The relief wells now being drilled are often portrayed as the last, best hope for permanently capping BP's underwater oil gusher. So the question is, are they 100 percent effective? The answer seems to vary according to which source you consult.
For instance, note the conditionality in a Discover magazine article quoting Jerry Milgram, a Massachusetts Institute of Technology professor the article describes as one of the world's leading experts in marine engineering.
"In this case it's kind of hard," said Milgram. "You are drilling through mostly rock. It's not going to be fast."
When the same oil-bearing rock is eventually reached by the relief well drill rig, the sealing of the broken well can finally begin.
First, seawater is pumped into the rock through the relief well. If all goes according to plan, that water should make its way into the lower end of the leaking well, displacing oil. If that succeeds, the next step is to pump in a mineral mud, which follows the sea water up the broken well. Once that mud fills the well, concrete can be pumped into the relief well.
"If you get a tall tower of mud, and if you're lucky, you'll stop it," said Milgram. "The next day you pump concrete and then the well is dead."
The mud and concrete work by their by density and shear weight, which is enough to counter the highly pressurized oil moving up the well.
There are a lot of "ifs" in that passage. Clearly, many pieces will need to come together if the relief wells are to work.
Last year, a blowout at an off-shore well in Australian waters was eventually stopped with a relief well. But that effort hit snags, as the New York Times reported earlier this month.
HONG KONG -- While BP tries various short-term efforts to plug a leaking oil well in the Gulf of Mexico, the company is preparing to drill a relief well as a backup plan. BP hopes to drill that well diagonally to intersect the original one below the seabed and then flood it with mud and concrete to stop the uncontrolled flow.
Although the idea sounds simple, the experience with a similar spill last year near Australia shows just how difficult it can be to execute the maneuver.
"It's like finding a needle in a haystack," said Rachel Siewert, an Australian senator who is a member of the country's opposition Greens Party and is critical of the oil industry.
The Australian accident, known as the Montara spill, began Aug. 21 with a blowout of high-pressure oil similar to the one in the gulf. With the well spewing 17,000 to 85,000 gallons per day, precious weeks passed before the relief wells were started. When efforts got under way, the first four attempts -- drilled on Oct. 6, 13, 17 and 24 -- missed the original well.
A fifth attempt finally intersected the original on Nov. 1, and about 3,400 barrels of heavy mud were pumped through the relief well into the base of the original well. The spewing oil finally stopped Nov. 3 -- more than 10 weeks after the original explosion.
The "Blowout and Well Control Handbook" by Robert D. Grace, published in 2003, makes a relief-well operation seem like a fairly straightforward project.
"... Relief well technology has advanced in the past 10 years to the point where a relief well is now a viable alternative. Modern technology has made intercepting the blowout a certainty and controlling the blowout from the relief well a predictable engineering event."
Given what we've seen in recent weeks, it's an open question as to whether there will be any "predictable engineering" events related to the Deepwater Horizon disaster.