But for a beginning modeler, changes in the "resource base" add a slightly higher degree of complexity, so I am only suggesting that 128shot, start with a simple case and move on from there.

Hopefully he or she will pretty soon improve the model to include an adjustment for discoveries and/or "reserve increases" as well as depletion, as some have.

Some consider the lack of credible data from the oil majors one of the strongest arguments that many/all of them are unable to grow production & are actively concealing this fact.

I personally think all modeling should come in 2 flavors.

1) Based on stated figures from the majors.
2) Adjusted to remove the fictitious OPEC reserves doubling in the 80's.

I suspect that reality will fall between these 2 estimates.

We should start calling a spade a spade. "Reserve increases" is a euphemism for "bad estimation correction" covering up for the fact that some engineer or bureaucrat screwed up on the original estimate. And to blame it on us, as the other guy claims to, is simply projection on his part.

The word "reserves" means something very specific, particularly in the US, and I'm hardly "confused" about the SPE definition considering I do them and have to defend them to auditors. The poster I was quoting said "reserves", not you, so its hardly ME being confused. As to what you THINK other people believe, you give everyone too much credit.



And if you didn't read my last provided example about the prototype engineer sitting on a California heavy oilfield 75 years ago, then you should either A) go do so and at least TRY to understand the point I was making or B) accept your cluelessness and stop saying things which contradict specific and known industry examples which provide insight into why petroleum engineering and reserves has NOTHING to do with using electrical circuits or whatever other formula's amateurs want to dream up to model hydrocarbon production.



You assume the same thing Campbell does, and he's already been wrong how many times? And is there any reason why you assuming the same wrong facts should lead to a CORRECT conclusion this time around?

I provided a perfect example as to why your theory on bean counting is wrong, myself, I would think that the fact that bean counters can't DO reserves is itself a giveaway...under your faulty scenario the accountants should just add up all their barrels and mcf's and go home happy....it doesn't happen that way and claiming it over and over while pretending to be clueless about examples which show why it doesn't does nothing except prove my point, amateurs should just stay home and play with their weeji boards.

Last I looked the SEC reserves from oil company majors was

A) reasonably solid
B) has been for the last couple of deacdes
C) is defined in a specific and reasonable way
D) exhibits reserve growth in measureable and real fashion.

Otherwise, this wouldn't be possible.

http://energy.cr.usgs.gov/WEcont/chaps/RG.pdf

Campbell uses numbers that get supplied from the oil companies as conservative estimates and then accuse him of being wrong?

There is a psychological term for this. It is called projection. Simply put, the patient that suffers from this malady "projects" his/her inadequacies onto someone else, thus displacing the blame for their own undoing. If the projectionist does this early enough, it serves to innoculuate or vaccinate theselves against somebody else coming along and accusing them of the same failures they accuse others of. So Campbell gets accused of being wrong, this fact gets echoed over and over, while no one bothers to look at the man behind the curtain running the projection machine.

I don't suppose you've even actually READ anything Campbell has published over the years?

And pointing out that he is wrong in his published predictions is hardly an accusation, it just IS. Pick up a book and pay attention already, some of this is so obvious I'm amazed I can use it as a bludgeon on people who should know better.

Oh...PetroConsultants isn't an oil company...maybe you should include some credit for them when referencing where Campbells estimates of WHATEVER come from.

Here is something to think about when we, the dirty masses, try to make objective sense out of the reams of crap laid down on us. This post from Glenn Greenwald speaks volumes if you replace "journalists" with "professional oilers" and "politics" with "oil depletion" ...


So unless you can address the merits or shortcomings of a model objectively, you don't have a chance to halt the flow of ideas, whatever your credentials may be.

Thanks everyone for your help.


The real bump in the road, as you could say, for me at least, is I have to have a form of calculation, since I have to present that as well. I can't just give a speech without some kind of data, calculated by myself or otherwise.


Otherwise this task would be a hell of alot easier..

Oh...thats all you want? Sure....Campbell called the first peak when? Late 80's? Early 90's?

And it didn't happen. Objective fact #1.

Campbell says reserve growth CAN'T happen when you backdate reserves to discovery point of a field. He then backdates reserves and quantifies those numbers. Years later, he then alters them upward. Presto....reserve growth from the guy who says there isn't any. Objective fact #2.

I ask again, have you ever READ anything this guy has published, or do you just enjoy PRETENDING to know what he has published?

OK

Try using the verhulst function. It is considered to be too simple by some on this board, but for your purposes should be fine.

[web]http://arts.bev.net/roperldavid/minerals/crudeoil.htm[/web]

ref: the "peak mart" thread on this forum where we did this awhile back.



For the "updated" model we did last year, we used the following:

n 0.149454128
qinf 2,461.82
t(1/2) 106.8467633
k 0.039628337

So the way this works is that you want to compute Q-sub t, which is the quantity of oil remaining in the ground for any given year t. So, for the first year, T=1.

You put in the above constants into the equation and it computes a Q-sub-t amount. You then do the same thing for t=2. That gives you the second year's quantity of oil remaining in the ground. You subtract the first Q sub t from the second Q-sub-t and that gives you the first year's pumping.

You repeat for t=2, etc. etc. and compute all of the individual years by subtraction from the following year's amount left. If you know how to run a spreadsheet program, no problem getting this to happen all at once.

Q-inf is the "ultmate remaining reserves" estimate. reservegrowthrulz will tell you that this number is not really a constant--it shifts over time, depending on new discoveries, etc. but for now, we will pretend it is a constant so you can get your assignment done.

t-1/2 is a "guesstimate" of the peak year. In this case, it's 106 years after year 1, which is 1900 therefore it's 2006 for the purpose of the model.

The other numbers, k and n, are just fudge factors. when I did the original calculation I used MS-Excel to compute it. Once you have your spreadsheet set up, you can experiment with different n and k values, and see how it changes the shape of the curve.

When I do these, I do one step at a time per cell when setting up the spreadsheet. It is uglier, but less chance to screw up.

This narrative describes the virtual flow of oil through the system from discovery to production, providing a problem domain description to how an oil peak occurs.

Precondition: Oil sits in the ground, undiscovered.
Invariant: The amount of oil in each of its states cumulatively sums to a constant.

The states of the oil in the ground are fallow, construction, maturation, and extraction.

1. A discovery is made and an estimate of the quantity extractable is made.

2. The oil sits in the fallow state until a decision is made or negotiated as to what to do with it. Claims must be made, etc. This time constant we call t1, and we assume this is an average time with standard deviation equal to the average. This becomes a maximum entropy estimate, made with the minimal knowledge available.

3. A stochastic fraction of the discovery flows out of this fallow state with rate 1/t1. Once out of this state, it is ready for the next stage (and state).

4. The oil next sits in the build state as the necessary rigs and platforms are constructed. This time constant we call t2, and has the once again an average time with standard deviation equal to the average.

5. A stochastic fraction of the fallow oil flows out of the build state with rate 1/t2. Once out of this state, it is ready for the next stage.

6. The oil sits in the maturation state once the rigs are completed. The maximum flow can not be achieved initially as the necessary transport, pipelines, and other logistics are not 100% ready. This time constant we call t3.

7. A stochastic fraction of the built rig's virtual oil flows out of the maturation state with rate 1/t3. Once out of this virtual state, it is ready for the next stage of sustained extraction.

8. The oil sits in the ready-to-extract state once the oil well is mature.

9. The oil starts getting pumped with stochastic extraction rate 1/t4. The amount extracted per unit time is proportional to the amount in the previous maturation state.

Post-condition: All oil eventually gets extracted at time=infinity. But because of the proportionality extraction rate assumed, this decline only asymptotically approaches zero at long time periods. Also, the cumulative amount extracted at time=infinity equals the total discovered. However, since we never achieve infinite time, cumulative extraction never matches cumulative discoveries.

Each one of these states can be considered as a reservoir with a capacitive time lag associated with the time constant set for each stage. In stochastic terminology the flow approximates a Markovian Process.

The extraction from the final stage gives the average production level. Since Markov processes have well-behaved linear properties and are conditionally independent of past states, an entire set of discoveries can be applied as forcing functions to this process flow and the result will be a convolution of the individually forced solutions.

The final production profile over time approximates the classic Hubbert curve. This narrative explains in very basic terms how and why the peak gets shifted well away from the discovery peak. However it is not symmetric, as the nature of time causality rules long negative tails out.

Equations here:
http://mobjectivist.blogspot.com/2006/0 ... place.html

Yes, and the oil companies refer to it with the euphemism "reserve growth" instead of the more correct "badly made estimates revised upwards". In politics, that is called framing the issue. So who made the bad estimates in the first place? Ever heard of the expression GIGO?

Give us non-garbage input and we will make hay. Remember, this is just an exercise in bean-counting. I suppose you can denigrate Campbell as being a misguided bean-counter, but eventually someone will take his place that understands how the "yanking of the chain" occurs, and can argue his way out of a paper bag when faces with the continuously revised bad data courtesy of the oil companies.

The current framing is this : "Campbell is wrong because he ignores the magical property of reserve growth"

As if reserve growth is some intrinsic property of an oil field, instead of errors compounded by the humans behind the wheel.

The correct counter-framing should be: "The oil companies continuously provide bad estimates of discovery yield, making it hard for anyone to predict oil depletion"

I have no problem equating reserve growth with "continuous bad estimates of discovery yield" because really, thats what it is.

Myself, I don't care what you call it, what I care about is it exists all over the place, its net positive, its a big amount of addition to reserves relative to new discoveries, and Colin hasn't even been able to guess at the correct numbers 10 years in the PAST yet.....so what kind of incompetence is that? 10 years after the fact and he STILL doesn't know what the size of the discovery is. Yup...I'm sure them accountants could do alot better....

Now that you won't even admit to having read anything Colin has written, I don't suppose you are familiar with Arringtons work? Attansai and Root at the USGS, particularly the Enigma paper in 1995?

And pup, those were some really nice graphs you provided. I like'em. Seems like it makes the point better than I ever could that trying to fit Hubberts curve to production data is a pretty useless exercise.

Regarding the US crude oil production curves, Staniford was able to make a very good fit over a few orders of magnitude using a gaussian:
http://www.theoildrum.com/story/2006/1/8/25235/83999


As for temporal properties of this curve over time, as SS noted it has the property that


where t0=PeakTime. This relationship reads that the production increase slows down over time linearly, but also scaled by the amount of production at that time. At t=t0, the production increase turns into a production decrease.