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

Wednesday, January 04, 2006

Norway offshore depletion

I gave a shot at modeling the Norway depletion curves starting from the Laherrere discovery data.

In the oil shock simulation, I used means of 5 years for the fallow, construction, and maturation periods, and a 10% depletion rate for years up to 1992. After 1992, like for the UK, I doubled the extraction rate over a 10+ year period. The fit is decent and it gives much more insight than the questionably derived logistic curve formulations.

Unshocked -- 10%

Shocked -- 10% to 20%

Note how close the profile of the shock perturbation approaches that of the UK North Sea model (chart to the right). In both cases, the increases in extraction rate occurred right around 1992, and essentially targeted the same 3 MBls/day sustainable production rate (competitive pressures perhaps?).

Like in the UK, the Norway model shows how the offshore areas can suffer rapid depletion. From the range of the parameters, the Brits developed and matured their rigs much more quickly than the Norwegians. The necessary increase in the extraction rates as the production curves started leveling off in the early 1990's became quite obvious; this basically forced the hands of each of the producers to pump harder. Without new discoveries, and continuous hammering on the extractive technologies, they will certainly see a steep decline before they put the expensively maintained rigs into mothballs as the North Sea oilers cut and and run and forever minimize their losses.

I got motivated to run this simulation from a thread on the PeakOil.com message board.

Update: Apparently, those in the oil industry call this falloff by the phrase pump and dump because of the high cost of extraction [link].


Professor Anonymous Anonymous said...

An interesting question might be what other sorts of inferences can we gain from the shock model? I agree that your work has been interesting and appears more accurate than the logistics curve, but the curve (regardless of reasons) has produced some reasonable fits to data thus far. It has also produced the derived "Hubbert Linearization" which has also been useful to some extent in giving us a means of estimating how much oil is left.

Since one of the more important questions facing society today is how much time we have left to transition ourselves off fossil fuels, such forward estimates are of great utility. Obviously, if your shock model could be used to produce better forward views then it would be of great benefit to society as we strive to deal with this issue.

12:34 PM  
Professor Blogger @whut said...

The logistics curve has only worked as a heuristic. The Hubbert linearization has got some unfortunate asymptotic characteristics that push the time component completely out of the picture. So you can get the URR, but how do we temporally approach that point? How do we correct for new discoveries coming on line?

The only (and I mean only) way that we can get any kind of predicted power in the time domain is to use the shock model. You are screwed if you start using the Logistic curve to do any subtle temporal dynamic studies. So I agree with your bottom line.

10:31 PM  

Post a Comment

<< Home

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