From EIA website and also Enno Peter's www.shaleprofile.com, you could find the production change from major US shale basins like this one below.
The observations are:
- Bakken started first & increased steadily and dropped slowly
- Eagle Ford started late but increased and dropped fastest
- Permian increased slowly but steadily and still not dropping
- Niobrara increased slowly and dropped at speed between Bakken & EF
What are the reasons for these differences?
When oil price was high in the 100, shale basins ramp up fast. But it took Bakken 8 years to go from ~100K BOEPD to over 1million BOEPD; whereas it took Eagle Ford only 3 years to go from ~100k to over 1.5 million BOEPD.When oil price dropped significantly the past year and a half, the Bakken actually dropped only over 10%, although it suffered the disadvantage compared to Eagle Ford of having to pay a much higher differential due to transportation costs; while Eagle Ford production actually dropped well over 20%.This is really puzzling, isn't it?
I took a look at the averaged well production decline curves for these 4 basins from Ennon's website, and the data are quite surprisingly confirming the observations above on the total production decline in the 4 majors.
For EF, the initial IP is the highest among the 4 major basins, but drops off quite fast, and essentially below 1/50th of the highest IP or 10 BOPD by year 6. Looking at the accumulated production, the 90% EUR of a EF well ends at about 4 years after IP, or >50% in one year and >70% in 2 years.
For Bakken, the initial IP is 2nd highest among the 4 majors, and also drops off quite fast, but was able to hold on after 6 years or longer. Looking at the accumulated oil production, in the first year, only ~30% of the EUR is generated.
Niobarbra has similar behavior has Bakken albeit all IP and flattened long term production are about 1/2 of Bakken.
The Permian has slower decline rate for the first 1-2 years when compared to the rest 3, but decline rate is still faster than other 3 majors after 4 years, and almost approaching end of life based on the available limited production data.
These averaged well decline curves explained, as oil price cycle increased and then reduced the number of new wells, why EF goes up fast and drops fast while Bakken goes up slow and drops slower.
Thanks to Enno's advice, here is the more convincing plot,
The geology behind this could be that EF has oil produced from shale while Bakken is essentially producing from the mid-Bakken nonshale tight formation.
Production from nonshale tight formation benefit from the so called Klinberg effect or Knudsen diffusion promoted oil and gas migration.
Such flattening production curve is already demonstrated in Barnett shale gas wells which might give a second lifeline for shale gas well if it works elsewhile. This is particularly a blessing for realizing the unproven EURs many shale oil company, if it also works for shale oil wells, and now Bakken is proving it.
But, Eagle Ford and Permian obviously follow a different curve, and how do we explain that. One possible explanation is related to the difference between shale oil and tight oil formations with minimal shale organics. Production from shale or shale mixed formation had the adsorption effect by shale organics trapping the remaining oil, e.g. EF and Permian. Permian might also suffer from water cut problem, while EF is known to have little water problem.
Bakken production mainly happens in the Mid-Bakken which has minimal amount of shale organics. This new STACK in Oklahoma's Anadarko basin should behave like Bakken as a nonshale low organic formation, and therefore we might have to double the EUR compared to EF or Permian.
If we use Ennon's website further, more interesting and alarming plots are generated.
Something weird is going on with production in Midland core Spaberries. Could this help Einhorn to finally blow away mother Fracker, PXD? And make Hamm the only hero in shale oil?
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.