And then of course as a counter argument to your comments (Luis) about the US HL, we have the two legged UK HL.  The twin peaks of UK production are in fact related to the oil price crash of 1986, followed by the Piper Alpha oil rig explosion.  But production had started to turn down before Piper Alpha - and effect of low investment / low price.


However, with respect to my comments in post above, we are probably talking detail here.  Prolonged life of off shore platforms must lead to higher recovery - but late on in field life we are talking about extending dribbles of oil.  The bulk of field production is already in the bag - we are just talking about a few more percent - but it all counts.

The UK is, in my opinion, a very good example of HL working perfectly, not the other way around.

You have two distinct discovery cycles. If you model production with two logistic curves you get the double peak perfectly.

On the €: no matter what the price is, a resource recoverable with positive energy yield will eventually be extracted, if not sooner, later. That's why the Hubbert curve is successful.

I guess I got to disagree then Luis, if you were sitting with your UK HL back in 1986, you would draw the wrong conclusion.  That is not to say that the technique doesn't work, just that you need to be aware of limitations.

I would argue the twin UK peaks are a function of oil price - impacting both new field development and exploration in the mid 1980s.  Resulting in production falling and a bi-modal discovery pattern.

The graphs:

I think a second cycle of discovery was already clear in 1986.

Gentlemen,

IMO, the key problem with the UK data set is the earlier data were just too immature.  As I have pointed out many times, the overall North Sea crude + condensate production peaked right in the 50% of Qt range (EIA data).

BTW, note that Stuart's HL estimate is quite close to lower Kuwait estimates that are being used.

Luis,

You did very good work on the oil export article.

Re:  Laherrere's HL plot

Note that Texas, the Lower 48 and Saudi Arabia showed the same type of inflection right before they started showing lower production.

Its a good thing most oil production in the world is done on the basis of maximizing URR, and not on shareholder wealth.  Your going to see oil peak significantly past 50% Qt because a greater portion of oil was both discovered and produced outside of NA and Europe 'the shareholders wealth'.
Isn't this a circular argument?
The production didn't follow a Hubbert curve because the field was immature.
We judge the field to be immature because it failed to follow the Hubbert curve.

In June 1989 1t looked like UK Qt would be 12.5Gb and cumulative production was 8.7Gb, 70% of Qt as judged then. If Qt and hence maturity can only be judged in retrospect the Hubbert technique loses all its predictive power.

If there is clear prior reason to fit a double Hubbert curve from examination of existing discovery data then this is possibly very valid and useful, but there is a great danger of just trying to fit the historical data with multiple curves. With sufficient free parameters it is possible to  get a magnificent fit to any data but as Stuart Staniford pointed out historical fit does not prove predictive ability. He described over fitting as 'pure epicycles' in reference to medieval attempts to force apparent planetary movements to fit a geocentric cosmos.

"Isn't this a circular argument?"

IMO, extrapolating from a P/Q intercept of 30%, is, in a word, ridiculous.  Again, we have people obsessing over a small sub-basin,the UK, with a complex producing history becauase of the Piper Alpha accident, when the overall basin, the North Sea is showing a perfectly predictable linear HL pattern.

I personally see no reason to use the UK as a model for anything.


" Again, we have people obsessing over a small sub-basin,the UK, with a complex producing history becauase of the Piper Alpha accident, when the overall basin, the North Sea is showing a perfectly predictable linear HL pattern."

I have often wondered if a similiar argument should not be made about Saudi Arabia, the "Neutral Zone", Qatar, U.A.E., Kuwait and Iraq.  It can be argued that there are at least.  All the above are political partitions placed over a much more interconnected geological area.  The reports (repeated even by Simmons) that when the Kuwait oil fires were ignited there were pressure drops in fields in Saudi Arabia indicated that the Western Persian Gulf area may be much more interconnected that once believed.

Is it possible that we are letting the Persian Gulf countries get by with a "shell game" (whose country has the oil under it, watch 'um close and step right there and put your money down....") when what we are talking about is a small number of big pools with a lot of extensions....any geologists want to take a shot at it...., remember, the oil and gas don't care whose country it is under, it still does what it does....

Roger Conner  known to you as ThatsItImout

On the €: no matter what the price is, a resource recoverable with positive energy yield will eventually be extracted, if not sooner, later. That's why the Hubbert curve is successful.

It really is not as simple as you make out. For example, you need to consider upfront capital costs. There are projects that have positive EROEI, but because the payback time is too long, no investor will consider them worthwhile.

The Hubbert curve says that the low hanging fruit are picked first, but it does not guarantee all the fruit will be picked.

Thats for sure. I could show you a bunch of heavy oil reservoirs in Texas that have never been exploited because the economics were not right when they were discovered. Now things have changed with the increase in the price of oil, but they're still a hard sell. Any good geologists out there that want to shoot craps with me? My Email is on my info at the oil drum file. Bob Ebersole
In parctice, I noticed that the HL result is usually unreliable when the P/Q values are above 10% (i.e. very immature production) which is the case for the first line in the UK but not for the second line.