Part Two –The cause
What, then, did cause the BP ‘Deepwater Horizon’ Mexican Gulf oil leak? Part One explained what did not. This is what The Lad has learnt about what the engineers involved believe so far in October 2010.
This drilling did not succeed: it failed. This time there were failures on gigantic scale. This cannot and must not be denied. The even bigger job of recovering from the blow out and the vast leakage needed even bigger teams of engineers and took 87 days; but the engineers did, in the end, succeed.
In most of the great engineering undertakings the teams succeed. Not because they make no errors but because of a combination of good practice, planned operational checks and balances and, if and when occasional potential errors still get through, other team members pick them up. There are thousands or hundreds of thousands of such projects at any one time that proceed to a satisfactory completion using the built in checks and balances and the judgements of the engineers.
There is a simple, descriptive model that I think casts some light on the tragedies of real world engineering.
The likelihood of one error in a job occurring has a probability and the checks and balances correct it. Two errors in the same job have a smaller probability and they are also corrected. Likewise, for a job to have a larger and larger number of errors, there is a smaller and smaller [called monotonic] probability. That is, they become less and less likely to occur. In this model, if even just one of the errors are spotted and corrected, the chain of error is broken and problems averted.
Consider venturing even further into very, very unlikely events, that is of very, very low, low probability. We could arrive at a job where there are more errors than there are existing checks and balances or some are not applied or without any team members picking them up. What might happen then? Catastrophe, that’s what.
In a big job there is a possibility everywhere and all the time of one or more errors lurking due to human mistakes or by malign circumstance. It is the enemy that he or she has to fight as part of the engineer’s professional task. That’s one of the things that engineers do.
There are the previous, rare cases where such multiple errors occurred and none are picked up. Take the Piper Alpha, UK, North Sea gas production platform accident; the Flixborough Nypro plant explosion; the West Gate Bridge collapse in Australia or even the horrifying Baby P case. Each one had multiple errors all in the same project and none put right. BP ‘Deepwater Horizon’ Mexican Gulf oil leak was undoubtedly another such. A chain of uncorrected errors dragged the rig and a group of its engineers into death, flame and pollution.
To repeat, all great disasters that I know of are due, not to one error great or small, but to a chain of mistakes. A chain where one correct decision would have averted a catastrophe or at least resulted in failure made small. It seems to me to be a valid, general rule.
I believe that the general outline of what went wrong is here, as described in the BP early analysis.
- Bad choice of concrete mix
- Seals in the bottom-most section leaked
- The pressure tests that should have shown that the concrete and seals were inadequate were misinterpreted as showing them to be adequate.
- The instrumentation in front of the controller’s seat showed a small whisper reaching the surface which was the quiet first stirring of the fierce, violent power of the hydrocarbons hurtling from the bowels of the earth. Those sitting in that chair should have fastened onto the whisper and recognised it. Then they could probably have acted to at least mitigate the problem. None of the crew Involved in the drilling operation recognised the whisper.
- In the last terrifying moments as the catastrophe took its gigantic form on the rig, the team could have sent the gas into one of two directions. One was overboard and the other was into a tank. The choice of an effectively infinite volume overboard could have given time to respond and the consequences lessened. Someone chose the tank. The finite volume tank was soon overwhelmed and gas flooded the rig. A rig power supply engine fed on the flammable gas rather than air went into howling over-speed. The gas entered parts of the rig with unprotected electrics and somewhere here a spark turned the gas cloud into a massive incendiary bomb.
- The Blow out Preventer failed entirely to close off the well. Not one of the three independent ways of it doing its job.
As the BP ‘Deepwater Horizon’ Oil Spill ground its way towards catastrophe, there was certainly no one engineer who ran the operation minute to minute and whose word was law. There weren’t even 2 or 3 such engineers. There were several corporate teams of engineers each answerable to their own management with their own team philosophy. There were three principal teams [there were others] working on the drilling.
There was BP as the operator of the lease on this part of the sea bed. It itself had a subsurface team as well as a well engineering team that was presumably based on land.
There was the Transocean team who owned and crewed the floating rig and had been operating and working it for BP for nine years.
Then there was the Halliburton team providing cementing services.
We will see that each team had or will claim to have had a different view on responsibilities. The drilling task was extremely elaborate and ever-changing hour-by-hour or even minute-by-minute. Despite this the costs involved mean that this argument is almost certain to grind through the courts for years. In the real world it isn’t the end when the engineers agree about the errors. Then; the blame game blossoms. This game features the financial specialists and the lawyers are at point.
We cannot yet be sure that the early analysis has got the details exactly right and one or more may not have occurred at all or in exactly this form. Longer and more detailed investigations that will take place stemming from the expenditure of more money and investigatory manpower will almost certainly fill in the detail. What, by the way will drive this further investigation. It is simple – more money. But why will this be spent and from where will the treasure come from? Why from several players. One is of course an extremely angry government. The others are, of course, the engineering protagonists who are extremely large and very rich. Because they are very rich this means that they can spend a lot to justify their actions and, they hope, show that the actions of the others were to blame. Because they are very rich also means that should they be shown to be at fault then their large treasure houses will be pillaged by fines and damages, Powerful driving forces indeed.
Whilst ever the global consumption of hydrocarbons runs, as it does, at such a febrile rate, men will continue to reach down under the ocean and ever deeper in the rocky strata. The economic pressure is still there so the engineers will work to successfully overcome the problems and, as usual, accommodating the problems.
The US President’s Commission has, at the time of writing November 2010, seemed to consider that the concrete failed to do its job and played a part in the accident. There are also arguments that the depth at which a single concrete plug was set is significant. The Lad recommends that you visit their website [link following] and you will find excellent graphics that explain and clarify the engineering of this and, by extension, other oil wells. For completeness after that, I have added the BP website that leads to the BP report.
All engineers will agree that if they are so responsible for so much of our modern world, then engineers must also bear proper responsibility for when there are failures. …But… Edward Phelps, a lawyer turned diplomat, observed back in 1899 that,
“The man who makes no mistakes does not usually make anything”
It is as true today as it was then.
“Edward John Phelps.” The Columbia Encyclopaedia, Sixth Edition. 2008.
In summary, I repeat that there is no sign that the accident was caused by the drill site being in deep water or not on land. Neither was the depth of the drill bore below the sea bed. The cause was a series of errors that occurred in an unbroken chain. These errors could have occurred in drill site on land or an ocean bore that was shallow.
It is the error chain overseen by teams of engineers that needs to be studied. The causes must be discovered and never repeated by future engineers. Engineers seek to make things happen and to bend natural forces to convenience of humanity. This aim must never be allowed to fall victim to hubris or a gung-ho attitude or a cavalier attitude that risks health and safety. That is the lesson that has to be continually re-learnt by BP, its contractors and all engineers everywhere.
[P.S. – I do not work for BP and never have. Nor have I ever had any connection with it or any other organisation in the oil industry.]
[P.P.S. – I do buy petrol for my car though.]