One who asks a question is a fool for five minutes; one who does not ask a question remains a fool forever.
     —Chinese Proverb

Google news searches often turn up editorials, opinion pieces and analysis written by people who don’t buy into this “peak oil” theory. Most of these writers don’t understand what “peak oil” means.

Preparing for ASPO-USA’s Houston conference held last October, the author wrote an Introduction to Peak Oil (Word document, click to open) that was included in the press packet for the gathering. Those new to the issue may want to read this background material. Let’s analyze a recent editorial, Peak No Evil from the Financial Times, to identity some of the confusions it contains.

The slim hope is that if antagonistic commentators actually understood what they were talking about, it might help focus the debate about the liquid fuels problem facing large oil consumers now and in the future.

Some Background Material

Before delving into details, it is necessary to clear-up a meta-misconception about those concerned about peak oil. The Financial Times assumes that the “peak oil” community is monolithic, that so-called peakists all speak with one voice. As in any diverse, loose group of researchers, opinions vary widely within the world-wide peak oil community. There is one thing they all do (or should) agree about and here it is: the world appears to be running out of the ability to increase oil flows. The world is not running out of oil in the ground.

Specifically, “oil flows” refers to the global volumetric flow rate = ν, which is conventionally measured in barrels per day. Peak oil is max(v), taken together with the assumption that the maximum output may vary within a small range (± a few percent) for some years before monotonic declines set in. Taking oil as crude, condensate and natural gas liquids, the world produced 81.291 million barrels per day (= ν) in September, 2007. The average daily flow for 2006 was 81.334 million barrels per day (= μ). In this example, which is made for illustrative purposes only, ν < μ. This supply data is from the U.S. Department of Energy (DOE EIA).

No one disputes that oil production will peak and decline some day. Peak oil is the hypothesis that max(v) will occur between now and 2012, to pick a date. Some analysts believe the peak for crude + condensate already occurred in 2005. Time will tell if this crucial observation, which is borne out by the EIA’s data, is correct.

Most peak oil antagonists are confused about the difference between reserves and flows.1 There is no straightforward mathematical relationship between the two, so endless arguments put forward to refute the peak oil hypothesis concerning the size of the world’s remaining oil reserves do not bring clarity to the debate. Conflicting reserve size claims often serve those who want to muddy the waters further. Confusion reigns.

Whereas the size of the reserves may be very large, the size of the volumetric flow rate may be relatively small. The Canadian tar sands presents a classic example (ASPO-USA, January 2, 2008). The recoverable reserves are estimated to be 175 billion barrels, but the current flow rate is only 1.157 million barrels per day. Russia’s remaining recoverable reserves are probably much less than 175 billion barrels, but they produce almost 9.9 million barrels per day. Examples abound. The world’s oil supply appears to be getting heavier over time as measured by its API gravity. Highly viscous oil does not flow at rates as high as those achieved for light, sweet crude. Miscible carbon dioxide gas injection used in enhanced oil recovery (EOR) increases the reservoir recovery factor over time but can not restore an oil field’s peak flow rates. That’s why it’s sometimes not economic to use this tertiary recovery technique.

The existence of recoverable reserves are thus a necessary, not sufficient, condition for estimating volumetric flows rates. When a peak oil researcher hears about discoveries like Brazil’s Tupi field, the only real questions of interest for the analyst are what will the oil flows be? What is the time frame for achieving those flows?

Confusion in the Media

Here’s the Financial Times2.

Production and Reserves Oil resources are finite. “Peak oil” theorists posit that about half of all the world’s crude has been used and that output will soon peak prior to an irreversible decline. Such thinking has helped propel crude to the $100 per barrel level it touched yesterday. Conventional oil fields are like champagne bottles: once “opened”, pressure forces out some of the contents. Eventually field pressure drops and, barring using such techniques as re-injecting gas, output inevitably declines. Back in the 1950s, Marion King Hubbert, a US geoscientist, correctly forecast – to within a few years – when output in the US’s lower 48 states would peak (it was 1970). The “Hubbert curve” is a totem of peak oil theorists…

One difficulty is poor data. Modeling the mature US oil sector – with its huge sample size today of over 500,000 working wells and more inactive ones – is relatively easy. In contrast, Saudi Arabia has only 2,000 producing wells and large unexplored areas…

Even if the world’s total amount of oil can be established – estimates vary wildly – better technology means the proportion that can be pumped out increases over time. Since 1980, this has risen, on average, from a fifth to more than a third, boosting recoverable reserves. In spite of rising consumption, the ratio of oil reserves to output has been pretty constant since the late 1980s. Today’s high oil prices also make complex sources, such as oil sands, viable and damp consumption.

The Times says “peak oil” theorists posit that about half the world’s crude has been used… No, they don’t. The relationship of “peak oil” to recoverable reserves has been explained above. The “Hubbert curve” is a totem of peak oil Haida_totemtheorists… No, it isn’t. For the record, a totem (example pictured left) is “an animal, plant, or natural object serving among certain tribal or traditional peoples as the emblem of a clan or family and sometimes revered as its founder, ancestor, or guardian, or a representation of such an object.” By contrast, a “Hubbert curve” (shown below left for Norway) is the mathematical model of oil production over time shown in equation (1) from Early new field production estimation could assist in quantifying supply trends by Rafael Sandrea (Oil & Gas Journal, Volume 104  Issue 20  May 22, 2006).

(1) dQ/dt = ro Q (1 – Q/K), where Q is the cumulative oil production, dQ/dt is the annual production rate, ro the initial production growth rate, and K the reserves.

Equation (1) defines the logistic function that yields a “Hubbert” curve. As Sandrea notes, Norway Hubbert Curve“the production potential or peak production of an oil field occurs at the half-life (Q = 1⁄2K) of its recoverable reserves.” The peak oil theory does not state that conventional oil production will peak and decline when exactly half the assumed global endowment has been used up. That notion assumes that we know with some certainty what the world’s recoverable reserve volumes actually are, and that the producing countries will extract their oil in an unconstrained way. Some peak oil analysts use the logistic function in their analyses, others do not. The peak oil community is not monolithic.

The Times says that in spite of rising consumption, the ratio of oil reserves to output has been pretty constant since the late 1980s… Has it? Since the Times insists that we talk about the size of the remaining proved reserves, let’s venture down that perilous road.

In their graph above3, the Financial Times get its reserves data from BP’s Statistical Review. Where does British Petroleum get its data? They get their OPEC reserves numbers from the OPEC countries. Are OPEC reserves numbers reliable? According to the Statistical Review, Saudi Arabia’s proved reserves were 261.4 billion barrels in 1996, 264.2 billion barrels in 2005, and 264.3 billion barrels in 2006. You will notice immediately that Saudi Arabia’s proved reserves never go down. Again, how do we know what the Kingdom’s reserves are? Saudi Arabia tells us what they are. No one may audit Saudi Aramco’s reserves claims.

You be the judge. Isn’t it at least a little bit curious that produced oil is never subtracted from Saudi Arabia’s reserves tally? Isn’t it a little bit suspicious that additions alway match or exceed produced oil?  In 2006, Saudi Arabia produced an average of 10.579 million barrels per day (includes natural gas liquids) according to the U.S. Department of Energy. That works out to 3.86 billion barrels over the entire year. Assuming an approximate average of 3.5 billion barrels a year since 1996, the rough tally for all oil produced by the Kingdom equals 30.5 billion barrels over the entire decade.

If the rough calculation made here is correct, the Saudis have added 33.4 billion barrels to their proved reserves since 1996. No matter how much oil the Kingdom produces, there’s always more oil left to produce! This conclusion is hard to believe. The Financial Times says that one difficulty is poor data. It is hard to disagree.

The Times states better technology means the proportion that can be pumped out increases over time. It has already been explained that there is no direct mathematical relationship between an oil field’s recovery factor—the percentage of the oil-in-place that can be recovered, i.e. the size of the recoverable reserves—and the volumetric flow rates achieved over time. Peak oil is concerned with flow rates, not recovery rates. Rafael Sandrea’s Oil & Gas Journal paper (link op. cit.) attempts to use apply the logistic function to establish a link between recoverable reserves sizes and flows. (See Anatomy of an Oil Discovery, ASPO-USA, May 16, 2007, for more details.)

Let’s examine the logic of the Financial Times’ argument. They first assert that “peak oil” theorists posit that about half the world’s crude has been used. They then associate the totemistic, all-powerful logistic function, which assumes that peak production occurs when half the recoverable reserves are gone, with all peak oil researchers. Having established the bogus premise that all peak oil researchers claim to know precisely what the world’s recoverable reserves are, The Times then chastises these analysts, saying that the world’s total amount of oil can [not] be established – estimates vary wildly (a paraphrase). The Times then goes on to cite BP’s Statistical Review, which is apparently a sacred document, to tell us that those very same proved reserves whose size can not be known have remained more or less constant as historical production has increased. Finally, they assert that whatever the size of the world’s proved reserves—whose size has remained constant heretofore but whose true extent can not be known—turns out to be, these recoverable volumes can only go up anyway because better technology means the proportion that can be pumped out increases over time.

Whew! You would not believe how hard it was to write up the preceding paragraph—it is not as easy as it might appear to summarize a self-contradictory, incoherent text. It’s a pity that this gibberish has almost nothing to do with volumetric flow rates, i.e. peak oil, as discussed above. It’s more akin to the apocryphal Medieval scholastic argument about how many angels can dance on the head of a pin, a form of “ontological argument … which cannot be proven [because] it has no basis in scientific fact.”

The Financial Times sums up their not-quite-cogent exposition as follows—

Oil output is not just a function of geology. “Surface” factors such as Opec have a huge impact. Indeed, geopolitics and environmental concerns provide enough reasons to curb dependence on oil for transportation. If the noise generated by the peak oil debate adds to the sense of urgency in addressing this, it will serve some useful purpose.

Peak oil analysts can certainly agree that oil output is not just a function of geology. One only needs to look at Nigeria or Iraq to understand that. Geopolitics and other “surface” factors are definitely a concern. A subtler point the Times could have made would have wondered whether stated proved reserves estimates are also not just a function of geology. Consider Saudi Arabia, whose unaudited reserves calculations are swallowed hook, line and sinker by the undiscriminating and uninformed editors of the Financial Times. It appears that Saudi Aramco believes that it would be bad for business if they admit that they have actually produced some of their oil.

Editors at the Times could have made the effort to investigate peak oil, but they did not. Instead they chose to flail about, huffing and puffing about an issue they don’t understand. This convoluted analysis would have been made easier if the Financial Times had simply said we don’t want to take peak oil seriously, it makes us uncomfortable. Such a statement would have been honest and understandable, inviting sympathy, not ridicule. It is a shame that peak oil commentary in much of the mainstream media is so debased that the wide gulf between what is said and what is real appears insurmountable. Similar confusion preceded the acceptance of the reality of anthropogenic climate change. One can only hope that peak oil will receive the attention it deserves before it is too late.

Contact the author at .


1. For example, Saudi Aramco vice president of Marketing and Supply Planning Adil A. Al-Tubayyeb recently told the Japanese that

… the abundant supply of conventional oil resources will provide sustainable energy for a long time. “The amount of conventional oil-in-place worldwide is estimated at between 6 and 8 trillion barrels,” he said. “The world does not have to worry about (peak oil) for many decades to come.”

6 to 8 trillion barrels! That sounds like a lot until you break down the numbers. Including 2007, the world has consumed about 1.1 trillion barrels of oil so far. Most estimates put the remaining proved reserves at about 1.2 trillion barrels—but see footnote 2 below. Al-Tubayyeb is talking about oil-in-place, not recoverable reserves. The worldwide recovery rate from all reservoirs is said to be about 33%. Using a far-too-simple rough calculation, add 1.1 to 1.2 and then multiply by 3. The answer is 6.9 trillion barrels, almost at the mid-point of Al-Tubayyeb’s range. He’s not Saudi Aramco’s vice president of marketing for nothing!

2. The text contains the sentence since 1980, this has risen, on average, from a fifth to more than a third, boosting recoverable reserves. It is not possible to source this claim. Footnote 1 alludes to a 33% recovery factor (rate) in the world’s oil fields. It is not clear how that number is derived either, but it appears in various publications. This is a complex issue. For example, EOR will boost an oil field’s recovery factor over time even as production flows from the field continue to decline.

3. Note also that the Financial Times‘ graph includes the suspicious reserves jump that took place in OPEC nations during the late 1980’s. Read The Little Sheikdom That Could (ASPO-USA, June 6, 2007) to better understand the reserves issues.