I know that post might be drowned out by the death of Sen. Byrd (R.I.P Senator, your service will be remembered for a long time) and the start of confirmation hearings for Elana Kagan (Release the Kagan!) but given the intense pressure that short-sighted politicians are putting on the administration for the resumption of deep water drilling I thought it is worth looking at what we might be able to do to decrease the chance of another of these accidents.
BP and the other oil companies rated the chance of a major blow out as statistically zero. This was a huge mistake in and of itself. The odds of picking all six numbers in a 42 number lotto drawing are 184 million to one. You have a better chance of being mauled by a polar bear and a regular bear in the same day than doing picking those numbers. Yet they are picked all the time. On average there is a jackpot won every 4 weeks. 184 million to one are effectively and statistically the same thing as zero chance, but as the example shows zero does not mean no chance.
"Originally posted at Squarestate.net"
There is also a difference between reliability and safety that seem to have slipped away in our discussion of what is appropriate for off-shore and deep water drilling. When we are talking about reliability what we are looking at is the chance of occurrence. This is what BP and its engineers were really talking about. They said the drilling plan was safe because it gave a low chance of occurrence of a blow out. This means it had a high reliability but that is not the right measure when we are talking about safety.
In safety we have to look at the magnitude of regret involved in a failure. It is the difference between a airplane door seal failing one time out of 10,000 in a test facility and it failing one time at 38,000 feet and causing a crash, killing everyone aboard. As the BP Oil Disaster has made clear, any drilling plan should be focused on the magnitude of regret, and not the reliability of the systems. If we had this standard there would have been a focus on multiple redundancy and testing which may have resulted in a another whole set of actions and perhaps an avoidance of this catastrophe.
While it is clear there is nowhere near enough R&D on clean up, the best case is always to avoid the accident or defect in advance. The cost of prevention is always orders of magnitude lower than the cost of remedy whether we are talking about oil spills or care accidents. The question is what kinds of actions can be taken that would put the focus on the front end prevention in a way that will allow us to drill more safely while new and more effective clean up technology can be developed?
To me the best place to look is NASA. There are strong similarities between deep water drilling and space exploration. Both environments are very hostile; work in both environments are often done by remote control; and the cost of failure is high in both environments. The big difference is that, as a rule, NASA will actually call a no-go if systems are not in the accepted parameters. There have been notable failures, but the culture at NASA has been one of safety over production. This seems to be the opposite for BP and other oil companies.
In general a Blow Out Preventer is a classic case of multiple redundancy. It has what are called shear rams, blind rams and an annular preventer. The shear rams are designed to cut the drill line, the blind rams are designed to push the drill pipe closed and the annular preventer is another way to choke the pipe closed. The problem here is that they don’t work very well in deep water. The cold and pressure of 5000 feet plus make metal brittle and rubber brittle. When we are talking about a system that works on high hydraulic pressure that is very bad news indeed.
One way to get around that, short of redesign is to use multiple sets of BOP’s. By having more than one of these (ideally three) it would increase the chance of one of them operating as designed in an emergency. However on its own this would not be enough. One of the known issues with Deepwater Horizon well is that the BOP was not in working order. There has been testimony that there were known problems (rubber from the annular preventer coming up the drill shaft, hydraulic leaks, a dead battery in the dead-mans switch, incomplete drawings of the actual package) with the BOP prior to the disaster. This can not be allowed to happen in the future.
Again we have to deal with the magnitude of regret. Going forward we must have regulations that require more rigorous testing of BOP’s and on a tighter schedule. There is a problem in that the MMS was a captured agency and so enforcement was increasable lax. One way to get around this is to require remote sensors on all BOP packages and that the data be transmitted daily to the Department of Interior. This needs to be combined with a new clause in drilling permits that make the permit expire if the BOP is out of specification. This puts the incentive in the right place for the oil companies, as they will have to shut down their operation if they do not have a working safety system.
Another area where we need to reset the incentives is in cheating the system. Any kind of inspection fraud, if discovered, would mean the automatic revocation of all drilling permits for that company. They would have reapply for each one, through a full process. Both of these measures would be incredibly costly for the company, so their best course would be to follow the rules the first time instead of cutting corners.
We know there was a concern on the part of BP about delays in finishing the Deepwater Horizon well. There were costs in the millions of dollars per day. In order to balance that kind of cooperate interest we have to make the cost of failing to do the safe thing as high or higher than what is saved by cutting corners. More BOP’s are also expensive, as would remote monitoring sensors and the transmission equipment. However it will never be as expensive as a single disaster like Deepwater Horizon.
The fact is we must move away from oil, sooner rather than later. It is a finite resource and is killing our environment. However it can’t be done in the flick of a switch. It is clear that our politicians will want to keep using our Outer Continental Shelf reserves as we move to new sources of power. This being the case, we must put in place the systems that will make the likelihood of such an accident as this as unlikely as possible.
For those who would whine that we are putting too much burden on the oil companies, I would like to point out that we subsidize these companies to the tune of to the tune of 30 billion dollars a year. There is also the issue of the cost of oil. It is a floating market and if the cost of drilling safely in our waters is too high, then they will not do it. The oil that is drilled in the Gulf does not automatically go to the United States, it is sold on the world market and can wind up anywhere. Since we bear the cost of an ecological disaster (even if the company pays for so-called clean up) then the burden to keep such a thing from happening falls squarely on the oil companies.
By creating as system where constant proof of working safety measures is a requirement for continued operation, at all times, and putting the appropriate incentives and negative consequences on the oil companies, we can increase our level of confidence that everything that can be done to avoid cutting corners on safety will be done. By having a system where we shut down drilling operations at the first hint of trouble we will increase the cost of drilling but might be able to avoid the cost of disasters. The key is making the magnitude of regret so high that they do it willingly.
The floor is yours.