Wednesday, March 5, 2008

IP Decline Rates

Thanks to Larry for putting this together and who wrote:

The following is a spreadsheet showing the IP and the first production for the wells. I based first production on the well producing at least 25 day during the first month it produced. If, for example, the well only produced 20 days in the first month, I then combined the first and secon month production.

Unless I am missing something there is not much correlation between IP and the production in the ensuing month.

IP vs.

1st bbls
Well Name IP Date IP BBLS Days BOPD prod.

H. RICE 1-26H 11/22/06 20 2092 40 52.3 261.5%
ENGET 4-11H 7/14/06 46 974 26 37.5 81.4%
NELSON 14-33H 8/23/07 53 1244 25 49.8 93.9%
LALIM D. #34-11H 12/24/06 76 5478 47 116.6 153.4%
RICE 10B-2-1H 2/20/07 193 7621 48 158.8 82.3%
NELSON FARMS 11-19H 1/8/07 198 743 31 24.0 12.1%
JEAN NELSON 1-35H 11/1/07 266 6133 35 175.2 65.9%
PARSHALL 1-36H 6/6/06 463 9154 29 315.7 68.2%
GEVING 1-09H 8/12/07 783 31332 50 626.6 80.0%
ZACHER 1-24H 6/13/07 870 49822 49 1016.8 116.9%
PARSHALL 2-36H 9/27/06 883 16960 31 547.1 62.0%
EHLERT 1-35H 4/29/07 918 45079 33 1366.0 148.8%
C & B 1-31H 5/15/07 922 43149 47 918.1 99.6%
FLORENCE 1-04H 7/25/07 1015 35048 38 922.3 90.9%
PEERY STATE 11-25H 5/13/07 1081 24985 44 567.8 52.5%
N&D 1-05H 7/3/07 1285 33567 29 1157.5 90.1%
RALPH 1-32H 9/2/07 1329 27815 30 927.2 69.8%
PATTEN 1-02H 3/15/07 1487 14565 31 469.8 31.6%
WARBERG 1-25H 1/19/07 1553 26102 41 636.6 41.0%
WENCO 1-30H 9/23/07 1670 47696 39 1223.0 73.2%
HOFF 1-10H 6/24/07 1675 45721 38 1203.2 71.8%
BARTELSON 1-3H 11/17/06 1800 30195 45 671.0 37.3%

Brigham Hynek

142.0 24.3%
Brigham Bakke

310.0 81.6%
Brigham Bergstrom

119.0 58.9%

Update: Larry updated the IP decline chart to include cumulative production thru January 2008, which indicates that the data confirms the average production over a longer period is about 60% of what is produced in the first month.

1st bpod

Cum Cum
Well Name IP BBLS Days BOPD cum bpod

PARSHALL 1-36H 463 105933 579 183.0 58%
ENGET 4-11H 46 13272 573 23.2 62%
LALIM D. #34-11H 76 30554 487 62.7 54%
H. RICE 1-26H 20 8243 461 17.9 34%
PARSHALL 2-36H 883 132500 440 301.1 55%
BARTELSON 1-3H 1800 174373 404 431.6 64%
RICE 10B-2-1H 193 33217 349 95.2 60%
NELSON FARMS 11-19H 198 41604 346 120.2 502%
WARBERG 1-25H 1553 140613 331 424.8 67%
PATTEN 1-02H 1487 144251 303 476.1 101%
PEERY STATE 11-25H 1081 111261 289 385.0 68%
EHLERT 1-35H 918 156502 250 626.0 46%
ZACHER 1-24H 870 118415 217 545.7 54%
HOFF 1-10H 1675 126544 216 585.9 49%
C & B 1-31H 922 116437 210 554.5 60%
N&D 1-05H 1285 108570 183 593.3 51%
RALPH 1-32H 1329 91546 173 529.2 57%
FLORENCE 1-04H 1015 88919 168 529.3 57%
NELSON 14-33H 53 5895 163 36.2 73%
GEVING 1-09H 783 71715 160 448.2 72%
JEAN NELSON 1-35H 266 19043 127 149.9 86%
WENCO 1-30H 1670 97724 123 794.5 65%


Anonymous said...

By IP do you mean induced polarization?

Anonymous said...

Initial Production.

(Note: to put some of these numbers in perspective, Warren Buffet was recently quoted as saying average well production across all operating wells in America is 11 barrels/day.)

Anonymous said...

Larry, you did an very good job putting this info together.. Glad someone finally pointed out that the IP and than the actual bbl's per day thereafter are never even close. The IP is usually a drill stem test, (open hole),without production equipment in place...

Anonymous said...

I thank you for digging out and posting this information. It is greatly appreciated.

It is clear the Brigham wells have fallen off much more than the average of the wells you have researched. On the other hand the three wells ahve an average production rate of 190 barrels per day and that will certainly pay the bills.

The Hynek is clearly the problem child of the 3 and has overly impacted the average. Obvoiusly a bigger sample from Brigham will help. It would be nice to know how the well was fractured compared to EOG.

It sounds like the current well in the Mowry is using the latest technology.

Anonymous said...

The production “life expectancy” of a new Bakken well is around 40 years. The resource is significant with approximately 4.5 million barrels of original oil-in-place per section of land in the greater play area.

Anonymous said...

Oil company only expects to recover 10-15% of that oil with the present technology at hand. I suspect that 10 years down the road we will see the recovery nearing 50% or better with new advances, + $200.00 per barrel of oil.

Anonymous said...

Adding some comments about decline curves....

I am not a reservoir engineer but can offer some comments based on observing Williston Basin production
over the last 50 years.

Initial production is always a good indication of performance, but there are great variances. One well which
IP'd for 162 barrels per day has produced 1,463,000 barrels so far; it's still producing. Another with an IP
of 480 barrels per day has produced 2,356,000 barrels to date and continues to produce. I believe this is
the granddaddy of all North Dakota oil far. (Both of these are vertical Madison wells producing from
relatively thin pay zones, 164-85.) Another extreme is a Nesson Anticline well with an IP over 1,000 barrels
per day that was plugged after only a few months (vertical Interlake).

More to the point, I believe you need 18 to 24 months of production to get a good feel for what a well is
going to ultimately produce. Since only a few horizontal Middle Bakken and Three Forks wells have more
than a year of productive history, it is difficult for me to draw any firm conclusions. It appears that
if you average the first 2 to 3 months of "flush production", the typical well might be producing 50% of this
average amount in 10 months to a year. After 15 to 18 months in appears production has leveled off at a rate
of about 25-30% of the first 3 month average (with little regard to the IP rate). Hopefully the decline from this point
forward will hold at about 10%-15% per year.

The obvious exception to the scenario is the Petro-Hunt USA 2D in the Charlson area. It's reported IP was
700 barrels per day. It's 16 month total production is 378,536 barrels and the most recent month production
was 1000 barrels per day! This was its best month ever; it is "inclining" not declining(?).

Other cautions on every well: did they stay "in zone" while drilling; did the zone get damaged while drilling;
did the direction of the lateral section optimize natural fracturing, did the frac job get into the intended zones,
and on and on. We'll all be wiser in a few years as this data base grows and learning curve goes higher.

Anyway, regardless of Williston Basin history, the future looks extremely bright! The Bakken play has
been nothing but fantastic. Also, we have to remember, this is a play brought on by technology: horizontal
drilling and fracing. Both of these will only get better and we've just scratched the surface of the Middle-Bakken
potential. Who knows where the Three Forks will take us.

Anonymous said...

Does any one know the production rate of STENSETH TRUST 11-5H?

Rufus said...

Thanks to the poster that provided their insights on decline curve analysis. An Excellent post! I hope he will consider adding it again, on the blog discussion board, as a new thread for discussion and questions specific to his very detailed remarks! Would enjoy discussing and learning more about this... So, a cordial invitation, I hope he/she accepts!

Anonymous said...

anyone have a play wide(100+ wells), post 2006,horizontal wells only, decline curve to share?

Anonymous said...

Who knows where to download XRumer 5.0 Palladium?
Help, please. All recommend this program to effectively advertise on the Internet, this is the best program!

Anonymous said...

The comments above do a good job on the subject. Here's soome other info- Typically the IP is the initial production that is a rough early indicator of the potential performance of the well (bopd & EUR/CUM). It is also helpful to get a general idea of reservoir pressure.
Wells typically decline as the pressure in the reservoir is depleted. Most Bakken wells are horizontally drilled and hydraulically stimulated so their initial decline tends to be more extreme- following a hyperbolic shape. Eventually thru depletion of the reservoir boundaries will be reached which lead to a more stabilized decline (exponential/x% per year).
If you want to get a rough idea of what exactly a decline rate would be at a particular time you would need to create a type curve; which one way to be accomplished is by normalizing the production of some reasonable wells that have similar reservoir properties.
(There's alot more exceptions to the above but hopefully that helps shed some light on the topic)