For Newbies Only: About Those Pesky Decline Rates

For newbies: there is a horrendous decline rate in Bakken and Three Forks Sanish wells in the Williston Oil Basin.

By "decline rate," I am referring to the announced initial production (IP) of a  well and subsequent production. This decline rate has been talked about from the very beginning and can be seen graphically on corporate presentations, Whiting and Continental Resources, having among the best presentations.

I  have opined frequently that the drillers will figure this out and over time the decline rate itself will decline. It's already begun to happen, as evidenced by the drillers themselves and their presentations.

My gut feeling is that the decline rates themselves are pretty unreliable (the trend is not unreliable; the trend is real; what is unreliable are the actual numbers, in my mind).

Last week there was a well with a typical Bakken initial production number, but by the time the well had come off the confidential list, the subsequent production numbers were significantly lower. One cannot generalize from these individual results.

Let's say a hypothetical well has an initial production of 3,000 barrels in a an early 24-hour period. But then two months later it's down to 300 barrels/day.  What spurious events might have aggravated the decline rate?

If the well was not hooked up to a pipeline, the accompanying natural gas was being flared off. North Dakota state law places restrictions on how much oil can be produced from a well that is flaring natural gas. That's why one occasionally sees requests to waive those restrictions come to the NDIC.

If the well is not hooked up to a pipeline, and there are only four tanks on the pad (which is not unusual in the Bakken), only 4 x 400 barrels of oil (1,600 barrels) can be accumulated before the tanks are full. And then trucks have to come in to haul the oil away.

When there were only 100 rigs operating in the Bakken in the early days of the current boom, there was a shortage of trucks and drivers. Now there are a record 165 active rigs in the Bakken. In addition, the wells are producing more oil per well than they were a year ago. Does one really think that there are that many more trucks and truck drivers in the Bakken now than there were a year ago. There's no housing available.

Even if the shortage of trucks and drivers is resolved, the roads are still a significant problem. The wells are in remote locations and one just doesn't go out and empty those tanks when one feels like it.

If a truck is scheduled to go to a particular well, and a gusher comes in elsewhere, the oil company now has to decide how to redeploy those trucks.

There were days last winter when all trucks were stopped for several days due to blizzard conditions.  For wells not on a pipeline, the tanks will fill up and the well will be turned off (from a remote location, or automatically if the tanks are full).

If the price of oil drops, the operator might find it better to stop the pumping until the price of oil goes back up.

If the pipelines are full, a particular well might not be able to get its oil into the pipeline. This was a significant problem earlier this summer when Enbridge experienced not one, but two, oil spills completely shutting down major pipelines. Whiting even announced it would miss its third quarter production targets due to the shutdown.

In addition, wells are occasionally taken off-line for "work-overs," processes to clean out the wells to make them more efficient again, or to fix problems that might have developed.

To the best of my knowledge, it is a rare well that is flowing at maximum rates 365 days a year.

I have no idea how the statisticians sort out decline rates that are affected by man-made factors noted above, but I know that for me, I cannot simply look at production numbers of a given well to sort out whether it is a great well or an uneconomical well.

This has not been proofread; I need to go over it, but I need to pick up my granddaughter from school, first.