Peak Oil — Where Do We Stand?

On October 14, 2010, Aaron Task interviewed Chris Martenson at Yahoo's Tech Ticker (video below). Martenson, fresh off of attending ASPO-USA's annual peak oil conference, told Aaron that "conventional oil" peaked "around 2005." Aaron had no way to assess this statement, so he ran with it. It was then that I decided to write this post in a futile attempt to get everybody on the same page about future oil production.

IMHO, anybody with a vested interest in the world's future oil supply—that would be almost everybody on Earth—should read this post. Realistically, my expectations are somewhat lower

First things first. Conventional oil refers to crude oil plus lease condensate according to the Energy Information Agency (EIA). That's as good a definition as any, and I will use EIA data today.  All liquids refers to conventional oil taken together with natural gas liquids, refinery gains and biofuels, but today I will stick with conventional oil. In July, 2010 the world produced 73.691 million barrels per day (b/d) of conventional oil. The all-liquids total for that month was 86.474 million b/d. You can see that conventional oil by far makes up the largest share of the total liquids supply. When I refer to just "oil" in the text below, I mean conventional oil as defined by the EIA.

Chris' claim that conventional oil peaked around 2005 is false. If anything, conventional oil production peaked in July, 2008 or the first half of 2008 (see the graph and caption below).

EIA data for conventional oil production, from Table 1.1d (xcel spreadsheet). The average world production through July stands at 73.426 million barrels per day in 2010. Oil production averaged 73.944 million barrels per day in the first 6 months of 2008 before the demand crash. Without the crash, which was obviously predictable given the price, 2008 would have been the peak year so far for conventional oil production.

The queston thus becomes: did world conventional oil production peak in mid-2008 or are we looking at a local maxima (high point) in production that may be surpassed in the future? To answer this question, we must first understand what this production graph is telling us.

During the "demand shock" that caused oil to spike to well over $100 dollars per barrel in 2008, world oil production actually declined year-over-year in 2006 and 2007 even as the price was rising. To a lot of us, this looked like an actual 2005 peak in the world's oil production. Conventional wisdom says that as prices rise in an environment of unrelenting demand, supply should increase as a result of new investment in production capacity. Prices had been rising since 2003. Where was the new supply that would rebalance prices? It wasn't there!

By mid-2008, the "Great" recession was well underway—it began in the 4th quarter of 2007—and our economic woes were further accelerated by the surging oil price. The nominal price rose into the $140s in June and July, which is not a price for oil people can afford. As a result of the recession and the oil price, demand crashed. Thus, conventional oil production after July, 2008 reflects much diminished demand, not an inability to increase supply—OPEC slashed production dramatically. The world economy hit bottom in early 2009, and oil demand started to recover in mid-2009. You can see in the graph that oil demand is rising in 2010.

If one looks at conventional oil production on a yearly basis, 2005 is still the peak year in the EIA data. However, we can now see that concluding that oil's all-time production peak occurred "around" 2005 as Martenson said conflates supply with demand. I have found that this is a common mistake.

Let's get to the 64,000 dollar question: has conventional oil production peaked? The correct answer is most likely not. There are two unknowns preventing us from coming to any definitive conclusions about timing:

• We don't know how the global economy is going to fare in the future. It certainly does look shaky. (I won't cover this today.)
• We don't know exactly how much spare production capacity the world has to meet future demand if the world economy grows steadily in the next few years.

It is this second unknown that makes all the difference from a supply point of view. Oil production outside of OPEC (non-OPEC) has "peaked" in the sense that it has flat-lined. In fact, the EIA forecasts that total non-OPEC supply will fall by 240,000 b/d next year due to the inability of other non-OPEC producers to offset declines in Mexico and the North Sea. Generally speaking, we can depend on the assumption that non-OPEC production will never rise much—if at all—above its current level and is more likely to stay flat or decline as time goes on.

That is why the world looks to OPEC to solve its future oil supply needs. And here we run into a host of problems. The EIA estimates that OPEC's spare (unused) capacity is about 5 million b/d right now in 2010 (and this doesn't change in 2011). Any increase in demand requires that new supply be taken from this capacity. The main problem is that Saudi Arabia has almost all the spare capacity!

From John Williams' Why The Saudis Hold The Cards Right Now. "The US government EIA now estimates OPEC spare capacity at 4.96 million barrels per day. Only Saudi Arabia with 3.75 million b/d excess capacity has any significant power. Kuwait and UAE each have an estimated 300,000 b/d of spare capacity and Qatar has 260,000 b/d. With Saudi Arabia controlling 76% of the spare capacity the combined power (16%) of Kuwait, Qatar and UAE is barely enough to influence prices."

World policy-makers and analysts believe the Saudis do indeed have about 3.75 million b/d in spare production capacity. A small but vocal minority of "peakists" do not believe the Saudis have anything close to this much spare capacity. If they don't, oil production might as well have peaked in 2005. If they do, and we throw in Kuwait, the UAE and Qatar, the world is capable of producing as much as 78+ million barrels per day of conventional crude oil. Once the 860,000 b/d the other Persian Gulf countries have is exhausted, everything rides on the question of Saudi production capacity and intentions.

Well, you might say, of course the Saudis have what they claim. Or at least they have some large part of it. And I agree—I think OPEC's spare capacity is at least 2.5 million barrels per day at the low end, and may be as high as 3.5 million barrels per day at the high end. This range puts Saudi spare capacity at between 1.7 million and 2.7 million barrels per day. Theoretically, the world can produce as much as 77 million barrels per day.

On the other hand, those saying we've peaked like to point out that Saudi Arabia has never produced 10,000,000 barrels per day of conventional oil on an annual basis since 1970 (EIA data). And it's true. The most they've ever produced in any given month in the last 10 years was 9.7 million barrels per day in—you guessed it!—July of 2008. Thus the pessimists argue that Saudi capacity is largely a reflection of Saudi production. However, absence of evidence is not evidence of absence. This is another common fallacy.

The world has put all of its future oil production eggs in one risky basket—the Saudi basket. Turning the argument around, they've never produced more than 9.7 million barrels per day in any month in the last 10 years! And they finally reached that level when the oil price was over $140/barrel and the economic damage had already been done. Realistically, when push comes to shove in terms of global demand, how much oil is Saudi Arabia willing to produce?  We really don't know. This is an extremely dangerous position for the world to be in.

A vaunted part of their large spare capacity now comes from the Khurais development which was put on-stream in mid-2009. This large field is supposedly standing by ready to produce 1.2 million barrels per day as needed. In order to achieve this output, the Saudis must pump millions of gallons of seawater from over a 100 miles away and inject it into the field to force the oil out. Khurais has little of the natural reservoir pressure that new (light) oil fields typically do. (Khurais was discovered about 50 years ago and produces light oil.) Has this field ever produced 1.2 million barrels per day? Well, no! I'm sure the Saudis did the required well tests, but as for turning this field up to full volume, and sustaining full production for months on end, that's never been done.

It seems clear enough that the world can produce at least another 2.5 million barrels of conventional oil per day beyond the 73.462 (monthly average) it has produced in 2010 (through July). The question becomes whether producing 75-76 million barrels could be sustained for any long period of time. The "natural" decline rate in Saudi Arabia's oil fields (taken as a whole) is about 8% per year. In order to mitigate this, they must constantly invest in field maintenance. Moreover, most of their mainstay producers, including the mighty Ghawar field (~5 million b/d capacity), are many decades old.

And then there is the trust issue. Clearly, the Saudis have exaggerated the size of their recoverable reserves as I demonstrated in OPEC Will Never Run Out Of Oil. Can we trust their spare capacity numbers?

Where do we stand? Peak oil does not mean the end of the world as we know it in the medium-term—the next 5 years or so depending on how the global economy fares—but in the longer term things don't look so good. Some say Iraq will increase its output by several million barrels per day in the next decade, but that's a very risky proposition.

Our utter dependence on the Kingdom of Saudi Arabia to meet our future supply needs poses a significant threat to the global economy. Moreover, it's a disgrace that the world has allowed such dangerous risks to pile up without adequate preparation for a future it was not hard to see coming.

Here's the Martenson video.

 

Comments

[Bill Hicks]

While your points are certainly valid, Martenson is not exactly way out on a limb with his statement. As your graph demonstrates, crude oil production has been essentially fluctuating in a range between 72 to 74 mbpd since 2004--going on 7 years now. A slight one month bump up above that range hardly indicates that we are on anything other than what Peak Oil theorists call the "undulating plateau" at the top of Hubbert's Peak.

You are probably right that Peak Oil will likely not reach crisis mode for about five more years. In fact, many Peak Oil writers, including the very astute Tom Whipple, have cited 2014 as the year the world may begin to see world oil production decline at the beginning of the back side of Hubbert's Peak. After that, it all depends on how fast we decline as to how bad the effects will be on the world economy.

But no matter what it won't be pretty.

[Dave Ranning]

So, 2005 still is the high in conventional (c+c) production on a yearly basis?
How does this negate Matenson's analysis?

But the question is, when do we fall off this plateau--

[Remi]

I concur with the other comments, that whether we poduce a little bit more oil in the next year or two, the fact is we have reached the plateau. The only thing that would disprove peak oil theory is if production started to increase year over year like it did pre-2005. I know you are not trying to disprove peak oil theory, but arguing over the year or date when it occured (or will occur) is semantics. Is it when the plateau was reached, the peak production over a day, a month, or a year?

[Dave Cohen]

I suppose comments like the ones I've seen so far will keep coming in, so I might as well respond now.

If the world produces 75-76 million barrels per day in some future years and then never exceeds that total, those will be the peak years. The timing of "peak oil" can only be determined by looking in the rear view mirror.

There is a considerable difference between 73.7 (2005) and 75-76. What constitutes a plateau? If we got to 77, would we still be on this "plateau"? This is simple math. How hard is it to understand? Apparently, very. And then there are gas liquids and biofuels to consider, but I didn't get into all that today.

These details I've discussed DO matter. Martenson DID overstate the case. There are people growing vegetables, and buying canned goods or shotguns because they've been told oil production is going to fall off a cliff any day now and thus the world is going to go down in flames. Some people I know MAKE A LIVING telling people the End Is Near.

I want people to have realistic picture of our likely future. I guess some folks object to that. Oh, well. With human beings, nothing surprises me anymore.

[Guy McPherson]

Dave, I agree with Martenson on the timing of the peak. C+C peaked in 2005, but the EIA covered it up by adding tar sands to conventional oil in their reporting (effecting January 2008). The result is slightly more than 1 million bpd of "crude" that isn't C+C at all. The Pentagon's 2010 Joint Operating Environment indicates we fall off the "undulating plateau" by 2012, and that by 2015 we'll have a gap between supply and demand of some 10 million bpd.

I'm not making a living telling people the end is near. By most imperial criteria, I'm not making a living at all. But I am making a life far removed from American Empire. That decision is a moral one, not a financial one.

[matt picio]

I'm not sure Martenson did overstate the case, despite your clarifying remarks. Until 2005, each production peak is higher than the one before. Afterwards, each peak is less than the 2005 peak, with the exception of a single spike in 2008, which I would argue was a last-ditch attempt at a production increase at a time when the per-barrel price was so high that all projects were cost-justified. The only other time this pattern of peaking has ever occurred was during the oil shocks of 73-74 and 79-80.

It's entirely possible that this is a localized peak, and that the actual peak remains in the future, but given the status of current and planned megaprojects and the depletion rate of existing fields, a higher peak is unlikely. In any case, it really comes down to how one defines "Peak Oil" or "Peak Production". If you are looking strictly from a numerical perspective, then yes, 2008 is currently the peak. If one defines "Peak Production" as the point at which sustained production increases become impossible, then 2005 seems to fit that definition quite nicely.

Really, though, this feels a lot like debating which piece of the iceberg that hit Titanic pierced the hull the deepest, and at which point on the hull was the flooding most severe? The world economy and oil industry has already hit the iceberg - like Titanic, it's some time before the boat actually sinks - but the sinking is inevitable. What remains is the transition from the deck of the ship to the water. It can be gentle, or catastrophic, and the decisions made now will determine how many are saved, and how many freeze to death.

[jdl75]

I think (to be rechecked) that if you do the computation on a yearly basis, 2005 is the current max in crude oil over 2008, whereas on a monthly basis july 2008 is indeed the max.

[Walter]

People get a false read from the current peak oil bell curves. The usual bell curve represented corresponds more closely to n=15 then n=30 or more. Since people process what they see in a picture or graph more than text, it would be a good idea to get the curves right. The "bumpy plateau" on top of the Hubbert curve is actually a shorter time than people realize. Depending on how "linked in" economies and cultures are with their physical environment, the falloff could be quite dramatic.

Culture gives us "slop" in the system, so we can weather the vicissitudes of the environment from which we are filtered. Replacing actual cultural modes with growth based on a cheap finite commodity has made us more vulnerable, not less. Here in the OECD countries, we won't be able to go back to cultural modes as quickly as folks in Africa, for instance. For both these reasons, I expect the falloff (and the dieoff, for that matter) to be more rapid than most "peak oil experts" predict. The only saving grace is the incredible amount of energy we have locked up in things that can be reused and recycled.

[KLR]

Matt Simmons gave the last documented production figures for Khurais in his book, if anyone's interested. Ca. 150 kb/d. I was always curious if any fields had received such a Extreme Makeover as the Saudis are attempting, and how the outcome matched expectations.

I was always irritated by people trumpeting 2005's yearly average vs. 2008's absolute peak. Attempting to factor in demand swings overheats such simplistic models that make for good scary oneliners. Now, one point in their favor is the strange phenom of KSA throttling back production as the price rose inexorably upward. None of the explanations proffered - insufficient refining capacity for heavy oil etc. - seemed worth a damn. Surely someone would be willing to load all that slop onto floating storage, at the least. Inventories overall still had a way to go before hitting their ceiling. The possibility that they began to decline and pulled out of a nosedive courtesy Khurais is worth entertaining at the least - let's look at some historic examples of them, or Texas, prorationing, with storage and price levels in mind.

I've looked at yearly averages for production as far back as I can find them; we had a corresponding crash in demand at the beginning of the 80s, of course, courtesy fuel switching, increased efficiency, move away from oil for heating, etc. About the only other time demand has contracted more than a single year is early 90s, when the USSR fell apart. So we can perhaps base projections for the near term on these historic examples. US demand seems to already be rebounding, and never contracted anything like it did 30 years ago, so bet on spare cap shrinking up quite quickly. In 1983 it was something like 9 mb/d - I've never found real numbers for those days, and plan to extrapolate them from graphs provided by sources like wtrg.com, for yucks.

Suggested listening for this post: [url=http://www.youtube.com/watch?v=xatqCiiXGEw]Joni Mitchell - I Don't Know Where I Stand.[/url]

[BR]

Martenson frequently talks about energy and claims to be an expert without knowing what he's talking about. He also likes to make vague doomer-y predictions that I'm sure he's hoping will come true if you squint at them right. (I'd like to see whether he's ever made a specific projection on energy or economics that's actually come true - not some vague "things will unravel faster than you think" crap.)

[Bill Hicks]

Simple math? Simple math is that the difference between 74 mbpd and 77 mbpd is less than 5%. How is economic growth going to continue even if we do manage to squeeze out that additional 3 mbpd for a sustained period?

We're essentially on the plateau now. There is no other way to explain how prices could skyrocket so dramatically with little additional production being added. And note that I did NOT saying this is the imminent end of the world, only that whatever happens will be a rude awakening for most people.

[mark robson]

Thanks for drawing up the chart of conventional only production. I have been hoping someone would sift this out for some time. Artic, deepwater, ethanol and tar sands are all just side issues - total liquids supply will peak when conventional oil goes into decline. Similarly, conventional will fall off the plateau when Saudi oil declines, and Saudi oil will go when Ghawar declines.

Considering they've supplied about 12% of the world's all-time total oil production I think the Saudis have been pretty fair in sharing their black gold with the rest of the world.

You correctly point out that Ghawa provides half of the Saudi production. You discuss a filed than may produce one million bpd when Ghawa produces five. How much water do they pump in there now? There's no need to watch 47 countries producing deepwater fields, just watch Ghawar.

When Ghawar goes like Cantarell into 9% decline then is the time to buy your electric pushbike.

[KLR]

You can't find the data for C+C? Just searching Google Images for 'Crude and Condensate Production' will yield innumerable examples, if you lack the ability to create one yourself. EIA has data sets for a handful of streams, if you want to break things down further. This is basic stuff. Really doubt Ghawar will crash ala Cantarell - haven't examined production numbers post-decline for many big fields myself but even big North Sea fields didn't keel over and die like Cantarell did, despite the companies wanting to maximize profits by running the fields hard. Pemex is a pretty boneheaded outfit all told - deeply corrupt, too. Aramco by contrast is a world class outfit.

[Tony Weddle]

I have to agree with those who acknowledge a yearly peak in 2005. Regardless of what one imagines might have been the case without the recession, the figures show 2005 to be the peak year so far. However, to claim that conventional oil has peaked, with the inference that it will never reach a higher yearly level is a bit premature.

I'm inclined to think that it won't get higher than 2005 but I don't know. If we get any kind of recovery worth that name and still don't reach a higher production number for conventional crude then we might be able to say that conventional oil peaked in 2005.

[Jay Clark]

I am just a newbie at this, but it seems to me that the real point is going to be when production on the downside collides with demand on the upside.

This does not require any decrease in production, just a continued increase in demand. Which seems pretty sure given increasing demand on the part of places like China and India. (Brazil seems to be out of the equation due to their efforts at alternatives).

Demand seems to be pretty elastic and directly tied to overall economic conditions. When gas hit almost 4 bucks a gallon the economy went into a tail spin and demand slumped, if gas in the US were to hit 7 bucks a gallon I think we might be pretty sure to see another slump in demand.

Looking at production numbers alone seems to me to be kinda like the sound of one hand clapping, where can I go to get some demand numbers in the same format as the production ones?

[MrEnergyCzar]

I've been preparing my family for Peak Oil for 4 years now and made some videos to help people who wish to do the same. I attached one of them here....

http://www.youtube.com/watch?v=hHmXhgBhtWk

MrEnergyCzar

[xraymike79]

Too many unknowns brought up in this essay to make anything close to an accurate timing of "peak oil production." All we can say for certain is that continued growth of Human consumption will outstrip production within our lifetime just as it has, for example, for all the large predatory fish we extract from the ocean:
http://blogs.redding.com/redding/dcraig/archives/2010/09/youve-heard-of.html

Not too hard to understand that we are hitting a wall against continued exponential growth. Simple as that. My 6 year old could understand this.

[Tony Weddle]

Jay's right, production numbers alone doesn't tell the whole story. However, we can be fairly certain that for economic growth to continue, we need a continuing increase in oil production. In a BAU situation demand will continue to increase. So if the production numbers are telling us that the peak production year was 2005, then it might be that conventional oil is, at best, on a rocky plateau and the "hoped for" rising demand will not be met. Remember that there were two years between 2005 and 2008, with oil prices slowly rising the whole time, yet there was no annual increase in those two years (as well as 2008 and 2009).

It took a while (two years) for conventional production to respond to rising prices with a new monthly peak which suggests to me that such responses will become more and more difficult, producing more and more expensive oil.

[Mike]

This is the most words I've seen used to SPLIT HAIRS.

"Would have" doesn't count: The plateau CAUSED the oil price spike, which CAUSED the demand collapse. World oil supply maxed out in 2005, and has been bouncing around on a plateau for five years. The minor spike upward in July 2008 is a fly speck.

This says nothing about "net energy."

This says nothing about "peak exports."

This says nothing about what happens when we fall off of the backside of the peak.

So it was a plateau instead of a peak. So what.

[ozzy]

I tend to agree with Mike, above. I think any discussion of peak oil which neglects to include the concept of 'net energy' necessarily presents a distorted view, if the goal is forward looking, rather than static, analysis.

The export land model is a bit different, as it depends largely not on geophysical factors, but geopolitical ones. That is to say, yes, it will have an effect - I personally think probably a large one - but it's much trickier to predict.