Hmm ...

I think I've heard of someone mentioning this possibility :)

Note it probably was not really 4mbd drop in six months. Higher chance is the 2008 production numbers are bogus and its been a fairly smooth decline to this point since 2005. We are I believe dropping fast now but closer to 4mbd every 12 months. So the drop itself was not real. The real decline rate is steep but I think half of what the current numbers are showing. However the drop is accelerating how much is effectively unknown.

I've guessed anywhere from 4-8mbd in 2009. 2.5-3 is probably not a bad guess. But given our current situation with probably ample storage. I don't believe its anywhere near as high as claimed but this does not mean that OECD countries are not currently well supplied at least in the middle of the five year range if not higher.
My best guess is overall we are really between the middle of the 5 year range for storage and the top. I don't believe we are actually over the top any more than we where in 2007.

However for a lot of reason I actually believe that the current production numbers being reported are real since they are used to support the lie of OPEC cutting back.

I give the chance of OPEC announcing another 2mbd cut after having "achieved" its goals at on of the next meetings a very high probability.

So the bad news scenario is not actually as bad as whats being presented its just at the moment for political reasons we happen to be getting good numbers out of OPEC. non-OPEC however is now questionable.
The rapid build given what we know about underlying decline in non-OPEC fields is questionable.

One of my predictions was that if my theory is right that GOM production would collapse in 2008 so far reported production is showing that the new fields that came online completely covered collapse. In addition given the timing we somehow managed to make a perfect production recovery from the Hurricanes despite the financial turmoil. And the fact that after all other hurricanes some production was not brought back online.

So we now seem to have OPEC telling the truth and US production looking questionable. GOM numbers seem to high even with the new fields coming online. Here I think the new projects are close to peak production so even if the absolute value is wrong one can watch with interest how US production changes over the next several months. If there is a real underlying decline then one would expect that once the new fields are fully operational we would again see a steady decline set in. Probably from the wrong absolute value and maybe even with the wrong rate but if we are entering fast collapse and its real the GOM production overall should be crashing and the rapid ramp up of the deepwater fields will quickly be swamped by overall decline.

So you still have a mini regional collapse model that can be proven or disproven without looking at the whole model. If collapse is happening then GOM must be one of the first regions that collapse.

Of course the model could be totally wrong but given the current numbers we are rapidly reaching the point that we can start discarding lots of models about future production. The idea of fast collapse could well hit the dustbin in the near future. If we don't see continued collapse in production and high oil prices by the second half of 2009 at the latest I'd suggest its highly doubtful that this concept is correct. Given that it sets the real peak back in 2005 we just can't go much longer without fast collapse becoming obvious if its real. You can get a bump in production in the fast collapse model but the production from 2000-present is treated as the bump at the end. The bump in 2008 is not real you don't get that sort of brief increase with any stable production model you have to see a multi-year bump if you will from real production. Thats not to say that you might not get one from some seriously unsustainable production modes but that makes it even less relevant to a larger model and even more likely to have caused additional damage canceling out the brief rise.

So for now watching US production closely over the coming months will be interesting. How the GOM production changes once the large new fields have hit maximum production is of EXTREME interest.

Outside of my prediction of and announcement of a additional 2mbd cut from OPEC I expect after this if prices do rise that OPEC production numbers will again become unreliable but with US GOM production becoming a bit more reliable in its direction of change and even its rate with the absolute value off a bit.
With luck we get good reporting and can get a reliable slope. Given the way things are going I'd not be surprised in the least to see US production go up 500-800kbd and hit a flat line regardless of the price changes in oil. Same for US storage levels for that matter :(

We just have to hope that at various points in time the need to tell lies results in people publishing the real data as the easiest way to support the current lie. I think that periodically as you look at the political situation that different areas do report real numbers. Russia and Norway are two that generally are very consistent having basically no ulterior political driving force to slant the numbers.

The nice thing is these two act as proxies for a lot of the worlds production so you can believe that whatever is happening in Norway and Russia is probably the truth and that the rest of the world is not all that different. This suggest that most of the worlds fields have now entered decline at the minimum.

One of the interesting things coming out of the forthcoming update to the our (Khebab/Brown) top five net oil exporter paper is that there is not much difference between the middle and high cases regarding the post-2005 50% depleted mark (when about half of post-2005 top five cumulative net oil exports have been shipped). It's between the end of 2012 (middle case) and the end of 2014 (high case), or seven to nine years after hitting the apparent top five net export peak.

Indonesia hit the post-1996 50% depleted cumulative net oil exports point in only a little over two years after their final net export peak, and they were 100% depleted in 8 years (assuming that they don't get back into the net exporter category).

It would be nice to have some accurate net export data for the US in the 1930 to 1948 time frame (from discovery of the East Texas Field to zero net oil exports), but assuming that the Second World War was the peak for US net oil exports, the US became a net oil importer only three years later, in 1948.

Wonder if Jonathan meant IEA instead of EIA as the IEA data is (initially) behind a paywall. However most of the data is released to the public 10 days after the subscription release at http://omrpublic.iea.org/ with archived tables at http://omrpublic.iea.org/tablessearch.asp

Unfortunately only the current "Table 1" of the IEA's OMR is available in spreadsheet form. Everything else is PDFs. Would be really nice if someone knows of somewhere all the historical OMRs are archived in one big spreadsheet.

Rembrandt, do you know if there's access to more usable data downloads with the subscription version?

"...if they don't [crash] it's probably because the cost of production of the last barrel of oil (from these projects) is probably about 50 USD/BBL..."

I don't agree that the sunk cost of getting some oil into production keeps the price up. It may well be that the high costs incurred cause a company to become bankrupt as they can't fund these liabilities with lower income. These risks are the same for any long term capital intensive project, in fact there are lots of mining companies in trouble and the assets will just be taken over when they go bust.

Talking of long term, China seems to be stocking up on copper.

Thats assuming a constant capacity or constant energy manifold. If the energy level is decreasing rapidly and say EROEI is also declining rapidly then the predator prey model is running on top of a fast decline base energy level.

The system would then rapidly dampen to head strait up or down depending on how you look at it.

Given the magnitude of the recent price flux and economic change and if a fast collapse in oil production is true on this sloping playground if you will you would I think see.

1.) Energy prices up && production down && economy down.
2.) Economy down && energy prices down && production down.
3.) Economy down && energy pries up && production down.

Some small swings in prices because of three but in general the system dampens rapidly and collapses.

More of a Easter Island than a predator prey. I'm not saying you don't get one and two but they dampen rapidly and the system spends most of its time in state three with minor excursions.

Now if the real peak for oil was actually say 2010-2015 then yes we could get several more waves like what your saying but implicitly this means we at least retain capacity very close to peak production to provide the level playing field if you will for classic predator prey model.

I prefer to see it as the predator prey model being played at through a massive meteor strike :)

You have a numerically higher number of predator and prey species then wham and the natural model is played with a lot fewer players taking part in the relationship and headed toward extinction. The natural system is so disrupted and broken that its natural oscillations rapidly cause system break down and extinction as the feed back into effects of the strike. I.e at first there are number of dead animals to eat and the top predators appear to be fat living as scavengers ( Goldman Sachs ). But then they rapidly have to eat all remaining animals and even their own young ( The American consumer parasite ) to survive.

Goldman Sachs and Morgan Stanley peak oil aware.

Think about the implications.

Well the math is which ever chaotic model fits the system.

The closest is probably the logistic map.

http://en.wikipedia.org/wiki/Logistic_map

But the math or at least chaotic math is not predictive because its extremely sensitive on the value of the initial conditions. A small change in initial conditions can lead to exponential divergence in the values
the system takes over time.

http://hypertextbook.com/chaos/43.shtml

However my toy theory differs from traditional choas in the sense I'm also asserting that the initial conditions are ill defined they themselves have noise !

So its a noisy chaotic system.
I actually did a lot of work on something similar called stochastic resonance.

http://en.wikipedia.org/wiki/Stochastic_resonance

Generally studied with bistable systems but not restricted to those.

Next of course the systems are complex by definition this means they are capable of computation which at its most basic level means the system itself is capable of changing depending on the value of its input and some of the outputs are also inputs. At the minimum they are finite automata.

Now we start hitting recent research but you can see some disagreement in the literature here.

http://mulhauser.net/research/tutorials/turing/index.html

Observations About Finite Automata and Chaos

A few features of finite automata are worth keeping in mind when it comes to claims sometimes made about chaos:

* The state space of a finite automaton consists wholly of fixed points and orbits.
* Strange attractors do not exist in the state space of finite automata.
* Finite automata cannot display sensitive dependence.
* Finite automata cannot display chaotic behaviour, as formally defined.

Some stuffy professor wants to sweep this concept under the rug by moving the goal posts.

And in response :)

http://adsabs.harvard.edu/abs/1993chao.dyn..3006K

And even better here.

http://books.google.com/books?id=MVqVAgeZPFkC&pg=PA191&lpg=PA191&dq=chao...

Now understand after all that math the truth is that all this cool math has zero predictive capability.
Especially if you include my conjecture that the system itself is also changing its governing equations
over time as the inputs change. Its reprogramming itself.

This is actually the key feature because its how the system reaches a new stability regime.
Its complex exists in some state has noise and is itself changing. The net result is to actually capture the state i.e think about a system thats moving chaotically it can change its governing equations to capture this orbit is a new stable region the best way is to see that the new equations are completely regular.

As a simply example consider the traditional 3 body problem but driven i.e energy is being added to the system it can be distributed randomly to the three bodies and all kinds of cool stocastic resonance sets up.

However in this case we can bet that at some point one of the bodies will be ejected from the system as the sum total amount of energy climbs past that needed for one of the bodies to take on a escape orbit.
I've actually done this simulation its easier to treat it as a scattering problem and have one body come in from infinity and interact with a two body system and vary the velocity of the incoming body to get different total energy values. Its a subset of the random distribution of energy case but probably does not lose any real information.

The ejection of the third body transforms the remaining two bodies to a simple two body stable problem.

But now you must understand all of this is utter and complete bullshit in the sense that the mathematics are of now value to actually predict complex system behavior for real systems. Can I chase gobs of mathematics and right tons of papers and go to conferences yes. Does it really have any meaning or utility ?
No. Its pure mental masturbation. It just gives you more detail with zero understanding. There is no comprehension here no new laws no real understanding. At best if you have a good model you have a manifold of possible outcomes. Given the nature of the system several are extremes so BFD.

Now of course we have other types of math relevant to oil is of course invoking the central limit therom and at best expanding it via perturbation theory. I.e the shock model.

http://en.wikipedia.org/wiki/Perturbation_theory

This what WHT does however the problem with this approach is its mathematically fragile. This type of mathematics averages out the higher order factors and intrinsicly is incapable of detecting its own failure.

It works till it does not. My own work was effectively the same trick.
http://en.wikipedia.org/wiki/Quantum_chaos

The problem of course is its a trick not one with a strong fundamental basis and thus as I stated subject to failure if your underlying assumptions turn out to be false. And of course you have to actually resort to experiment to see if you got the right answer :) You can never know if a dropped higher order term actually blows up. And of course if the governing equations themselves are changing the perturbation approach itself quickly collapses into a mathematical mess of epic proportions. Don't even go there I have its beyond ugly.

A simply example is to take the three body equation and apply a drumroll... shock to the system by having the mass of one or more of the bodies in the system vary via a independent equation.

Or you can consider a 2D equation thats really 3D and vary the strength of a constant governing the third dimension but restrict yourself to only measuring the 2D variables. The chances of you being able to deduce the real system given the measurements in X,Y are slim. Throw in a chaotic equation and good luck.

However noise in the system and hidden variables are closely related and I'm aware that hidden variable theories are routinely bashed and dismissed but I think we are discarding them to early.

http://en.wikipedia.org/wiki/Hidden_variable_theory

Bohm's (nonlocal) hidden variable is called the quantum potential.

Obviously I'm of the Bohmist persuasion thus unsurprisingly most people who think they understand mathematics think I'm a complete fool.

I of course think that many mainstream approaches are just producing drivel counting the distribution of trees in a number of hypothetical forests tells you absolutely nothing about the forest.

Somebody is really wrong and someone is right. My new approach is to obviously try and see if I can come up with a set of valid observations of complex systems without doing the "math" because the math is bullshit.

This toy theory is completely based on the injection of a non local quantum potential into a quantum computer or computational engine.

http://en.wikipedia.org/wiki/Qubit

Now yes I've not written the great treatise that provides a Bohm variation of the Qubit.
Give me some cash and some time and I'm capable of it :)

Regardless thats where I think the real math that needs to be pursued lies. For now until I get some
free time I'm happy enough to make guess based on simple or toy Bohm qubits and see if I'm right.

I'm not the only one going off in this direction and it looks like you tend to sound like a kook.
http://www.quantumconsciousness.org/overview.html

I believe that somewhere and somehow when you really deal with complex systems that the predictions of the various views of quantum mechanics and classical correspondence and this classical physics diverge.
And Bohm wins :)

In the PDF, Rembrandt mentions that most recent month's production is probably biased high.

I believe there is a significant upward bias in the latest month for world production in every Oil Market Report of the International Energy Agency. Because in the consecutive publication by the IEA production estimates for several countries are nearly constantly revised downwards.

http://tonto.eia.doe.gov/country/index.cfm

http://www.eia.doe.gov/ipm/

TSK,

I know you asked specifically about EIA data but you might also consider perusing the charts at the Energy Export Databrowser which uses data from the 2008 BP Statistical Review.

-- Jon

Wow serious doublespeak sure you did not snag that out of 1984 ?

The steep drop in excessive profits from the credit bubble may lead to future steep drops in excessive profits making it difficult to determine when we will see and expansion of excessive profits greater than the current increase in the reduction of the excessive profit level.

Or the equity levels in a lot of houses is less that the current abnormally low sales prices well below market rate. This is caused by the abnormally large number of houses which the owners chose to return to the banks before their equity could be captured at real market rates. But its never a better time to buy since these below market sales are not expected to last long as more homeowners choose to hang onto their large equity gains.

Maybe the sharp drop in surplus capacity could actually gasp lead to lower real production which gasp might cause you to question the actual existence of this phantom capacity and for that matter profits and equity.

Maybe its not real.

There is oil stored at ships at sea maybe they buy some oil a little bit earlier and count it as storage at sea before it arrives?

I've tried to find overall storage for oil in the US without success.

What's wrong with the EIA data? It's broken down by type of stock and with/without SPR.

Despite the claims of high storage levels it looks like refinery storage capacity has fallen.

The additional US storage is largely in tank farms (up 50k bbl - nearly 30% - from Sept-Jan).

I have a hard time believing refineries where trimming storage and that third parties where furiously building tanks elsewhere.

Why?

Refineries have no need to increase storage, as they're producing a lot less (net inputs were down over 10% YOY in Jan), meaning their stored days of supply has increased substantially even without adding more physical oil. Why would they want to compound this increase further? What economic benefit would it have for them?

If somebody's going to be stuck storing oil, it's more likely to be the importers or producers than the consumers, which is why the increase in storage mostly being in tank farms make perfect sense.

storage levels have reached values
not seen since 1990 we would have to deduce that outside of refineries the empty storage tanks remained operable and waiting for almost twenty years.

Tank farm storage in Jan 2009 was only about 2% above May/Jun 2007 (EIA), so it's utterly misleading to suggest that these tanks haven't been used in decades.

There does not appear to be anything particularly surprising or unusual about the US storage/consumption data.