Enno Peters
This is an older blog post, you will find one on more recent data here
These interactive presentations contain the latest oil & gas production data, from all 9,508 horizontal wells that started production in Colorado and Wyoming since 2009/2010, through October.
Oil production in these 2 states set a new high in October, at just over 550 thousand bo/d. Gas production also came in at a record level, at close to 3 Bcf/d.
The year over year growth rate dropped however, compared with the previous year, despite that more wells were completed in the first 10 months of 2018 vs 2017. More wells were needed to offset the decline from wells that came online in 2017, and well productivity also fell a little, based on preliminary data (see the ‘Well quality’ tab).
The DUC count has remained steady in the past year, as you’ll see in the ‘Well status’ dashboard if you only select the DUCs (using the well status selection on the top).
Anadarko, the largest producer in this area, showed a drop in production in the previous 12 months. The numbers 2 to 4 (Noble Energy, Extraction Oil & Gas, and PDC) did break their previous records in October.
The ‘Advanced Insights’ presentation is displayed below:
In this “Ultimate Recovery” graph, the average cumulative production is plotted against the production rate. Wells are grouped by the quarter in which production started. This time I only selected Weld County (using the ‘County’ filter at the bottom), as it is good for almost 80% of total production, and I wanted to highlight some interesting things happening here.
The first observation is that well productivity appears to have fallen since 2016 Q4 / 2017 Q1, as wells from later quarters are trending towards slightly lower ultimate recoveries.
The second, and probably more important one, is about the terminal decline rates that you can see here. As you follow the curves from wells that started between 2011 and 2015, you’ll see that they start to accelerate downward as lower production levels are reached. You’ll see the same effect if you select the natural gas stream from these wells (‘Product’ selection). That doesn’t bode well for long-term recovery estimates.
So how big are these terminal decline rates actually? We’ve just added a new dashboard in our Professional Analytics service, which aims to answer these kind of questions. Here you will see a screenshot of this dashboard, in which all the horizontal wells in Weld County are selected, that started production since 2012. Only wells are selected that fell below a production rate of 40 bo/d, from which they never fully recovered, before November 2015.
Terminal decline rate in Weld County
(if you click on the image, you will see a larger version of it)
You can see 2 graphs here. The one on the top shows the average flow rate of all the 1,354 horizontal wells that met these criteria, versus time (the number of months after they fell below 40 bo/d).
The graph on the bottom plots the average terminal decline rate of all these wells. I recommend ignoring the results up to month 20 or so, due to the inherent bias of this selection. However, you can see that a relatively steady state has been reached after 24 months. Between 24 months, and 36 months, which contains data for all these wells, you will find an average annual decline rate between 25 and 30%. This, I believe, is a far higher terminal decline rate than is commonly assumed when making ultimate recovery estimates.
In this dashboard, you will have many more options. For example, you can look at all the other shale basins, or at the terminal decline rate of the gas streams. If you want to analyze how these terminal decline rates have changed with newer completions, you can group these wells by e.g. the year in which they started. Other basins I’ve checked didn’t show the same high terminal decline rate, but also there they were significant.
Later today in our show at enelyst, at 10:30 EST, we will take a closer look at the latest data from North Dakota, in which we will also examine some findings of this new dashboard. You can join this event here: Enelyst ShaleProfile Briefings channel. If you are not an enelyst member yet, you can sign up for free at www.enelyst.com, using the code: “Shale18”
Next week we will have updates on the Eagle Ford, and also the Permian if new data for New Mexico has been released by then.
Production data is subject to revisions.
For this presentation, I used data gathered from the following sources:
- Colorado Oil & Gas Conservation Commission
- Wyoming Oil & Gas Conservation Commission
- FracFocus.org
====BRIEF MANUAL====
The above presentation has many interactive features:
- You can click through the blocks on the top to see the slides.
- Each slide has filters that can be set, e.g. to select individual or groups of operators. You can first click “all” to deselect all items. You have to click the “apply” button at the bottom to enforce the changes. After that, click anywhere on the presentation.
- Tooltips are shown by just hovering the mouse over parts of the presentation.
- You can move the map around and zoom in/out.
- By clicking on the legend you can highlight selected items.
- Note that filters have to be set for each tab separately.
- The operator who currently owns the well is designated by “operator (current)”. The operator who operated a well in a past month is designated by “operator (actual)”. This distinction is useful when the ownership of a well changed over time.
- If you have any questions on how to use the interactivity, or how to analyze specific questions, please don’t hesitate to ask.
News, trends and data for the US upstream industry
Novi Energy Newsletter
6 Comments
Enno-
Excellent insight on terminal decline. That, or market price of oil at $80+, is the ballgame.
Jim
Hi Enno,
Why choose only the lower flow wells? Seems to be cherry picking, just as the oil companies cherry pick the best wells as typical for their investor presentations. From a statistical perspective it makes sense to look at all the wells from Weld county. As in plot below, we see there is a kink in the well profile at about 54 months probably because some of the lower quality wells reached their economic limit as oil prices declined in 2015 and 2016. The economic limits of these wells will be affected by the price of oil, in many cases they might be temporarily shut in and brought back online when oil prices rise, though that may not be possible with horizontal fracked(SP?) wells, I know it is often done with vertical wells in conventional reservoirs for stripper wells.
For chart above (weld county 2012 wells) annual decline rate from 36 to 60 months is about 14%.
Hi Dennis,
Thanks for chiming in on this very important topic.
I am a big fan of looking at the total population and using averages. But sometimes it is helpful to look at what a subset of wells is doing, in order to understand their effect on the overall population.
If wells at low levels of production rate decline faster than the average, that is a concern. Eventually all wells will get to that point and this would then start to drag down the average decline rate.
This appears to be what we’re seeing. Let’s look at the average decline for the 2012 vintage in Weld County after 36 months. In the 2 years thereafter, the annual decline rate is not 14% as you stated, but 21% (from 22.3 bo/d, to 13.9 bo/d), which is still lower than the terminal decline rates that I have mentioned in my post. By year:
36 -> 48 months: 17% (22.3 bo/d to 18.4 bo/d)
48 -> 60 months: 24% (18.4 bo/d to 13.9 bo/d)
60 -> 72 months: 27% (13.9 bo/d to 10.1 bo/d) (I used our Analytics portal, which has more recent data, so I am confident about this last number. See screenshot below)
Why is the annual decline rate increasing for the whole 2012 vintage? A big part of the answer I believe is because more and more wells are getting to low levels of production rate, at which the decline rate is higher (probably for economic reasons).
I have not often seen such an analysis, or such a consideration, when EURs are created, but it appears to me that it has a very big effect.
To conclude, I am not saying that the terminal decline rates I mention in the post should be used for the whole average. But they are important because eventually the average decline rate will start to converge to these numbers. Therefore, this effect should be considered when creating EURs.
Thanks Enno,
Probably use a 24 month period before the kink at 54 months so month 30 to month 54 and fit trendline to nat log of output to get annual decline rate. Probably 18 to 20% would be the terminal decline before the reverse survivor bias giving you the excess decline after month 54.
Enno,
If we consider all Niobrara wells in 2011 from month 35 to month 71 there is about a 19.6% annual decline rate over that 3 year period. For all 2012 and 2013 wells from month 35 to 59 (fit trendline to nat log of output in this case only) the exponential annual decline rate is 26%.
Rune Likvern tends to look at the Bakken as a single project to consider the overall play’s economic viability, which is an excellent approach. To use that methodology it makes more sense to look at the well profile of the average well rather than single out the lower quality wells. Generally the low economic cutoff output rates will be highly sensitive to wellhead oil prices.