by WebHubbleTelescope » Sat 28 Jan 2006, 22:19:10
$this->bbcode_second_pass_quote('ReserveGrowthRulz', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')As for Verma, here is a telling quote (
USGS):$this->bbcode_second_pass_quote('', 'V')olga-Ural and West Siberia provinces show most of their reserve growth in the first 5-7 years after discovery and little or no growth thereafter. It is difficult to compare the growth in Russian fields with those of the U.S. fields where growth continues even after 90 years, because in Russia oil fields are first evaluated over a 5-7 year period before being produced whereas in the U.S. both the evaluation and production of fields start shortly after their discovery.
Could it be that the Russians actually do real honest-to-goodness predictions while the USA does very conservative measurements? The Russians actually "evaluated" over a 5-7 year period. Apparently this is enough time for them to estimate the full life-cycle of a field.
So let me paraphrase....after professionals who study a future field for the better part of a decade start producing the SAME field, and then have to highly alter their original estimates over another period of time spanning the better part of a decade, this all strikes you as a perfect example of what?
How smart this group of people was that it only requires what, 15 years or calculating and producing to minimize reserve growth? That reserve growth does in fact exist even when people try to avoid it? That its effects can be minimized if only exploration and developement staff "study" something for years before actually DOING something? A habit which might be great in a highly structured environment, which isn't necessarily the environment most prevalent in the rest of the non nationalized oil and gas industry?
It sure sounds to me that while you don't "like" the idea of reserve growth, you sure aren't talking about it not existing any more, but rather complaining about its size and scope.
So are you ready to talk about quantity of reserve growth rather than its very existence?
I don't understand what you are trying to say. I do however enjoy your criticism of amateurs. Like average people that invest their money in the stock market, treasuries, and bonds, nothing says that they can't become knowledgeable about the field. They don't actually have to know the details of printing money, etc to know how to look at the numbers and figure a way to maintain their investment. I don't think any amateur investor has ever been coyed by the thought of Alan Greenspan or Warren Buffet lecturing him about his amateur credentials.
Saying I don't like reserve growth? I am only trying to discern the truth. I am certainly using discovery curves that are backdated, am able to model the USA production profile, and certainly don't see how reserve growth can turn back the hands of time to 1970 and remove the USA Hubbert peak.
I still haven't a clue as to whether reserve growth is a book-keeping ploy designed to attract investments over time or something truly related to a technical advancement. I can appreciate that as you drill deeper and explore more difficult areas, that it can contribute to a parabolic growth law. What I don't see is why you can't figure this out ahead of time.
Let me give you an example that should blow everyone's mind. Take the case of the Alaska National Wildlife Refuge. Say we have estimates of how much oil is available there. I think I have heard estimates of
6 billion barrels on the low side (the number does not really matter). Now with the amount of political contention surrounding this issue, you wouldn't suppose that this actually should be
60 billion barrels, all due to reserve growth? Wouldn't this help sell the idea to drill there?
You do believe in reserve growth don't you?
So put your money where your mouth is. Tell me the low-end estimate is actually 10x higher than what the current Alaska National Wildlife Refuge number bandied about is. And if the low-end 6 billion does actually include the reserve growth correction factor, then I have to ask why we don't routinely do this on all our estimates, before we even start drilling? Like stock market rules, it benefits everybody and establishes a level playing field. Would it be too sarcastic for me to say how much I appreciate the numbers game that the Saudis and Kuwaits play?
You say I complain about the size and scope of reserve growth. Now that I understand it a bit more, I think I can incorporate it into my model quite readily. Right now, I include a phase called
maturation into my model. This occurs after the construction phase and before the extraction phase. What maturation establishes is a time constant to describe pumping to full capacity. Ordinarily I put an exponential function of time constant around 5 years for this phase. Replacing the exponential with an inverse square root distribution to spread the maturation period out may actually be the right thing to do as it replaces a guessed at variable with an experimentally observed relationship.
by ReserveGrowthRulz » Sun 29 Jan 2006, 00:11:29
$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')I don't understand what you are trying to say.
Well then I'll go back to the basics. What is your measure of reserve growth and how do you account for it within your own models? I have mentioned this before.
$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')I do however enjoy your criticism of amateurs. Like average people that invest their money in the stock market, treasuries, and bonds, nothing says that they can't become knowledgeable about the field. I don't think any amateur investor has ever been coyed by the thought of Alan Greenspan or Warren Buffet lecturing him about his amateur credentials.
If you can mention any particular amateurs in the oil and gas industry who have made contributions to this reserve growth problem? I've got a bibliography of most of the experts on reserve growth on the planet sitting back at my desk, pick out your favorite amateur and I'll go review his work. While your "industry shopping" for examples is amusing, lets try and stick to what Peak is related to?
$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')Saying I don't like reserve growth? I am only trying to discern the truth.
Excellent! So what is your measure of reserve growth and how would you like to best determine it, taking into account your desire for truth?
$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')I am certainly using discovery curves that are backdated, am able to model the USA production profile, and certainly don't see how reserve growth can turn back the hands of time to 1970 and remove the USA Hubbert peak.
by WebHubbleTelescope » Sun 29 Jan 2006, 17:10:19
$this->bbcode_second_pass_quote('ReserveGrowthRulz', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')I suppose I should have known this but I really thought that a scientific method was at the heart of all this, not some bureaucratic shenanigans.
I would respond in the following way....your position that the SEC itself causes reserve growth is contradicted by your own reference to the Verma and Ulmishek paper.
I might also mention that Lynch's work examining reserve growth in Colin Campbells numbers using international data sources would ALSO have nothing to do with the SEC.
I suppose when one is committed to zealotry, its difficult to remain internally consistent?
How about this as a trick question for someone who says they have the A&R paper but doesn't consider it "scientific" enough for his tastes...because A&R and Verma used a similar method ( Arringtons, modified slightly by the authors of both papers ) to calculate their reserve growth numbers, one inside an SEC regulated environment and one outside, is it possible to calculate the "effect" of SEC regulations on reserve growth?
If you aren't familiar with the "trick" in this question, I will of course use the opportunity to bash amateurs again.
I don't think everyone understands the term speculative. In the most conservative terms, speculative is anything that you can't see or touch at the current moment. So apparently they can't report much beyond what they have extracted -- excepting the initial estimate they make, which is still quite conservative. In which case the reserve growth does not look much different than the creaming curves and cumulative production curves that we also see reported. And we all know what happens when those start going horizontal. In the case of offshore, they get shutdown. But onshore they keep pumping because of the lowered costs and the use of stripper wells.
And I know the "trick" to which you refer; it is called
backdating. I have it covered in my model, and with my maturation phase, I can account for this kind of behavior.
I read the synopsis of the Russian studies, as done by Verma, who said that the reserve growth effects were much less in Russia, presumably because they allow speculative estimates. Let me quote
again what Verma said:
$this->bbcode_second_pass_quote('', 'V')olga-Ural and West Siberia provinces show most of their reserve growth
in the first 5-7 years after discovery and little or no growth thereafter. It is difficult to compare the growth in Russian fields with those of the U.S. fields where growth continues even after 90 years, because in Russia oil fields are first evaluated over a 5-7 year period before being produced whereas in the U.S. both the evaluation and production of fields start shortly after their discovery.
And another point, the approximate parabolic growth law can't go on forever as hard limits do exist. In my analogy to silicon oxidation, the parabolic growth law is infinite because the thickness of the silicon wafer is much greater than the oxide. I have not gone to quite the degree of zealotry to those who think the parabolic growth law can continue indefinitely for fossil fuels. It is clear that the hard limits do exist, otherwise we should be going back to Pennsylvania and Ohio to get the infinite supply that remains.
So it is easy to see why USA growth continues even after 90 years. It is because we are still pumping oil, and because of the SEC reporting rules to report only non-speculative results we have to continually go back and backdate the discovery numbers. Under these conditions, not seeing a reserve growth would be surprising. It would mean we were completely out of oil! Think about that one.
I will continue to enjoy playing this game of 20 Questions, as I practice this stuff every day at work. I heard one bigwig call it "stump the chump".
by ReserveGrowthRulz » Mon 30 Jan 2006, 19:52:47
$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')And I know the "trick" to which you refer; it is called backdating. I have it covered in my model, and with my maturation phase, I can account for this kind of behavior.
Sorry, thats a Campbell "trick", not the one I was thinking of. The reason its a Colin "trick" is because it immediately shifts the cumulative growth functions to a point farther out along the curve which then drops year over year growth to much smaller multipliers. A handy "trick" if your main game is to influence the data away from things you don't like, a tactic which you appear to disapprove of only when others besides Campbell try and use it. Verma bumps into this effect as he toggles back and forth between using discovery year and production start year as his starting point. ( 2nd paragraph of the abstract in "West Siberian Basin" paper is the example I'm thinking of related to what you consider a "trick" ).
I might also mention his conclusions point up something I've mentioned and you want to use as a complete and total excuse to do away with reserve growth, which is the effect on booking requirements and lack of capital investment in countries other than the US. Oh...and you forgot to mention, what were the results from Verma's studies on reserve growth in these areas of poor capital investment and non SEC booking requirements? Thats right....3X minimum? 6X? Using discovery year? 12X 15X?
Seems like your desire to use the SEC as an excuse doesn't work in Siberia either....and then when Verma compares these basins to the North Sea as far as their reserve growth?
Any other quick theories I can dispense with out of hand for you?
Also, I don't know if you referenced it, but SPE Paper # 62616 is where I got more V and Ulmishek information on the Volga/Ural Province in Russia.
$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')I will continue to enjoy playing this game of 20 Questions, as I practice this stuff every day at work. I heard one bigwig call it "stump the chump".
You have any other amateur "reserve growth doesn't exist "experts who's published work I can check out at the office or do you want to continue using names and references which I furnish which show reserve growth planet wide? And then continue insisting its somehow all the SEC's fault?
by ReserveGrowthRulz » Tue 31 Jan 2006, 14:53:44
$this->bbcode_second_pass_quote('WebHubbleTelescope', '')$this->bbcode_second_pass_quote('ReserveGrowthRulz', 'o')ops
Save that thought, here is
the latest on reserve growth modeling
COOL! So you are now agreeing that reserve growth is a parabolically limited type thingie/effect, approximately 10-15X of initial discovery size estimate in scope?
Let me throw out another factoid real quick, and you can tell me if it appears to fit in with your theory....this is from the abstract of the Verma and Ulmishek paper, last paragraph. "models for the conterminous United States predict a growth of 0.54-0.75% per year." Considering the way you have modelled reserve growth from the A&R paper, does this sound like a number reaching towards year over year growth of 0%, or not? Considering the age and size mix of US onshore and non continuous fields of course.
by WebHubbleTelescope » Tue 31 Jan 2006, 21:11:06
$this->bbcode_second_pass_quote('ReserveGrowthRulz', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '')$this->bbcode_second_pass_quote('ReserveGrowthRulz', 'o')ops
Save that thought, here is
the latest on reserve growth modeling COOL! So you are now agreeing that reserve growth is a parabolically limited type thingie/effect, approximately 10-15X of initial discovery size estimate in scope?
Let me throw out another factoid real quick, and you can tell me if it appears to fit in with your theory....this is from the abstract of the Verma and Ulmishek paper, last paragraph. "models for the conterminous United States predict a growth of 0.54-0.75% per year." Considering the way you have modelled reserve growth from the A&R paper, does this sound like a number reaching towards year over year growth of 0%, or not? Considering the age and size mix of US onshore and non continuous fields of course.
Yes, I agree that the growth of the original
non-speculative estimate likely grows that much. However, if we had any balls at all, we would be making the speculative prediction right away. In that case, the growth would be much smaller. Going back to the weatherman analogy, how many 100 degree days in a row will a weatherman keep predicting 72 degrees before he gets a clue to start raising his prediction? The way that Bayes theory works is that you use previously obtained data to adjust to a probabalistically better prediction. After a while the 100 degree days will stop being 28 degrees above average and turn into
the de facto average.
As far as the 0.54-0.75% growth is concerned, if it follows the square root progression, it has to getting progressively smaller with time. The only question is whether it gets smaller faster than 1/sqrt(t). You really need the whole curve to look at the trend. The number quoted looks like a snapshot in time.
by WebHubbleTelescope » Wed 01 Feb 2006, 02:37:50
$this->bbcode_second_pass_quote('ReserveGrowthRulz', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')Yes, I agree that the growth of the original non-speculative estimate likely grows that much. However, if we had any balls at all, we would be making the speculative prediction right away. In that case, the growth would be much smaller.
As far as the 0.54-0.75% growth is concerned, if it follows the square root progression, it has to getting progressively smaller with time. The only question is whether it gets smaller faster than 1/sqrt(t). You really need the whole curve to look at the trend. The number quoted looks like a snapshot in time.
Well, it looks like, because of the speculative nature of the intial guess, the primary cause which you keep siting, versus my comfort with some of the more conventional reasons listed in the papers we've been referencing, we can agree that this thing is going to go on for awhile?
I might also mention that Hubberts measure of reserve growth was similar to your basic assumption/explanation on your weblog, his being an exponential function which approaches an asymptote. Not being a mathmatician, I don't really care what its called, but it sure sounds like your description of a parabolically limited function.
Do you have a functioning model of your version of peak, with reserve growth layered in through time based on some sort of discovery/production profile? I don't mean simply going back and 5X the initial discovery size estimate, I'm thinking more along the lines of something which layers in the addition volumes through time? I think you mentioned a way to model it in before but I'm not certain I was convinced at the time the suggestion you had was valid.
If you've already done this, or something preliminary, point me at it and I'll go check it out.
My model includes a few phases so that the extraction doesn't track the discovery size estimate and reserve growth initially, but eventually merges asymptotically into it at longer times. The reason it doesn't match at all initially is because it takes time to lay claims, build the rigs, build the pipelines and other things for the field to mature so you can start pumping. And this average time lag is a stochastic variable. So when you see that the estimate shoots up right away, I assume that comes about from test drills and not from the actual pumping. I hope that this is pretty apparent from the following chart:
The green curve is generated from my USA-lower48 model with a normalized backdated discovery from 1899 using the same parameters as I used for the entire fit. You can see that it takes
on average around 10 years for any amount to flow at all, and then the average lags behind until it eventually merges after a number of years. The 10 years is an average, of course some are less and some are more, but this is what happens in a statistically distributed population. So at the moment you start pumping, on average, you have only 2x actual reserve growth left, not the 11x.
In summary, the estimate is not the cumulative extraction. It couldn't be, otherwise every field would start pumping immediately and everyone understands that just doesn't happen. Even today, it takes years for many rigs to come online, and so they have plenty of time to improve the initial estimate. Unless I am missing something and the discovery date is not the actual "discovery" and it actually establishes the date at which the fields come on line. But I kind of doubt that. So I still think it is safe for me to use backdated data or the 11x pro-rating to establish an eventual cumulative maximum. The model includes an extraction rate which is proportional to the amount remaining so everything works out in the end.
By the way, the A&R reserve growth heuristic is also a stochastic model based on a statistical population of sampled data. So we are comparing apples and apples by overlaying these curves.
