September 08, 2006
Big Oil takes on a Big Job
Many press reports (for example, CSM) described this as potentially increasing U.S. petroleum reserves by 50%. The 50% figure appears to come from taking the upper end of an estimated range of 3 to 15 billion barrels of ultimately recoverable oil. That 3-15 billion barrel estimate in turn is not found in Tuesday's official press releases from either Chevron or Devon, but instead apparently comes from an article by Chevron engineers published in World Oil in May 2005. This week's production test is evidently being interpreted as justifying some of that earlier optimism.
What one does find in the official press release is some very impressive details of how much was required from Chevron to make this find. The well is located 270 miles southwest of New Orleans, or about a third of the way across the Gulf of Mexico. It begins under 7,000 feet of seawater, and proceeds from there another 20,000 feet into the earth's crust. That's a total of over 5 miles down, setting all kinds of records for pressure, depth, and duration.
Meanwhile, at the ocean surface, Chevron and its partners can basically count on the fury of future Gulf storms to pound savagely on the billions of dollars of equipment a full-scale production operation will require. 180,000 barrels of production of oil from the Gulf of Mexico was still shut-in as of June 2006 as a result of last summer's Hurricanes Rita and Katrina, at which point MMS decided to give up on reporting the lingering damage estimates from last year's storms. BP's massive ocean rig Thunderhorse, found listing after 2005 Hurricane Dennis, is still not expecting to see operation until 2007.
Some observers have also noted the parallel with an announcement last March that Mexico had discovered another 10 billion barrels of oil in the Gulf of Mexico. That announcement has subsequently appeared to have been unwarranted.
But let us suppose that Chevron's optimism turns out to be fully justified, and they have a new oil supply to rival Alaska's Prudhoe Bay. Finding and producing new oil, even in these quantities, need not mean that world oil production will increase. The reason is that, after a sufficient quantity of oil is taken out of the ground from the reservoirs we've found earlier and are currently exploiting, less oil starts to be produced from those older fields each year, a phenomenon known as oil depletion. Harry J. Longwell, a retired senior vice president for Exxon-Mobil, estimated that average global depletion rates are around 4-5%. The reality may prove to be greater or smaller than this figure, but if we use 4.5% as a benchmark, that means that enough new oil fields need to come on line each year so as to add an additional 3.8 million barrels per day of production just to keep global production from falling. If we accept, for example, the estimate of 750,000 barrels per day of production from the Chevron field that some have proposed, and if all we did globally was to bring on line 5 new fields like this every single year, all that would accomplish would be to keep global production from falling.
Of course, it's a bigger and more important contribution if we compare the potential find just with the U.S. market rather than the world as a whole. Even so, when we look at the U.S. in isolation, the reality of the logistical challenge raised by depletion becomes all the more stark. Basically U.S. production has been in decline for the last 35 years, despite a huge increase in drilling effort. Production from Prudhoe Bay in the 1980s produced a temporary blip up in the long-run downward trend.
A specific contribution to U.S. production from the new find is at least a half-dozen years away. If you extrapolate the slope of -129,000 barrels/day in U.S. production since 1970 another 6 years into the future, new production of 750,000 barrels/day coming on line in 2012 would in fact put us exactly where we are in terms of U.S. production right at the moment.
The real grounds for optimism here are presumably the fact that this is the first time we've explored at such extremes, and the payoff looks to be large. That certainly raises the possibility that there will be more related good news that we may expect in the years to come as even these new drilling records come to be broken. Nevertheless, some of that "good news" is already incorporated, for example, in the forecasts of Cambridge Energy Research Associates, whose optimism in their report from last summer was based in part on the expectation of significant new production from "ultra-deepwater" oil. In any event, I would suggest that greeting the Chevron announcement as heralding the end of the end of oil or signaling that sky high oil prices may be a thing of the past may be a bit premature.
None of which is to take anything away from the technological achievement represented here, for which I think the only reasonable reaction is one of awe and respect. Celebrating this success is fully appropriate. I'm just not prepared to join those who are taking this as reassurance that what we've seen in oil markets over the last three years will now just fade away like some forgotten bad dream.
Posted by James Hamilton at September 8, 2006 03:47 PMdigg this | reddit
I somewhat disagree with one aspect of what you say. You talk about production rates from existing fields declining at 4-5% per year and imply that it is a tremendous challenge to bring enough new oil online every year just to counter these declines. And in that context this new discovery might be seen as just one more such new field.
However, this underemphasizes the fact that this condition is not new, it has been going on for decades. It has long been the case that existing fields are declining and we have to bring on substantial new production just to counter it. And yet, over the decades oil production has grown enormously. The oil industry has repeatedly met and exceeded the challenge of supplying enough production to more than compensate for declines of existing fields.
Some of that new production comes from new fields, some comes from new wells in existing fields, and some comes from enhanced recovery techniques applied to existing wells. Most of this time, the new production has not come in the form of major new discoveries like this one may turn out to be.
Such discoveries are rare and are deservedly newsworthy. For the most part, oil producers have met their production goals without the benefit of new 15 billion barrel discoveries. In that sense, this additional supply of oil is over and above the routine discoveries which have sufficed in the past to maintain and grow production levels.
Posted by: Hal at September 8, 2006 07:16 PM
How exactly would a discovery the top end of which adds 1% to world reserves be "major"? Newsworthy, yes, especially for DVN shareholders. For the rest of us, eh, not so much.
FD: actually short the oil complex the last months, but not because of this.
Posted by: wcw at September 8, 2006 07:44 PM
Thank you JDH for this clear-headed assessment.
Posted by: Avo at September 8, 2006 09:28 PM
I would agree that we shouldn't say this is the end to the end of oil. But only because peak oil isn't correct. We have plenty of oil. There's never been an end of oil coming in terms of literally running out of it (at least not anytime soon). As your blog illustrates, it's a matter of how expensive it will be to extract. But we know this because you've addressed this in the past, too. Great work!
Posted by: allen at September 9, 2006 11:26 AM
(Hi all, I'm a long time lurker on this great blog and I've finally been coaxed out of the shadows.)
Allen, many of the theories and predictions made by people who consider themselves "peak oilers" will be proved to be incorrect in the coming decades but that does not mean that the concept of "peak oil" itself is not correct.
While it's true that some people over-dramatize the situation and say that we are running out of oil (which you rightly point out is not true), that's not what peak oil predicts. At its simplest, the peak oil theory says that at some point in the future global oil production will reach a peak and then begin an inexorable decline. This is an inevitable situation because 1) there is a finite amount of oil in the ground and 2) we have naturally been extracting the easiest to get oil first and the remaining amounts are becoming harder and harder to extract.
Posted by: Paul at September 9, 2006 01:19 PM
JDH & Paul I cannot disagree with anything that you say.
While this find in the GOM is good news, there is much work to ascertain proven reserves, optimal production flows and the like. This will take many months. It will be interesting what the overall reserves will be.
The reserves number that is being talked about is what is called all liquids. Barrels of oil equivalent. This includes liquid natural gas, gas dissolved in oil, butane, propane as well as crude oil and other liquid codensates. All very useful in their own right. It is an all encompassing number.
One could also look at Knotty Head also in the Gulf. Knotty Head holds the record for the deepest find at over 34,000 feet. However the reserves were less than first expected. Still impressive never the less.
Even in peak oil models such as ASPO's there is the assumption more oil will be found, the problem is that the rate of finds has not been able to keep up with the current rate of production. One only has to read their newsletters to see this in their models.
Deep water and Polar seem to be the last frontiers but the cost of exploration and production from these areas is not cheap.
These types of discoveries suggest from an exploration point of view is we have moved along the marginal cost curve where these investments can only be justified by higher oil prices.
The fact that US production has been in decline for 35 years despite being the country with the best access to technology in the world tells us that once the peak is reached there is a decline. Technology can help but will not sustain or reverse the effect of the peak occurring.
What we all need to think about is how does the economy transition to higher priced oil on based on reduced availability in a manner that does not cause undue pain from inflation, higher interest rates and higher unemployment?
Posted by: David at September 10, 2006 04:50 AM
Let's see. The first oil well in Pennsylvania was, what, 75 feet deep? The Jack well is about 28,000 feet deep. Do I detect a trend here?
Posted by: Lawrence at September 10, 2006 05:53 PM
What energy source will they use to extract the oil from that deep, deep well? If they use oil, how much oil will they have to burn to get a barrel? I wonder if it would be possible to save money by using ocean thermal power or floating windturbines to gain the energy required? I have to admit that the thought of an oil rig powered by renewable energy does amuse me.
Posted by: Ronald Brak at September 10, 2006 11:05 PM
The fact that US production has been in decline for 35 years despite being the country with the best access to technology in the world tells us that once the peak is reached there is a decline.
"Once the peak is reached there is a decline." Well, yes, that's a tautology.
However, it's also true that cheap oil was exhausted in the US but still exists elsewhere in the world. US production could still well increase if oil prices were to remain at $100 a barrel or more. (Also, perhaps not incidentally, if the US decided to allow drilling in areas which have been restricted for understandable environmental reasons.)
Posted by: John Thacker at September 12, 2006 01:45 PM
The Jack discovery is seen as confirming the potential of a region, not a single field. This is a pretty important distinction - in general, although the capital costs of developing a single large offshore field are spectacularly large, the risk/reward ratio is much better when you have a single well delineated field than when you have a set of much smaller accumulations.
This also significantly impacts the timing of production from the region. Although oil companies do a lot of studies for large fields, and first production schedules may slip by a few years after the initial forecasts, in general project economics dictate that optimal profitability requires a rapid rise to peak production, followed by a shortish peak or plateau, then steady decline.
If you have a scattering of fields of different sizes and characteristics (water depth, subsurface depth, hydrocarbon/contaminant composition, rock characteristics, etc), the appraisal and development schedule becomes necessarily much longer and more stretched out.
In other words, while good news for companies which have recently found attractive discoveries harder to come by, this isn't an instant panacea for a nation with a big thirst for oil.
For more, see www.theoildrum.com/story/2006/9/8/11274/83638#more
Posted by: Andrew at September 17, 2006 08:10 AM