GUY RAZ, host:
We're back with ALL THINGS CONSIDERED from NPR News. I'm Guy Raz.
As the oil continues to gush into the Gulf of Mexico, BP engineers are scrambling once again to find a way to stop it. This time, they'll try to put a cap on the blowout preventer where that oil and gas are shooting out. A pipe from that cap would then divert oil to a ship.
Meanwhile, BP is still drilling two relief wells that they hope might be able to kill the spill deep underwater, but that could take up to two months to complete.
NPR's science correspondent Christopher Joyce is one of our reporters following the engineering side of the story.
CHRISTOPHER JOYCE: Hi, Guy. How are you doing?
RAZ: So BP started out trying to cap this gusher with a containment dome, that didn't work. And then they stuck up a pipe into the well, that didn't work. And obviously, pumping - drilling mud into it didn't work either. How is this cap going to do better?
JOYCE: Well, hopefully, they're learning from their mistakes because, obviously, the big containment failed because of unforeseen circumstances. They got caught up with crystals. This one is going to fit much more tightly. It fits right over the top of the so-called blowout preventer.
If you (unintelligible) bottled beer or anything, you know that - you take a bottle cap and you squeeze them on the top of it and you can pretty much seal it off. This won't be quite that type, but it's the same sort of idea. And hopefully, that'll divert enough oil so that they can then buy time. If that doesn't work, then the next option down the line is to put yet another blowout preventer on top of that and do the job that the first blowout preventer didn't do. I mean, it's Rube Goldberg (unintelligible).
RAZ: Chris, I want to ask you about those relief wells. These are being held out as a kind of last resort, right?
JOYCE: That's the eventual way to kill this whole thing. I mean, all of these things are stop-gap until they really kill it.
RAZ: But I mean, they have to hit what is essentially a seven-inch target a mile under the ocean. Explain how that works.
JOYCE: Finding the needle, the one needle...
JOYCE: ...at the bottom of the haystack. Actually, I'm told by petroleum engineers it's not quite as horrifically difficult as it sounds. In the last dozen or 20 years or so, they've developed something called 3D seismic imaging with vast computer power. They can image in three dimensions miles and miles of area underground. You look (unintelligible) computer, it looks like a big box, a big cube. And they know right - pretty much where the pipe is. They know all the sandstone layers and all that. So they're drilling down - the drill beat is basically an intelligent object. It knows where it is at all times in the X, Y, Z axis, like a 3D compass.
So you navigate them. And it's kind of like navigating a boat. If you know where you started from, the dock, you know how fast you're going and how long you've been going, you can pretty much know where you are at any one time. So they get close. Then they - what they do is they create a magnetic field out from that drill beat. The sandstone is invisible to the magnetic field. Hah, but what shows up, pipe. So they know with that magnetic field exactly how to hit that pipe.
RAZ: And that visibly sucks the oil into the well.
JOYCE: Well, they got to drill a hole in that pipe...
JOYCE: ...and know - what they'll do is they'll now do the same sort of forcing mud and cement into the original well bore. But again, this is really tricky. They have to hit it at the right angle. If they come in from the side, they're not likely to get enough pressure to force the mud and the cement to block up that hole.
RAZ: Could they conceivably make a mistake that would cause the spill to become even worse?
JOYCE: Well, people don't know a lot about that well bore, the original well bore. It's supposed to have cement casing around it. There's a metal casing and cement around it (unintelligible) the cement set properly. If not, you could cut into the well bore, but then the oil and gas could then divert up into what should have been the cement casing. So it's really hard to tell.
RAZ: That's our Christopher Joyce from NPR science desk.
JOYCE: My pleasure.
NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.