I was talking to a friend of mine today who knows about pumps and asked him what he reckoned the velocity of the BP blow out to be from the video feed. He guessed about 3 km/hour (1.88 mph). (In these sorts of things, I value opinions from practical guys – I don’t think that you need to be a professor of fluid mechanics to guess at flow velocities.)
Let’s assume that his guess is right within an order of magnitude/ The diameter of the riser pipe is apparently 21 inches =21*.0254 m= 0.5334 m. The volume of the blow out would therefore be: 3000 (m/hr) *pi *( 0.5334/2)^2 = 670.374 m^3/hour = 670,374 liters/hour.
One bbl oil = 158.987 liters. Thus 670,374/158.987 = 4216 barrels/hour
I.e. 101,184 barrels/day. (Phil Worley of Purdue estimated 70,000 bbl/day.
In order to get 5,000 barrels/day, you would have to have a discharge velocity of 0.1 mph instead of my friend’s guess of 2 mph or a smaller effective pipe diameter. [See update below as this latter seems to be the case, though not down to 5,000 bbl/day. Looks like 15-20,000 is more probable.]
I wonder how they arrived at their estimate of 5,000 barrels/day. Maybe their Group Vice President, Research and Engineering should have spent more time trying to figure this out and less trying to hide the trick to hide the decline.
Update – good news. It appears that the “top kill” was successful and the blow out is sealed. The article contained new estimates of discharge rates by the US Geological Survey about 3-4 times higher than BP’s, but not 10-20 times higher:
Marcia McNutt, the director of the US Geological Survey, estimated that the flow ranged from 12,000 to 19,000 barrels per day.
Up to now, BP estimated the leak at 5,000 barrels (210,000 gallons) per day, but has said that figure is unreliable.
Update: Re-doing these calculations with suggestions made by readers – 9 7/8″ pipe and 50% gas and the same discharge velocity (3 km/hour) yields a bit over 11,000 bbl/day.
Pipe area is pi *(0.250825/2)^2 # 0.0494119 m^2
Volume discharge is 0.0494119* 3000 # 148.2357 m^3/hour
i.e. 148.2357 *1000 /158.987 #932.3762 #bbl/hour
Oil discharge is 24 hours *932.3762 bbl/hour *.5 # 11188.51 bbl/day
Update June 16 – here current estimates are about 60,000 bbl/day
Climate Audit 
114 Comments
Now thats big oil!
Really, Steve? Really? You’re going to try and shoehorn everything into “hide the decline” – even if, as in this case, it’s an oil company trying to suggest the quantity of the leak is less than it is?
According to this article (May 19), BP believes that about half of the discharge is natural gas, and that they were collecting 3000 barrels and 14 million cubic feet of natural gas a day. 3000 barrels is ~126K gallons, which is approx 16,800 cubic feet of oil. That’s nearly 3 orders of magnitude more volume of natural gas at the surface, which makes me think their “half gas” comment is off… it’s actually more than half gas. The pressure down there is ~150 atm, and the temperature is maybe 10% lower (in degrees K) than at the surface. So if it were really 50/50, I’d expect them to see 16,800 * 150 * 1.1 = ~2.8 million cubic feet of natural gas at the surface…
The biggest hide the decline story is oil depletion. Nice to see everyone analyzing the throughput of a single tube, when the global decline situation meets with indifference.
Some additional oil flow measurements that are much higher than BP. However, BP says the drilling pipe which fits inside the riser is only 9 in diameter.
http://www.npr.org/templates/story/story.php?storyId=126809525
As a petroleum engineer I can finally comment on this site with more confidence.
You forget the natural gas content. From what I’ve read, BP is estimating a Gas/Oil ratio of 4000 scf of gas per bbl of gas. That can take up a lot of volume for the part of the gas not dissolved in the oil (at least 1/2).
Plus, there is a second pipe (drill pipe) inside the riser that takes up a fraction of the cross section. I suspect the outside of the pipe is round 4 1/2 inches plus or minus, depends on what pipe type was in the hole at the time. If it was a drill collar, then it could be 9-10 inches in diameter.
Plus, from a practical perspective, a 100,000 BPD well is a stupendous well, even in blow out conditions. (It isn’t a full blowout as there is a 5000 foot column of water on the outlet,which is about 2200 psi). I’ve never seen a well in the world capable of 100000 bpd, especially for 30 days with 2200 psia at the well head.
Additionally, the volume of oil in the ocean doesn’t seem to be 30 days at 100,000 bpd, which is 3 million barrels. The slick doesn’t seem that bad yet.
Lastly, I think the expansion of gas and oil (oil has a lot of gas in it) at the outlet masks the true velocity of the fluid).
The spill is a serious spill, but it isn’t a 100,000 BPD.
Concerning the natural gas, it’s an OK observation but Steve’s calculation of the volume per day is still valid. Volume is a volume – cross section times velocity times time. It’s a volume of a mixture of natural gas and oil, rather than pure oil, but it’s still the volume that Steve wrote. 😉
Comparing the result of the calculations to “5,000 barrels/day” is wrong then. You get to pick 🙂
Lubos,
A bit pedantic but true. But no one cares about gas volume as it is released to the atmosphere (and a bit may dissolve into the water). I assumed that Steve’s calculation assumed 100% oil volume in responding to it.
Steve’s analysis included the diameter of the pipe at 21 inches. If it is only nine, or four and a half, then his calculation overstates the flow.
Steve – I had noted that the effective pipe diameter might not be 21″. With a 9 7/8 inch diameter, one can get to flows on the order of present USGS estimates with an elementary calculation. A question – are we seeing flow from the 9 7/8 pipe or the 21″ pipe? If the choke point is at 9 7/8″ but we’re seeing flow from the 21″ riser, wouldn’t the velocity at the choke point be greater than from the riser pipe??
Flow isn’t quite as simple as velocity times diameter. For laminar flow, average velocity is about half the peak velocity. Turbulent flow would be much higher. In any event, all this is thrown to the wind given that it’s a non-newtonian fluid mixed with a gas. I think the original calculation achieved it’s aim – a rough estimate of the volume flow.
I’m not sure its that non newtonian. Viscosity/shear force relationships for the oil gas and any water would be pretty well understood and be newtonian.
I read that the oil density is 35 API, which is a light to medium oil. Given teh reported gas content, i tprobably has a viscosity around 1-3 Centipoise, 2-6 times the water at reservoir temperatures.
Leonard-
Since crude oil is only newtonian at high temperature (hot enough to keep the parrafins and asphaltines dissolved), I think it’s safe to say this is a non-newtonian fluid. Since apparent viscosity at lower temperatures depends on temperature, pressure, shear rate, and the crude oil composition, I have no way of getting closer to a good answer. The available data tends to be geared pipeline conditions. You can look some of this over by searching on crude oil rheology. None of this takes into account the dissolved gas coming out of solution as the oil trvaels down the pipe – my guess is this is likely to be tripped at the pipe surface by imperfections, but it’s just a guess.
When I looked at this originally, I couldn’t get closer than a factor of 2 (laminar vs turbulent flow), and I think the answer still lies in between these two situations, but can’t really say one or the other – it’s much too complex
Also, BP has a siphon in the opening and they say they are getting about 20-40% of the fluid flow and they are getting 2000 BPD in the drill ship up hole.
Whatever they are getting, they aren’t getting only 2% of the volume.
I confirm your estimates and calculations, Steve. After all, you’re not the only one who guesses about 100,000 per day, see e.g.
http://www.examiner.com/x-41082-Public-Policy-Examiner~y2010m5d26-Oil-spill-in-Gulf-of-Mexico-is-much-larger-than-BP-admits
The drill pipe downhole is about 9 7/8 inch at the oil sand level in the reservoir. It is NOT 21 inches. The RISER going UP to the platform may be , but not the drilled hole. The hole steps up to 12-14 inches at the TOP, but the max flow is set by the bottom hole diameter where the flow originates.
Here is a link to an internal BP video of the whole operation and the logistics of what is going on.
http://bp.concerts.com/gom/kentwells_update24052010.htm
yes, but on his approach he’s using the estimated velocity emerging from the larger diameter portion (the max flow down hole doesn’t affect his calculation the way I see it). So, you get a volumetric flow rate assuming Vel * Area; then to get pure oil volume you multiply by the percent oil (by volume) of the emerging fluid.
Also this crude is the Light Texas Crude and not the heavy tar like you see in some places. A lot of it has evaported already in the hot water of the Gulf and wave action. In this hot water and enviro it will be eaten up by bacteria much quicker than the Alaska spill.
And for some perspective, there have been over 35,000 wells drilled in the Gulf over the last 40 years. That is part of the reason safety likely got lax….not many incidents.
I know the question will never get asked by the media with all the politicians making noises etc, but it needs to be pointed out or asked…..Why are we drilling in this harsh area to start with? Because we are not allowed to drill in the close in shallow water areas that could be fixed immediately if something went wrong. The more they shut out areas, then we go elsewhere. Such as war zones, deepwater, dictator regimes, etc. This well is only in 5000 feet of water. There are wells out there that are in 10,000 feet of water in the Gulf.
I have located a US government report dating to roughly the year 2000 that claimed 35,500 wells had been drilled in the Gulf of Mexico by the year 1998. That number is probably closer to 40,000 now.
I found a UK site that lists rig incidents. I didn’t count them but the total was at most 100 around the world with maybe 30% of them in the Gulf of Mexico. This one will likely be one of the most expensive, but will not crack the top 10 in deaths, thankfully. It might crack the top 10 in spill volume.
Writing this from memory at the moment, I am struck by the rarity of these types of incidents, and that the claims of ecological disaster are overblown. Significant temporary impact, yes; but not the end of the world as we know it for those with eyes to see and ears to hear dispassionately.
My own money is that this incident is a combination of not knowing the actual gas content combined with design decisions based upon prior experience and limited prior experience with the real conditions at the site.
Why are we drilling out there? The P word comes to mind.
I find the following:
IXTOC I blowout: 30,000 barrels per day (BPD)
Ekofisk Bravo blowout: 28,000 BPD
Lucas gusher: 80,000 BPD
Lakeview Gusher: 100,000 BPD
Barroso 2: 100,000 BPD
Baba Gurgur: 95,000 BPD
Wild Mary Sudik: 72,000 BPD
Qom, Iran on August 26, 1956: 120,000 barrels per day, supposedly the largest ever. Reached a height of 170 feet, which means the exit velocity was ~ 115 km/hr, and the pipe was about 4″ …
I note that in all cases, there was a comment that the rate decreased markedly after a few days. Hmmm … as always, more questions than answers.
In light of the recent changes to the more diverse and political nature of this site I believe Steve may be correct in flaging its demise. I think it would be a great loss and would suggest that Steve might wish to establish another site to run in tandem, as “Climate Audit” does not seem to reflect the new tone.
This is not ment as a critisism, as I agree with the need to raise the issues that have been addressed, but a political site is, by nature, biased and should be known as such.
This topic is more in the “audit” category than “politics”. It’s about making the right observations and the right calculations and questioning them if they appear to have inaccuracy.
Personally, I strongly hope that Climate Audit continues in whatever form Steve likes. It will take time for his substantial contribution to be accepted. I hope that he has the patience to make more contributions. But it’s his call, 100%.
Well said.
I have kind of been waiting to see if Steve was going to make some examinations of all the code contained in the Climategate release to see what kind of mishmash they had made of those.
Steve – there’s only so much that I can do personally. It looks like it has to do with papers on tree ring density, but it takes quite a bit of work to decode precisely. You’d think that one of the four inquiries would analyse it, but it doesn’t look like they will.
Some background information on historical oil spills and oil in the ocean is available here:
http://oceanworld.tamu.edu/resources/oceanography-book/oilspills.htm
Interestingly, North American waters see 160,000 MT/yr in natural oil seeps from the ocean floor, mostly in the Gulf of Mexico. That’s about 1.2 million barrels/year. It is widely dispersed in both time and space so that it is essentially invisible and, practical purposes, doesn’t make it to shore.
Aren’t they saying 5,000 gallons per day?
I would park up a few big oil tankers there, and siphone/pump off near the puncture?
That cannot be rocket science can it.
There is plenty of 1million barrel oil tankers around , we are in recession.
5000ft tubes cannot be the holy grail either I see good chunks of that laying around high ways.
Too many professorships and 1000pund an hour contractors around there who never were exposed to anything.
The catastrophe was caused by a BP middle manager btw who wanted “to put his mark” on the rig crew who advised to do otherwise.
A Dilbert story.
I ceased to believe anything from BP when it became clear they couldn’t even tell how fast their tankers were filling up from the collection tube they put into the pipe.
I showed my daughter the videos of the gusher before and after they inserted the collection tube and asked, “do you think it made a difference?”
“No.”
Estimating the discharge rate of pipe from video footage is equivalent to measuring temperature data from tree rings.
Please, please, please. Audit the facts. Don’t get caught up playing the other-sides game.
Good point – but who is the other side in this case?
“The biggest hide the decline story is oil depletion.”
It always puzzles me why the “market is not pricing it right we have almost run out” scaremongering is always about oil. Oil is relatively cheap because it is relatively plentiful and when it gets expensive there are replacements which take the slack – gas, coal, nuclear, biofuels. The later you put off the replacement the cheaper the replacement is. If I had to panic about something soil depletion would be infinitely higher up the list. It is sad that the environmental lobby has become an expensive one trick pony, and nowadays when you hear somebody professing to be trying to look after the environment you tend to try and work out where the scam is.
thanks for the comments from petroleum engineers and others. To be clear, I wasn’t presenting my own estimate of the discharge volume – if I were doing so, I would consult appropriate engineers. I was simply trying to understand the figure of 5000 barrels/day that has been out there for about a month against my friend’s ballpark estimate of exit velocity as BP never presented details of their calculation.
This doesn’t mean that their calculation is wrong or their estimate is wrong, but no one knows where it came from. Obviously other estimates have been made that have been an order of magnitude higher.
One commenter observed that the natural gas expansion may well give a too high visual impression of liquid exit volume. That’s an excellent point. Nonetheless, the elements of the calculation are understandable enough that it seems to me that someone at BP would have been well advised to lay out the calculations to the public in a clear and understandable way. Maybe their Group Vice President, Research and Technology could spend a little time on the problem.
If memory serves, BP’s original estimate was 1,000 b/d. The 5,000 figure came out of NOAA and BP adopted it.
In a CNN interview the BP CEO said that the spill estimates were government numbers. This doesn’t make sense to me, but that is what he said without challenge from the interviewer.
The end of the riser was capped. Thus, the oil flowing is coming out of cracks. There is also the issue of optics (curved lenses) to contend with.
I don’t know the exact geometry, neither do you, but BP has ROVs with radiographic equipment, so they probably do. And they have a much better handle on the gas content.
That Purdue professor who estimated 50K-90K bpd had his 15 minutes of fame, now he can have a lifetime of teaching his students how not to make extraordinarily wrong estimates. Lucky for him he wasn’t actually in a position for his numbers to be relied on so he couldn’t be sued for professional malpractice…
Sense-check: This is supposedly a cased hole that had been cemented (at least sort of cemented). Remnants of cement tend to “clog” the pores in the rock. Furthermore with all the partial obstructions (drillpipe, partly-engaged shear rams etc) and a substantial water column anything above 50,000 barrel/day seems far fetched. I have not seen the videos but the expanding and then rapidly cooling gas might distort the picture. Not impossible to get 100k barrels/day but there needs to be evidence for this claim (remind you of sth?).
This blowout risk is a bit like flying an airplane. Multiple redundancy. Once human error meets machinery failure it comes out of the sky. or blows out. Most of us accept this – but it is easy and fun to bash big oil. Of course BPs lawyers have told them to shut up – this is the US where a jury of housewifes will eventually vote on the “true” flow rate in court.
BTW Anyone notice/care about the month-long burning rig north of Australia recently???
One important factor in measuring oil volumes is that oil in the reservoir has gas in it so it “shrinks” as it is produced to lower pressure/temperature. So, a volume of 5000 barrels per day of reservoir volume of oil, may only 3000 barrels per day of actual liquid on the surface in a tank, the rest is gas. When BP say 5000 barrels per day, I assume (but I’m not sure) they are referring to what is called a “separator” volume or surface volume under specific conditions. That could be 8000 barrels per day at the pipe at the sea floor.
The specific thermodynamic/phase properties of the oil as well as the reservoir fow capacity will determine the gas/oil split at the outlet. This is not easy to predict without good oil and reservoir properties and flow conditions data.
“Maybe their Group Vice President, Research and Engineering should have spent more time trying to figure this out and less trying to hide the trick to hide the decline.”
That comment deserves a gold star! Well said!
Well whatever it was it thankfully now appears to have reduced to zero
http://news.bbc.co.uk/1/hi/world/us_and_canada/10174861.stm
From a few minutes ago, a report that the US Govt will announce a revised estimate of oil flow later today.
“[U.S. Coast Guard Adm. Thad Allen] also said that, later Thursday, an interagency team would release a revised estimate of how much oil had flowed from the well into the gulf before the “top kill” effort began. The Coast Guard had estimated the flow at 5,000 barrels a day, but independent estimates suggested it was much higher, perhaps tens of thousands of barrels a day.”
latimes.com/news/nationworld/nation/la-na-oil-spill-top-kill-20100528,0,5782115.story
The BOP had apparently partially closed off the pipe, so said BP, just not fully. So it’s unknowable what diameter is passing oil. Therefore the velocity is coming out of a cut down orifice. Using the two together, the intact pipe diameter and the cut down orifice velocity, is an error.
good news. It appears that the “top kill” was successful and the blow out is sealed. The article contained new estimates of discharge rates by the US Geological Survey about 3-4 times higher than BP’s, but not 10-20 times higher:
As of 5pm PST, it was still unblocked…
Update: Re-doing these calculations with suggestions made by readers – 9 7/8″ pipe and 50% gas and the same discharge velocity (3 km/hour) yields a bit over 11,000 bbl/day.
Pipe area is pi *(0.250825/2)^2 # 0.0494119 m^2
Volume discharge is 0.0494119* 3000 # 148.2357 m^3/hour
i.e. 148.2357 *1000 /158.987 #932.3762 #bbl/hour
Oil discharge is 24 hours *932.3762 bbl/hour *.5 # 11188.51 bbl/day
And coincidentally just as success is announced, prematurely according to both BP and the WH, NOAA comes out with their 2010 hurricane season prediction: severe. If we have to stop worrying about oil at least we now have a replacement. Yay!
Good, now I can plan my late September Hatteras vacation without any worries of Hurricane activity.
Fom the NY Times
11,000 to 19,000 barrles per day but very uncertain
URL for NY Tiems Report is
This does seem to agree with SMc’s latest estimate — just goes to show you. SMc vs. the team
BP can claim they “had warned that its estimate was preliminary and not very reliable,” but they had repeatedly spurned offers for direct measurement.
Steve: The calculation was not BPs. It was the NOAA’s, and used by the Coast Guard, based on slick area and thickness.
I calculated oil per day, at 3700 bbls/day, based on slick size, and using the Coast Guard numbers for average slick thickness (.1 to 1 micron for 93% of slick, 10 micron for 7%).
With 30% evaporation (estimated for this oil), my number is close to the NOAA number. Is it correct? Probably within a factor of 2 or 3, and probably at the low end.
I would like to add that trying to calculate flow rate, using particle velocity measurement, is totally useless when there is 3 phase flow. Unless you know the gas/oil/water ratio, and pre and post choke pressure, then any number you calculate will be the absolute maximum.
BP has stated that gas has been as high as 6000 standard cu ft (SCF) per bbl oil back to the ship, using the RITT. Assuming the gas does not go supercritical, then the gas needs 800 SCF to occupy 1 bbl of space at sea bottom pressure.
This translates into the calulated oil flow being 8.5 times too high. 85,000 bbls/day becomes 10,000 bbls/day.
This is just for volumetrics.
Using E=m*v^2, with E being pressure at the wellhead and constant; the more gas, the higher the velocity, as the density comes down. This could increase velocity 30% or more.
As well, I noticed that there was gas at the upper side of the horizontal riser. This means segregation of gas/fluids is occuring. As the bottom of the riser was not visible, it is likely that water was venturied into the bottom of the riser, and ejected at the upper section with the oil. This water entry reduces producing cross sectional area, and increases volume. Take off up to 25%
Lastly, almost every well in the world produces water with the oil, especially at high rates. Water cuts of 10-20% are likely. Some older wells produce up to 90% water.
Use the velocity as an upper base line only, then start calculating the other effects.
But, unless BP releases wellhead pressure reading, we will likely never know the true oil rate.
My guess? 5000 – 10,000 bbls/day.
For what it is worth, I have seen, up close and personal, 10,000 bbls/day blow outs in Iran.
They were much more impressive than what I saw of this one.
Of course, those were to atmosphere, while this into water.
Buyer beware.
Incident News from the Office of Response and Restoration, NOAA’s National Ocean Service bears this warning:
Documents are posted chronologically and early reports likely contain factual errors. These errors may be corrected in a later report.
Click to access FRTG_Fact_Sheet_5_26_1.559427.pdf
The oil would slow down relatively rapidly when expanding full-flow from a 9 7/8″ pipe, a partially closed orifice, etc, to the 21″ riser pipe.
It seems the images I’ve seen show a very stable flow for the first foot or so out of the pipe (smooth edges, flow coming out at the full pipe width) before getting very unstable (puffing out into clouds – eddying and whatnot). That suggests to me that the flow has transitioned pretty well to the full diameter of the exit pipe.
If the pipe in question in the images and videos is 21″, then the flow would seem to be signficantly higher than what a 9 7/8″ pipe could output.
Mike, look closely at the video. The oil/mud “jets” emanate from cracks in the pipe, not the end (that was capped a couple weeks ago). The cross-sectional area of those cracks is nowhere near equivalent to a 21″ circular diameter, or appx. 300 sq. inches. Nor, does it appear to me, close to a 9″ diameter.
My guess is more like five or six 2″ diameter orifices, about 19 sq. inches.
Hello
If one were interested in the harm done to the environment, it seems the important metrics would be the amount of oil on the sea surface and plumes under water rather than the amount released by the well. It seems that the important metrics should be able to be measured more accurately than estimates of oil/gas released from the well head.
klee12
BP spill eclipses Exxon Valdez, says government
Reuters, 5/27/10, 12:09 pm
If/when this is sealed, I assume the ROV will get a good view of the openings in the riser through which the oil was flowing and correct the calculations for gas, ROV camera distortions (since this is a wide-angle lens), and the like. I won’t be surprised if it is in the 8-15K bpd range.
One other consideration before anyone erects a scaffold for “lying” BP estimates- since there is sand/concrete chunks in the oil that’s been flowing for a month, it’s very possible that seals and cracks have been sandblasted larger, and oil was flowing at a higher rate than before.
RE: BP discharge rate.
I value your information on the Climate discussions, but the kind of calculations you have shown on the oil rate don’t cut it. First, in addition to solution gas, the fluid exiting the pipe may contain significant amounts of salt water that also occurs in most oil reservoirs. Second, end of the pipe velocity calculations are notoriously inaccurate for pollution estimates. The US EPA (or our authorities in Canada) would not usually accept such an estimate in stack gas measurements, for example. The kinds of estimates floating around by various groups (including the USGS)are mere political escort music that help nobody except people trying to profile themselves. I suggest you leave this matter to people who actually know something about petroleum engineering, understanding that the BP estimate is no more than a ballpark estimate either. It simply can’t be, but that does not imply an attempt to mislead.
You captured my thoughts exactly. From the beginning, BP/Coast Guard cautioned that the estimates were ballpark at best. Not until some of the flow was captured into tanks could oil gravity,GOR and other factors be obtained which the new estimates presumably take into account.
Additionally, BP warned the public and media that they would not know results of the Top Kill operation for at least 24 hours -probably more, yet conclusions are prematurely arrived at and posted here.
David Smith comments: “Interestingly, North American waters see 160,000 MT/yr in natural oil seeps from the ocean floor, mostly in the Gulf of Mexico. That’s about 1.2 million barrels/year. It is widely dispersed in both time and space so that it is essentially invisible and, practical purposes, doesn’t make it to shore.”
The fact is that there are a lot of sub-ocean floor oil seeps around the world. It appears that the ocean is well-equipped to deal with these oil spills, and it may be that the oil is actually a major primary food source – for krill, plankton and the like.
It is likely that naturally occurring bacteria (and other organisms?) have the capacity to deal with oil spills. The issue is, how long does it take for bacteria to do their work. ie, if BP did nothing to ‘clean-up’ the oil spill, how long would it take for the natural processes to deal with the spill?
It is worthy of note that in relation to the Exxon Valdez spill, apparently the only damage to the environment these days is where chemicals were used to ‘clean up’ the oil spill. Apparently the chemicals caused serious problems. In more inaccessible locations, where chemicals could not be applied, natural processes dealt with the spill relativly quickly, and the coastline is as pristine as it ever was.
No doubt some smart people, somewhere, have researched these issues?
good point mondo
who has? always wondered how many times this has happened in geological time & how nature can adapt/move on.
still does not mean we should avoid this kind of man made disaster, but may put in context.
oops
read – still means
Stack gas measurements are a different story. For one thing, gases expand and contract readily based on temperature and pressure. Compress the gas, and you get more mass out per unit volume. If the gas is expanded, you get less. A mass balance is impossible just based on instantaneous volume and velocity alone.
Oil is considered an “incompressible fluid.” The flow rate is dependent on velocity and cross-sectional area only for full-pipe flow.
Thanks to Les Johnson, Harold Thimm and others for contributing your comments. Your insights are all things I should have thought of while watching the video of the flow on TV, but I didn’t. I just saw lots of oil.
I would like to point out that fluid flows at different speeds in the pipe. A layer near the walls of the pipe is almost stationary. The speed increases nonlinearly towards the center. Since a layer near the walls both has the slowest speed and the largest area. This means your estimate can be quite far off. Only Harold has pointed that out so far.
I don’t know if Harolds claim of average velocity being half that of peak velocity is correct. But if so that would halve your last estimate of 11188.51 bbl/day to 5594.22 bbl/day. Which is quite close to the 5000 bbl/day estimate.
Why has no-one pointed out that the natural gas in this flow will kill all the polar bears?
It’s Natural Gas, not anything nasty like Methane.
Now who have guessed that we would end up closer to BP’s estimate once the experts all chimed in?
That’s the power of a thoughtful blog such as CA.
Les’s comments on gas oil ratio and phase segregation are spot on. I’m far from certain it is 3 phase flow however. If the reservoir was producing significant water volumes it may be starting to load up with water and I doubt that BP had the water zone (if any) open to the well at the reservoir depth. I suspect it is producing clean out from the bottom, but possibility of water is not zero. Pretty low, but not zero.
However, there may be dead spot of no velocity at the bottom of the pipe lying on the sea floor as the fluids segregate vertically inside. Hard to say. Doesn’t look like it on the video.
A climate scientist would have used E=mc^2 instead of E=1/2 mv^2 for energy. That would have been a better calculation regardless of scientific validity.
Top kill has restarted. I hope it works, like everyone else.
My (subjective) eyes tell me that the flow has diminished this Thursday evening (US central time zone). I think that top kill is at least a partial success.
So its kind of enlightening that with a relatively well known problem (amount of oil flow from a measurable pipe diameter, making a very educated guess about two variables (flow rate and oil vs methane ratio) and then doing the simple math that smart people can come up with an answer that is an order of magnitude off (perhaps – could be the velocity or mix is way off and the initial hypothesis is closer).
Why? To me, it’s enlightening in that, considering the number of variables that we’re making educated guesses about wrt climate forcings, and the math involved to analyze them, that so many can be so sure about so much, and with such a degree of precision. I’d guess that the number of known variables we make educated guesses at wrt climate is in the dozens, and the number of unknown or unguessable variables is in at least the thousands. Just a thought.
Anyway, interesting post – thanks Steve.
TerryMN-
At least as important is the discussion here has had a lot of “why the estimate is wrong” input, refining both the estimate and it’s uncertainty. This is one factor that Climategate shows was severely lacking. I think the peer review process is fallible (at least on some time scale) even when conducted to the highest standards. Compromise the standards, and it’s a disaster.
Dear Mr. McIntyre,
In one day the Climate Audit posters have done a better job of calculating the flow rate from the BP MC 252 gusher, and in explaining the uncertainties around that calculation, than the mainstream media have done in the five weeks since the blow-out.
Congratulations to all involved. The uncertainty in the gas-to-oil ratio is, and will remain, the key factor on why the range of flow rate estimates is so broad. Since there are no calibrated flow meters on the broken riser, that uncertainty is not likely to be narrowed any time soon.
Let’s hope a successful top kill soon renders the question moot.
Gospel of Luke, Chapter 16, Verse 6.
Heh.
David Smith at 6:15am
quote Interestingly, North American waters see 160,000 MT/yr in natural oil seeps from the ocean floor, mostly in the Gulf of Mexico. That’s about 1.2 million barrels/year. It is widely dispersed in both time and space so that it is essentially invisible and, practical purposes, doesn’t make it to shore.unquote
Not entirely invisible. There a nice image at:
http://petesplace-peter.blogspot.com/2009/02/oil-seeps-in-gulf-of-mexico.html
Does anyone have the figure for how much waste oil is coming down from the Mississippi? Worldwide we throw enough sump oil, washings etc down the drain to cover the entire ocean surface with an oil sheen every two weeks, so one might expect the Gulf to show the effects of western civilisation.
JF
I was wondering what those numbers were. Getting a handle on the scale is important.
I remember reading a few years ago about the seeps off southern CA and those numbers being compared with a number Exxon Valdez spills. Anybody remember what the CA rates are?
I found a reference here: http://www.sciencedaily.com/releases/2009/05/090513130944.htm
They say its between 8 and 80 times the Valdez spill, but don’t give a seep rate for that number.
The article in my original comment estimates that 160,000 MT/yr of oil enters “North American waters” from natural sources. If my early-morning math is correct then that is 3,000 bbl/day. If the Gulf is the major oil-bearing region and has, say, 25% of the total then that’s 500 to 1,000 bbl/day. But, it is probably well-dispersed so that it is rarely noticed.
“Update – good news. It appears that the “top kill” was successful and the blow out is sealed.”
Wait a minute. According to the link (in the original “good news” is a link to this BBC story: http://news.bbc.co.uk/2/hi/world/us_and_canada/10174861.stm)
“BP has slowed the flow of oil and gas from a ruptured well into the Gulf of Mexico, a US official told local media.
The company’s “top kill” effort has “stabilised the wellhead”, Coast Guard commander Admiral Thad Allen said.
But he cautioned it was too early to declare success. This is the first step in BP’s plan to seal the well for good.”
AP just reported this:
” By GREG BLUESTEIN and BEN NUCKOLS, Associated Press Writers Greg Bluestein And Ben Nuckols, Associated Press Writers – 4 mins ago
ROBERT, La. – The chief executive of BP PLC says it will be about 48 hours before they know if pumping heavy mud into a blown-out well is successful in stopping the massive Gulf of Mexico oil spill.”
http://news.yahoo.com/s/ap/20100528/ap_on_bi_ge/us_gulf_oil_spill
Thus your statement is more optimistic than the evidence suggests.
USGS announce flow rate at 11,000 bpd. he he
“Update – good news. It appears that the “top kill” was successful and the blow out is sealed.”
Your assessment is a bit of an over statement. Even the link to the BBC you have says:
Top kill’ method ‘slows BP oil leak’ in Gulf of Mexico
Page last updated at 17:35 GMT, Thursday, 27 May 2010 18:35 UK
BP has slowed the flow of oil and gas from a ruptured well into the Gulf of Mexico, a US official told local media.
The company’s “top kill” effort has “stabilised the wellhead”, Coast Guard commander Admiral Thad Allen said.
But he cautioned it was too early to declare success. This is the first step in BP’s plan to seal the well for good.
Ixtoc I oil well: Gulf of Mexico; June 3, 1979–March 23, 1980; 3,328,000–3,518,000 barrels
http://en.wikipedia.org/wiki/Oil_spill
How long did this oil pose a threat to the area? Is the oil still there?
What about the natural leakage from the oil field?
http://sarsea.com/natural_seapage.html
I would like to pick the brains of the many people with oil industry experience contributing to this blog.
I understand that many countries insist on having a backup blowout preventer that is independent of the main one and can be triggered acoustically.
Would that have killed the blowout?
if so:
Why didn’t they have one?
Bryan Leland,
The simple answer to your question is that it wasn’t required.
It appears that the BOP was not properly maintained. If that is the case, then one can surmise that an acoustically actuated BOP would likely also suffer from the same neglect in maintenance. Would a broken acoustic BOP have prevented the blowout?
There seems to be a lot of misinformation about this.
There is no “backup blowout preventer.” The acoustic control is just another way of communicating with the same blowout preventer that failed.
Only two countries require an acoustic trigger (not “many”) – Norway and Brazil…and Brazil only has since 2008. A 3rd party study done on behest of the US gov’t (think it was 2002 or 2003) found that the acoustic devices were expensive and unreliable. The report did not recommend their use.
The BOV in this accident seems to have partially closed based on tests that were done after the accident. So either someone on the rig had tried to shut it, the BOV tried to shut itself as it is supposed to as a failsafe, or the manual shut-off attempts using unmanned subs activated it.
All an acoustic system would have provided would be an alternate means for communicating with the BOV in addition to the three above. Communication wasn’t a problem. The issue appears to be that the valve didn’t completely shut for whatever reason. Possibilities included the inability of the shearing ram to go all the way through the pipe due to the presence of a joint or thicker-walled pipe, blockage from the concrete work Halliburton had done, and leaking hydraulics (the latter has been verified to have occurred, although it’s not yet determined for sure how significant it was at the time of the accident).
The best way to think about it is this: if your TV is blown due to a power surge and won’t turn-on when you hit the power switch, how is having a remote control going to help you turn it on?
the total volume of the oil spill is quite small to previous events.
The gulf war oil spill was about 10 million barrels size, the 10th biggest still with 1 million barrels.
Additionally the leak is far from the coast and oil certainly decomposes in the warm carribean ocean much more quickly than, for example, in the arctic ocean
Again, this event was massively overhyped and instrumentalized for politics.
http://en.wikipedia.org/wiki/Oil_spill
If it wasn’t required wouldn’t a prudent operator have installed one anyway? Or was it that, as many overseas engineers (including me) have discovered, many North American engineers regard complying with regulations more important than good design?
But if they had two, and they were different, the chances of both failing simultaneously would be substantially reduced.
I have just read in a New Zealand newspaper that operators wanted to trigger the preventer but “didn’t have the authority”. So there was a delay that might have been significant. The same thing happened at a major failure at a Russian hydro station last year. As a result, 75 people lost their lives.
Prudent operator? I thought we were discussing BP …
http://pipeflowcalculations.com/
Here use this site.
Brian Leyland: An acoustic switch, as Michael said, would not have worked.
It is only a way to communicate with the existing BOPs. It is not a separate system.
These BOPs had 2 different hydraulic systems and two different electrical switches, all to control the valves on the BOPs.
On top of the 4 way redundancy, there is also a “failsafe” system, which shuts the valves if communication is lost with the surface.
Having an acoustic switch would only mean you now have 6 ways of not being able to control the valves.
The only advantage to an acoustic switch, is that if communication is lost, and you still have a functioning BOP,, then the acoustic switch will work a few hours quicker than sending down ROVs.
Note that this is the first thing BP did: send down ROVs to try to shut the valves.
The fact that the ROVs can’t, means an acoustic switch couldn’t either.
The fact the ROVs can’t close the valves, means the power source or the valves themselves failed.
Imagine an acoustic switch as a light switch. If the power is down, or the bulb is burnt out, no matter how many times you flick the switch, its still going to be dark.
Had the Blow Out Preventer worked, would this mess have been prevented?
Is the present design of these devices adequate? Are they serviceable in situ at these depths?
Should there be design, testing, certification requirements for these devices specific to their use at these depths?
If a BOP is triggered from the surface by mistake and it functions properly, is it difficult (ie. very expensive) to get the well reopened?
If you had the authority, are there changes to the requirements for these devices that you would impose?
What would these changes include?
J. Ferguson:
Had the Blow Out Preventer worked, would this mess have been prevented?
Yes. The oil is comimg up through the inside of the BOPs, and not outside it, so the loss of control is in the BOPs.
I find it nearly inconcievable that these failed. There were 4 seperate systems on the BOPs (shear rams, pipe rams, blind rams and hydril) that could have prevented this. Also, there were 5 seperate control systems for each ram, plus they can all be operated by the ROVs.
In addition, there were at least 3 more systems outside the BOPs that should have contained the pressure, and as many 5.
Even after a month, I am shocked that so many systems, with so much redundancy, all failed.
Is the present design of these devices adequate?
We will need to wait for the final report, but in my mind, yes. The failure had to be in the human interfaces, simply because there were so many failures.
Are they serviceable in situ at these depths?
To a limited extent, yes. However, the system is designed to be lifted back to surface, for servicing, while the well is contained with other methods. But you need control of the well, for these other systems to be used.
Should there be design, testing, certification requirements for these devices specific to their use at these depths?
There already is. In spite of the noise the government is making, they legistlate, monitor and enforce adherence.
If a BOP is triggered from the surface by mistake and it functions properly, is it difficult (ie. very expensive) to get the well reopened?
No. Best case is a few minutes lost, with no damage. Worst case is loss of drill pipe, and a few days spent fishing it out.
If you had the authority, are there changes to the requirements for these devices that you would impose?
I need to see the final report. Beforehand, I thought current procedures were way over the top.
One thing I would insist on, is a kill system already in place, so that ROVs and other ships could quickly hook up, instead of having to build everything from scratch. All the existing kill systems assumed the rig would still be there.
Again, there must of been multiple human failures, which is subject to training and monitoring. Hardware changes will not address these shortcomings.
As I am fond of saying, you can foolproof a system all you want, but nature is already working on a better fool.
Mr. Johnson,
thanks much for your informative observations. Do I understand that the connections for the “kill” plumbing had to be added to the apparatus already there? If so, what more specifically needed to be there and would it have been plumbing attached to the BOP or in addition to it?
It’s kind of hard to accept that this spill is a result of operator screw-up at the time and not multiple defects in the equipment – redundant as it might be.
j ferguson: Your
Do I understand that the connections for the “kill” plumbing had to be added to the apparatus already there?
Yes. The kill lines were there, but were hooked to the rig lines. New connections had to be manufactured, and the old ones disconnected.
I would probably also want a sensor package inside the BOPs, that could be routed by the ROVs to a new monitor.
If so, what more specifically needed to be there and would it have been plumbing attached to the BOP or in addition to it?
Manifolds, piping, quick connects and hoses could be already connected to the BOPs, in addition to the rig kill lines. It would just require ROVs to switch valves to external kill, and to quick connect to another ship with pumps.
This assumes, however, that no wreckage falls on the BOPs or kill manifold system.
It’s kind of hard to accept that this spill is a result of operator screw-up at the time and not multiple defects in the equipment – redundant as it might be.
Its probably both. Some equipment probably failed, and some mistakes were made. But I cannot see how 7 to 10 different hardware systems failed. Any hardware failure was probably compounded by human error.
As I said, knowing the redundancy in deep water work, I am shocked, literally, that there could be this much failure.
Les, I agree. the failure of the BOP is actually the biggest mystery here, but I don’t think even shear rams can cut drill collars. You may know better. With drill pipe in the well, the blind rams are useless as well.
j ferguson, was the Blow Out Preventer any good to start with? Was it a known rejected Blow Out Preventer waiting for a time like this, or set up to fail?
Seems very convenent this happens right after Drill Baby Drill by you know who!
Prior to starting the top kill, BP had inserted a siphon tube into the end of the fallen 21″ riser. This siphon was collecting several thousand bbls of oil each day. Once the siphon was established, flow would preferentially move in the direction of the least amount of hydraulic/friction head. There are lots of variables here.
The oil/gas has to choose between flowing out the several cracks in the 21″ riser that formed when the riser bent over the top of BOP (and against the water column of the 5000′ of sea water pressure) or heading down the 5000′ of riser to the siphon tube which has a column of a much lower density oil and gas flowing up to the surface.
No oil has leaked in the past few days due to the efforts of the top kill. If BP needed to, they could just continue to pump mud until the relief wells are completed. Doubtful though, considering the cracks would eventually wash out.
I think when this whole event is over, the amount of oil lost will be much less than many people here are predicting. I also predict the ramifications of this tragic accident will ultimately be felt by all the oil companies (and to the pocket books of the average citizen) and that these ramifications will be largely negative and unnecessary.
This is a bit away from the mechanics of this thing, but I wondered if any of you had felt as I did, that the guys at the Senate interrogation were not at all trying to shift blame, but genuinely didn’t yet know, in detail, what had happened.
And they were browbeaten for their uncertainty.
There is reference from time to time, here, to quantitative approaches vs qualitative approaches. I suggest there is another – maybe not in the same family, making things – might be same as engineering.
My ire is raised by people who would be hard pressed to make a blt sandwich berating guys whose project is having problems.
the public outrage that something that is hard to do is hard to do really offends me.
Making things does require aptitude, I think, one that is not found in everyone – not to diminish the aptitudes they do have, but….
The people who are in the business of fooling all of the people some of the time etc. should recognize that the people in endeavors in which fooling mother nature is impossible all the time should cut them a little slack.
And no, Russ, I don’t believe in conspiracies of the type you suggest although I agree that this mishap was about as untimely as it could be.
Leonard: Your
Les, I agree. the failure of the BOP is actually the biggest mystery here, but I don’t think even shear rams can cut drill collars. You may know better. With drill pipe in the well, the blind rams are useless as well.
Yeah, thats my thought, that a collar is across the shear rams. Or a pipe joint. Its bad procedure, though, to have only collars at the rams. It would mean the shears and pipe rams would be useless. It would be bad luck to have a pipe joint at the shears, but also bad management, as the crew should know the position of the pipe joints in realtion to the rams.
There is also some speculation, that the new materials they are using to make the drill pipe, is too strong for existing shear rams.
The blinds were apparently useless, but as they were circulating the riser to sea water, that means they were at or above the BOPs with the bottom of the drill pipe. A few meters of pipe movement, perhaps, and they could have engaged the blind rams.
Lots of second guessing, so we will need a report to look at.
I am thinking along the same lines….you have multiple devices to kill the well with multiple ways of communicating this and all fail….looks like either a collar, a joint or harder-than-heck alloy in the drill pipe. Did the guys engineering the BOP communicate to the guys designing the new pipe that needed to operate in frigid waters? Wouldn’t want to be the systems engineer who had that job.
From the 60 minutes report, the seal (pipe ram??) was allegedly damaged when the pipe was moved while the “BOP” was closed. So, perhaps the seal is all blown out by now.
Its all speculation I suppose. The next move is to cut the damaged riser and lower a new package on top. Seems to be a pretty reasonable cahnce of success.
Just for the clarification of those who are mystified by the language:
Hydril (or the bag) – a rubber sack that inflates with hydraulic pressure, to seal outside any size pipe. This is the lowest pressure containment.
Pipe rams – hydraulic rams pincer the pipe from two sides, and forms a high pressure seal around the pipe.
Shear Rams – hydraulic rams that cut through pipe, to form a seal inside and outside the pipe.
Blind rams – hydraulic rams that close on the open hole, with out pipe in the hole.
I’m learning more here in a few minutes than I learned in a month of legislative hearings, newspaper reports and TV. Why is this so hard? Thanks Les! I’m now reading The Black Swan by Nassim Taleb. Very interesting.
Mr Les Johnson receives my vote for quote of the week:
“You can foolproof a system all you want, but nature is already working on a better fool.”
CNN is breaking the news that BP says the Top Kill procedure has failed: http://www.cnn.com/2010/US/05/29/us.gulf.oil.spill/index.html?hpt=T1&iref=BN1.
“Venice, Louisiana (CNN) — BP’s “top kill” attempt to stop the flow of oil from a ruptured well in the Gulf of Mexico failed, the company’s chief operating officer said Saturday.
The oil giant has tried for days to stop the the largest oil spill in U.S. history by pumping heavy, mudlike drilling fluid into a ruptured oil well, a method known as “top kill.”
The next option is to place a custom-built cap known as the “lower marine riser package” over the leak, the company’s chief operating officer, Doug Suttles said. BP crews were working Saturday to ready the materials for that option should it become necessary, he said.
“We’ve been prepping that all along in case we need to move to that option,” he said. “People want to know which technique is going to work, and I don’t know.”
And if “lower marine riser package” were to fail, he said, BP engineers would try placing a second blowout preventer on top of the first, which failed to cut of the oil flow after the April 20 explosion of the Deepwater Horizon rig. The failed blowout preventer is a 48-foot-tall, 450-ton apparatus that sits atop the well 5,000 feet underwater.
Meanwhile, teams in Louisiana were working Saturday on a clean-up project aimed at protecting coastal marshes while BP continues its efforts to stop oil from gushing into the Gulf of Mexico.”
…
…
…
Steve, you’re a clever guy. So how do you explain your lack of knowledge re “significant figures”.
Your last calculation includes “3 km/hr”, ONE sig fig. But the result, 11188.51 bbl/day, has SEVEN??!!
What’s up with that?! (The correct result is 10^4 bbl/day.)
This is terrible, I feel this is going to have a lasting effect on this ocean for years to come. There is already reports of marine life dying far away from the oil spill due to the disperant possibly, which is causing nearly as many problems as the oil itself.
More info on nalco BP and corexit on my blog.
Well, as of June 8th they’re capturing 15,000 barrels per day of the spill, and do not have it contained yet. BP claims it will increase the capture by 5,000 barrels to 20,000 barrels per day, and that may not be the limit either. We can now determine that the spill rate is definately greater than or equal to 20,000 barrels per day.
So it looks like your estimates were pretty good, better than most others.
Steve, it looks like your initial flow rate calculations were right in line with what BP calculated.
“June 20 (Reuters) – An internal BP Plc (BP.L) (BP.N) document released on Sunday by a senior U.S. congressional Democrat shows that the company estimates that a worst-case scenario rate for the Gulf of Mexico oil spill could be about 100,000 barrels of oil per day.
The estimate of 100,000 barrels (4.2 million gallons/15.9 million liters) of oil per day is far higher than the current U.S. government estimate of up to 60,000 barrels (2.5 million gallons/9.5 million liters) per day gushing from the ruptured offshore well into the sea.
The document, which is undated, was released by U.S. Representative Ed Markey, chairman of the energy and environment subcommittee of the House of Representatives Energy and Commerce Committee.
The amount of oil actually gushing from the well has been a matter of considerable controversy since the spill began on April 20, with critics saying BP has deliberately understated the flow rate.
“This document raises very troubling questions about what BP knew and when they knew it,” Markey said in a statement.
“It is clear that, from the beginning, BP has not been straightforward with the government or the American people about the true size of this spill. Now the families living and working in the Gulf are suffering from their incompetence,” he added.
The document was posted on the Internet here: http://globalwarming.house.gov/files/WEB/flowrateBP.pdf.
(Writing by Will Dunham, Editing by Sandra Maler) ”
http://www.reuters.com/article/idUSN2020003220100620
Two months after the Deepwater Horizon explosion, the New York Times has published information about the innerworkings of a blowout preventer (BOP).
href=”http://www.nytimes.com/interactive/2010/06/21/us/20100621-bop.html?ref=us”>Inside the Blowout Preventer
Part of a larger article.
href=”http://www.nytimes.com/2010/06/21/us/21blowout.html”>Regulator Failed to Address Risks in Oil Rig Fail-Safe Device
Like with airline disasters, it will be many months or years before the hardware is surfaced, disassembled and inspected to determine the cause of failure. Right now we have educated guesses, uneducated guesses and speculation.
The American public needs to stop being so critical of BP Oil. They are doing the best they can to bring an end to the oil spill disaster. Why not look at Transocean for the blame? It can’t be all BP’s fault.
Am I the only one sceptical about these ever escalating numbers? According to Steve’s calculation above, the well would have to be spewing oil at the rate of up to 17 mph for the figure of 100,000 bpd to be in the right ballpark. Is this feasible? Have any of these published estimates shown their calculations?
NPR today reports on the source of the estimated flow rates.
http://www.npr.org/templates/story/story.php?storyId=130390111
National Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling, The Amount and Fate of the Oil
http://www.oilspillcommission.gov/document/amount-and-fate-oil