A Williston Basin Primer
In my previous article Addressing the World’s Flare Gas Problem, I discussed my current project, which recently took me to the Williston Basin in North Dakota and Montana. Today, I will discuss the region’s shale oil boom in greater detail. In Part 3 of this series, I will conclude by delving into the economics of shale oil production.
The Williston Basin underlies parts of North and South Dakota, Montana, southern Saskatchewan, and southwestern Manitoba. Within the Williston Basin is the Bakken Formation, which first produced oil over 60 years ago. It was on North Dakota farmer Henry Bakken’s farm in 1953 that Amerada Petroleum — later acquired by Hess (NYSE: HES) — discovered oil at a depth of about 10,000 feet. The Bakken Formation is to date the source of most of North Dakota’s rapid oil production growth, but underneath the Bakken Formation is the Three Forks Formation, which has also begun to produce oil:
Source: US Geological Survey
*AU = Assessment Unit and TPS = Total Petroleum System
There are billions of barrels of oil in the Bakken Formation, but the crude is trapped inside a fine-grained rock called shale. This shale is porous, which means it has tiny spaces that hold the oil and gas. But the permeability of the rock is very low. In other words, the rock holds a lot of oil, but it doesn’t flow out readily. Thus, 50 years after the discovery of oil in the Bakken, North Dakota was still a minor producer at less than 100,000 barrels per day.
That would change dramatically with the marriage of two technologies — which only became economical when oil prices began to escalate in 2005.
The Two Game Changers
The technique of hydraulic fracturing, or “fracking” had been around since the late 1940s and has been used extensively to promote higher production rates from oil and gas wells across traditional production regions like Texas and Oklahoma. Fracking involves pumping water, chemicals and a proppant down an oil or gas well under high pressure to break open channels (fractures) in the reservoir rock trapping the deposit. The proppant is a granular material like sand designed to hold those channels open, allowing the oil (or natural gas) to flow to the well bore.
Hydraulic fracturing rectified the permeability issue of shale. But there were wells being fracked in North Dakota in the 1950s. Why did it take another 50+ years before oil production took off in the state?
There are expenses involved in fracking a well, so the increase in oil production has to make the extra expense worthwhile. But the Bakken is only 100-150 feet thick. You might imagine that if it were thousands of feet thick you could frack many times at widely-spaced vertical intervals. However, that wouldn’t work in a formation that is under 150 feet thick. If the fractures are too close together you get diminishing returns. You don’t want the fractures from one stage to overlap another stage. So, a vertical well in the Bakken might only support a single frac stage, which would increase oil production but not dramatically so.
Like fracking, horizontal drilling was invented decades ago, and has been widely used in the oil and gas industry since the 1980s. As its name implies, horizontal drilling involves drilling down to an oil or gas deposit and then turning the drill horizontal to the formation to access more of the deposit. These horizontal “laterals” can be 5,000 to 10,000 feet in length. So now instead of a well being able to access maybe 130 feet of the Bakken Shale, a single well could access more than 50 times this distance. It was the combination of these two decades-old techniques — hydraulic fracturing and horizontal drilling — that inaugurated the U.S. shale boom.
Source: American Oil & Gas Historical Society
North Dakota’s oil production began to take off in 2008, with natural gas production not far behind. Today North Dakota is the 2nd largest oil producer in the U.S., behind only Texas:
Getting to Market
Logistically, it was a challenge to get this new oil and gas to market, since North Dakota wasn’t a major traditional oil producer. Because pipelines take years to construct, the early surge of production proved a boon to railroads. The rails already criss-crossed the area, and ramping up was simply a matter of adding the right kind of rail cars and loading terminals. In less than three years railroads including Berkshire Hathaway’s (NYSE: BRK-A) BNSF Railway increased crude shipments by over 700,000 bpd.
This year, however, the volume of oil being moved by rail has fallen by about 200,000 bpd. Some of this can be explained by flattening oil production as a result of the price crash, but the other major factor is that pipeline infrastructure is finally beginning to catch up. (Of course natural gas isn’t suitable for shipping in a rail car, so much of the associated gas production was flared as discussed in my previous article.)
Source: North Dakota Pipeline Authority
Resources and Reserves
So how much oil are we really talking about? In 2013 the US Geological Survey (USGS) estimated that the Three Forks Formation contains an estimated mean resource of 3.73 billion barrels of oil and the Bakken contains another 3.65 billion barrels of oil for a total estimated resource of 7.4 billion barrels of undiscovered, technically recoverable oil in the two formations. The two formations were also estimated to contain a mean of 6.7 trillion cubic feet (tcf) of undiscovered, technically recoverable natural gas and 0.53 billion barrels of undiscovered, technically recoverable natural gas liquids (NGLs).
Of course “technically recoverable” is an estimate of what could be recovered with existing technology, and therefore vastly underestimates the amount of oil in place in the Williston Basin. Most of this oil isn’t recoverable with existing technology, but the total oil in place has been estimated to be as high as 500 billion barrels. The state of North Dakota estimated a more modest 167 billion barrels of oil in place, which is still far more than the amount that is technically recoverable.
But what is “technically recoverable” isn’t necessarily economically recoverable. To be classified as a proved reserve, oil or gas has to meet both qualifications at prevailing prices. In a December 2014 report, and based on oil prices still hovering around triple digits, the Energy Information Administration estimated that the proved oil reserves in the Williston Basin were 3.2 billion barrels. To put this in perspective, this is equal to just under 6 months of U.S. annual oil consumption which is presently about 7 billion barrels of oil per year.
What about Economics?
I started to include a section here on shale oil economics, but ultimately decided that this topic can stand alone in next article. There is a school of thought that shale oil isn’t economical, because the wells deplete too quickly. The truth is a lot more nuanced than this. I will get into this in the next article, and rank the region’s top producers.