Christopher Brownfield, who I debated on NPR a few weeks ago on my pet topic — nuking the Gulf oil leak — has shown up in the New York Times arguing for a conventional detonation:

the Navy could focus on stopping the leak with a conventional demolition. This means more than simply “blowing it up”: it means drilling a hole parallel to the leaking well and lowering charges to form an explosive column.
Upon detonating several tons of explosives, a pressure wave of hundreds of thousands of pounds per square inch would spread outward in the same way that light spreads from a tubular fluorescent bulb, evenly and far. Such a sidelong explosion would implode the oil well upstream of the leak by crushing it under a layer of impermeable rock, much as stepping on a garden hose stops the stream of water.
It’s true that the primary blast of a conventional explosion is less effective underwater than on land because of the intense back-pressure that muffles the shock wave. But as a submariner who studied the detonation of torpedoes, I learned that an underwater explosion also creates rapid follow-on shockwaves. In this case, the expansion and collapse of explosive gases inside the hole would act like a hydraulic jackhammer, further pulverizing the rock.
The idea of detonating the well already has serious advocates. A few people have even called for using a nuclear device to plug the well, as the Soviet Union has done several times. But that would be overkill. Smartly placed conventional explosives could achieve the same results, and avoid setting an unacceptable international precedent for the “peaceful” use of nuclear weapons.
At best, a conventional demolition would seal the leaking well completely and permanently without damaging the oil reservoir. At worst, oil might seep through a tortuous flow-path that would complicate long-term cleanup efforts. But given the size and makeup of the geological structures between the seabed and the reservoir, it’s virtually inconceivable that an explosive could blast a bigger hole than already exists and release even more oil.
The task force could prepare for demolition without forgoing the current efforts to drill relief wells. And even if the ongoing efforts succeed and a demolition proves unnecessary, the non-nuclear option would give President Obama an ace in the hole and a clear signal that he’s in charge — not BP.

As I said at the time, I think there is very little indication that a conventional blast would work. To do it the way Brownfield suggests, by drilling a new hole parallel to the wellhead, would take longer than completing the relief wells and drawing off the pressure. Moreover, it now seems clear that the concrete casing surrounding the BP well is fatally compromised, and the upward pressure from the well is beyond anything we’ve seen before. This means that the rock surrounding the well bore could be full of fissures, and those fissures could be full of oil, or worse, methane gas. A conventional explosion might not just fail to seal the well bore, it might ignite those pockets of oil and gas, leading to a potentially catastrophic situation where we have not one big leak, but a thousand little ones.

Consider that Brownfield is talking about a detonation of “several tons” of conventional explosives. A nuclear blast, even from a small fission bomb, would be on the order of 20 or 30 thousand tons of explosives. An underground blast of that size, and the attendant million-degree temperatures, would both create the kind of rock implosion necessary to seal the well bore and vaporize incidental oil or gas in the surrounding rock.

The nuclear option brings its own risks, to be sure, but it also brings a much higher chance of success than the conventional option. But I have no hopes that it will be seriously considered. In fact, I received this e-mail, from an engineer working at a Department of Energy national lab, just yesterday:

On the nukes to plug the hole . . . if my experience in a federal lab can be extended to other government operations, nothing that risky will ever be done, even if it had a high probability of success. This is because any government led effort is extremely risk-averse. Something that had a 90% chance of success but a 5% chance of making things much worse will never be attempted. No one will go on record recommending it, because theirs is the a$$ that will be kicked if something goes awry. And in something this high-profile, it won’t just be the loss of a job, but public disgrace forever. Once the government is involved, the only solutions to be tried are the safe ones (with much smaller probabilities of success) — unless there’s leadership at the top that is willing to take any heat. That is obviously missing here.