situations  that may affect them adversely. Responsible reporting does not engage in cheerleading. Perhaps it would be more newsworthy if Nigeria, another West African producer, were Topic A. According to EIA data, oil output there declined from an average of 2.627 million b/d in 2005 to 2.440 million b/d in the last year due to rebel activity by MEND in the Niger delta, despite new production from large offshore fields such as Bonga. As it turns out, the Times' Mouawad did cover Nigeria in Shell Expects Full Output From Nigeria. Using the single IOC source quoted below, the article takes an optimistic tone —

“We see this year as a year when we start clawing back our production,” Basil E. Omiyi, Shell’s managing director for Nigeria, said in an interview on Tuesday in Abuja, the Nigerian capital. “I don’t see anything standing in our way of restarting our production.”

Nigeria's oil production is still down and remains precarious, with pre-election violence clouding future output. Unlike Angola, however, Nigeria's woes would not be a "good story for consumers."

To be fair, the Times' Elisabeth Malkin took on Mexico's falling production in Output falling in oil-rich Mexico, and politics gets the blame (reprint from the International Herald Tribune). As the title indicates, geologically determined declines at Cantarell set the stage but are not the main story. The Mexican government, which takes much of Pemex's revenue and thus prevents Pemex from making sufficient investments in new oil exploration and production, gets the blame. While there is a lot of truth to Malkin's story, the failure of nitrogen gas injection to prop up Cantarell's production any longer is not even mentioned in the article. No one at the Times seems to be able to connect the dots  — in Mouawad's Oil Innovations article, gas injection technology will save the day. However, such enhanced oil recovery eventually failed to sustain production in Mexico — Cantarell is in permanent decline now and output is falling fast.

The persistent pattern of selection bias in the Times' articles on the global oil supply makes its reporting on this subject suspect. Furthermore, there is direct bias in the use of loaded language to characterize peak oil issues and those studying them, e.g. it is a fading theory. Anonymous IOC sources state that so-called peak oil theorists do not take old oil recovery technology — which is presented as new — into account. This accusation is simply false. Neither of McChesney's criteria for competent, responsible reporting are met in recent New York Times articles on the future global oil supply. Their journalism in this area does a disservice to a public who they are supposed to inform.

Peak Oil -- The Power of Declines

This is the first in a series of weekly columns I will be writing on this website. I wish to thank the ASPO-USA Board of Directors for the opportunity to keep the public informed on peak oil and other related energy issues — Dave Cohen

The peak of the global oil supply refers to the uncontroversial observation that the world's incremental production of conventional oil, a finite resource, will reach a high water mark and then decline thereafter. What is more controversial is when the peak will occur. In its usual usage, the phrase "peak oil"  refers to the hypothesis of a near-term peak in the oil supply. Additional complications regarding what "conventional" oil is muddy the waters. Analysts such as CERA will tell you — for $1000 — that the peak will not occur until the 2035 to 2045 period, which, to most people, seems too far in the future to matter much. However, those studying the near-term hypothesis worry, and have good reason to believe, that the peak will occur anytime between now and 2015.

Since 1870, the health of the world's economies have hinged more and more on a secure, dependable and growing flow of conventional oil. The consequences of a near-term peak could irrevocably damage global economies unless mitigation steps are taken to smooth the transition to a world that must consume less oil with each passing year. Because petroleum provides the bulk of the world's liquid fuels, the immediate effects on transportation will be the most pronounced. Disconcertingly, the list of all the other products we consume that are made (in part) from oil is very long, as even those in school grades 3 to 7 can understand. Ailing, contracting economies would have harmful effects on people's lives. So, the debate about the timing of the peak is not "academic" — much depends on a correct analysis of the future oil supply.

There has been a lot of rancor and obfuscation in the debate about a near-term peak in the oil supply, largely because so much depends on the outcome for both producers and consumers. A highly placed representative of Cambridge Energy Research Associates (CERA), who characterize the hypothesis as "a myth", has called the near-term peak notion garbage. Others in the popular press often give the impression that "peakists" are claiming that the global oil endowment is almost depleted. The media uncritically cite huge recoverable reserves numbers taken from CERA or ExxonMobil, amounting to trillions of barrels, to "prove" their point. However, you can not put oil reserves in your car's fuel tank until they have been produced, transported and refined.

In the near-term, what is running out is the world's ability to increase the oil supply to meet growing demand, not the oil itself. The peak oil hypothesis, properly understood, is about constraints limiting the conversion rate of oil reserves into useable flows. One important consequence is that cheap oil is a thing of the past.

Future columns will discuss constraints limiting future oil production in some detail. Here, it suffices to expand our understanding of why the near-term peak hypothesis is plausible. As it turns out, the biggest problem is that many optimists considering the problem know how to add, but have forgotten how to subtract. Look at the figure below, which is revision of Chris Skrebowski's Oil-a-Gator.

The graph describes the global oil supply over time. The alligator's jaw is defined by the projected demand on the top and declines in the current oil supply from existing reservoirs on the bottom. The middle line (the x-axis) is the breakeven point -- the increasing amount of oil that needs to be replaced year-on-year to compensate for these declines. The red line defines the amount of oil that is actually produced and consumed starting at time t0. As such, it represents additions or subtractions with respect to the breakeven point. If CERA is correct, the red line will rise steadily over time to both make up for ongoing declines and meet future demand — at least out to about 2035 or so. However, the graph reveals that increasing the oil supply to meet growing demand becomes more and more difficult over time, as the 'gator's gape gets wider and wider. At some point — t1 in the graph — this Herculean, ultimately unsustainable, task will become impossible and the world will no longer be able to breakeven. This is the time of peak oil.

So far, nothing has been said about the initial time (t0) and the peak date (t1). Assume that the starting point is 2006 and further consider that we've got some ability to look ahead at the future, as described in the MegaProjects list compiled by Chris Skrebowsky, editor of Petroleum Review. This database is a schedule of large (> 40 thousand barrels per day) oil projects coming onstream between 2007 and 2012.  Because there is a long lead period (several years) between the time an operator decides to go ahead with a project and the time the first oil is produced, we can be confident that almost all large developments out to 2012 are accounted for.

First, we must learn how to subtract i.e. establish the decline rate. Using a baseline of 2005, the Oil & Gas Journal (OGJ) data indicate that the world consumed 80.099 million barrels per day (b/d) that year, where "conventional" oil is defined as the sum of crude oil, condensates and natural gas liquids. The OGJ data further indicate that usage in 2006 was 80.594 million b/d, an increase of 0.495 million b/d. According to Skrebowski's data, 3.219 mb/d of new oil was produced during 2006. Given the smaller net increase, a straightforward calculation shows that the "visible" decline rate was 3.4% for existing oil production, i.e. this is the rate that could be seen in the raw production data. There are dangers in assuming such a rate going forward in time because historical data shows that "visible" declines may underestimate true declines based on new production capacity additions. Next week's column will discuss the subject in detail.

It is informative to consider a specific country, Brazil. The OGJ data tells us that oil production in some countries declined in 2006. Examples include Norway, Indonesia, the U.K. and Egypt. These countries are in what Skrebowski terms Type-3 Depletion  — they are past peak and so will only decline year-on-year from here on out. Others, like OPEC members Nigeria, Saudi Arabia and Venezuela, declined, but it is not clear whether their peak production has been reached. However, declines in existing production are also happening in oil suppliers that increased production in 2006, as in Brazil or Russia. In these cases, new oil production has not only offset declines, but has also been sufficient to provide net gains. Looking more closely at Brazil, Skrebowski's data indicates that for 2007 projects, some of which will reach full capacity in 2008, this large offshore producer is scheduled to add 0.610 mb/d to the world oil supply. However, according to Rigzone's  Brazil Petrobras Plans to Bring 2 New Platforms On Stream in May, PetroBras indicates that they expect a net production increase in 2007 of only 0.141 million b/d on average for the year. Why?

The explanation lies in the problem posed by ongoing declines. Much of the new production Brazil will put onstream will  compensate for declines in their existing offshore fields. Beyond that, new production will result in only modest net increases. The underlying problem is that declines in existing fields are often invisible to the oil & gas analyst, whereas there is usually ample information about new projects coming onstream.

Moreover, there are often delays in new production. For example, the Rigzone article indicates that —

Production at the P-50 rig which Petrobras took on stream last April [in 2006] now has reached 170,000 b/d, Eduardo Molinari, coordinator for strategy and portfolio management of exploration and production at Petrobras, said. The P-50 originally was supposed to reach its full 180,000 output capacity last October [in 2006]. But problems with drilling equipment had delayed the ramp-up schedule.

So, delays happen, even as declines over time continue. The Brazilian example illustrates the power of the 'Gator graph as we anticipate the future global oil supply. Reading the media accounts, one might be forgiven for thinking that world oil production is continually increasing at a rate fast enough to meet projected demand. However, those worried about the peak oil hypothesis have remembered an important elementary school lesson that the press does not report — knowing how to subtract, as well as add.

Next week, I will explore decline rates and the near-term future of the non-OPEC oil supply.

Decline Rates and Non-OPEC Supply

Every year, baseball starts up in the Spring and there are rosy forecasts for supply growth in the non-OPEC oil supply. 2007 is no exception. Wood Mackenzie's Non-OPEC Increases to Continue in 2007 announces the good news.

The strong upward momentum for non-OPEC supply, seen in the fourth quarter of 2006 [when Dalia in offshore Angola came onstream] is expected to be maintained in 2007. The rate of increase is likely to accelerate in the fourth quarter of the year, when production is expected to be 1.6 million b/d higher than in the fourth quarter of 2006.

Total non-OPEC oil/NGL production, including Angola, is forecast to average 50.2 million b/d in 2007 according to Wood Mackenzie, up 1.5 million b/d from 2006. Patrick Gibson, Principal Oil Supply Analyst for Wood Mackenzie said "Our analysis shows that there will be significant increases in the FSU states, North America and Africa.  The main areas that will experience decline are the North Sea and the Asia Pacific region.“

Last week's column discussed the power of declines in existing oil production using the Oil & Gas Journal (OGJ) world oil supply data along with Chris Skrebowski's new projects database, as layed out in New capacity fails to boost 2006 production -- delays or depletion? from Petroleum Review (henceforth, PR database). In accord with Wood Mackenzie's forecast for 2007, Skrebowsky's schedule, shown in the figure taken from PR, indicates that new non-OPEC production will increase in 2007, and each year until 2009. Thereafter, new oil production will decline steadily out to 2014.

The putative surge in new non-OPEC production in the 2007 -- 2009 period is crucial to the world oil supply. ExxonMobil's 2005 Edition of  The Outlook for Energy -- A View to 2030 forecast that the non-OPEC share of world supply will plateau after 2010. (snapshot shown left). The IEA's World Energy Outlook 2005 made a similar prediction. Any increases made between now and 2010 will most likely determine the high-water mark of non-OPEC oil production forever!

The apex of non-OPEC production may not be the global peak yet, but the importance of this event can not be overstated. From that point forward, the "Call on OPEC" must increase if world oil production is to show any net gains. Consequently, the security of the world's oil supply is in greater jeopardy should OPEC production falter for any reason. Given ongoing events in Nigeria, Iraq and Iran, along with declines in Venezuela and Indonesia, there are plenty of reasons to worry. As usual, Saudi Arabia is the wildcard. A future column will address the OPEC issues.

From the perspective of eternity, delays in new non-OPEC production may not be a bad thing, in so far as postponed production will be available in the plateau or mild decline period after 2010. However, a perception problem arises if Wood Mackenzie makes overly optimistic estimates of the near-term supply additions from the non-OPEC producers. Policy makers and the public are misled into thinking that there are no serious problems with the oil supply, an atititude which leads, in turn, to complacency in the face of a unique crisis in the history of industrial civilization which is only 3 years away. Even worse, declines in existing non-OPEC oil production imply that new supply delays move the peak forward in time and lower the peak rate of production, although it may not matter at this point — it is late in the game.

The decline rate in existing production underlies the forthcoming peak (or plateau) of non-OPEC oil production. Despite Wood Mackenzie's optimistic forecast of an average addition of 1.5 million b/d, with a surge in the 4th quarter, the Centre for Global Energy Studies (CGES)  disagrees. The executive summary of their report Non-OPEC Production: Rising Slow Output Growth — the title is a bit misleading — is worth quoting in its entirety, both for accuracy and clarity in identifying the issues.

The oil industry is finding it harder and harder to expand upstream capacity. Development costs are up sharply, essential equipment and skilled labour are in short supply and host governments want a bigger share of the proceeds. As a result, projects take longer to complete and output is growing more slowly than predicted. In 2006, non- OPEC oil production rose by about 450,000 bpd and although this was better than 2005 — when a myriad of unanticipated problems kept output virtually flat — it was still less than expected by most industry forecasters. Once again, project delays, adverse weather, equipment failure and oil field problems combined to trim nearly 1 mbpd off non-OPEC supply growth projections for 2006 — leaving the market much tighter than expected.

While there is no doubt that the industry is working flat out to expand its production capacity in response to high oil prices, the problem is that new projects must first compensate for output declines at existing fields before they can add to overall capacity. With underlying oil well productivity declining at an average rate of 5% worldwide, the industry needs to drill enough wells to replace more than 2 mbpd of ageing non-OPEC capacity every year just for production to stand still. Any delay in starting up a new project therefore widens the gap that needs to be filled before overall non-OPEC production can begin to rise.

In recent years, a ‘rational’ approach to forecasting based on planned capacity additions and observed output decline rates for mature producing areas over-estimated the actual growth in non-OPEC supplies by an average of more than 400,000 bpd. Using the same publicly available information about new projects and oil field performance, both the CGES and the IEA made errors of the same magnitude, suggesting that the real problem is not accuracy but uncertainty.

For 2007, a ‘rational’ forecast indicates that non-OPEC oil supplies (including Angola) are expected to rise by [1.07 million b/d]. Like last year, the biggest growth is expected in the FSU, closely followed by Africa. Significant increases are also expected in North America and Latin America, and modest gains in the Asia Pacific region. Production is expected to decline in Europe and the non-OPEC Middle East countries. The CGES believes, however, that this rational forecast should be cut by 400,000 bpd to reflect the impact of uncertainty on output growth, implying a likely gain of only 700,000 bpd in non- OPEC supplies this year.

The summary highlights worrisome uncertainty about declines (and delays), casting doubt on using a "rational" approach to estimating the 2007 non-OPEC supply addition. A concrete example of such over-estimation comes from the EIA, whose analyst Michael Cohen predicted that non-OPEC supply growth would be 0.900 million b/d in 2006. The table shows some early release numbers for non-OPEC production increases for 2006 (in million b/d) made in 2007.

Centre for Global Energy Studies	0.456
Energy Information Administration	0.345
Oil & Gas Journal	≅ 0.622
Average	0.474

The IEA has also acknowledged past over-estimates in a special section of the current Oil Market Report (March, 2007) called "Clarifying the Call on OPEC".  At long last, the IEA, "adjusting for a known trend in supply and the miscellaneous-to-balance" category, has explicity noted that non-OPEC supply additions have been later revised downward by an average of 0.350 million b/d each year. Current predictions are flagged with a warning to this effect.

CGES states that a rational forecasting approach is based on "planned capacity additions and observed output declines". Using the EIA production data and the PR database additions, let's do some number crunching to see what the rational approach predicts for new non-OPEC supply in 2007. Last week, a simple calculation based on the OGJ data revealed a "visible" global decline rate of 3.4%, with an overall additon of ≅ 0.495 million b/d (for crude + condensates + natural gas liquids).  Using 2005 data as the baseline and Skrebowski's gross addition of 3.219 million b/d in 2006, the EIA data, which currently shows that global production was down 0.053 million b/d in 2006, yields a decline rate of 4%. This result is 0.6 greater than the OGJ percentage. Clearly, the fairly large discrepancies in the data sources can not be accounted for. But, let's move on.

The PR database new capacity addition is 2.844 million b/d in 2007 from the non-OPEC side of ledger (including Angola, now an OPEC member). Again using the EIA data, it is possible to calculate the "visible" non-OPEC (as opposed to global) decline rate. Skrebowski indicates that 1.559 million b/d of new capacity was added to the non-OPEC supply in 2006. This works out to a rate of 2.56%. Applying this rate to the 2006 non-OPEC EIA production data, the rational approach yields a net gain of 1.662 million b/d in 2007 — this number is only about 10% higher than the Wood Mackenzie forecast of an additional 1.5 million b/d on average until the 4th quarter, when new supply climbs to 1.6 million b/d.

A reasonable working hypothesis postulates that what distinguishes the CGES forecast from Wood Mackenzie's more optimistic prediction is the choice of a decline rate, although there may also be differences in the planned new capacity numbers. The underlying problem is that the decline rate in existing oil production can not be measured directly. CGES makes the reasonable inference that past over-estimates using a rational approach require an increase in the assumed decline rate, which they set at 5%. Remarkably, if a 5% decline rate in 2007 is applied to PR database capacity addition of 2.844 million b/d, the rise in supply is predicted to be 0.694 million b/d. The tally is essentially the same as the CGES estimate of 0.700 million b/d.

This number crunching exercise serves to demonstrate how analysts like Wood Mackenzie make their projections. It is not possible to determine precisely how Wood Mackenzie worked out the numbers internally, although they have left us some clues.

We have identified seven projects, led by BP's Azeri-Chirag-Gunashli development in Azerbaijan, which will add an average of over 100,000 b/d each; all but one of which are already onstream. With the top 25 projects adding an aggregate 2.1 million b/d of capacity there is a broad base to the production growth expected.

Wood Mackenzie does take declines into account. Their press release indicates that they expect major declines in the North Sea and the Asia Pacific region. Using a decline rate just a little greater than the 2.56% number used here, it is easy to obtain their stated estimate of 1.5 million b/d if the planned capacity addition is close to the estimate given in Skrebowski's PR database. This latter number represents an addition of about 0.744 million b/d above the figure given in the Wood Mackenzie press release. However, if Wood Mackenzie is using something closer to the lower planned capacity addition number quoted above, then they are also using a non-OPEC decline rate much lower than that which historical data would tell us is realistic.

So, we can conclude that Wood Mackenzie's forecast is likely wrong on the high side. On the other hand, 2007 may be a stellar year for new non-OPEC production capacity additions, defying the historical trend.

The world is fast approaching a monumental historical shift. The impending peak of the non-OPEC oil supply circa 2010 illustrates the Red Queen Problem as posed in Lewis Carroll's Through the Looking Glass and illustrated by last week's Gator graph — one must run faster and faster merely to stay in place, let alone increase production. As CGES points out, and as we have shown here, there is no longer any historical basis for using a "rational" approach to estimate new capacity additions. Wood Mackenzie's rosy forecast does not take the full power of declines in existing production into account. In so doing, they have sent an unrealistically reassuring message to the policy-makers and the public that peak oil is not a clear and present danger to the health of the world's economies.