Posted: Fri Nov 05, 2004 1:56 pm Post subject: "Experience Based" Decline Model
Per the interesting suggestion by Missinglink the other day, here is a model decline scenario based on the following methodology:
a. Use BP production data in the 2004 statistical review.
b. Throw out all nations which have not peaked yet, also those who have recently “peaked” to eliminate any nations with a “faux” peak that will throw us off. This is per the Simmons theory that peak will only be observable after some length of time after it occurs. For the purposes of this exercise, I used 3-5 years.
c. Index the “peak year” at 100, and the remaining years proportionally. The result is “production rate as a percent of peak year” for the decline curve for each of the years. This takes away the scale differences between big and little production areas.
d. Define “peak year” as year 1, and calculate the average percent of peak production for all “post-peak” nations for each year to determine the shape of the decline curve based on the production of nations that have already experienced it.
e. Assume the global decline will be close to that of the “composite decline average” , and compute the global curve based on this. For the purposes of this exercise, this assumes the Boone Pickens theory that PO is “now”., which some may or may not personally subscribe to.
I’ve included the model below, and plotted it with the actual historical data, plus a Verhulst curve based on the “actual” data up until 2003. By the way, “n”, the number of countries, was 24 which is not a very big sample.
Observations:
a. Once decline sets in, on average, it is really noticeable. The first year, post-peak, gave an average of 94% of the peak production, and by the third year, on average, production averaged 87% of the peak amount, a 13% decline. The standard deviation was 15% however, so obviously a lot of noise in the data. The extremes were Italy, which by year 3 post-peak was down to 77%, production, and the US which still managed to squeeze out 96% of production 3 years after the supposed “peak”. TSHTF for the US about 4 years after that when it was down at about 85%. UK is headed the way of Italy.
b. Once the decline actually sets in, the continued steep dropoff occurs until production hits a plateau of about 65% of peak production. Very few countries have made it out this far, though, and the results are thrown off by Venezuela, which evidently found another oil field at some point and started pumping at 90% as recently as 1999, and Iran, which never did recover all the way from the revolution. So, no way to tell if this is a real phenomenon or not. Maybe this represents some economic choices made by the US based on cost and the apparent availability of a backup supply.
c. Obviously if you had the data by field, rather than by country, you could get a lot better curve and do a lot better job of determining the shape of this decline.
d. Here is a scenario: You get to the peak, for a couple of years you use technology and frantic effort to keep production level, but at some point geology takes over. The resultant decline is steeper than it would have been if you had let things happen more or less naturally. A comparison of the actual data vs. the Verhulst curve suggests this might be the case.
e. Here is another scenario: You get far out after the peak and you find a way to keep your production at about 65% of the original for a few years, and at some point even these efforts fail also, and the resultant dropoff on the “far out” other side of this is even steeper. This has not actually happened anywhere yet, so speculation only.
f. There might be some guinea pigs in the data identified as “other middle east”, although small, and not broken out by country. The “other middle east” curve is almost exactly the same shape as the “average” curve, which means that if there is some geological variability associated with this decline phenomenon that is peculiar to the middle east, you might expect the “big middle eastern oil fields” to behave the same way as the aggregate of these small guys.
g. It would be interesting to know the “walk away point” the point at which production was so minimal that you just shut off the pump and walk away. It’s not on here yet.
h. In any case, the observed behavior does not exactly follow the nice, neat mathematical model, which is what we suspected.
Joined: Oct 16, 2004 Posts: 1422 Location: Appalachian Foothills of Virginia
Posted: Mon Aug 01, 2005 9:34 am Post subject: Re: "Experience Based" Decline Model
pup55 wrote:
Per the interesting suggestion by Missinglink the other day, here is a model decline scenario based on the following methodology:
a. Use BP production data in the 2004 statistical review.
b. Throw out all nations which have not peaked yet, also those who have recently peaked to eliminate any nations with a fauxť peak that will throw us off.
It is unclear to me why you are discarding seemingly important data. The results would be extremely limited at best.
The intent was to develop a "typical decline scenario" therefore reasonable to include the nations that were "in decline".
A secondary idea was to look at various decline curves and see how they do or do not conform to the Hubbertian "bell shaped" decline model.
What was found out of course was that there is no "typical decline" because the shape of the decline curve varied by location. Some curves were more like a "cliff" while others showed more gentle downslope.
This seems a very useful estimation of the general format of decling supply that we'll face.
Two things I'd ask, one that you switch the data table result into a graph,
(so much easier to consider)
and the other would be to graph the boe eqivalent that is missing, from, say, a 2.5% global growth rate starting at year 1/.
The latter would give some idea of the scale of development of replacement liquid fuels required to maintain the global economic growth required for that development.
pup55 may I have the Verhulst ?
The other ones scare me to death _________________ "Nuclear power has long been to the Left what embryonic-stem-cell research is to the Right--irredeemably wrong and a signifier of moral weakness."Esquire Magazine,12/05
The genetic code is commaless and so are my posts.
Its a very sobering view, for all its nothing approaching a detailed forecast.
Its inverse, that is a graph of the BOE growth required to maintain economic growth, should maybe be on the front page of the site, as it has to be a telling curve.
A lot of people seem inclined to clutch at techno-straws (with full media backing) rather than face structural change with its admission that the ideology of market-driven growth is a cock-up on a global scale. Maybe that curve would be of assistance to them ?
We had some previous discussion on this model in the above thread.
You are right. We had better root for the verhulst.
I was expressing a preference for the Verhulst model
Yeap I know about the discussion. Since running those numbers I consider Mathematica a wicked evil piece of software.
Backstop:
It is time for the structural change, but what we know?
We are a bunch of stupid POilers, environmentalists, most anti-capitalists who carry a message that the majority of the world does not want to hear: It is time to cut back.
Doomerosity Level 7: Roaches will inherit the earth. Meet the most intelligent species on earth in 50 years from now. One click away _________________ "Nuclear power has long been to the Left what embryonic-stem-cell research is to the Right--irredeemably wrong and a signifier of moral weakness."Esquire Magazine,12/05
The genetic code is commaless and so are my posts.
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
News
Discussions
Resources
Members
Forum FAQ
Search
Usergroups
Profile
Log in to check your private messages
Log in
