Update on the 2007 CERA projects

Here are news updates on the 2007 CERA projects.

High Production Scenario for CERA 2007 Listed Projects.

  1. No change on Plutonio so far. Still on track for the middle of 2007.
  2. HO suggested that Rosa will not result in any new production, rather just extend production at Girasol. However, Total says that while Rosa production is being tied back to the Girasol FPSO (as was production for the Jasmin field), this is leading to the FPSO capacity being increased to 250kbpd. That was previously at 180kbpd, so it seems fair at a minimum to count at least the additional 70kbpd as a new project. However, Rosa is expected to increase to 150kbd, presumably as Girasol declines and makes space within the FPSO capacity. So I'm going to account this as a slow ramp to 150kbpd, and we'll have to account for Girasol's declines separately.
  3. Lobito/Tomboco is the second phase of a development with the same production facility as used for the Benguela/Belize fields. According to this reference, the capacity expansion for Lobito/Tomboco is 80kbpd, so that's what I credited it at.
  4. Marlim Sul Mod 2 is actually not expected till 2008, according to Petrobras' business plan (hat tip cerqueira).
  5. Ditto Marlim Leste.
  6. But Roncador P-54 is indeed slated for 2007.
  7. Khursaniyah is claimed to be on track for 500kbpd by Q4 2007. There's a nice description at Hydrocarbons Technology. Greg Croft describes the history of these old fields which are now being redeveloped. Here's an interesting piece in Saudi Aramco world which discusses Khursaniyah declines in the 1960s. So this is definitely a zombie field being brought back to life. Costs are rising as the drilling program has turned out to be more complex than anticipated.

  8. The dispute over the alleged landmine problem at Azadegan drags on. Politics over Iran's nuclear program may possibly have a role -- explicit threats have been made. Ah, but at least some Venezuelan oil may come of it. Clearly this is one where the low case needs to take a big hit, given the political risks. It sure is a huge field, though -- 26gb estimated recoverable. In the high case, phase I has first oil at 50kbpd in Sep 2007, and reaching 150kbpd a year later, and 260kbpd by the time phase II is fully implemented in 2012. The reserve is going to be barely scratched at that rate (though some reports say there's only 9gb of oil (plus astronomical amounts of gas)).
  9. Kikeh is on track, and there's a little more Malaysian deep water oil on the way.
  10. I think HO might be misreading this reference - the 300kbpd refers to the capacity of the oil terminal, not the Vankorskoye field. Looks like there is close to a gb of recoverable oil. First oil is anticipated for 2008, and could go for 275kbpd. However, the schedule seems riskier than most, since they haven't yet determined how to transport the oil out of there (earlier, they were planning to build a pipeline and be shipping by 2007). So the low case adds two years delay.

Low Production Scenario for CERA 2007 Listed Projects.

Finally, here's the spreadsheet, for anyone who wants to see the work in progress.

Well Stuart, should we then view ourselves as having a decade of spiky prices before things settle down to constant increases? Do we have a decade or not?
It would seem to me that there are far more factors than just peak oil at play here in answering such a question. Right now it appears obvious that we are headed towards recession and we're clearly seeing "demand destruction" occurring. If this continues and we do see some increase in available supplies, then we might limp along in a recessionary state for years. Or we might bounce through shorter growth/recession cycles with lots of spikey economic behavior. This happened under different circumstances in the 19th century in some places. What we don't know is how bad the recession will be as well as how much demand is reduced and the demand reduction is the key other factor in trying to model how long existing oil is going to "hold out" before things begin sliding downward.
Good work, Stuart!

Now we can describe a few scenarios for 2005-6-7:

  1. Straight low, middle (average), and high cases for each year.  If there's a 50% chance of low or high, then the pure low or pure high outcome each has p=12.5%.  
  2. A "lagging high price" scenario, whereby delays in 2005 production are assumed to be technically insurmountable in the short time available (low-med case) but 2006 and 2007 production is assumed to be accelerated due to higher prices (med-high 2006, high 2007).  Given a year or two, more money can solve a LOT of problems...

Any other suggestions?

-- Silent E

Sorry guys: I'm still quite some way from having any clue as to the global picture over the decade. I'm turning up a lot of projects even in the 2005-2007 timeframe that CERA doesn't list explicitly. So I need to go through those and model them. Then I need to work out through 2010 (which seems to be about the limit of visibility into new projects). Then I need to work back a few years (not quite sure how many yet) - because I need to understand how quickly the base production declines when nobody does any projects, and the only way I can see to get a decent sense for that is to analyze all the projects for a number of years for which there are production data. After all that's done, I'll have some kind of basis for saying whether there are enough projects going forward or not. There's probably over 150 projects to go (I've done 30). So don't hold your breaths. Hopefully, some kind of useful partial conclusions will emerge as I go.

At the moment, probably the only thing I can say with any confidence is that CERA's explicit project lists are of about the same size as they said, but there are some significant delays in some cases. Peak production from these 2005-2007 projects doesn't come until 2010-2011 (somewhat because of delays, and somewhat just because there's usually a significant ramp-up time from first oil to peak production). Of course, delays per se may not be significant if they are normal for the industry and the market has allowed for them. Only a trend of increasing and unexpected delay is likely to cause real problems.

Silent E:

The minimal time I've run across for a big project so far is about 3 1/2 years from commitment to first oil (with another maybe 12 months to peak production). Thus no amount of money can produce much increase in the oil supply in less than four years, unless there is spare capacity. Thus the situation through about 2008 is largely set by decisions already taken. The situation we are in today was decided in the 1999-2001 timeframe.

Stuart, thanks for slaving over the details.

Do all projects come on line at the same rate, around 3.5 years?  Or do the more difficult ones (at sea, tight rock, heavy sour) take more time  to develop (get delayed)and is this built into the projections?  

Do you have a feel yet if the past will be a good indicator for timelines, when many of the new fields were bypassed earlier for some negative economic reason, i.e. is there any basis to expect a higher delay rate moving forward?

I would say my subjective impressions are roughly as follows:
  • The bigger the project, the longer from commitment to first oil.
  • The bigger the project, the longer from first oil to peak production
  • The bigger the project, the longer production will get spread out over. Ie small projects ramp up fast, have a very brief plateau, and then decline fast. Big fields are likely to be produced with a longer depletion profile.
  • Delays are as much or more about political issues as about technical ones. If the industry has free rein, it is a pretty well honed machine for deciding how much oil is in a place and exploiting it. But the great bulk of the new projects are in politically challenging environments. The industry faces bureaucratic restrictions, arbitrary changes in rules, corrupt governments, environmental restrictions, and requirements to use home country personnel and contractors who may be less experienced than internationally competitive subs. Technical issues primarily seem to arise when the industry is first trying to do new kinds of projects, or encounters conditions it hasn't seen before.
  • It does seem to me delays and difficulties are likely to get worse. The current generation of oil projects is coming primarily from deep water off the coasts of Africa, Brazil, etc, as well as Russia, the Caspian, and the Middle East. Once the deep water is mostly produced, it seems like everywhere else is going to be either politically or technically very challenging - there's a lot of oil in the Russian Arctic, but there's no infrastructure up there, lots of ice, and the Russian political environment seems very challenging. Russian projects seem particularly likely to incur delay in the recent past. There may be oil in deep water off the US coasts, but the industry faces the considerable influence of the US environmental community. There's a lot of oil in Iraq, but good luck producing it.
  • Doing this analysis is definitely given me a slightly awed sense of respect for the oil industry. They deliver dozens of new projects a year and each of those is a massive multi-billion dollar engineering project in enormously challenging conditions. And all they get from the rest of us is bitching and whining if they screw up in any way. We are very spoilt.
  • The gap between discovery and peak production seems to be going down. I will analyze this statistically at some point, but my sense is it's now under a decade from discovering a field to reaching plateau production of it (unless it's particularly undesirable to produce for some reason - tiny or lousy oil/rocks, or it's in an especially inhospitable political environment (Sudan, Iraq, Iran).
  • In general, hardly any of our new oil is coming from a politically stable place with clean government and tolerably happy and free citizens.
Doing this analysis is definitely given me a slightly awed sense of respect for the oil industry. They deliver dozens of new projects a year and each of those is a massive multi-billion dollar engineering project in enormously challenging conditions.


Oh yeah. Kashagan (in the Kazakhstan part of the Caspian sea) is going to be a textbook example for years to come of the complexity of these projects (not sure in which year you put it in, but 2007 is possibly it)

  • extraordinarily high field pressure
  • very high sulfur content
  • located in shallow waters that are alternatively frozen, liquid (10m or so) or mud
  • the nearest town is 400 km away, and the nearest international airport 1,500 km away; winter lasts 10 months and summer is worse (mosquitoes and stifling heat)
  • you need to cross at least 2 countries to get the oil to any market
  • the sponsors include Shell, Exxon, ConocoPhillips and Total (plus others) as minority non-operating shareholders
  • China, Iran, Russia and the USA (well, the West) want the oil to come to (or via) them.

Should be fun to watch.
Yes, and then there's the political environment in Kazakhstan itself. I have Kashagan for 2008 but the risks are considerable as you note. It is enormous, however.
Re: "the nearest town is 400 km away..."

It's not for nothing that the whole region is sometimes called pipelanistan....
Superb Analysis Stuart, im very excited by this!
Using the spreadsheet, and assuming the "most likely" case is Low+0.4(high-low), and taking a decline from existing fields of 2 Mb/d/yr and adding another 1.5 Mb/d from refinery gain plus NGL plus tarsands, and adding contribution from 2004 projects, 2010 total production would be up 2.1 Mb/d in the best case, down 1.3 Mb/d in the likely case and down 3.5 Mb/d in the worst case. Even going back and adding increases for 2006/7/8 from projects that started producing in 2002/3/ we are down every year from 2006 in the likely and worst cases. It seems like these projects are not enough to offset declines.  Murray
I tried to redo the above a little more carefully, and to average with an independant analysis I did previously. The result is practically flat from 2007 through 2012. Any year in that group could be the peak, but would be little dufferent from the others.  Murray