Thank you, oilaholic, for putting the question in such a concise way.

This is exactly the argument that I have been trying to make and somehow not been able to clearly convey,  Despite my crude efforts at wordsmithing, I seem to have managed to make people annoyed or downright angry, when I say the biggest danger may not be immediate peak (again, it MAY be, we simply cannot know), but that we are running completely and totally in the blind.

The changes in technology and the changes in the exact definition of what is:
OIP
URR
production
production capability
oil
all liquids
are now occurring so fast as to make the HL (Hubbert Linearization)  model more and more difficult to use as a predictive tool in the old way, that predicting peak becomes a complete crapshoot, and runs the risk of undermining credibility of those trying to do so based on the old models.  

There is possible good news in this, if one thinks a delayed peak is good news, in that it "may" give us more time to attempt the "Hirsch Report(s)" type mitigation plans to reduce the chaos and suffering when peak does come.

There is bad news though, in that when, as you point out, and Simmons and Hirsch himself and others try to point out, the "peak" itself may come with virtually no warning whatsoever, and the post peak drop in production may be much more severe than expected  (the "cliff" instead of the "slow squeeze")

We cannot even be sure of the above however, because the changing definition of "oil" per se, and "all liquids" further cloud the picture.  We simply cannot dismiss the effect of NGL (Naturual Gas Liquids or Condensates), "GTL" (Gas to Liquids) different than GTL, although the two are often confused in the popular press, and the introduction of heavy and sour crude in a much bigger way  (notice I am leaving out tar sand and ultra heavy oils, simply because they are too speculative to even think about, but many popular commentators gaining widespread press do not leave them out (Yergin and Lynch comes to mind, and I am not discussing the ethanol issue, which I see as  a giant fuel switching operation, from natural gas to liquids, but again, many commentators now throw this in, thus making the statistical picture murky beyond all comprehension even to the people who spend hours studying these issues....to the average citizen, it must appear more esoteric and complex than subatomic physics!  We cannot, repeat, CANNOT fault the man or woman on the street for being unable to separate the wheat from the chaff!)

What's it all mean?  I have insisted that it means that someone should be getting the message to the public that they should be laying contigency plans for sudden and massive changes, some of them less than pleasant, possibly outright catastrophic to them and their organizations.  Interruptions of energy supply and massive price moves  in any direction are possible.  The long held faith in the stability of energy sources and supply should now be viewed as a dangerous myth.  The complexity of the fuel delivery system means that you and your organization can no longer rely on the simplicities of a "mono fuel" crude oil based world....there may be energy available but it may be the wrong kind, and you may not be able to make use of it.

And for the "peak aware", it should be given as a caution:  Sudden disruption/interrruption/price movement and chaos may NOT in and of itself be proof of "peak" per se, depending on how "peak" is defined  (peak oil, peak all liquids, peak light sweet crude, peak usable liquids, etc)  

Even being aware of the threat above gives us some slight advantage in vision over the general public, but, and I don't say this to be in any way rude, and let me say again that the work here is extremely valuable, useful and informative, I cannot say enough about how highly I regard what I have learned from folks like Standiford, Robert R., and Westexas, and the others, and have NEVER disparaged in any way the efforts and sheer clever and perceptive knowledge of this group), but....we are running, due to the definitional problems cited above, with only slightly more vision than the public at large.
Humility, caution, is the order of the day.....but we can take comfort in the old proverb, "in the kingdom of the blind, the one eyed man is king!"  :-)

Caution.  Strategic and Contigency Planning.  Energy Diversity. Conservation and reduction of waste. Transport alternatives.  The acceptance of workable and usable alternatives and renewables, and the acceptance of something close to "Mitigation options" of the type that Hirsch and others who see the enormous challenge in front of us.  Perhaps not all of the options.  But use the best first, and work outward.
This is the concise, coherent planning that I am accepting, and I am encouraging folks to spread to others and to their organizations.  

"I cannot do a lot, but I will do the little I can do."  Helen Keller

Thank you.
Roger Conner  known to you as ThatsItImout

The volume of oil that is there (Oil in Place) remains the same.  What IHS, which is also CERA, are doing in part is saying that the amount of that oil that can be recovered (the Recovery Factor) can be increased, due to advances in technology.  However in the field that is closest to exhaustion they are now backing away from that conclusion.  Also I am not sure, as I noted above, that it will be possible to reach even the numbers that Jean Laherrere has estimated as the amount that will be recovered.

Bear in mind, however, that as the oilfield production drops, it is not going to be how much, in absolute terms, we get out that is the main concern.   It is the rate of production, since demand is going up, and we are looking to see where we can meet that demand.  There is no question that the field is declining, and I have tried to explain why that is irreversible.  It has implications for the future of Ghawar, which is close, and the Northern parts of which are close behind Abqaiq in relative amounts of oil produced.

Controlling the pressure, to control water movement, also controls the amount of oil that is drawn from an individual well, and the rate at which it is produced.

Thanks.  One question, does the water injection extraction technique totally obliviate the possibility of using stripper type wells for extraction in a resevoir's production.  Having a cursory knowledge of well function for water production it seems if you have rest a well liquid level settles back from a cone shape would it work for oil as well in some resevoir shapes?  
The only difference between a stripper well and an ordinary oil well is the fact that a stripper well has a high water cut. Given enough time, all oil wells in a waterflood turn into stripper wells. Not sure if there is an official definition but probably anything over 95% or so (meaning, nineteen barrels of water for every barrel of oil produced). Large-scale water injection makes high producing watercut more likely, not less likely. So, yes, stripper wells and water injection are very commonly found together. Once the oil has been separated out, the produced water is often re-injected through other wells anyway.

As far as reservoir management is concerned, the ONLY purpose of water injection is to maintain reservoir pressure by replacing the volume of produced fluids. Some folk will say that water injection is "designed" to sweep oil towards production wells, but that is as much a matter of hope as it is of engineering. You try to inject the water in such a manner (i.e. in such a place) that it doesn't immediately channel through to your producers - this is mainly common sense, e.g. you don't normally want to put a water injector immediately updip of a producer.

Yes, water cones can collapse given enough time and the right rock and fluid properties. This sometimes makes a difference to how wells are managed. Another approach is to produce below the critical rate at which the cone will reach the production well. As always, it depends.

Re: Another approach is to produce below the critical rate at which the cone will reach the production well

Could you expand on this remark a bit, plucky? When, where applicable? does it extend production into the future at lower flow rates? Etc.

OK, briefly: oil floats on water in the rock. Horizontal interface between the two phases is called the oil water contact (ignore the transition zone). Production wells are drilled down to somewhere above the contact. When production begins, the area of reduced pressure (drawdown) around the wellbore causes the contact surface to be perturbed upwards until it reaches the production well and you start producing water. This is "coning", so called after the shape of the perturbed contact. If you stop production then the pressure disturbance goes away and the cone collapses, in certain rather complex to define circumstances (has to do with "relative permeability hysteresis", which I can't describe compactly, or maybe at all).

Really hard to interpret diagram here http://www.glossary.oilfield.slb.com/search.cfm and search for "coning"

Coning is promoted by:
High production rates
Narrow interval between contact and bottom of well
High oil viscosity (=> high drawdown & adverse mobility ratio)
Low horizontal permeability (ditto)
High vertical permeability (gas moves faster)
High net:gross (no shale stringers to stop water moving vertically)

If you know enough about the rock and fluid properties then you can estimate the "coning critical rate", i.e. the rate at which the cone will have grown to the point where it just reaches the bottom of the producing well. You can then choose to limit your production rate to just below this level to avoid water production. It certainly extends DRY production into the future at lower flowrates, but if you are willing to cycle enough water through the reservoir then you'll eventually get all the moveable oil out anyway, up to the economic limit. And sometimes the coning critical rate is so low that you just have to shrug and accept that it will happen at any economically realistic well production rate, and manage the water on the surface.

Remember that if you are injecting water below the oil, or if an aquifer is influxing into the reservoir, then the contact will be moving upwards anyway so the coning critical rate will change with time as the contact gets nearer to the wells and eventually breaks through independent of production rate.

Coning can happen with gas as well, this time coming from above not below - think Cantarell (but unlikely there due to very high horizontal permeability).

Horizontal wells spread production out through a larger volume of reservoir (hand wave) and so reduce drawdown and the propensity for coning. Again this delays water breakthrough but once it happens, it happens. See "cresting" in the Schlumberger glossary linked above.

Let me wave an arm back, but in appreciation, for the clear verbal and graphic explanations in this and other threads. I'm certain those contributing explanations on topics like this know the underlying equations and models that are so pleasingly (for me) missing from the posts. Thank you.  
Re: Bear in mind, however, that as the oilfield production drops, it is not going to be how much, in absolute terms, we get out that is the main concern. It is the rate of production, since demand is going up, and we are looking to see where we can meet that demand

This seems to be constant confusion at TOD but I think it is due to many others (eg. IHS/CERA) spreading it publicly, which I'll call "the appeal to URR" argument. Field F has a URR of 10 Gb. You have recovered 4 Gb. If you have a recovery efficiency (factor, rate) of 50% instead of 45%, you can extract 1 Gb over time instead of 0.5 Gb. At what flow (production) rate will you do so?

In other words, what is the shape of the tail? How are you managing it? You might maximize flow rates now and near-term followed by a steep decline (this is the strategy I have called Extreme Production Measures). You still get your extra 5% of URR but then you are hosed--to use a technical term :) The proximate cause for this strategy is economic but there is a deeper reason: a system built on continuous growth must deploy such a strategy and it is always seen as necessary and good.

Or you can manage the tail differently, reduce flow rates from the field by delaying in-field drilling and try to maximize yields over a longer time--again assuming that the recovery efficiency remains unchanged, you will still get your 50%. This strategy assumes that you have a finite resource, better to live a bit more humbly now and longer than burn the candle at both ends. You will meet demand by demanding less and implement other strategies for living.

OK, as David Byrne once said, I will stop making sense now.

best --

In other words, what is the shape of the tail? How are you managing it? You might maximize flow rates now and near-term followed by a steep decline [...] Or you can manage the tail differently, reduce flow rates from the field by delaying in-field drilling and try to maximize yields over a longer time.

Dave, whatever about the ethical dimensions of accelerated versus protracted extraction, I don't really see how speedy oil recovery affects the overall depletion curve except that it might bring on the peak a few years sooner than otherwise.  How can rapid extraction of individual fields impact the overall decline rate? Not all oilfields were discovered at the same time.

If there are lots of oilfields, all of which matured at a different date, and presuming their individual nose-diving decline is staggered over time (say, 20 years), won't the cumulative decline rate be much the same as if all had declined gradually?

In other words the 'slow food' as opposed to the 'fast food' extraction approach will only postpone the hour of reckoning by a couple of years at best. And it won't change the shape of the decline curve to any notable extent.

Copelch,

I am not an oil guy but I disagree with this logic for the following reasons.

  1. I have experience with water wells for irrigation in Texas and New Mexico.  They also have shown decline over time from pumping.  In many places the aquifer recharges so this is not applicable to oil fields which don't recharge.
  2. However there are some aquifers in West Texas near Lubbuck and in much of New Mexico that are not recharging, at least not in the last 50 years, this is water from 10,000 years ago or more in underground lakes.
  3. A lot of places near Lubbuck used to have one well per quarter section that used to pump say 750 gallons per minute out of a lobe of the Ogallala aquifer.  These wells were drilled in the 1940's and 1950's mostly, and for 30 years supplied all the water to grow crops on that quarter section.
  4. Irrigation pivot equipment is often sized for 400 gpm in many of these locations. Over time the well declined until it was pumping below 400 gpm and wouldn't feed the pivot.  Might only produce 350 gpm by the 1970's.
  5. So they "infield" drilled another well between pivots some years ago that also initially delivered 350 gpm.  They were back to 700 gpm (very slight decline) but out of two wells now, not one.
  6. In about another 10 years both of these wells declined rapidly until the net was again below 400 gpm.  So now there is often 4 wells pumping 125 gpm or a total of 500 gpm for the quarter section.  They didn't get back to the original 750 gpm and they have 4 wells now not the original single well.
  7. What is scary is that no matter when the original wells were drilled, or farms developed in those 1000's of square miles, everybody has drilled lots and lots of new wells in the last 20 years.
  8. So at present everybody is still farming using lots of water.  Some farms have more or less water under them but all wells have declined very rapidly in the last 20 years compared to the first 20 years of operation.

So in summary these water wells have maintained a near constant production via lots of new wells.  But they are reaching a point when doubling the number of wells isn't really going to help that much.  And the cost is going through the roof.  So even though these individual pockets of water were developed at different times they are all going to collapse at about the same time.  The reason for this is that new water wells always needed to be pumped at the maximum rate to make up for declining original wells.  People moved more of their high water requiring crops to where the most water could be pumped.  No spare capacity like the original wells had.  

Very late in the day people have come to realize that using irrigation equipment that only uses 250 gpm or less, instead of 400, can work just as well if applied efficiently.  But now wells are dropping below 100 gpm and there just isn't very much water left.  If the same efficient systems had been put in place 30 years ago there may not have been the need to drill more wells and there would still be a lot of water left to use.  But in aggregate all the water has been found and all the aquifers depleted to the same low level because the easiest water is always removed fastest.

So even though these individual pockets of water were developed at different times they are all going to collapse at about the same time.

NC,

Thank you for your highly instructive reply. Until now I had always believed that the US still had a long way to go before groundwater shortage became a problem.

You've certainly given me food for thought -- so I've put my thinking cap on.

Groundwater depletion is even scarier than `peak oil'.

BTW has anybody ever constructed a Hubbert curve for `peak groundwater'?

Whilst your argument is probably valid for individual wells or fields, it fails to address copelch's thesis, which considers the production profile for the world as a whole. Your view reflects an opinion which by my reading, has become quite popular among drummers of late.

Many people on this site at the moment seem to think it self evident that if individual fields all show sharp declines, then the sum total of those fields must show a sharp decline also.

This is not the case.

If this seems counter-intuitive to you, please dwell on it.

I think this idea has become popular mainly because it supports the general doomeristic framework.

Hubbert didn't use a symmetrical curve to model oil production because he thought production from individual fields was symmetrical! This isn't typically the case, and as far as I know, never was. Rather, he knew he could use a symmetrical curve because summing large amounts of skewed distributions will likely produce a more or less symmetrical curve.

It is possible that world oil production will have a significant skew post peak. If so, this would likely be caused by such things as political events. The fact that individual production wells are getting more skewed is unlikely to be a candidate.

If you (ie. anyone) think the post peak production curve will show a skew, you should be able to explain why you think this will be the case without appealing to the fact that individual fields show a skew.

I would think that, if the major oilfields are badly skewed (i.e. fast to catastrophic decline) this would skew the whole data set. We have repeatedly seen data here that the (many) smaller fields tend to decline much more rapidly than the bigger ones, thus contributing their 'collective skew' to the statistical mix.
Re: Hubbert didn't use a symmetrical curve to model oil production because he thought production from individual fields was symmetrical!

You are right. But see my remarks below. Look at the graph for North America from the Hirsch article cited there. I am interested in production curves. Look at Stuart's Extrapolating World Production. Consider this.

Again, it's very important to remember that the high and low models are error bars on where the center of the model would go, if it was truly logistic, and if extrapolation of this last region of linearity in the linearization is valid. Annual production can have significant noisy excursions above and below whatever the true line turns out to be.

The Gaussian peak is in 2024 at 95mbpd, but I don't trust that extrapolation. The linearization-based logistics are a lot more likely to be correct in my opinion. Notice that the last couple of year's production are a big spike above the logistic models. But that's ok, because the spike seems to be ending in the recent production plateau.

Same graph below except with Stuart's SWAG in yellow.

However, please don't take the specific numbers on that yellow curve too seriously - it's just intended to illustrate a general qualitative idea of what might happen.
What might happen. That's what I'm looking at.

I am trying to explain generally the "big spike" (1st graph) and consider the plausibility of Stuart's guess (2nd graph) which does not display a "bell curve" shape. My theory is that it has to do with advanced EOR/IOR applied to old giant existing fields and also the spike in deepwater production. I've called these strategies extreme production measures. I think that yellow line, which shows a sharp decline, may be our future. I've tried to explain why. I've called for production slowdowns because there's no other good explanation for what's happened in the last few years. It's not because we are swamped with new production but we have encountered a demand shock. The world is pumping like crazy to meet it.

It's not because I'm some doomer.

Implicit in your comment (it seems to me) is the idea that, if the world produces XX million barrels in a given year, it doesn't matter which fields it comes from, or whether or not any particular field is squeezed hard.

But we must ask ourselves why individual fields are being squeezed so hard. Globally, the answer seems to be a lack of replacement fields coming on stream to take up the slack. So, individual field profiles will show, instead of a bell-curve decline, a plateau followed by a steep decay.

If it turns out that a high proportion of existing producing fields are past their peak, and are in this artificial post-peak plateau phase (big and small fields alike), then it seems likely that the aggregate production numbers will also be deformed in the same way.

If it turns out, over the next couple of years, that world production numbers are indeed in a plateau, whereas we know that large numbers individual fields are being squeezed, then that would seem to indicate to me, a strong likelihood of a steeper downward slope once the plateau ends. Given that the area under the curve is constant, you can't have your cake and eat it.

Re: If there are lots of oilfields, all of which matured at a different date, and presuming their individual nose-diving decline is staggered over time (say, 20 years), won't the cumulative decline rate be much the same as if all had declined gradually?

There are lots of oil fields but not lots of giant oil fields (like Abqaiq) and these produce somewhere between 40% and 50% of the world's oil daily depending on you how define "giant". If you define it as a field with a URR of greater than 0.5 Gb, then the percentage is closer to 50%. Most of these fields are old (pre-dating 1980 at least, most are much older). See Giant Oil Fields of the World (pdf file).

If a significant number of these fields go into rapid decline close to each other in time--say, within a decade--then due to the large percentage of the world's oil we get from them we will alter the shape of the global decline curve--it will be sharper. Even a few percentage points makes a difference here. Abaqaiq is producing 400/kbd and peaked some time ago. If you manage the tail end production of this field (and others) to promote longevity and not rush to suck them dry, you are partially managing the global decline also. These really big fields are in some sense "priceless" -- they can not be replaced. The discoveries trend shows this clearly. See Hirsch's Shaping the peak of world oil production. Do we want to see a global peak that looks like North America?

Indeed, the trend is toward a greater share of world production from smaller fields in countries like Chad, Mauritania, Malaysia, offshore Brazil, et. al. What you say would be true if all oil fields were created equal. But they are not. If you lose 200/kbd within a few years at Abaqaiq because of the way you produced it, you need to put 20 10/kbd fields onstream just to replace that. At Cantarell, the situation is much worse. Ghawar could go that way.

I might have to post on this again. Concerning the ethical implications of my position, my answer is:

  1. Shall we mortgage our children's oil future?
  2. Shall we start mitigating the peak now?

No to #1, Yes to #2. Why can't humans make reasonable extrapolations about the future and act accordingly? Unfortunately, that's just the way it is.

Erratum

Should read

these produce somewhere between 40% and 50% of the world's daily oil flows depending on how you define "giant"

Thanks for your clarification.  It had indeed occurred to me that most of the elephants might be about to collapse together within the same decade. Well, it occurred to me just after I pressed the 'post' button.

Still, there might be enough relatively immature 'rhino-sized' and 'water buffalo-sized' oilfields (in Chad, Brazil etc.) to compensate and stave off the grand finale just a little bit longer.

Though it takes an awful lot of rhinos and water-buffalos to outweigh half a dozen elephant.

I am far more fatalistic than you are, I suppose -- it's the irreversibility of the oil crisis, not the date of its eruption or even the shape of the downward curve, that is the most depressing. What are ten or twenty years reprieve for the automobile compared with an eternity of the horse and cart?

Except they are not rhinos. The average field size discovered lately is soemthing near 34 million barrels _total_, as I recall. You're trying to cover the death of a dozen elephants with thousands and thousands of bunny rabbits... except these ones are sterile and don't reproduce.
We are not running blind!

Well, you and I are not running blind anyway, the problem is everyone else is running blind because they are wearing blinders. It is like we were part of a huge herd of buffalo running full steam toward a cliff. Now you and I can see the cliff ahead. We both know we are about to run over the cliff. But everyone else seems totally oblivious to our predicament. And there is not one damn thing we can do about it. The vast majority do not wish to see the cliff so therefore they do not see it.

And by the way Roger, I must emphatically disagree with you as to how much we can know about the future of oil resources. When you say "we simply cannot know", you are simply dead wrong. What an insult that is to the hundreds who have studied the problem without their blinders on. What an insult that is to the good people who run this list who have gone to great lengths to try to describe the problem and to explain why we do know! Now I do not mean to say we know the exact day of the peak, but we do know it is imminent. To say that "we simply cannot know" is, in effect, to say it could be thirty years from now. If that is all we know then we might as well quit worrying and join the herd in running headlong toward the cliff. All the data we have gathered concerning the nature oil reserves, the nature of water injection and other methods to enhance recovery is not for nothing.  If we simply cannot know then what the hell are we doing here. If we simply cannot know then Heading Out's thesis on the life of a large oilfield was a useless exercise. He, and all the other good folks who help run this site, are simply wasting their time because we simply cannot know. The hell you say, we do know.

One more point. There is nothing you can do to turn the herd. They are running toward the cliff and are destined to run off it. What you can possibly do is somehow keep yourself, and perhaps a few others, from running off the cliff. But if your dream is to spread the word and save the whole damn herd, forget it, it is already way too late.

Ron Patterson


Darwinian,

You say,
"Now I do not mean to say we know the exact day of the peak, but we do know it is imminent. To say that "we simply cannot know" is, in effect, to say it could be thirty years from now. If that is all we know then we might as well quit worrying and join the herd in running headlong toward the cliff."

Well, I am about to get in deep trouble, but it is not fair not to be honest and straight up.....sorry, but I do believe it could possibly be 30 years or more away, if you are referring to what I call "geological peak" of the true Hubbert type.  I do not dismiss that this whole current situation is purely logistical/geopolitical/economic in nature, and has very little to do with geological peak.  The peak could indeed be 30 years away, or it could have occured yesterday.  I am sorry, and I intend to be insulting to no one, and I have great respect for the work they do, and consider it useful to inform us of possible danger, but....I simply feel that the graphers and the stats makers have no good and valid imput information to work with.

Why are we here?  Because even if peak happens to be further away than we think, nothing changes the fact that oil and gas consumption has gone to mad and wasteful levels, and the even if we don't reach peak oil soon, we are running the risk of hitting peak money and logistical rescources very soon, in other words, we are hemorrhaging money, effort and resources with or without peak, and must attempt to regain control of out of control waste and consumption before it bankrupts us.  I am going to say something a bit odd, but I think true:  some who dismiss "peak oil" say, "oh, don't worry the oil is 'out there'....."
out where?  The fact is the one thing that may be worse for America than the oil not being 'out there' is if it really is 'out there', and I do see that as a real potential danger. I repeat, I cannot KNOW that peak is not already behind us, but I cannot KNOW that is not decades away, and with technical advances, may go by essentially unnoticed.

So my goals stay the same.  Strategic planning,  Case harden America through reserve planning, fuel diversity, acceptance of conservation and workable renewables, transportation alternatives.  The biggest enemy is bing single sourced, get loose from the "mono-fuel" past.
There is plenty to do.  :-)

Roger Conner  known to you as ThatsItImout