The US Petroleum Supply Monthly just came out with production data for every state and territory. US supply was up 168,000 bpd to 8,864,000 bpd in September. The biggest gainers were North Dakota, up 53,000 bpd to 1,185,000 bpd and Alaska up 79,000 to 477,000 bpd. Alaska was way down in both July and August and are just recovering from that. There was only one big loser, New Mexico, down 18,000 bpd. Texas was up only 9,000 barrels per day which was surprising. The Gulf of Mexico was down 3,000 bpd.
The Choke theory and why I ain’t buying it.
North Dakota publishes a Daily Activity Report Index of all permits and other well activity in the Bakken as well as the rest of North Dakota. In this report is a list of all producing wells completed as well as wells released from confidential (tight hole) status. Wells usually stay on this list from a few days to a few months, but the average is only a few weeks.
I have collected this data from October 2013 to present and found some startling results. But some have said this data means nothing, that wells are usually choked off by the driller so therefore we can gain nothing from the data. But looking at the individual wells that just doesn’t make any sense. No, I agree that the driller chokes but that he would not gradually choke more according to increasing well number.
Below I have posted the first 24 hour data for all 122 wells reported by North Dakota for the first 25 days of November. The first 24 hour production ranges from over 3000 barrels of oil per day to a low of only 10 barrels of oil per day. Barrels of water range from a high of 6663 bwpd to a low of 48 bwpd. And the percent water cut ranges form a high of 94.15% water to a low of 12.75% water.
It just seems incredible to me to claim that these numbers are meaningless. Throughout all the almost 14 months of data I have gathered there are lots of very large producing wells and a lot of small producing wells. The point is as the well number increases the number of very large producing wells seems to decrease while the number of small producing wells seems to increase. And I just don’t believe this is due to the many drillers, after checking their well number, decides what size choke to apply.
Just below the list of all wells I have averaged the production according to well number. The sample however is not large enough to really mean a lot. The number of wells in the sample are: Below 26000, 11 wells – 26000s 35 wellw – 27000s 54 wells – 28000s 23 wells.
But over the last 14 months we get a trend. Phil Scanlon has analyzed the data and sorted it into percentiles. This is his chart. He did not think he had enough 28000s to get a representative sample.
The gradual drop off in production is more pronounced at the higher production numbers than below the 25th percentile.
Here is my updated chart after adding October 2013 and a couple of days this week to the data. I have made it zero based in order to get a truer look at what the data shows. There are 2,390 wells in the sample but only 46 are above 28000 so that might not be a good representative sample.
And here is my updated 24 hour chart with the October 2013 data. Do you see a trend?
Thanks to Freddy W. for this chart. I consider this chart to be very important because it shows that the early water production, though clearly higher due to the frack water coming back, is nevertheless proportional to the total water cut later after the frack water is gone.
The water cut in all wells is clearly increasing as the well ages. Some have said this would not happen in a LTO reservoir but this chart clearly shows they are wrong. But more importantly it shows the water cut increasing in new wells each year with 2014 the highest at about 44% after 7 months and 2009 the lowest at about 24% after 7 months. This leaves no doubt that the water cut is increasing as they move further from the earliest drilled sweetest of the sweet spots.
Rockman on the first 24 hours of well production, in response to my last post.
The first 24 hours of production is far from being the average first years production. And though all wells are different I am relatively sure there is an average conversion rate but I have no idea what it is. I would guess it is somewhere between one quarter to one third of the first 24 hours of production.” Very nice data gathering. But I suspect you won’t like hearing my answer. First, I drill in Texas and La but I suspect procedures aren’t much different in the Bakken.
So I drill, complete and test a well in Texas. Conventional or frac’d unconventional…doesn’t matter. A simple but common example: the oil flows out of the well head. Within the head is the assembly that holds the “choke”. The choke is a small plate with a hole thru which all the oil flows. The chokes are designated by the diameter of that hole expressed in inches: an 8/64″ choke, a 12/64″ choke, etc. Yes: might be difficult to imagine but wells with flow
Now I test the well by flowing it for maybe several hours thru different chokes. Maybe 900 bopd thru a 14/64″ choke. And then 750 bopd thru a 12/64″ choke. And then 500 bopd thru an 8/64″ choke. And in addition to those rate changes the flowing pressure changes with choke changes. I flow it thru different chokes because the variations help me to better understand the completion quality.
Now what rate do I report that the well testing at: 900 bopd,750 bopd, 500 bopd? I can report any one of those numbers I choose. There is no public reporting standard required by the Texas Rail Road Commission. The state does get all the engineering details and the are available to the public: all you have to do is go find them in the massive data base.
So here’s a simple question: how many times have you seen someone post an oil flow rate AND included the choke size and the flowing pressure? There’s a huge difference between two wells flowing 700 bopd if one is on a 16/64″ choke and the other on an 8/64″ choke…a hole half the diameter.
If that’s not bad enough I can flow a well on a 12/64″ choke at 900 bopd and see that rate slowly drop or increase. I can stop the test at anytime and report a rate that hasn’t stabilized.
You want more complications? It’s not uncommon to let a frac’d well (hz or very) flow at a lower initial rate for weeks: you don’t want to “pull the well too hard” initially for fear of damaging the frac job.
Now here’s the real killer: the initial flow rate of a well in a pressure depletion drive reservoir, as all the shale plays are, has no implication of how much it will be flowing in 12 months. I can test two EFS wells at 900 bopd…same choke and same pressure. And 12 months later one is producing 600 bopd and the other 100 bopd. The rates will be determined by the volume of the reservoir being drained. Simple analogy: you have two steel tanks with one containing 20,000 gallons of water and the other 50,000 gallons. Both are at 9,000 psi. You punch a 12/64″ hole in each tank and they shoot out water at the same rate. But as time passes the pressure in both tanks declines and thus the flow rate decreases in the smaller tank faster. That rate decreases faster in the smaller tank faster because the pressure decreases faster. I’ll skip the physics but this is part of the science behind hyperbolic decline.
Essentially ever EFS well at the same depth has roughly the same reservoir pressure. But how quickly that pressure (and flow) declines will depend upon the volume of the pore spaces being drained. Which is why there are EFS wells that initially produced at 500+ bopd with some ultimately producing 300,000+ bbls and others less than 100,000 bbls.
And if that isn’t confusing enough there are a variety of logistical issues that can prevent a well from flowing at the higher it might be producing several months later. After 3 or 4 years of producing an EFS you can make a pretty good guess of the URR. But doing so during the first few months (let alone from the initial tent rate) of production? Not so much. LOL.
This is all very well but It clearly does not explain the gradual decrease in production as the well number increases. Also there are always outliers. That is there will always be wells that produce little the first 24 hours and will be found to be producing even more one year later. But these are the rare exceptions rather than the rule.
What Rockman is talking about here is something random, or should be random. The driller decides to choke and how much to choke. And there are many drillers with many companies. There should be no discernable trend in their choking actions. If we do see a trend then there has to be some cause and that cause must have some meaning.
Other News of note:
The story of a marginal producer:
SandRidge Energy: You Don’t Know What You Don’t Know
And this is BIG NEWS: Russia says will keep oil output steady in 2015
Russia is hoping to keep output stable by exploiting new areas such as the Bazhenov oil formation and Arctic offshore.
Got that? Russia hopes to keep production from falling next year by exploiting the Bazhenov Shale and the Arctic Ocean. Lotsa luck with that one.
Active drilling rigs in North Dakota stands at 183. 182 if you don’t count the one drilling a salt water disposal well.
Note: I send an email notice when I publish a new post. If you would like to receive that notice then email me at DarwinianOne at Gmail.com.