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

Tuesday, October 17, 2006

Khebab plots

Khebab at TOD (linked via Big Gav) presented a range of peak estimates all on one plot.

I placed the Oil Shock Model data as stars on the graph. My estimate -- based on discovery data -- remains on the pessimistic side, agreeing largely with the ASPO data from October 2004 and Bakhtiari from 2003 (both using conventional plus NGL as the production total). ASPO has since upwardly corrected their estimates, something that I have really no basis for, as discoveries have not gone up much since late 2004.

8 Comments:

Professor Blogger Big Gav said...

I'm glad you posted that follow up - one of my issues with the original is that it didn't include your model (or Laherrere's for that matter).

12:35 AM  
Professor Blogger @whut said...

Hmm, Laherrere also likes to use discovery data as input. Do I detect a bias here?

Good catch.

10:39 PM  
Professor Blogger Big Gav said...

Don't all the peak oil modellers (from Hubbert onwards) use discovery data as input ?

Anyway, Khebab recognised he'd omitted Laherrere and said he add him to the graph next time round - maybe you should request rectification of his overlooking the oil shock model too :-)

4:16 AM  
Professor Anonymous Anonymous said...

nice site


http://exportersinfo.com

10:51 AM  
Professor Blogger @whut said...

Big Gav: One of the interesting features of the Logistic model and other URR-based models is that they intentionally do not use discovery data. It is curve fitting on production data alone. I broad-brushed this a bit but only to make a point.

8:41 PM  
Professor Blogger Big Gav said...

Ah - there you go - demonstrating my ignorance once again.

Because the ASPO newletters always show discovery and production data I assumed Colin Campbell used both to create his depletion model curve.

I don't see how you can use past production data alone - surely thats a function of policy and price and an inaccurate guide to future production (as long as we are pre-peak anyway).

4:54 AM  
Professor Blogger @whut said...

Especially on the ones that use the URR asymptotic curve fitting, like the Hubbert linearization in fashion these days. These simply use the inflection point and slopes on the cumulative production curves to map to a Logistic equation. I always wondered what happened to the discovery components and other dynamic inputs.

9:16 PM  
Professor Blogger Khebab said...

WHT,

I was planning to add your forecast in my next PO update (November).

Could you send me the production profile predicted by your model, ideally in two columns (year, daily production).

my email: Khebab@theoildrum.com

2:07 PM  

Post a Comment

<< Home


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