[[ Check out my Wordpress blog Context/Earth for environmental and energy topics tied together in a semantic web framework ]]

Sunday, October 09, 2005

Shock Model Applied to USA lower-48

The Minnesotans for Sustainability compiled a bunch of interesting oil depletion charts on a single page. I decided to test out the oil shock model on discovery data from the USA lower-48. This contains good historical information of an oil depletion profile past the peak (if you haven't paid attention, our own!).

I used exactly the same base parameters as I used for the fit to the world data, namely each of the fallow, build, maturation phases set to a mean of 8 years, and the average Markov extraction rate set to 0.07 of volume/year (~14.2 year 1/e time). The "unshocked" fit to the data turns out to match the observed production quite nicely.

But the fit can get better. First, ignore the poor fit during the 1930's (depression era). Next, as Laherrere has noted, we can likely account for the perturbations. The first one occurred during the late 1950's:
The initial "voluntary imports quota" of 1957 did not worked and President Eisenhower made them "mandatory" in 1959. The US were disconnecting their domestic oil market from that of the rest of the world, leaving to the large foreign producers the responsibility of managing the market, what they did by creating OPEC the following year, in September 1960.
So that clearly a quota-driven reduction in extraction occurred in the late 1950's, in other words, the first local USA shock. During the 60's, with the quota in place, the oil industry carefully modulated extraction (note that a big dependency on foreign oil did not exist), yet with a peak quickly approaching, extraction rates had to increase to make up for the continuing economic expansion at the time (and to feed the Vietnam War/Great Society guns&butter machine). Once the peak hit, the prorationing reached 100% and the oil industry moved to supplant domestic oil with that from foreign sources.
In March 1971, the balance of power shifted. That month the Texas Railroad Commission set proration at 100 percent for the first time. This meant that Texas producers were no longer limited in the amount of oil that they could produce. More importantly, it meant that the power to control crude oil prices shifted from the United States (Texas, Oklahoma and Louisiana) to OPEC. A little over two years later OPEC would through the unintended consequence of war get a glimpse at the extent of its ability to influence prices.
Actually, 1970 marked another turning point:
The transition. It started in 1969 when the Santa Barbara oil spill triggered some important US environmental laws (Clean Air Act, Clean Water Act). They slowed down the development of domestic energies (mostly coal) at the very time when the US indigenous oil production was going to peak (1970) before decline (see graph 1416). US net imports, which since 1959 were kept at about 20% of US consumption, were relaxed and grew by 25% each annum from 1970 (3.15 Mb/d) to 1973 (6.02 Mb/d), a growth perfectly matched by Saudi net exports. World oil market became tight, as revealed by the successive upwards price revisions in Tehran, Tripoli and Geneva before the October 1973 oil price explosion. The "Club of Rome" was right, oil was unable to fulfil all energy needs, but it was wrong because oil was only scarce in the US 48 lower states but abundant elsewhere.
[...]
Transport demand in the US was hardly affected by the first oil shock because the oil quota put in place by President Eisenhower in 1959 isolated the US oil pricing system from the international oil market However, demand fell strongly at the time of the second oil shock because its timing coincided with the liberalisation of the US oil prices.
I used the following perturbation profile to try to fit to this evolving scenario.

Which gives this shock fit:

What does this prove? I don't have all the answers, but it does show that the oil shock model demonstrates good scaling properties. If we can use the same properties on USA data as I did for the global data, it suggests that the rates have a universal property. It makes a lot of sense, if we consider that a few companies own the state-of-the-art in oil drilling technology and use the same expertise worldwide.

It also gives substantiating evidence for us to once again reconsider the traditional Logistic model to understand oil depletion.



This presentation (PDF), provides some good background on the interaction of the quota system with energy economics. And this one gives some history on the politics behind the Texas oil men and the rise of Halliburton.

6 Comments:

Professor Anonymous Anonymous said...

This is interesting - thanks again!

And I found here again something: the period of 1900 - 1930. It belongs really to an another curve and resembles a lot of the world oil production curve before 1972. The production grows very fast, exponentially, and extrapolates to an earlier peak than the actual one.

The US production grew rapidly up to the beginnings of the '20s, had some trouble int he middle of the decade but resumed relatively rapid growth for the end of it. Then comes a crisis and a permanent trend break. The production never returns to the same path.

So what does this mean? Was the depression of '30s an energy crisis after all? People living then could not see it this way. There were an oil glut and the decreasing production was attributed to that. Only afterwards we see the fact that there was also a permanent trend break. And we know for sure that in the Europe there was at the same time a complex coal crisis.

I am inclined to think that the energy supply problems are connected to economic recessions and depressions but the connection is indirect and very difficult to see. Energy problems - mostly turning points in growth rate - can induce recessions and these recessions mask the energy problems. And the other way round - an economic boom increases energy demand and masks the fact that there is also the supply side - that there must also be an ability to meet the demand and sustain economic growth.

1:33 AM  
Professor Blogger @whut said...

The 1930's also saw the rise of prorationing via the Texas Railroad Commission and aborted trials on establishing quotas. The last link has some historical details on that. I would like to study that history some more, and agree with you that this era may provide some more insight into how economics influenced oil and vice versa.

9:51 AM  
Professor Anonymous Anonymous said...

Interesting indeed. Has anyone attempted to do this kind of analysis before? If not, maybe you should consider writing an actual paper on this subject and try to publish it in a journal.

5:06 AM  
Professor Anonymous Anonymous said...

I have one more question about the US oil production history. This time it is the years 1965 - 1970 and spare capacity / overproduction. We are here at the causes of the "upward shock".

The 48 states production made a last jump during that time. This made the peak and the consequent fall rather sharp. What really happened? Was it so that in the end the quotas was increased so that all the existing spare capacity was used up and some extra production was brought online at the same time? Did the producers shift to overproduction after that?

This is important because on world scale we might have a similar situation. The oil supply increased exceptionally during 2003 - 2004. Did OPEC countries use up there spare capacity and was Russia overproducing? The oil markets seem to show that there really is not much spare capacity left.

The latest Peak Oil forecasts seem to agree that we should expect a rather smooth and flat peak that would be difficult to predict exactly and even to discern. But the US peak was not like that. The possible combination of overproduction and exhausting spare capacity could lead to a sharp peak also globally.

7:02 AM  
Professor Anonymous Anonymous said...

Shouldn't it be no surprise that the US production data fits so nicely in your model? After all as far as i understand most of the principal concepts for Peak Oil Analysis were derived from US production data, including the strong correllation between discovery and production you use for your analysis. Maybe another prodction area, for example the North Sea, would be better suited to test your model.

8:26 AM  
Professor Blogger @whut said...

The 48 states production made a last jump during that time. This made the peak and the consequent fall rather sharp. What really happened? Was it so that in the end the quotas was increased so that all the existing spare capacity was used up and some extra production was brought online at the same time? Did the producers shift to overproduction after that?

This is important because on world scale we might have a similar situation. The oil supply increased exceptionally during 2003 - 2004. Did OPEC countries use up there spare capacity and was Russia overproducing? The oil markets seem to show that there really is not much spare capacity left.


You nailed it.


As for North Sea, I was just looking at that data yesterday and will give it a shot later tonight. I won't guarantee a fit but that discovery data is so interesting in its detail that I can't pass it up.
http://www.mnforsustain.org/images/oil_lisbon_laherrere_uk_prod_discv_fig17.jpg

10:42 AM  

Post a Comment

<< Home


"Like strange bulldogs sniffing each other's butts, you could sense wariness from both sides"