[Pops]

Hats off to Ron Patterson at POB for continuing good work, this is from a post by David Archibald

$this->bbcode_second_pass_quote('', 'T')he seven years of production of tight oil in the US has produced enough data to
enable estimation of the amount of oil that will be recovered from these systems and
the timing of peak production. Based on data to May 2014, the four main tight oil
basins will produce a total of 7.7 billion barrels with a peak production rate of 3.9
million barrels per day in mid-2015. Following that peak, production is predicted to
decline as rapidly as it rose. That in turn is expected to cause a re-assessment of the
ability to produce sufficient transport fuels based on current policies.

The Bakken in North Dakota

Jean Laherrere has plotted monthly oil production from the Bakken Fm in North
Dakota using Hubbert linearization:




...

[MD]

Nice!

Doomers need not worry about this. It's irrelevant.

[Tanada]

Grand timing, right at the start of the 2016 North American silly politics season we hit LTO peak and go into decline. I predict all sides will be pointing fingers and calling names instead of doing something useful.

[ROCKMAN]

No problem if folks want to call this a peak in production in US oil production. But so far it isn't THE PO of the US. There have been short term peaks in US oil production in 1938, 1958, 1971, 1985 and the current peak. Of course, the current peak (for now) is just 80% of the 1985 peak. It could happen but we still have to increase current production to match THE US PO of 1971. Time will tell. Of course we're till waiting to see if a plateau in oil production develops as it did for about 15 years beginning in 1971. The current surge has lasted only 1/3 of that time so far. Again only time will tell.

[Pops]

I'm not a mathematician, not even much of an arithmetician but the linearization plots in the article are interesting in that they have already settled down to a pretty straight line. Barring another price bump or technology leap or whatever the ultimate amount gettable at this price and tech seem pretty cut and dry.

The big difference between this guess and the EIA guess is the far side of the curve. The EIA thinks the rate of depletion will catch up to new production as new locations dwindle so production will peak at about the same time as the OP.

But the EIA predicts a long flat tail. I'm not sure why. If a lack of new well locations to replace rapidly declining wells causes a peak - how come the same problem doesn't cause a rapid decline? I'd guess there will be a tail of some sort, as long as lifting costs allow a profit but this curve just seems pretty flat.

[Pops]

BTW, that chart came from the EIA "Table Viewer"

--
And I also just found an excel spreadsheet that uses the Hubbert Linearization for DIY plottage! This is courtesy of Dr. Thomas Huber. It is current to 2011 and uses EIA forecasts from that date. I'll update the table if I can figure it out.

It has a facility to what-if additional amounts but not in a curve fashion. I would be cool for some excel guru out there to modify the sheet to allow the addition of multiple curves instead of the simple addition of a set amount for a set period. I might try but it may be beyond my ability ...

[Pops]

Here is another chart by Ron from back in June: decline rates:

$this->bbcode_second_pass_quote('', '[')img]http://peakoilbarrel.com/wp-content/uploads/2014/06/Shale-Decline-Rates1.png[/img]

This is the most important part of the story, shale oil decline rates. As you can see the EIA says Eagle Fore is declining at about 7.75% per month and the Bakken is declining at a little over 6.6% per month. And notice the EIA has decline rates rising, not falling. They say the Permian is declining at about 3.2% per month. That is because, in my opinion, the Permian is mostly conventional oil.

[Pops]

LTO decline rate is important because it could compound the shock of our underlying decline. We (US) could experience a greater decline over the next 10 years than we have in the preceding 40 if shale declines as advertised. Like so:



So here we are with Diane Sawyer saying we're going to be the world's largest exporter of oil, watching the messages from the Oil & Gas Industry on the tube and feeling all warm and fuzzy and then all of a sudden the bottom drops out about 2016 with 8-10% per year declines.

The idea that decline rates would never be that large has always been the thing that kept me from being an Overnight Armageddon type. I've pointed to shale as the only thing that has kept global decline at bay but RU has been increasing as well and KSA & Iraq to an extent.

Need I mention that global exports have already peaked, RU is now looking pretty flat and some say peaky, KSA is building rigs right and left to keep up production, the N Sea has been experiencing huge, steady declines as has Mexico - oh, and US nat gas is increasingly coming from those same shale wells that are producing the LTO ...


The only thing worse than hubbert's chart is the shark fin plot

da nuuuu ... da NUUU ... DA NU DA NU
.

[ROCKMAN]

Pops - It reminds me the joke about plotting the daily happiness of a turkey: it just keeps getting better and better. One could project a nice linear increase in turkey happiness for many years. But the next November comes around and that plot ain’t worth turkey sh*t. LOL.

Same thing with shale well locations. Same thing with every geologic trend bearing oil/NG reserves. As long as the economics hold one can just keep drilling more locations. But only as long as there are locations left to drill. Every oil/NG developed in the world either has or will run out of a meaningful number of locations to drilling. As I’ve pointed out many times the horizontally drilled Austin Chalk was THE hottest oil play on the planet in the 90’s. The play eventually covered several times the areal extent of the current Eagle Ford Shale play. There a bit of AC drilling happening today but not very much given oil is selling for more than 3X what it was when the play was red hot.

One cannot plot the past production trend of any play to predict future drilling activity. The only chance one has to make that estimate is to map out the remaining number of viable locations left to drill. Unfortunately that map is a lot more accurate after you’ve drilled enough wells that should not have been drilled. By then you have a pretty good idea of the extent of a play since you’ve gone beyond its economic.

Not to take anything away from the man but as far as doing a Hubbert style analysis one needs to remember that his projection was based on the analysis of US oil trends that were very mature and had already developed the vast majority of wells that would eventually be drilled in all of those trends. Which is why some for wrongly criticize Hubbert for an incorrect projection: he was only projecting production from the trends he had in his data base. And that didn’t include the shale plays of the DW GOM. For the system he analyzed Hubbert has been proven to be quite accurate. Just as one will certainly be able to make an equally accurate projects of the EFS or Bakken once 70%+ of all the viable locations have been drilled.

[Subjectivist]

$this->bbcode_second_pass_quote('ROCKMAN', 'S')ame thing with shale well locations. Same thing with every geologic trend bearing oil/NG reserves. As long as the economics hold one can just keep drilling more locations. But only as long as there are locations left to drill. Every oil/NG developed in the world either has or will run out of a meaningful number of locations to drilling. As I’ve pointed out many times the horizontally drilled Austin Chalk was THE hottest oil play on the planet in the 90’s. The play eventually covered several times the areal extent of the current Eagle Ford Shale play. There a bit of AC drilling happening today but not very much given oil is selling for more than 3X what it was when the play was red hot.

One cannot plot the past production trend of any play to predict future drilling activity. The only chance one has to make that estimate is to map out the remaining number of viable locations left to drill. Unfortunately that map is a lot more accurate after you’ve drilled enough wells that should not have been drilled. By then you have a pretty good idea of the extent of a play since you’ve gone beyond its economic.

[Pops]

$this->bbcode_second_pass_quote('ROCKMAN', 'O')ne cannot plot the past production trend of any play to predict future drilling activity.

[Graeme]

Aren't we in the green zone for global oil production?

[vtsnowedin]

$this->bbcode_second_pass_quote('Graeme', 'A')ren't we in the green zone for global oil production?

[ROCKMAN]

Sub – “…Hey Rockman, care to make a guess how many old plays will get the horizontal multi completion treatment?” You might be able to answer that question yourself: with oil being $90+/bbl for some years now how many other plays, besides the Bakken, EFS and Marcellus, do you see booming? At the moment I count zero. And it hasn’t been for a lack of trying: $billion have been spent evaluating the dozens of other fractured reservoirs plays in the US and still 80%+ of the unconventional oil is coming from just two reservoirs. Hundreds of wells have been hz drilled and frac’d in those other formation. Some small economic successes but nothing big popping. The only other area where there’s a boom is in the Permian Basis. But almost all of that gain is coming from previously producing reservoirs in known fields. That activity isn’t so much finding new reservoirs but just accelerating the recovery of proved reserves. Which is a good thing from a production rate standpoint but isn’t adding as much to the nation’s reserve base as many assume.

In my project I’m doing the same thing. Almost all the fields in my trend still have a significant number of the original producing wells. But typically those wells are each averaging several bopd and 100 to 200 bbls of salt water/day…that oil stained water folks refer to. My well is doing 125 bopd and 400 bwpd. But I’m not developing new reserves but the ones that were already proven more than 50 years ago. But if I drill another 50 to 100 wells the rate gain will be invisible in the national oil rate stat. Good for me but nothing for the rests of you. There’s only one other company trying to do what I’ve done: their first wells is doing 40 bopd and 2000 bwpd. And that’s in the same field I’m drilling in. They thought they understood what I was doing when they tried to copy me. They were wrong…very wrong. LOL.

[Tanada]

$this->bbcode_second_pass_quote('Pops', '')$this->bbcode_second_pass_quote('ROCKMAN', 'O')ne cannot plot the past production trend of any play to predict future drilling activity.

[ROCKMAN]

Pops - So based on that chart how many Eagle Ford Shale wells will ultimately be drilled? How many future Bakken wells? Do those numbers represent 100% of the potential physical locations in the play? 50%? 150%? And what price assumption is the basis for that projection? IOW what would that chart look like 10 years ago if one assumed $30 oil today? $90 oil today? Do you believe that chart if oil drops to $40/bbl. Increases to $160/bbl?

This has been my point all along: any of these projection plots are meaningless unless one understands the price assumptions behind them as well as the actual number of potential locations left to drill. It doesn't matter how good the economics of the previous average EFS wells looks at $X/bbl if most of the potential drill sites have already been developed...the boom is over.

[Pops]

$this->bbcode_second_pass_quote('Graeme', 'A')ren't we in the green zone for global oil production?

[Pops]

$this->bbcode_second_pass_quote('ROCKMAN', 'P')ops - So based on that chart how many Eagle Ford Shale wells will ultimately be drilled?

[Subjectivist]

Rockman thanks for the comprehensive answer. You burst my bubble, but at least I know counting on multi completion reworking of old field is not a panacea that will delay decline significantly.

[Pops]

You're way more of a mathematician than I am ROCK, so you know that the linearization bit is simply modeling the ultimate extraction/production of X from the past extraction/production of X. IOW it has nothing to do with the manner of extraction or any other "above ground" influences, merely that it will be done as fast as is possible.

Of course there is no way for it to forecast extraction of Y or Z, merely X.

And definitely there is no way for it to forecast a change in technology or area although it seems to have worked pretty good up till now (well at least till 2007, I can't find a more recent fit):