As BP works with its partners to contain the oil spill in the Gulf of Mexico following the explosion and sinking of Deepwater Horizon, E&T examines the task at hand and the measures being deployed.
It all went wrong for BP in the Gulf of Mexico late on the evening of Tuesday 20 April. At the time of the incident, BP contractor Transocean was drilling an exploration well on Mississippi Canyon Block 252, where BP has a 65 per cent interest.
The first hint of impending risk was when a fire was reported at 10pm. This was followed by a huge explosion, and two days later the rig sank, setting in motion an extensive spill response operation involving more than 2,500 people, a fleet of vessels, aircraft, dispersants and booms.
Of the 126 people on board the rig at the time of the explosion, 11 are missing feared dead, and 17 were injured, three of them critically.
The exact cause of the disaster remains unclear, and full details will only come out in time when the remains of the well are recovered from the sea bed. The US Government has already subpoenaed all the subsea equipment, particularly the blowout protector - designed to seal the bore in the event of such a failure - that failed so disastrously to operate.
What has become abundantly clear in the four weeks since the disaster is that, in trying to plug a well 5,000ft below the surface, the industry is reaching into the unknown. BP CEO Tony Hayward likened it to the Apollo 13 accident or a jumbo jet falling from the sky - neither of which, he says, would cause an industry to crumble.
'We are learning all the time,' he says. 'It is the first time that the industry has had to confront these issues in these water depths, and there is some real-time learning going on which is informing the intervention effort.
'The ongoing investigation will undoubtedly throw up things that we should be doing differently. The investigation should focus on the original incident, but also the failure of the blowout preventor (BOP) as well as the inability of the industry to intervene at the subsea environment at 5,000ft.'
It must be remembered that the oil and gas industry has been exploring in deep waters for over 20 years and this is the first time that it has suffered an incident of this nature. Drill ship Deepwater Horizon had operated in 10,000ft of water and drilled to a depth of 35,000ft, far beyond the 5,000ft of water and 18,000ft of drill depth in this instance. That aside, it has become apparent in the efforts to plug the leak that the industry is pushing new technology boundaries to their limit and, to date, have come up short.
'The industry has drilled over 5,000 wells in water deeper than 1,000 feet and has not hitherto had an issue of this sort to contend with,' Hayward continues. 'The reality is that you can have all sorts of theoretical capability, but until you actually get to use them in anger it is difficult to figure out what is necessary. There is an important amount of learning going on here because we are doing it for the first time and the learning will, I am certain, have profound effects for the industry.'
Biggest ever response
All the operations to plug the leaking well in the subsea environment and contain the oil slick at sea and on the American coastline are managed from the control centre in Mobile, Alabama under the ever watchful eye of Keith Seilham, BP's incident commander. 'Most important is that you have to have excellent planning if you are going to respond to an incident,' he says. 'We have an operations section that is primarily concerned with responding with equipment E F and resources; we have legal and finances, and most importantly logistics and other various support organisations.
'This incident is the largest that we have ever done. In recent history we have only initiated oil spill response with two oil spill vessels. To put that in perspective, in this instance we have a response with 16 oil spill vessels. We are employing a safe chemical to disperse the oil and in this incident so far we have used more dispersant than has been used in the cumulative response to all oil spills in history.'
Above the waves over a million feet of boom has been deployed, with another three million feet in reserve, and chemical dispersant has been deployed from ships and aeroplanes, with relative success, but it is 5,000ft below the waves that the real engineering challenges have been taking place.
There were always two strands to the subsea approach. The first was to minimise and contain the oil pouring from the well, and the second, and ultimately more important task, was to permanently seal the well.
The first option on the agenda, soon after the response was on station, was to try to manipulate the BOP to seal the bore with the use of eight remotely-operated vehicles (ROVs), an operation that Bob Fryar, senior executive vice president of BP likened to open-heart surgery at 5,000ft. Those attempts ultimately proved unsuccessful.
At the same time BP was injecting dispersant directly into the oil leak at the seafloor, pumped from the surface and applied by ROVs with wands. 'This is something that has never been done at this sea depth before,' Fryar explains. 'What this involves is pumping the dispersant down a pipe and injecting it with a wand on top of the out-flow of oil coming from the pipe. What happens then is that the oil and dispersant mix and the oil is broken down into smaller droplets.'
One of the more challenging solutions that BP pursued was the subsea collection and capture of the oil. This initially involved lowering a huge dome - a 'cofferdam' - over the outflow of the oil and to move the oil from the dome via a pipe connected to the dome up to Enterprise - an oil platform sitting on the surface. This has been achieved in shallow water before, but never at this depth.
The attempt ultimately failed because of the formation of hydrates. 'The cofferdam was 40ft tall; it was massive,' Kent Wells, senior vice president of exploration and production at BP, explains. 'We felt at the time - and remember that we have this broad array of technical people that look at all our options - that this was the best option available to us.
'We were aware of the problem of hydrates. We knew that if there were a large amount of hydrates it could do what it did. As soon as we deployed, the top of the dome filled with ice and it shut off the flow. Our belief was that we had a 12in opening there, and that would be enough to allow the hydrates to escape. We knew it would be a problem going up the pipe and that is why we were going to allow the oil to go up the centre while we pumped warm water down the outside to try and keep it warm, but there were more hydrates than we expected.'
Failure is all part of the process of engineering innovation at 5,000ft. 'We wanted to approach this in a multi-faceted way, with parallel approaches,' Wells adds. 'At every stage we want to have multiple options going simultaneously and, as we learn something, that will bring that option to the forefront. We are designing every option to be successful but we are planning for it failing, so that if it does we know immediately what we are going to do next.
Wells explains that they learned a lot from the failure of the cofferdam and the next attempt to capture the escaping oil was to be on a much smaller and manageable scale, although by definition it was likely to be less productive. 'We're going to go from this big 40ft structure to one that is four feet in diameter and five feet tall, so we've gone from one extreme to the other,' he says. 'The purpose of that is to keep the amount of water out.
'If we can keep the water away from the gas then the hydrates can't form. We are also going to pump methanol into it so that even if water gets in there it will stop the hydrates from forming. We obviously couldn't do that in a large vessel because we would be pumping too much methanol, but in the small vessel we believe that will work.
'We will once again have the warm water pumped down the outside to prevent hydrate forming as it flows up. And what we are also going to do is have it all connected before we install it. One of the things we learnt when we set the big dome over the well was that we had problems before we had it operational. We will bring it in operational. When we place it over the well it will already be connected back to a drill ship we call Enterprise that will be handling the oil and gas. It will be trickier for us operating the smaller dome, but we will make that work.'
However, much as Wells explained, theirs was a multi-pronged approach and before the small dome - dubbed a top hat - could be installed they attempted another option - a riser tool insert that was placed in the broken riser pipe that, at least initially, appears to have been a success.
At midnight on 15 May ROVs inserted the tool into the riser and, after an initial set-back, the scheme is moving ahead slowly. It is certainly not the 'eureka' moment that the BP staff are looking for, but it is certainly a step in the right direction. 'We were able to collect oil for about four hours and we were just starting to get oil to the surface when the tool became dislodged from the riser,' Wells explains. 'We were able to displace the fluid that was in there, which is important because we were always concerned that if this tool became dislodged it would hydrate over. All the modifications we have made have allowed us to avoid that.
'Subsequently we have reinserted the tool again; it is working as planned and we are very slowly increasing the rate at which the oil is coming through the riser tool and going to the surface. So we now have oil and gas coming to the ship, a flare is burning off the gas and the oil is going into a surge tank. We will be looking to optimise this over the next couple of days and start putting the collected fluid through the oil separator and at that stage we can calculate how much oil we are collecting.'
Sealing the bore
With the oil leaking from the riser being mitigated, focus could move back to the real task of permanently sealing the bore. 'Our number one priority has always been what we are doing to shut off the flow,' Wells says. 'We have spent a lot of time trying to shut the flow off at the BOP and we have been unsuccessful. I have been intimately involved with this and have seen what people have tried to do with ROVs and it has been absolutely stunning.'
Although almost three weeks of high-intensity effort has failed to close the BOP, BP have gleaned some important information that has given them confidence to attempt further operations on the seabed. 'We have come up with some diagnostic tools,' Wells says. 'Much like a doctor in a hospital would use X-rays, we had to find out how we could look inside and find out exactly what is going on and, as a result of some of that work, we have become more comfortable with our top-kill techniques.'
The first task was to disconnect and remove the control pods that sit on the outside of the BOP, take these to the surface and refurbish and reconfigure the electronics within them. 'We now have them ready to put back in place and reconnect so that it can operate the choke and kill lines,' Kent adds. 'We are not looking to try and operate the rams in the BOP, it's about what we are doing for the choke and kill lines.'
The choke and kill lines - one runs up either side of the BOP and before the rig sunk were connected back to the surface - were disconnected from the BOP, and a new pipe was connected, via a manifold on the seabed back to the Enterprise. 'With that ability to pump fluid down there and the ability to control the valves on it we believe that we can successfully complete a junk shot and what we call a 'top kill',' he continues.
'The well is currently flowing through the BOP; what we are looking to do is pump stuff into it that will plug it up. There is some science to this, I know you have heard people say, 'my God they are pumping golf balls and tyres into it' but they are right, because you [need] pieces of different sizes that will wedge into there and hold.
'We have been testing some different recipes. The intention is that from this manifold we will inject the junk shot and it will go into the BOP, ideally it is going to plug as soon as we go in, but we can give this several shots. If the first one doesn't work we can change the recipe and try a different one. If we get that to shut off we have the ability, which will follow immediately with what we call 'kill mud'. This mud will be pumped down the well bore, because it can no longer go up. We will have huge pumping facilities at the surface - 30,000 hp - to pump this down the well and through that increase of hydrostatic head it will push the oil and gas back into the reservoir and the well will be killed. As soon as we've done that we will follow it with cement and we will permanently abandon this well.'
The final attempt will be made at the end of May, and if the bore is successfully plugged then BP can move on to the clean-up and the industry itself can decide whether it has the tools available to safely continue deepwater drilling.
'The implications for the industry will only become apparent when the results of the investigation are known,' Hayward says. 'But I suspect that there will be significant implications in terms of BOPs, regulations, redundancy of equipment and safety. I would also expect there to be some significant findings in response to subsea intervention, our ability to intervene at depths of 5,000ft.'
Both BP and the oil industry at large will be anxiously waiting for the fall-out, but whatever the findings the future shape of deepwater drilling will never be the same.