After reading the responses in yesterday’s HL thread, I have decided to take a break from TOD. While some people clearly understood what I was doing, I felt that there was far too much aspersion casting, insinuations of dishonesty and cheating, and just generally nasty responses that had absolutely nothing to do with my argument. Some people reacted in far too personal a way. As I am in the middle of a very stressful job (this week has been particularly difficult and hectic), and I have been separated from my family now for 6 weeks (with another 10 weeks to go), I have enough stress in my life without requiring more. At this point, I am dropping the HL thread and won’t be responding to any more comments.
I want to make it very clear that I am not looking for praise, or for people to blindly agree with my conclusions. What I am looking for are some reasoned arguments to the examples I provided, and especially some real examples indicating when the HL would have worked to predict peak, and why the parameters would have identified peak at that time. While there was a fair amount of good discussion, here is a sampling of some of the vitriol directed at me:
“basically garbage”, “keep being unreasonable or start thinking”, “not that interesting”, “not even close to being the right way to critique HL”, “assumption you childischly refuse to mention”, “sad, silly, egotistical”, “pissing contest”, “arrogance, pigheadedness and perhaps even childishness”, “waste of time”, “absurd”, “clumsy and actually self-defeating”, “gross”, “way off base”, “contrived examples”, “get off your high horse”, “re-inventing the wheel”, “junk”, “deceitful”, “unrealistic scenarios”, “let me hand you a clue”, “over the top”, “not very useful”, “cheating”, “vindictive, and spiteful”, “constructed cases where it does not work”, “diatribe”, “obnoxious attempt”, “a guy with an aganda and a axe to grind”, and “quite revealing in an unflatering way”
Admittedly, I do not have thick skin and sometimes I bite back (although I don't bite first). Anyway, I have reached my quota of gratuitous insults for a while. Imagine that after you spent 6 or 8 hours writing an essay - as a volunteer whose purpose here is to promote awareness of peak oil and energy issues - that those were some of the responses you received. Yet this has been a pattern every time I challenge certain arguments.
Deceitful? Cheating? Junk? Some people react as if I had slapped them in the face and they just lash out without attempting an actual rebuttal of any of the arguments. That is why I liken some of these positions to religious faith. If you have staked out a position in which you have on blinders, allowing entrance only to evidence that supports your position, you have the makings of a faith-based position. There are many here who fall into that category, and those are the ones who respond with the most vitriol.
Anyway, hope everyone has a great weekend. I am going to take some time off and re-evaluate what it is that I actually think I can accomplish here, and whether my efforts might be better served doing something else. And please, I am not looking for people to respond with flattery and such. I am in fact as uncomfortable around praise as I am around insults. I would rather just discuss the data.
Sad to see you off on what I hope is only a TOD sabbatical. FWIW I thought your article on HL was excellent (indeed I was one the few who "Dugg" it) and I was disappointed by some of the responses from people who I would have expected more of, based on their contributions to date on TOD. They know who they are.
I hope we will see you back soon, and certainly no later than the end of summer, by which time I expect answers to have been given regarding Saudi production. For what it's worth, I expect them to ramp up to 9.5 million or so in response to a demand-driven price spike. It will be interesting to see how long they can go at that rate if the price does not drop again. We shall see.
Re: That is why I liken some of these positions to religious faith. If you have staked out a position in which you have on blinders, allowing entrance only to evidence that supports your position, you have the makings of a faith-based position. There are many here who fall into that category, and those are the ones who respond with the most vitriol
I think of it as a selection bias which, in turn, supports a faith-based position. When you argue with people about their religion, then you are likely to get called names — or worse. For example, people have been known to nail others up on a cross, if they've got one handy.
Just my excuse, then. Sorry I was out of line there, I should have thought before writing. I'll think it over, and if I have something clever to add, I'll post that. Next time I'll keep the wordings more courteous! Have a nice weekend!
I've found the posts over the last few weeks really fascinating, though was disappointed with the personal nature of the HL discussion. Have a good break Robert. Seems to me the points to be taken from these discussions are:
HL should not be used to predict URR
HL, with informed guesses on URR, may be used to predict peak
There is no situation which will fit HL perfectly, all curves must be worked to fit available data
We do not know the URR of KSA
The decline in output of KSA, combined with increase in drilling, combined with increase in horizontal and advanced drilling techinques are potentially very worrying
Of course, I'm just a lurker, and not as knowledgeable as most of the posters here. But thanks to the Oil Drum for my understanding as it is.
As a frequent reader and infrequent contributor, I think it's very sad that one of the most distinguished contributors should be subjected to such abuse. Constructive criticism is always valuable but what Robert has quoted above does not fit that definition in any way. I do not blame him for reacting as he has - I would probably do the same myself.
I think it is only appropriate to extend the same thoughts to WT and Dave Cohen. Robert's work was not judged any harsher than the others. When you put yourself out there, you need to expect critics.
First I stand by every statement I made about the post no problem this is a crappy post. It has obvious problems. I've said the same in private and make no apology in public.
Good arguments questioning HL can and have been made this is not one of them.
I didn't read your comments on Robert's "crappy" post. In fact, I didn't even read the post.
Hubbert modelling provides a non-physical view of the oil supply. I've always thought that if one was going to write about the oil & gas business, one should know something about it, including the oil price, which depends on the economics of supply & demand, or various "aboveground" events or factors. Many things can disrupt a Hubbert curve. Never let reality get in the way of a beautiful theory. Furthermore, reserves growth does happen. The US is the best example of that, being the most mature oil province in the world.
There is a problem with assuming a normal distribution (as in Hubbert modelling) for global oil production, which can only follow from a mathematical result called the Central Limit Theorem. Not the least of these problems is what one defines as "oil". Another problem is what one defines as the domain for the analysis in cases where one is dealing with something smaller than the whole Earth. But if one is talking about our happy, little Planet, the factors I talked about in the paragraph just above are paramount.
I was in the habit, when I posted here, of talking about pertinent and important details about the oil & gas business, and mentioning specific fields or trends. That was not (apparently) a popular thing to do, at least with the "regulars" who comment on this site. I could care less.
If you or others insist on carrying out this mediocre, simplistic mathematical analysis, so be it. But, that's all the worse for The Oil Drum, which becomes more marginalized as a result. Try a creaming curve, sometime. Look at how the people whose livelihoods depend on this (in the oil business) do things. Reflecting a bit, this is a civilization in decline. I can always find examples, like much of the discourse on this website, that are "close to home".
"If you or others insist on carrying out this mediocre, simplistic mathematical analysis, so be it. But, that's all the worse for The Oil Drum, which becomes more marginalized as a result. Try a creaming curve, sometime."
Dave,
Good, relevant points on HL. However, there's an important facet to the issue of HL and modeling that I haven't seen brought up much.
Is TOD an experts-only forum?
What I mean by this is that, if it's only for those with the background to really dig into the geophysics and assorted heavy-duty modeling approaches, than it should be stated as such. From what I gather, many here aren't experts in petrogeology or the operation of an oil company, but are people from many walks of life who are seriously trying to understand the underlying issues, and look forward to those moments when the experts communicate something meaningful to help them better grasp the problems at hand.
Robert stated that he's interested in educating people about the real problem of Peak Oil to hopefully give them what amounts to the "real story" (if you will) to help in making a well-informed decision about whether-or-not to act (these aren't his exact words--but I gather from what he said that it's something like this). Or, is his focus just for the experts, the ones who care about, and have the background for understanding, all the little details and whatsits involved with developing an iron-clad case for peak-now or peak-later? Or is it both?
"this mediocre, simplistic mathematical analysis"
You stated it yourself. Mediocre and simplistic might be the strongest features going for HL, because those words suggest that HL might be the easiest model for the greatest number of people to follow. Think about it. If many folk are having trouble understanding HL, then it's likely they'll even have more trouble understanding a model that's more complex. Would the general population be more convinced by the simple model, or more complex one? The answer, of course, isn't straightforward or necessarily obvious. People themselves are complex. But I think you can get the gist of what I'm suggesting here.
What's your audience? What will make the greatest number within this audience at least accept the potential dangers enough to give Peak Oil serious consideration?
Re: Good arguments questioning HL can and have been made this is not one of them
I don't mean to mischaracterize your position. Apparently, you have not drunk the "kool-aid" ala Jim Jones. Sorry if I misrepresented your views in my zeal to mock a mathematical, abstract simplification of a complex reality — world oil production.
Well then Dave lets see the "real deal". I can only assume that there are many other models out there, perhaps they are propriotery, but the debate on Huberts curve bugs you guys so much then lets see something better. I'm no petro guy but can follow most of what goes on. I thought fractional_flow was like a breath of fresh air and I also noticed very little debate from you senior guys. He said ghanwar(sp?) was 70% depleated, that is huge! and there was very little comments to the contrary...
I think it is very easy to understand oil is finite for most people, especially those in the US where we now import 60%+/- of our oil and have our youths involved in a nasty(oil) war. Most people can follow that oil must reach a peak in extraction and decline, just like the US.
I have tried baiting this question before with no response, so I will try again. I think alot of people are tired of the bickering, and I think this includes you. I have a solution...
IF YOU THINK HUBERTS CURVE IS A BAD MODEL THEN SHOW US SOMETHING BETTER.
A good model, in my view, would be along the lines of those developed to predict the economy. (Many of which currently say we are headed for recession). Of course, greater attention would have to paid to oil supply and demand issues than might be common at present.
There are many here who fall into that category, and those are the ones who respond with the most vitriol.
I think that's why you see so many negative comments. Because those people are the ones most likely to respond. I agreed with what you wrote but didn't respond because I didn't really have anything to add.
One thing your posts did do is to get me to take a closer look at HL. I had never really paid much attention to it before since I'm really not concerned with predicting a date for the peak. I don't think I'll be convinced of the date for the peak until after the fact. Then I'll be able to look back and say, yes there was the peak. But there's just too much uncertainty before the actual peak to be able to predict it. That's not to say that some people won't make good educated guesses.
I've taken a brief look at the HL graph and I don't see how anyone can think it can predict URR. The point where the line intersects the x axis increases with each new data point. It's an interesting graph to look at and think about, but it really isn't very useful.
I am new to TOD and want to thank you for your hard work that has led to a fabulous discussion in the last few weeks about peak oil. Your "bravery" has brought other experts into the fold lately, what a great education for the layperson like myself. I now realize peak oil is certain, we can look upon this as the end or a great new opportunity, I choose the latter.
While you are pondering the amount of energy you can devote to TOD, please keep in mind that there are a lot of us who read the debates, and the posted comments are not always a fair representation of the "community". If the critical mass of TOD veers away from data and analysis and is driven by something more evangelical, then its usefulness would be greatly diminished.
The HL thread was obviously an emotional marathon. I hope that the experience, as unpleasant as it was, will not drive you away permanently. Your articles and commentary have always been spot on, grounded in evidence, and are the strength of this site.
I guess Robert and I can now argue over who first decided to take a sabbatical from TOD. I was trying to at least take a sabbatical, when the HL discussion popped up again. Following is what I just posted over on the HL discussion:
The Role of Oil Companies in Post-Peak Regions
For the purposes of this discussion, crude oil = crude + condensate.
The Lower 48 and the North Sea started showing a strong linear pattern on their crude oil HL plots in 1954 and 1988, respectively. The Lower 48 peaked in 1970, the North Sea, in 1999.
The HL data from 1954 to 1970 and from 1988 to 1999 are consistent with the respective post-peak HL data for the Lower 48 and North Sea.
The Lower 48 crossed the 50% of Qt mark in 1970, the North Sea, in 1999. The post-1970 cumulative Lower 48 crude oil production through 2004 was 99% of what the HL model predicted it would be--using production data from 1954 to 1970 to generate the HL model.
Interestingly enough, according to Matt Simmons, the major oil companies working the North Sea in 1999 were predicting that the North Sea peak was at least a decade away. Sound familiar?
There were no material restrictions on drilling in either region (minor, but not material restrictions in the case of the Lower 48), and both regions were developed by private companies.
Despite the best efforts of private companies using the best available technology, the Lower 48 and the North Sea have both shown long term declines since peaking. Empirically, the role of oil companies in these two post-peak regions has been to slow the rate of decline of conventional crude oil production.
In my opinion, major oil companies are deeply threatened by mathematical models like the HL method, as applied to world crude oil production, because it suggests that they probably can’t replace conventional crude oil reserves, and they are (justifiably) deeply concerned about punitive taxation. I thought it was interesting that Matt Simmons said that he did not own any major oil company stocks, because they probably could not replace their conventional crude oil reserves (he is focusing on service companies and smaller oil companies).
The world started showing a strong linear pattern in 1983, and on Deffeyes' crude oil HL plot, the world crossed the 50% of Qt mark in 2005. Deffeyes predicted that the most likely year for a world crude oil decline was 2006, within a predicted peak range of 2004 to 2008. As Deffeyes predicted, world crude oil production declined in 2006, relative to 2005 (EIA).
So, the Lower 48, the North Sea and the world all started showing lower crude oil production after crossing their respective 50% of Qt marks.
Empirically, in the absence of constraints on production, e.g., Texas and Saudi Arabia, regions tend to peak in the vicinity of 50% of Qt.
I can’t get inside Robert’s head, so I will assume that he is simply trying understand the HL method, rather than attacking the HL method because of the unpleasant implications it has for major oil companies.
But I do have a couple of questions for Robert.
(1) Why do you continue to treat the Lower 48 and world HL plots as if they are radioactive--while going into extreme detail on the Texas HL plot?
(2) Why do you think that world conventional crude oil production will increase past the 50% of Qt mark, when the Lower 48 and North Sea did not? (Even as world crude oil production is declining).
FWIW I think you two should give it a rest for a while. You have given us plenty to chew on and a period of reflection and waiting for more data seems to be in order here. I am grateful for all of your work.
(2) Why do you think that world conventional crude oil production will increase past the 50% of Qt mark, when the Lower 48 and North Sea did not? (Even as world crude oil production is declining).
I don't think he ever said that the world conventional crude oil production will increase past the 50% of Qt mark. I guess the million dollar question is: How do you know when the world is at 50% of Qt mark? I think he just showed that HL can't be used to reliably predict Qt or URR until after the fact.
Qt is a mathematical estimate of URR. The pre-peak Qt estimates for the Lower 48 and the North Sea are both consistent with post-peak estimates of Qt, and in both cases the regions peaked right at 50% of estimated URR.
The world is showing the same strong linear pattern as the Lower 48 and the North Sea, before their respective peaks, and like the Lower 48 and North Sea, the world started showing lower crude oil production after crossing the 50% of Qt mark.
As further support for the world peak, it's a near certainty that 14 out of 14 of the super giants oil fields that are or were producing one mbpd or more are now in decline.
WT (or anyone else); do you have a link to an overview of the full 'export land' model? I have seen tantalizing hints about it, but haven't been able to get a complete picture. I'd appreciate any pointers to older posts that would lay out the basics.
Thanks!
CW
Global peak: 2007 - 2010
Global decline rate, Post peak: 2%
Economic response: Severe global recession, ~5 years, then slow recovery
The implicit assumption is that consumption in oil-exporting countries will not be affected by worldwide shortages. But that may not be true.
Here's a concrete and somewhat pressing example: Mexico. Cantarell production is declining, and while they are scrambling to get other fields going, they can't make up the shortfall. The country is facing production declines, yet oil export revenues have been one of their main sources of national income for years.
So here's the question: suppose total Mexican oil production falls X percent. Do you think that Mexican oil exports will fall (a) more or (b) less than X percent?
Export Land would predict the answer is (a), more than X percent. Mexico will preferentially divert its falling oil production to its internal consumption. This will give it less to export and so exports will fall even worse and faster than production is decreasing.
But it's very possible that the answer is (b)! Mexico could allow internal gas prices to rise, curbing demand, so that it can maintain its export level and keep that hard currency coming in that the country depends on. This would mean that Mexican oil exports would fall less, and slower, than its internal production declines.
This is the big flaw in the Export Land model. It assumes that this can't happen. Export Land says that exports will fall faster than production and doesn't consider the reality of regimes that would prefer to receive foreign capital than to keep internal gas prices low. It's not an easy choice for a government in this situation, but it is far from clear that there is only one answer. By neglecting to consider the full range of options available to oil exporters, Export Land builds its conclusions into its assumptions. It is an oversimplified and IMO basically useless model.
A quote from Spilling Oil in Mexico linked to in yesterday's drumbeat:
Meanwhile, Mexicans themselves suffer. They pay about a dollar more for gas at the pump than their illegal-immigrant cousins in Texas. You see, there is no competition. They can buy only from Pemex.
The Mexicans suffer??, not the way you imagine. They pay a dollar more a gallon than in the US?? In a pigs eye they do! I live in Mexico. The price of gasoline is set by the government, and is the same in the entire country with the exception of Pemex stations close to the US border, where the price is set to match the US price -- because it has been cheaper here than in the US, and my ex-fellow gringos were crossing the border in droves to fill their gas sucking hogs and straining the supply network down here provoking spot shortages.
I bought gasoline just today. Have the printed receipt right here. I paid 6.82 pesos per litre for regular unleaded 87 octane. That works out to slightly less than $2.35 per US gallon at today's exchange rate. A few days ago I bought premium unleaded 92 octane for $2.88 US per gallon. That compares pretty favorably to the US average price, and for the west coast in the US those prices would be to die for. They're payin' over 3 bucks for a gallon of regular that's adulterated with 10% ethanol. When I still lived up there a few years ago that crap fuel used to lower the highway mpg in my 5.0 HO Mustang from 21.5 to 15. But I digress...
The price of the fuel doesn't tell the whole story...Of course everybody at TOD to varying degrees realizes that anything under 3 bucks a gallon is ridiculously cheap for a non-renewable resource whose days are numbered. With respect to Mexico, you have to keep in mind that wages are -it varies- roughly 1/10 what they are in the US, so for the typical Mexican motorist, the 335 pesos or so that it currently costs him to fill the 13 gallon tank in his Nissan Tsuru represents a far larger percentage of his disposable income than 30 dollars and change does for an American. And think, of the 190 or so national economies tracked by the World Bank, Mexico ranks around 45th in the world in terms of personal income. Imagine what "pain at the pump" must mean to somebody in a country that ranks about halfway down the list, (out of about 190, remember) like Albania or Bosnia or El Salvador. Actually, you won't have to imagine it for very long, pretty soon now we're all gonna get the chance to find out first hand.
My point here is not to appeal for sympathy for the majority of mankind that has no access to the energy products that fuel the first world's consumer orgy, and the entire planet's environmental catastrophe. I really can't recall hearing people down here bitch and moan about energy costs. The fuel costs what it costs, and they just pay, and the price goes up a few cents every month like clockwork. My point, if I actually have one, is to illustrate how far out of it most of the "enlightened scholarly discussion" in the peak oil community really is. Americans especially, and to a lesser extent Canadians, Europeans, Japanese, almost half of Mexicans, etc. (anybody at the party or who aspires to join it, or seriously thinks it isn't going to end badly) are all living in a fool's paradise.
It's kinda' like...
Nero didn't fiddle while Rome burned, he was just waiting for more scientifically rigorous modelling on how long it would be before his beard got singed enough to yell FIRE! and call the fire department. And since when does a lucid, tightly reasoned, peer-reviewed argument mean diddley-dick to a continent or two full of debauched drunks. (ooops, I mean "developed world political leaders, CERA, Rush Limbaugh types & other assorted *oil* addicts")
Westexas, re, Export Land Model, see the chart in this 3/9 NYT story on Mexico by Elizabeth Malkin: www.nytimes.com/2007/03/09/business/worldbusiness/09pemex.html?ex=133109... . Just an anecdote, but chart shows roughly level exports to US while overall prod. drops from 2004-06. Not picking a fight here, just saw this and passing it on. Is the 2-yr time just too short for the export drop to kick in, or is something special going on, e.g., we sell them nat-gas, we refine and pass-back some of their product, etc.?
The US is presumably Mexico's best customer, for a number of reasons, so I assume that we would be the last to be cut, but they did start curtailing crude oil deliveries to Gulf Coast refineries in January.
I think that the best model for Mexico may be the UK, which went from exporting one mbpd in 1999, to being a net importer in 2005.
The pre-peak Qt estimates for the Lower 48 and the North Sea are both consistent with post-peak estimates of Qt, and in both cases the regions peaked right at 50% of estimated URR.
WT, but was it possible to know in 1972 that Texas was at 50% of Qt? In 1972 you would have thought that Texas was at 70% of Qt. Now, 35 years after the peak, you know that Texas was around 50% of Qt in 1972. So, you see, the % of Qt at peak retreated with time. So how can you be sure that the world or Saudi Arabia is at 50% of Qt now? How do you know that the % of Qt for world and Saudi Arabia will not retreat with time, a la Texas? This is the point that RR has been trying to make for several days.
As further support for the world peak, it's a near certainty that 14 out of 14 of the super giants oil fields that are or were producing one mbpd or more are now in decline.
The world may be past or very close to peak. But you can't draw that conclusion from HL alone. Now, if Saudis fail to increase production as inventories fall and price skyrockets, that is a different story.
I suggest that you compare the Lower 48 HL plot to the world HL plot. As I have said several times, the pre-peak Texas HL plot is the noisiest of the four that we looked at in the Texas/Lower 48 article. I simply used the Texas HL plot to get the Probable URR, so I could then determine at what stage of depletion that Texas peaked.
The Lower 48, North Sea and world HL plots all have solid pre-peak HL profiles.
Then why not take up Robert's thrown down gauntlet.
Similar to his modelling by stepping us through at 10 year intervals, the data modelled out on HL. Only include those data that would've been know at the time the HL plot would've been figured.
I don't see why none of the HL defenders are seemingly unwilling to do this. Robert has repeatedly asked to be shown an example in which HL would've shown a peak in Real Time, but nobody has done this yet.
Why not? The lack of a working counter example at this point is perhaps more damning than anything Robert ever wrote.
After reading the responses in yesterday’s HL thread, I have decided to take a break from TOD. While some people clearly understood what I was doing, I felt that there was far too much aspersion casting, insinuations of dishonesty and cheating, and just generally nasty responses that had absolutely nothing to do with my argument. Some people reacted in far too personal a way. As I am in the middle of a very stressful job (this week has been particularly difficult and hectic), and I have been separated from my family now for 6 weeks (with another 10 weeks to go), I have enough stress in my life without requiring more. At this point, I am dropping the HL thread and won’t be responding to any more comments.
I want to make it very clear that I am not looking for praise, or for people to blindly agree with my conclusions. What I am looking for are some reasoned arguments to the examples I provided, and especially some real examples indicating when the HL would have worked to predict peak, and why the parameters would have identified peak at that time. While there was a fair amount of good discussion, here is a sampling of some of the vitriol directed at me:
Admittedly, I do not have thick skin and sometimes I bite back (although I don't bite first). Anyway, I have reached my quota of gratuitous insults for a while. Imagine that after you spent 6 or 8 hours writing an essay - as a volunteer whose purpose here is to promote awareness of peak oil and energy issues - that those were some of the responses you received. Yet this has been a pattern every time I challenge certain arguments.
Deceitful? Cheating? Junk? Some people react as if I had slapped them in the face and they just lash out without attempting an actual rebuttal of any of the arguments. That is why I liken some of these positions to religious faith. If you have staked out a position in which you have on blinders, allowing entrance only to evidence that supports your position, you have the makings of a faith-based position. There are many here who fall into that category, and those are the ones who respond with the most vitriol.
Anyway, hope everyone has a great weekend. I am going to take some time off and re-evaluate what it is that I actually think I can accomplish here, and whether my efforts might be better served doing something else. And please, I am not looking for people to respond with flattery and such. I am in fact as uncomfortable around praise as I am around insults. I would rather just discuss the data.
Enjoy your time away...read some good books...watch the sun go up and down.
Rick
Robert,
Sad to see you off on what I hope is only a TOD sabbatical. FWIW I thought your article on HL was excellent (indeed I was one the few who "Dugg" it) and I was disappointed by some of the responses from people who I would have expected more of, based on their contributions to date on TOD. They know who they are.
I hope we will see you back soon, and certainly no later than the end of summer, by which time I expect answers to have been given regarding Saudi production. For what it's worth, I expect them to ramp up to 9.5 million or so in response to a demand-driven price spike. It will be interesting to see how long they can go at that rate if the price does not drop again. We shall see.
Enjoy your much-deserved break.
Re: That is why I liken some of these positions to religious faith. If you have staked out a position in which you have on blinders, allowing entrance only to evidence that supports your position, you have the makings of a faith-based position. There are many here who fall into that category, and those are the ones who respond with the most vitriol
I think of it as a selection bias which, in turn, supports a faith-based position. When you argue with people about their religion, then you are likely to get called names — or worse. For example, people have been known to nail others up on a cross, if they've got one handy.
Just my excuse, then. Sorry I was out of line there, I should have thought before writing. I'll think it over, and if I have something clever to add, I'll post that. Next time I'll keep the wordings more courteous! Have a nice weekend!
Uninformed lurker warning.
I've found the posts over the last few weeks really fascinating, though was disappointed with the personal nature of the HL discussion. Have a good break Robert. Seems to me the points to be taken from these discussions are:
HL should not be used to predict URR
HL, with informed guesses on URR, may be used to predict peak
There is no situation which will fit HL perfectly, all curves must be worked to fit available data
We do not know the URR of KSA
The decline in output of KSA, combined with increase in drilling, combined with increase in horizontal and advanced drilling techinques are potentially very worrying
Of course, I'm just a lurker, and not as knowledgeable as most of the posters here. But thanks to the Oil Drum for my understanding as it is.
I think that's an excellent summary.
Robert:
Just want to extend a note of appreciation for your posts, both the HL post and the prior work on ethanol. Your contributions will be missed.
Best Regards
As a frequent reader and infrequent contributor, I think it's very sad that one of the most distinguished contributors should be subjected to such abuse. Constructive criticism is always valuable but what Robert has quoted above does not fit that definition in any way. I do not blame him for reacting as he has - I would probably do the same myself.
I think it is only appropriate to extend the same thoughts to WT and Dave Cohen. Robert's work was not judged any harsher than the others. When you put yourself out there, you need to expect critics.
First I stand by every statement I made about the post no problem this is a crappy post. It has obvious problems. I've said the same in private and make no apology in public.
Good arguments questioning HL can and have been made this is not one of them.
If personal attacks have been made I'm sorry.
And I'm done with the post.
I didn't read your comments on Robert's "crappy" post. In fact, I didn't even read the post.
Hubbert modelling provides a non-physical view of the oil supply. I've always thought that if one was going to write about the oil & gas business, one should know something about it, including the oil price, which depends on the economics of supply & demand, or various "aboveground" events or factors. Many things can disrupt a Hubbert curve. Never let reality get in the way of a beautiful theory. Furthermore, reserves growth does happen. The US is the best example of that, being the most mature oil province in the world.
There is a problem with assuming a normal distribution (as in Hubbert modelling) for global oil production, which can only follow from a mathematical result called the Central Limit Theorem. Not the least of these problems is what one defines as "oil". Another problem is what one defines as the domain for the analysis in cases where one is dealing with something smaller than the whole Earth. But if one is talking about our happy, little Planet, the factors I talked about in the paragraph just above are paramount.
I was in the habit, when I posted here, of talking about pertinent and important details about the oil & gas business, and mentioning specific fields or trends. That was not (apparently) a popular thing to do, at least with the "regulars" who comment on this site. I could care less.
If you or others insist on carrying out this mediocre, simplistic mathematical analysis, so be it. But, that's all the worse for The Oil Drum, which becomes more marginalized as a result. Try a creaming curve, sometime. Look at how the people whose livelihoods depend on this (in the oil business) do things. Reflecting a bit, this is a civilization in decline. I can always find examples, like much of the discourse on this website, that are "close to home".
"If you or others insist on carrying out this mediocre, simplistic mathematical analysis, so be it. But, that's all the worse for The Oil Drum, which becomes more marginalized as a result. Try a creaming curve, sometime."
Dave,
Good, relevant points on HL. However, there's an important facet to the issue of HL and modeling that I haven't seen brought up much.
Is TOD an experts-only forum?
What I mean by this is that, if it's only for those with the background to really dig into the geophysics and assorted heavy-duty modeling approaches, than it should be stated as such. From what I gather, many here aren't experts in petrogeology or the operation of an oil company, but are people from many walks of life who are seriously trying to understand the underlying issues, and look forward to those moments when the experts communicate something meaningful to help them better grasp the problems at hand.
Robert stated that he's interested in educating people about the real problem of Peak Oil to hopefully give them what amounts to the "real story" (if you will) to help in making a well-informed decision about whether-or-not to act (these aren't his exact words--but I gather from what he said that it's something like this). Or, is his focus just for the experts, the ones who care about, and have the background for understanding, all the little details and whatsits involved with developing an iron-clad case for peak-now or peak-later? Or is it both?
"this mediocre, simplistic mathematical analysis"
You stated it yourself. Mediocre and simplistic might be the strongest features going for HL, because those words suggest that HL might be the easiest model for the greatest number of people to follow. Think about it. If many folk are having trouble understanding HL, then it's likely they'll even have more trouble understanding a model that's more complex. Would the general population be more convinced by the simple model, or more complex one? The answer, of course, isn't straightforward or necessarily obvious. People themselves are complex. But I think you can get the gist of what I'm suggesting here.
What's your audience? What will make the greatest number within this audience at least accept the potential dangers enough to give Peak Oil serious consideration?
-best,
Wolf
Re: Good arguments questioning HL can and have been made this is not one of them
I don't mean to mischaracterize your position. Apparently, you have not drunk the "kool-aid" ala Jim Jones. Sorry if I misrepresented your views in my zeal to mock a mathematical, abstract simplification of a complex reality — world oil production.
Well then Dave lets see the "real deal". I can only assume that there are many other models out there, perhaps they are propriotery, but the debate on Huberts curve bugs you guys so much then lets see something better. I'm no petro guy but can follow most of what goes on. I thought fractional_flow was like a breath of fresh air and I also noticed very little debate from you senior guys. He said ghanwar(sp?) was 70% depleated, that is huge! and there was very little comments to the contrary...
I think it is very easy to understand oil is finite for most people, especially those in the US where we now import 60%+/- of our oil and have our youths involved in a nasty(oil) war. Most people can follow that oil must reach a peak in extraction and decline, just like the US.
I have tried baiting this question before with no response, so I will try again. I think alot of people are tired of the bickering, and I think this includes you. I have a solution...
IF YOU THINK HUBERTS CURVE IS A BAD MODEL THEN SHOW US SOMETHING BETTER.
A good model, in my view, would be along the lines of those developed to predict the economy. (Many of which currently say we are headed for recession). Of course, greater attention would have to paid to oil supply and demand issues than might be common at present.
I think that's why you see so many negative comments. Because those people are the ones most likely to respond. I agreed with what you wrote but didn't respond because I didn't really have anything to add.
One thing your posts did do is to get me to take a closer look at HL. I had never really paid much attention to it before since I'm really not concerned with predicting a date for the peak. I don't think I'll be convinced of the date for the peak until after the fact. Then I'll be able to look back and say, yes there was the peak. But there's just too much uncertainty before the actual peak to be able to predict it. That's not to say that some people won't make good educated guesses.
I've taken a brief look at the HL graph and I don't see how anyone can think it can predict URR. The point where the line intersects the x axis increases with each new data point. It's an interesting graph to look at and think about, but it really isn't very useful.
Robert Rapier,
I am new to TOD and want to thank you for your hard work that has led to a fabulous discussion in the last few weeks about peak oil. Your "bravery" has brought other experts into the fold lately, what a great education for the layperson like myself. I now realize peak oil is certain, we can look upon this as the end or a great new opportunity, I choose the latter.
Robert,
While you are pondering the amount of energy you can devote to TOD, please keep in mind that there are a lot of us who read the debates, and the posted comments are not always a fair representation of the "community". If the critical mass of TOD veers away from data and analysis and is driven by something more evangelical, then its usefulness would be greatly diminished.
The HL thread was obviously an emotional marathon. I hope that the experience, as unpleasant as it was, will not drive you away permanently. Your articles and commentary have always been spot on, grounded in evidence, and are the strength of this site.
I guess Robert and I can now argue over who first decided to take a sabbatical from TOD. I was trying to at least take a sabbatical, when the HL discussion popped up again. Following is what I just posted over on the HL discussion:
The Role of Oil Companies in Post-Peak Regions
For the purposes of this discussion, crude oil = crude + condensate.
The Lower 48 and the North Sea started showing a strong linear pattern on their crude oil HL plots in 1954 and 1988, respectively. The Lower 48 peaked in 1970, the North Sea, in 1999.
The HL data from 1954 to 1970 and from 1988 to 1999 are consistent with the respective post-peak HL data for the Lower 48 and North Sea.
The Lower 48 crossed the 50% of Qt mark in 1970, the North Sea, in 1999. The post-1970 cumulative Lower 48 crude oil production through 2004 was 99% of what the HL model predicted it would be--using production data from 1954 to 1970 to generate the HL model.
Interestingly enough, according to Matt Simmons, the major oil companies working the North Sea in 1999 were predicting that the North Sea peak was at least a decade away. Sound familiar?
There were no material restrictions on drilling in either region (minor, but not material restrictions in the case of the Lower 48), and both regions were developed by private companies.
Despite the best efforts of private companies using the best available technology, the Lower 48 and the North Sea have both shown long term declines since peaking. Empirically, the role of oil companies in these two post-peak regions has been to slow the rate of decline of conventional crude oil production.
In my opinion, major oil companies are deeply threatened by mathematical models like the HL method, as applied to world crude oil production, because it suggests that they probably can’t replace conventional crude oil reserves, and they are (justifiably) deeply concerned about punitive taxation. I thought it was interesting that Matt Simmons said that he did not own any major oil company stocks, because they probably could not replace their conventional crude oil reserves (he is focusing on service companies and smaller oil companies).
The world started showing a strong linear pattern in 1983, and on Deffeyes' crude oil HL plot, the world crossed the 50% of Qt mark in 2005. Deffeyes predicted that the most likely year for a world crude oil decline was 2006, within a predicted peak range of 2004 to 2008. As Deffeyes predicted, world crude oil production declined in 2006, relative to 2005 (EIA).
So, the Lower 48, the North Sea and the world all started showing lower crude oil production after crossing their respective 50% of Qt marks.
Empirically, in the absence of constraints on production, e.g., Texas and Saudi Arabia, regions tend to peak in the vicinity of 50% of Qt.
I can’t get inside Robert’s head, so I will assume that he is simply trying understand the HL method, rather than attacking the HL method because of the unpleasant implications it has for major oil companies.
But I do have a couple of questions for Robert.
FWIW I think you two should give it a rest for a while. You have given us plenty to chew on and a period of reflection and waiting for more data seems to be in order here. I am grateful for all of your work.
+1
+1
Agreed.
Disagree. I'd like to hear what both gentlemen have to say about Ace's post http://www.theoildrum.com/node/2389#comment-171951
and I'd particularly like to hear Roberts analysis of Robertos http://www.theoildrum.com/node/2389#comment-171802 work.
I don't think he ever said that the world conventional crude oil production will increase past the 50% of Qt mark. I guess the million dollar question is: How do you know when the world is at 50% of Qt mark? I think he just showed that HL can't be used to reliably predict Qt or URR until after the fact.
Qt is a mathematical estimate of URR. The pre-peak Qt estimates for the Lower 48 and the North Sea are both consistent with post-peak estimates of Qt, and in both cases the regions peaked right at 50% of estimated URR.
The world is showing the same strong linear pattern as the Lower 48 and the North Sea, before their respective peaks, and like the Lower 48 and North Sea, the world started showing lower crude oil production after crossing the 50% of Qt mark.
As further support for the world peak, it's a near certainty that 14 out of 14 of the super giants oil fields that are or were producing one mbpd or more are now in decline.
WT (or anyone else); do you have a link to an overview of the full 'export land' model? I have seen tantalizing hints about it, but haven't been able to get a complete picture. I'd appreciate any pointers to older posts that would lay out the basics.
Thanks!
CW
Global peak: 2007 - 2010
Global decline rate, Post peak: 2%
Economic response: Severe global recession, ~5 years, then slow recovery
“Export Land” Article:
http://www.theoildrum.com/story/2006/1/27/14471/5832
Model:
http://static.flickr.com/97/240076673_494160e1a0_o.png
Thanks!
It was much simpler than I was expecting, but still sobering...
CW
Global peak: 2007 - 2010
Global decline rate, Post peak: 2%
Economic response: Severe global recession, ~5 years, then slow recovery
Export Land is way oversimplified if you ask me.
The implicit assumption is that consumption in oil-exporting countries will not be affected by worldwide shortages. But that may not be true.
Here's a concrete and somewhat pressing example: Mexico. Cantarell production is declining, and while they are scrambling to get other fields going, they can't make up the shortfall. The country is facing production declines, yet oil export revenues have been one of their main sources of national income for years.
So here's the question: suppose total Mexican oil production falls X percent. Do you think that Mexican oil exports will fall (a) more or (b) less than X percent?
Export Land would predict the answer is (a), more than X percent. Mexico will preferentially divert its falling oil production to its internal consumption. This will give it less to export and so exports will fall even worse and faster than production is decreasing.
But it's very possible that the answer is (b)! Mexico could allow internal gas prices to rise, curbing demand, so that it can maintain its export level and keep that hard currency coming in that the country depends on. This would mean that Mexican oil exports would fall less, and slower, than its internal production declines.
This is the big flaw in the Export Land model. It assumes that this can't happen. Export Land says that exports will fall faster than production and doesn't consider the reality of regimes that would prefer to receive foreign capital than to keep internal gas prices low. It's not an easy choice for a government in this situation, but it is far from clear that there is only one answer. By neglecting to consider the full range of options available to oil exporters, Export Land builds its conclusions into its assumptions. It is an oversimplified and IMO basically useless model.
A quote from Spilling Oil in Mexico linked to in yesterday's drumbeat:
The Mexicans suffer??, not the way you imagine. They pay a dollar more a gallon than in the US?? In a pigs eye they do! I live in Mexico. The price of gasoline is set by the government, and is the same in the entire country with the exception of Pemex stations close to the US border, where the price is set to match the US price -- because it has been cheaper here than in the US, and my ex-fellow gringos were crossing the border in droves to fill their gas sucking hogs and straining the supply network down here provoking spot shortages.
I bought gasoline just today. Have the printed receipt right here. I paid 6.82 pesos per litre for regular unleaded 87 octane. That works out to slightly less than $2.35 per US gallon at today's exchange rate. A few days ago I bought premium unleaded 92 octane for $2.88 US per gallon. That compares pretty favorably to the US average price, and for the west coast in the US those prices would be to die for. They're payin' over 3 bucks for a gallon of regular that's adulterated with 10% ethanol. When I still lived up there a few years ago that crap fuel used to lower the highway mpg in my 5.0 HO Mustang from 21.5 to 15. But I digress...
The price of the fuel doesn't tell the whole story...Of course everybody at TOD to varying degrees realizes that anything under 3 bucks a gallon is ridiculously cheap for a non-renewable resource whose days are numbered. With respect to Mexico, you have to keep in mind that wages are -it varies- roughly 1/10 what they are in the US, so for the typical Mexican motorist, the 335 pesos or so that it currently costs him to fill the 13 gallon tank in his Nissan Tsuru represents a far larger percentage of his disposable income than 30 dollars and change does for an American. And think, of the 190 or so national economies tracked by the World Bank, Mexico ranks around 45th in the world in terms of personal income. Imagine what "pain at the pump" must mean to somebody in a country that ranks about halfway down the list, (out of about 190, remember) like Albania or Bosnia or El Salvador. Actually, you won't have to imagine it for very long, pretty soon now we're all gonna get the chance to find out first hand.
My point here is not to appeal for sympathy for the majority of mankind that has no access to the energy products that fuel the first world's consumer orgy, and the entire planet's environmental catastrophe. I really can't recall hearing people down here bitch and moan about energy costs. The fuel costs what it costs, and they just pay, and the price goes up a few cents every month like clockwork. My point, if I actually have one, is to illustrate how far out of it most of the "enlightened scholarly discussion" in the peak oil community really is. Americans especially, and to a lesser extent Canadians, Europeans, Japanese, almost half of Mexicans, etc. (anybody at the party or who aspires to join it, or seriously thinks it isn't going to end badly) are all living in a fool's paradise.
It's kinda' like...
Nero didn't fiddle while Rome burned, he was just waiting for more scientifically rigorous modelling on how long it would be before his beard got singed enough to yell FIRE! and call the fire department. And since when does a lucid, tightly reasoned, peer-reviewed argument mean diddley-dick to a continent or two full of debauched drunks. (ooops, I mean "developed world political leaders, CERA, Rush Limbaugh types & other assorted *oil* addicts")
Westexas, re, Export Land Model, see the chart in this 3/9 NYT story on Mexico by Elizabeth Malkin: www.nytimes.com/2007/03/09/business/worldbusiness/09pemex.html?ex=133109... . Just an anecdote, but chart shows roughly level exports to US while overall prod. drops from 2004-06. Not picking a fight here, just saw this and passing it on. Is the 2-yr time just too short for the export drop to kick in, or is something special going on, e.g., we sell them nat-gas, we refine and pass-back some of their product, etc.?
The US is presumably Mexico's best customer, for a number of reasons, so I assume that we would be the last to be cut, but they did start curtailing crude oil deliveries to Gulf Coast refineries in January.
I think that the best model for Mexico may be the UK, which went from exporting one mbpd in 1999, to being a net importer in 2005.
The pre-peak Qt estimates for the Lower 48 and the North Sea are both consistent with post-peak estimates of Qt, and in both cases the regions peaked right at 50% of estimated URR.
WT, but was it possible to know in 1972 that Texas was at 50% of Qt? In 1972 you would have thought that Texas was at 70% of Qt. Now, 35 years after the peak, you know that Texas was around 50% of Qt in 1972. So, you see, the % of Qt at peak retreated with time. So how can you be sure that the world or Saudi Arabia is at 50% of Qt now? How do you know that the % of Qt for world and Saudi Arabia will not retreat with time, a la Texas? This is the point that RR has been trying to make for several days.
The world may be past or very close to peak. But you can't draw that conclusion from HL alone. Now, if Saudis fail to increase production as inventories fall and price skyrockets, that is a different story.
I suggest that you compare the Lower 48 HL plot to the world HL plot. As I have said several times, the pre-peak Texas HL plot is the noisiest of the four that we looked at in the Texas/Lower 48 article. I simply used the Texas HL plot to get the Probable URR, so I could then determine at what stage of depletion that Texas peaked.
The Lower 48, North Sea and world HL plots all have solid pre-peak HL profiles.
Then why not take up Robert's thrown down gauntlet.
Similar to his modelling by stepping us through at 10 year intervals, the data modelled out on HL. Only include those data that would've been know at the time the HL plot would've been figured.
I don't see why none of the HL defenders are seemingly unwilling to do this. Robert has repeatedly asked to be shown an example in which HL would've shown a peak in Real Time, but nobody has done this yet.
Why not? The lack of a working counter example at this point is perhaps more damning than anything Robert ever wrote.