The decline rates of individual wells are very steep, especially early in the life of the well (as much as 75% in the first year for the average Eagle Ford well), though the decline rates become lower over time and eventually stabilize at around 6 to 7% per year in the Bakken.
What is not obvious is that for the entire field (or play), the decline rates are not as steep as the decline rate for individual wells. I will present a couple of simple model to illustrate this concept.
Much of the presentation is a review of ideas that I have learned from Rune Likvern and Paul Pukite (aka Webhubbletelescope), though any errors in the analysis are mine.
A key idea underlying the analysis is that of convolution. I will attempt an explanation of the concept which many people find difficult.
I think the best way to present convolution is with pictures. Chart A below shows a relationship between oil output (in barrels per month) and months from the first oil output for the average well in an unspecified LTO play.
This relationship is a simple hyperbola of the form q=a/(1+kt), where a and k are constants of 13,000 and 0.25 respectively, t is time in months, and q is oil output. Chart A is often referred to as a well profile. The values for the constants were chosen to make the well profile fairly similar to an Eagle Ford average well profile. EUR30 is the estimated ultimate recovery from this average well over a 30 year well life.
Chart B shows the relationship between the number of new wells that begin producing each month and the months from the start of production for the entire field.
The convolution of the relationship shown in chart A and the relationship shown in chart B results in a third relationship between oil output and the number of months from the start of field output, which is shown in Chart C below . Output has been converted to kb/d from barrels per month.
It is indeed strange that two very different shapes (a hyperbola and a trapezoid) would combine to form the shape shown in chart C. A spreadsheet can be downloaded here, with the scenario above laid out.
What was surprising to me when I first tried this analysis was that a combination of the average well profile with the number of wells added each month reproduced the oil output data fairly closely.
To clarify this further, I have created a simple model. As before we have a hyperbolic well profile in chart 1 (slightly different than chart A above) and a the number of new wells added each month in chart 2, but in chart 2 this is over a short 6 month period. After that time no more new wells are added.
In the chart below I show the output for each group of wells that begins production in successive months. The output from all wells starting production in month 1 are labelled “month 1 wells”, there are 6 of these groups up to “month 6 wells”. The number of wells added each month is shown as a dashed line read off the right axis. Remember that 30 wells are added each month from month 1 to month 6 so output for “month x wells” will be 30 times month 1 of the well profile in month x and 30 times month 2 of the well profile in month x+1, etc.
The convolution of Chart 1 and Chart 2 result in Simple oil model 1 shown below.
This model is very simple in order to present how the principle works in a clear manner. When the annual decline rate for the “field” is compared to the average well’s annual decline rate, they are very similar for this simple 6 month model. More realistic models are presented later for comparison.
Note that month zero in the chart below is the month of maximum annual decline rate, for the average well the maximum annual decline rate happens in month 13 and for the field it occurs in month 18, the curves have been shifted to the left by 13 and 18 months so that the maximum decline rates match up at month zero for easy comparison.
The spreadsheet for simple model 1 can be downloaded here.
A second simple model with the number of wells added each month rising from 5 new wells per month to 30 new wells per month over 6 months and then falling back to no wells added by month 12 is shown below.
Note that the “month 7 wells” output curve is the same as the “month 5 wells“ output curve, but shifted 2 months to the right. Likewise month 8 is month 4 shifted 4 months to the right and this same symmetry is true for months 9 and 3(6 month shift right), months 10 and 2, and months 11 and 1 where the shift right in the curve is equal to the difference in the month when the well started production (8 months and 10 months for the last two cases respectively).
When all of these 11 curves are added up for each month (the convolution of the “well output of the average new well” chart and the “number of new wells added per month” chart) we get the Simple Oil Model 2 chart below.
Simple model 2 can be downloaded here.
I now present a different model with a higher EUR well profile (than in chart A) and a lower rate of addition of new wells (than in chart B). This model’s well profile is similar to the average North Dakota Bakken well profile.
The convolution of the two charts above results in the field output shown below.
How does the annual field decline rate compare to the average new well annual decline rate in this case? In the chart below we see that a slower decrease in the rate that new wells are added causes the annual field decline rate to be only 22% at most, about 3 times lower than the maximum annual well decline rate.
The spread sheet for the model above can be downloaded here.
As this result is rather counterintuitive, I will try another modification to the model. The well profile remains unchanged, but there is a steeper reduction in the rate that new wells are added to field production.
Such a scenario could occur if there was a steep drop in oil prices as in the early 1980s. It will also occur if there is a decrease in new well productivity which will reduce profits and the incentive to add more wells.
The well profile chart is unchanged, the other two charts are as follows:
Even in this case the maximum annual field decline rates are less than half the maximum well decline rate. This is because we have almost 15,000 wells added over an 11 year period and their decline behavior in the aggregate is much different than that of an individual well. See chart below.
Note that the field decline rate is very high, close to a 30% maximum rate in this scenario. If the rate that new wells are added drops to zero over a 1 to 2 year period and no further wells are added, we would expect the field decline to behave like the gray curve in the chart above. Spreadsheet can be downloaded here.
Earlier I mentioned that when I first tried this method I was surprised that such a simple model could accurately match output from the Bakken or Eagle Ford fields.
Using data from the North Dakota Industrial Commission(NDIC) on oil output, the number of new wells added per month, and individual well data(from Rune Likvern initially and lately from Enno Peters) I attempted to match scenarios initially presented by Rune Likvern at the Oil Drum.
Below I present the well profile and number of new wells added each month.
When the two charts above are combined (convolved) we get the output curve below.
Note that the sharp drop off in the number of producing wells added each month is not very realistic and is an artifact of the way I set up these simple models for illustration (they end at 130 months so the number of producing wells had to be ramped down very quickly).
Such a scenario would be more likely if there was a sharp rise in well costs, or a sharp drop in oil prices or new well productivity (EUR). The field decline rate is somewhat similar to the previous scenario, rising quickly to a 28% annual decline rate which falls to 10% after 5 years and to 7% in 8 years.
.
Spreadsheet for the scenario above is here.
A fairly realistic scenario for the North Dakota Bakken is presented now for comparison to the model above. This scenario has an ERR (economically recoverable resource) of 5.3 Gb where the more likely range is 7 to 9 Gb, based on USGS estimates. The average well profile and number of new wells added each month are below.
When we convolve the two charts above the following model output results. The match to the data is surprisingly good.
The annual field decline rate and well decline rate are shown below. In this case the maximum annual field decline is about 16% in 2021 and falls to 8% by 2026 and to 5% in 2031, the maximum annual well decline rate is 61%, the well decline rate is shown for a well starting production in Dec 2013.
The spreadsheet for the scenario above is quite large (21 MB) so if you have limited bandwidth skip it, download here.
For the Eagle Ford play I was able to collect data on single well leases from the Railroad Commission of Texas, data on the number of producing wells in the play and output data. I developed an average well profile (shown below) and combined it with the number of new wells added each month to produce an output chart. Note that the output chart is for crude only and does not include condensate.
The two charts above are combined (or convolved) to give the output chart below.
Note that about 20% of Eagle Ford output is condensate, when this condensate is added to the URR above for crude only we get a URR of 5.1 Gb of C+C. As in the case of the North Dakota Bakken/Three Forks the match between the model and data is surprisingly good considering the simplicity of the model and the complexity of the real world.
Eagle Ford spreadsheet is here.
Summary
Oil field output can be simulated with the convolution of the average well profile of newly added wells and the number of new wells added each month. I presented several simple models to demonstrate this concept. An obvious weakness of any attempt at forecasting is that the future average well profile may change over time and the number of new wells added in any future month is unknown.
The decline rate of a field of wells will be considerably lower than the decline rate of the individual well. The field decline rate depends on several factors: the decline rate of individual wells, the total number of wells in the field, the period of time over which these older wells were added (whether the period was long or short), and finally the rate at which the number of new wells added decreases as the field begins to decline.
Several models were presented showing how the field decline rate might vary under differing circumstances.
The concepts presented were applied to scenarios which simulated both the Bakken and Eagle Ford shale plays with fairly good precision.
Great work, DC.
ReplyDeleteAnother effect that a lot of people, cough, Ron, cough, don' understand is how average production per well will change as time goes on. If you imagine a field of identical Enno-curve wells being added, the average production per well will change (drop) just because of having more old wells in the group. (however those wells will also have lower decline curves). In ND, it's a little convoluted because you have all the old wells from pre-Bakken. But eventually, I think we see this patter of the production per well dropping because the impact of diluting pre-Bakken is done and now you have more and more old Bakken ones.
I read Rune's latest effort on his blog. Really, I think the guy is just confused. He repeats sentences, points. He muddles financial stuff with geological analysis. And then has a conclusion prediction at the end, which is not built up from his earlier discussion.
ReplyDeleteND Bakken crosses a million.
ReplyDeleteTell Enno I love him (again). Thanks for updating his cum curve. 2014 is looking pretty good.
ReplyDeleteNDIC Director said a month ago that JUN-JUL-AUG increases would be 5-6%. That translates to 50-60 M bpd/month. Call it 50 and you are still looking at another 100 over next two months. Seems like we will be at 1.1 after August comes in. SEP-OCT-NOV a little slower and then DEC the weather decrease. Probably 1.2 by year end.
Hi Nony,
DeleteA lot of this depends on how many new wells are added per month (if the average well profile remains about the same.) If an average of 200 new Bakken/TF wells are completed per month through Dec 2014, the model predicts about 1150 kb/d (call it 1100 to 1200 kb/d). The model has been lagging the actual output data so the well profile may be changing, though the well profile bounces up and down over time so the model is far from perfect.
Ron's latest analysis shows no comprehension of the idea that newer plays (EF, Niobrara) will have higher decline rates just because of the average well age. I don't think he really understood your explication from a while ago.
ReplyDelete1,047,000 from the ND Bakken. Rune's article and all the MSM attention to it looking more and more silly.
ReplyDeleteNony,
DeleteRune's analysis was correct except for underestimating how quickly the number of wells would increase.
He was off by about a factor of 2. He made a very near term prediction of 600-700 kbpd (IOW 650 +/- 50). Then he repeated it by rerunning his article a few months later. We passed his threshold within months of his last production and are knocking on 1.2 bpd now. And the only thing slowing it down now is price, NOT geology. IOW, overproduction from US LTO itself affecting world supply and demand.
DeleteHe was also very slow to say "I was wrong". And still really has not done so, explicitly. He had a lot of blather about how screwed up or conflicted other analysts were...and they were a lot more right than he was.
Hi Nony,
DeleteIn his articles he gave a forecast based on a certain number of wells added per year. Check his July 2013 post at the oil drum where he has 975 kb/d by Dec 2014 if 1800 new wells were added each year in 2013 and 2014. As I have told you many times he underestimated the number of new wells actually added, it was more like 2000 new well in 2014. In addition new well productivity increased, his forecast assumed the new well productivity would be unchanged. His forecast was wrong due to some incorrect assumptions (fewer new wells than actually added and new well EUR increased), but his forecast was only 17% too low, your 2 times too low is very far off the mark.
I'm talking about his initial prediction.
Deletesee: http://www.theoildrum.com/node/9748 with the comment "Now and based upon present observed trends for principally well productivity and crude oil futures (WTI), it is challenging to find support for the idea that total production of shale oil from the Bakken formation will move much above present levels of 0.6 - 0.7 Mb/d on an annual basis."
He never explicitly went back and said "scratch that". If he had, I would respect him. But "rowbacks" (look it up) are not honest.
It is a very simple thing to say "in SEP2012 and JAN2013, I published this article (link) with this prediction. It was low and now I revise it up." But he didn't! And he enjoyed all kinds of MSM attention on those initial posts. And he castigated people who disagreed with him (e.g. Mike Filloon).
He messed up and failed to admit it.
I find you much more straightforward.
He also had stuff about the average new well first year production trending down in that SEP2012/JAN2013 Red Queen article that got all the MSM attention. That has not held up at all! We've been stable to slightly improving. You know that from the Enno curves.
Delete[And it's been a major meme of the peakers when they talked down LTO as it rose. Either they thought the sweet spots would be used up or they were skeptical about downspacing and completion improvements.]
And he didn't go back with a simple clear, "I was wrong, here's what happened, here's what I learned in future...not just about the specific discrepancy but about perhaps his own negative bias.
But, honest. I have a paying job now. I should beg off of all the "somebody was wrong on the Internet" kerfuffle. What's in it for me? The thought of messing with someone like Futilist, just depresses me. He's not going to admit wrong and not going to learn anything. It would be a total waste of my time. Very different from interacting with you or Enno. And even you are not professionals. Take care, though. You are a gentleman. And the topic is interesting.
Hi Nony,
DeleteThere was a decrease in new well EUR in 2012 relative to 2011, then it rebounded in 2013.
If you read what he said, it was that the new well EUR was trending down, but that it might change.
If you read carefully it is clear what went wrong with his forecast, new well EUR increased and more wells were drilled than he predicted. In Dec 2014 he predicted 975 kb/d with 1800 new wells per year in 2013 in 2014, actual ND Bakken output was 1163, really not that far off especially considering there were about 3900 new wells added over those 2 years rather than the 3600 he guessed. If he was off by a factor of 2 as you suggest, Bakken output in Dec 2014 would be 1950 kb/d in Dec 2014, it is 800 kb/d lower than that.
He did a rowback (changed his prediction without saying he was wrong). Look at the SEP2012-JAN2013 article. He said in his summary points, 600-700 bpd.
DeleteHi Nony,
DeleteYes he was wrong, he underestimated the number of new wells that would be added in the future.
Not really a big deal, if you plug the actual number of wells added into his original model, the model was pretty accurate. The well profile changed in 2013 and 2014 so that his original model is off by a small amount, but you are nitpicking, nobody knows how many wells will be added in the future. Hindsight can be very accurate, most people have very little difficulty, "predicting" the past.
I wouldn't have a problem if he had forthrightly REVISITED his prediction and admitted it was wrong and explored why. He passed. That's a problem.
DeleteP.s. Also interesting what direction he was wrong. The direction all the peakers had. One of underpredicting production.
The bottom line is I see you as very different from Rune, Dennis. I see you as truth seeking and frank in revisiting your scenarios. That's the way to be.
Deletehttp://investors.clr.com/phoenix.zhtml?p=irol-eventDetails&c=197380&eventID=5164268
ReplyDeleteCheck out the CLR webinar. Especially OIP and TRR discussion at 50-55. But the other stuff on density testing, completion change are interesting too.
Hi Nony,
ReplyDeleteI doubt we will see a URR for the Bakken/Three Forks of 50 Gb. CLR has a tendency to take results so far and then extrapolate with the assumption that there will be no change in the average new well EUR.
Every oil field has sweet spots that eventually become saturated with profitable wells, then the decrease in new well EUR begins and eventually expected profits decrease to zero due to lower EUR and drilling stops.
The most optimistic estimate that is believable is the high end of the USGS April 2013 assessment which would be a TRR of 13 GB. If oil prices follow the EIA's AEO reference case and well costs do not increase (in real terms), the ERR will be 11 to 12 Gb.
Nobody knows what the OOIP is, most geologists would put it in the 400 Gb range. A realistic recovery factor is 2 to 3% that gets you 8 to 12 Gb for a URR.
Read the presentation. CLR has done some substantial work on OIP (not just relying on Price, but massive amount of data gathering). Of course it's still a guess, but it's based on lots of data and analysis. They say 413 BBL OIP. So same as usual.
ReplyDeleteWhat's interesting is that they see ~15% recovery as the end state.
Even EOG is saying 15-20 billion.
Maybe we need a new USGS study?
Hi Nony,
DeleteThe USGS study is fine. In the ND Bakken/Three Forks undiscovered TRR (mean) was about 5.8 Gb, in 2012 Reserves plus cumulative production was about 4 Gb, so roughly 10 Gb. Their high estimate (F5) was about 13 Gb. The 60 Gb estimate by CLR is way off, the low end of the EOG estimate(15 Gb) is a remote possibility, but not at $65/b. Not going to happen. You have seen that with all the increased frack stages and extra sand that well costs are going up rather than down (which was Maugeri's prediction). With all the excess rigs, well costs may fall a little, but so will the number of new wells drilled.
I have always said that a price crash would hinder shale production and felt it was the most aggressive aspect of Magueri's prediction (not the volume but that it could happen without the high prices). I think everyone saw year after year, at 100+ how US LTO kept kicking ass and exceeding forecasts. But we are seeing the rigs stack at 50! ;)
DeleteHi Nony,
DeleteWe will see if oil prices are at $65/b in 2012$ in 2017, and Bakken/Three forks at 1.5 MMb/d as he predicts, it may indeed reach 1.5 MMb/d, but if it does the decline will be steep even if there are 15 Gb of ERR. Maugeri's forecast will be far from the mark because he assumes new well EUR will never decrease, a bad assumption
I agree that he is likely too aggressive. However, he looks less insane than you all thought he would. Price at least has followed his thoughts.
DeletePeakers like Piccolo and Rune have been way off on their Bakken predictions. And the TOD commenters had a long history of talking down the US LTO. And it outperformed their expectations. Just surf old comments on the forums at peakoil or read the old comments on TOD.
Why should I look at these people as prescient for oil in general, when they were so off on LTO. And even so graceless in admitting it. (Not saying you...you and Enno seem graceful and truth-seeking. No one, cornie or peaker or in betweener can predict the future...but I get the impression you are intellectually honest.)
Hi Nony,
DeleteYou are correct that many (including me) underpredicted how quickly the Bakken would ramp up.
You should look back at some of my early posts on the Bakken, using Rune Likvern's analysis as my starting point. These are back in 2012.
Then I misinterpreted the USGS study when it first came out ( not realizing their prediction was undiscovered TRR), eventually this mistake was corrected by late 2013. Rune's analysis was excellent (Red Queen series).
http://seekingalpha.com/article/2510275-continental-resources-why-is-the-stock-down-after-an-impressive-analyst-day?uprof=45
ReplyDeleteInteresting analysis to show that CLR 603,000 type curve is not representative of their reserves.
How about that oil price? Not rising the way most of your scenarios predict.
ReplyDeleteAlso not what the EIA predicted, which is what most of my scenarios use.
DeleteFutures curves were closer...
Delete;)
In July 2014 at 4 months out they were at $90/b
Deletethey were backwarded. You and EIA were the reverse. Prices went down. Thus, they were closer.
DeleteP.s. Also, I think the exact shape is not so interesting. (your comment about straight ramp for Maugeri.) What is interesting is something like a 2 or 5 year horizon out. That gives you the market "bet" on the quasi-equilibrium supply-demand balance. IOW, does the market believe in "peak oil". My point to you was that the markets were showing long term oil at 80 when we were at 100 (this is longer than the 2-5 year window, but same point). So clearly in a world that is industrializing, markets did not see "peak oil". Conversely, looking at it now, the markets clearly project out to around 70 in the long term. so they don't believe in some sooper cornie dream either.
Hi Nony,
DeleteSo the futures market is saying roughly $70 to $80 in 2019 or so. You realize that as often as not the futures curve turns out to be very wrong. The EIA does a supply demand model to come up with its forecast, which goes to 2040, well beyond the futures curve.
So you think that oil will not peak? If that is the case (note that whether it is supply or demand is irrelevant, supply is equal to demand), you will be wrong. It is possible that the peak may not arrive until 2020, but I think 2017/2018 is more likely, a lot depends on the real price of oil. If the futures market is correct, possibly the peak could be 2019, but the futures market 4 years out often gets the prices wrong. I think the EIA's reference case will be closer to the mark in 2020. The futures market is predicting about $62/b in 2015$ in 2019, this will be too low. We will likely be over $100/b in 2015$ by June 30, 2019.
The price prediction from Leonard maugeri not looking so crazy now is it???
ReplyDeleteMaugeri predicted $65/b in 2017 and remaining there long term, note that prices too low would make his model not work because it depends on reduced well costs by 8% per year. Current prices make drilling unprofitable so we would expect output to flatten out or perhaps fall, depends on how many shale producers go belly up.
DeleteIf you look at the futures curve for 2017, it's eerily like Magueri. I think if current futures holds up, he will have done a stellar prediction on price. Much better than you or EIA or even the futures curve at the time he made his predictions (but the last is the closest).
DeleteI agree that LTO will likely have a hard time holding up with low prices. Look at Bakken rigs below 120 now. After all, we had already seen that play turn off in 2008-2009. No surprise.
I double checked Maugeri's paper on the LTO boom. The $65/b is nominal not real oil price. So assuming 3% inflation this would be $59/b in 2013 $ by 2016 with prices remaining at this level long term (he uses nominal oil prices so presumably the real price would decrease by 3% per year).
DeleteSo Maugeri's predictions are below the futures curve. The Dec 2023 WTI future is at about $73/b.
The futures market thinks Maugeri's oil glut fantasy is incorrect.
In his Oil: The next revolution, the extra capacity comes primarily from US LTO, Canada, Venezuela, Brazil and Iraq. So far, Canada and the US have delivered, Iraq(7.5 Mb/d), Brazil(4.2 Mb/d) and Venezuela(3 Mb/d) are unlikely to increase output to the levels Maugeri predicted by 2020.
Ron's questions on ND Bakken increase:
ReplyDelete1. You can't try to implicitly compare linear change in completions with linear change in growth. Because of decline. You could implicitly compare it to NEW OIL (ala EIA drilling productivity). This makes the "3 times greater growth", much, much smaller variation in production
2. He also doesn't seem to take note of adjustments in previous month production much. IN previous 2 months, the "small" increases were bigger than observed because last month had been adjusted up. This last big jump did not have that. So difference smaller than he thinks.
P.s. I love the oblique allusion to ND lying and such. So doomer tin foil conspiracy blogger...
Had an idea for a cool Enno analysis.
ReplyDeleteShow his cum curves for CLR Bakken wells by year. Should help to see if they are getting better/worse. Maybe provide some insight on this question of their EURs being too optimistic.
(WHT kerfuffle)
ReplyDelete*Shale plays have different characteristic "b". Therefore "aha b=1.4 in the Bakken, therefore diffusion physics" is not supported. Because what about the Barnett or the Haynesville?
*I did a Google scholar search. Several papers have been written on b factor for the Bakken (estimates range from 0.5 to 2).
*Even the Mason paper actually says 1.1 to 1.4 (if you read all the text, not just look at the figure caption).
*The Mason paper is not a very good paper (not what I'm used to for a good journal or a thesis). Covers a lot of ground, written by a consultant, doesn't drill in depth on the curve fitting.
*The USGS paper on Bakken depletion curves is much more substantial.
http://pubs.usgs.gov/of/2013/1109/OF13-1109.pdf
-huge scatter of b for the wells [see chart 3] Not what I'm used in semiconductors...even if you have some band gap differences, the basic 1/T plot is still linear. But here they are SO vastly different from well to well, it becomes sort of silly to try to have a micro model for the "average one". What is the model for the well at 0.5 versus the one at 2.0 versus the one at 1.0 versus the one at 1.5. They're all physical. (that's not physics to have a model that does not work for the different samples so wildly). You don't have a good microscopic explanation with power law changing exponent by that much.
-the average b's were lower than 1.4. More like 1. (see chart 4)
*I would be leery of someone really throwing a lot of fancy words out there and putting guys in the box who don't know all the math...when he can't explain the physics.
*I would ALWAYS BEWARE physicists who rely too readily on power law intuitions and who make broad interpretations of complex systems with them.
Hi Nony,
DeleteJust do some reading at context earth, the concept of dispersive diffusion is a useful one.
We are interested in the average behavior of the field, more of a macro model than a micro model, think classical thermodynamics rather than statistical mechanics.
(Cutting and pasting a comment left on James Hamilton's blog. He moderates my comments.)
ReplyDelete1. You had a VERY widely publicized article talking about 'hundred dollar here to stay'. And you covered the John Kemp criticism and Kopits counter on this blog. I think you owe it now to not just comment on current prices, but to remind the readers of your predictions and to revisit them, specifically. Not just the wrong call, but why. To link the discussions. There's a HUGE change now, even fore you, where you talk about prices rising sometime to the 80s. That's pretty different from 'hundred here to stay' in terms of the prevailing paradigm.
2. You have (in passing) mentioned that you think most of the drop in prices is due to supply. However you have NOT written in detail on the details of that. This is confusing given you say it is the MAJOR factor. Also, given that you have previously poo-poo-ed the impact of US shale. And there is one supply aspect (depletion) that you like to cover in extensive detail. So why not get into the meat of analyzing the new supply?
3. As far as understanding demand, I think it is useful to disaggregate current demand changes from expectation of future demand. The key metric here is CURRENT VOLUME. Thus in 2008-2009, there was an actual at the time, drop in volume coincident with the price crash. And right now, we don't have that, happening. This is a HUGE insight. This then brings up the interesting aspect of how much is the now low price based on expectation of China slowing or how much is the old high price based on an unreasonable expectation of China prolonged growth. (You could say this is semantic, but I think it's an important perspective to ponder. E.g. Should we think of the last 4 years as an unreasonable bubble.)
4. It would also be useful to revisit some of your comments on the impact of OPEC. Your previous discussion (sort of "analysis by lit review") minimized the impact of OPEC by referencing several papers that said it had minimal impact on prices. However, we did see a big drop in Thanksgiving coincident with a NON-INTERVENTION. And this was even after news stories estimated a 40% chance of an intervention. This implies that the market did/does price in OPEC actions. It makes one wonder, how much of the high prices during the "hundred here to stay" 4 years, might have been based on assumptions of OPEC impact.
Hi Nony,
DeleteNobody can predict oil prices with any accuracy. Do you think that Maugeri's predictions will be correct because of a short term decrease in oil prices. My oil price forecast was simply the EIA reference case, the EIA got it wrong, so I did as well. What do you think prices will be in Sept 2015? My guess would be $75 to $85 per barrel for Brent Crude, which is the benchmark that the EIA now uses.
Hi Nony,
DeleteThe difference today vs 2008/9 is that OPEC cut output in 2008/9 to try to stabilize prices and that explains the change in volume then vs now. In 2008/9 economic conditions were very different with a sharp drop in demand which caused the price crash. In this case supply grew more quickly than demand so prices fell. The expectation was that OPEC would cut output as they usually have done in the past, this expectation was not met so prices fell sharply in response.
Many including the EIA, and futures markets did not see this price drop coming. In fact Maugeri wa predicting a gradual price drop to $65/b by 2017, so he also was wrong (but closer), the EIA low price scenario had oil prices at no lower than $70/b. Though I imagine that you must have known what would happen to the oil price in advance.
Interesting to see Hamilton using my key insight about current versus expected demand. He didn't cast his blog post (or previous ones) in that lens, but it's a key, key aspect. Likes to delete my comments and use my points.
ReplyDeleteIn addition to expectation of future demand, there is an issue of expectation of future supply. I think it's an easy thing to consider that people's views of shale changed. Remember when it was "not going to have any impact". And then it beat predictions (even the EIA predictions that peakers said were too optimistic) three years in a row. Then it was the comment about it running out in a year or two. But more and more people saw the drilling location inventories in EF and Bakken producers presentations.
So you have an industry that consistently delivers 1MM bpd/year increases in volume. That has an impact. It makes 100+ unsustainable. It's basically a ceiling. And basically complete opposite of Hamilton's article that said 100 was a floor. And for a very interesting microecomic reason (supply at 100 is large). But he doesn't think with micro insights. He's a trend follower.
Hi Nony,
DeleteI did not expect the Bakken and Eagle Ford to produce as much as they have at first. USGS estimates early on were too low, but I think they may have it about right at this point.
In December 2012 I wrote the post at the link below:
http://oilpeakclimate.blogspot.com/2012/12/quick-update-to-tight-oil-models.html
The Bakken model was not that far off, but I didn't realize that the Eagle Ford would ramp up so much more quickly than the Bakken.
Keep in mind that not all petroleum liquids are the same so the key metric is C+C, not all liquids (you can't produce diesel and jet fuel from NGL or ethanol). For the World C+C has increased from 74 MMb/d to 77 MMb/d (12 month averages) from 2006 to 2014, that is less than 0.5 MMb/d per year.
For all liquids there has been about a 1 MMb/d per year, if demand grows more slowly than this then prices will fall. Most economists would expect the lower prices will result in higher demand, and lower supply of oil what would tend to happen to the oil price in that case. What is your expectation? Have you seen the oil rig count? Are they going to use shovels to produce the oil? :)
I guess on a world basis, you are right. I was just talking about US. I think we had 1 MM bpd in 2013 and 2014 (maybe even a little more last year) and then the NGLs on top.
DeleteI agree that C&C is the relevant metric. I think of condensate as an extralight oil. It really is an oil and has traditionally even been price advantaged over WTI...and still gets decent prices compared to heavy sour. It runs through a refinery and makes normal products. Just in the US, we have too much of it recently and the refineries were built for heavy sour. But it's "oil".
NGLs are more like "extraheavy gas". They are a limited substitute for oil. butane blending into gasoline or isomerization to isobutene for use alkylation units of refineries (for gasoline stock). Propane for LPG in the US mostly and separate markets (maybe a little heating oil competition LT, but not much near term substitution.) Ethane is crack stock and to some extent a naptha substitute. But even there we have so much that it gets put into the NG and burned now. So, yes, net net I DON'T confuse NGL with crude. I think of it more as gas. After all, it is gas at RT! ;-)
isobutane, not isobutene. Spell correcter made the change!
DeleteI agree the US increases have been impressive. I think it is possible that the EIA predictions for the shale peak may be realized, but my expectation is that the fields will decline more steeply than the EIA predicts, especially if Maugeri's oil price predictions of real oil prices at $65/b (in 2012$) over the long term are correct, higher prices might mitigate the decline a bit, but the overall ERR of the EIA's forecast is not consistent with USGS estimates and in my estimation the geologists at the USGS are more likely to be correct than the economists at the EIA. Time will tell.
DeleteLatest article on Ron's blog makes the tired mistake of not understanding how production/well just drops naturally based on population aging. You've pointed it out before (Ron didn't understand it). Don't you get sick of the lack of comprehension?
ReplyDeleteNo I think Ron and his blog are great.
DeleteIF I take 5 samples of gas, they will all follow PV=NRT.*
DeleteI have much more trust in just the phenomenological curve fitting than in some idea that there is any advantage in thinking that there is a microscopic explanation of the behavior that is useful from diffusion analogies or equations.
*at some decent distance above condensation temp/pressure.
That was meant for the WHT stuff. Anyhow, I'm repeating myself. You can have last word. Points stand.
DeleteWhat I meant to say here was "you are a gentleman". Glad you can get along so well with those you agree with, disagree with, and even agree/disagree partially with.
Hope all is well with you personally and thanks for all the attention to my posts. Real life is more important than Internet.
I may not answer much as I tend to get Internet addicted and have a pretty high paying project starting up. :-) [Need to behave.]
Thanks Nony.
DeleteYou are relatively polite here. You should pick another handle and post at POB, just get some temporary email account, it would be interesting to have you and Futilitist discuss stuff from opposite ends of the optimist-pessimist divide.
Be careful of WHT. He is a second rate physicist blathering a bunch of stuff about power laws. Anyone who has deep experience in science knows how easy it is to get correlatations from a semilog or log log stragight line fit. In addition, he ignores the MASSIVE scatter of individual wells even versus his power law argument.
ReplyDeletePlus on top of that, he likes to squirt squid ink with assumptions that others can't do Bessel functions.
He's a second rater. Watch out.
I disagree. There is always individual well variability, big deal. You take the average well and the field output can be modelled very closely. Your complaints about Rune Likvern's work are a little like criticizing Hamilton for not predicting oil prices correctly, pretty silly.
DeleteRune simply underestimated how quickly the number of producing wells in the Bakken/Three Forks would increase. Perhaps your crystal ball is perfect, most of us have difficulty predicting the future perfectly.
Sure, you will have differences in individual wells in terms of IP. But the difference is not just in that but in decline curve. Read some papers. the scatter on the exponent is massive. Now if I take some random pieces of metal, they will have different resistivities based on grain size or even on element. But the shape of the curve versus temperature will still be very similar.
DeleteThe point is that if you actually have a physical intuition for the exponent, then you should not have the spread that you have. Spreads from .5 to 2.0 are MASSIVE for a power law. You have much more case for just making a curve fitting argument and just picking an exponent than his claims on the square root of 2 BS. [And then leave in that many shales have an average number noticeably different from 1.4.
You're left with a dude who puffs himself up on the Internet and is full of hot air because he can initimadate a few people with some squid ink physicsy math.
The analysis is about macro behavior. Do you think every molecule of nitrogen in a container of air has the same kinetic energy? We are looking at average behavior of the average well. The dispersive diffusion argument makes a lot of sense from a physics perspective. You should do some reading at context earth. Or you can hand wave all you want.
DeleteAgain...you can have last word. But with WHT, one last thought. Realize he is a guy "publishing" blog posts and such. He is not in the mainstream of well estimation publishing in the best petroleum journals, driving textbook explanations, etc. What do you think of intelligent but not mainstream climate skeptics who "publish" on blogs? It's similar here...there is a discipline that you get from running through peer review. Doesn't make everything perfect...but catches a lot of crap. WHT has not run that check. And I suspect, real experts in reservoir mechanics would be pretty unimpressed by his stuff.
DeleteHi Nony,
DeletePossibly true. I don't have access to all the peer reviewed literature, much of it is not available for free. The explanation looks good to me, I am not an expert on reservoir mechanics. Either the dispersive diffusion model or a hyperbolic model can be used to approximate the field behavior.
If you look at the work of David Hughes or the Bureau of Economic Geology, the results are very similar to what I do, which is simply fit a hyperbolic to the average well data, with an exponential tail (which a petroleum engineer with 40 years experience described as boilerplate.) Then use the number of new wells added to model the output. The dispersive diffusion model also works just fine and has the advantage of an underlying physics justification. The hyperbolic is used simply because it fits the data, no physics explanation is offered, if that is your preference, its all good.
Really much ado about nothing.
As an economist, I find there is a critical and huge missing factor in any and all these analyses. Over and above all the complex and beautifully constructed physical modelling, as exemplified above, the number of wells drilled (and maintained) or more specifically the decision to do so is not linked to the prices of the relevant output. The relationship(s) between prices (appropriately defined) and well decisions has not been modelled. The challenges in achieving that are large:
ReplyDelete1)Given that prices were relatively stable for most of the ramp- up period, there may well not be enough data (even with good sources) to know the sensitivity of well decisions to the market prices as they rise and fall.
2)If the relationship between well decisions and prices is non-linear, which it probably is, and as we know the period of history has no price data anywhere near current price levels; modelling of any kind is hazardous .
3)Next, the decision-makers in many cases will be hedging the relevant prices. There may be no data on this and so we do not know the expected price of the decisions makers even if we managed to model future market prices correctly.
4)It gets worse. My gut feel is that activity has been phenomenally profitable and supply of physical resources has been the constraining factor, rather than the rate of return, for most or a significant chunk of the period that can be analysed. This makes it even more difficult to work out the sensitivity of drilling activity to the price; even if we knew what the expected price of the decision-makers, adjusted for hedging activity, was at any point.
Anyway, my suggestion is that the focus shift to trying to model this aspect.