Discuss as:

How safe is the shuttle?

The shuttle Endeavour stands on Launch Pad 39A at NASA's Kennedy Space Center
in Florida, in preparation for its Nov. 14 launch to the international space station.

Both of this year's presidential candidates - Barack Obama as well as John McCain - have called on NASA to look into the idea of flying the space shuttle fleet past its scheduled 2010 retirement date.

Now the space agency is providing some sobering estimates of the costs and the risks that would be involved - leading one seasoned space observer to wonder whether the shuttle program should be throttled back rather than extended.

NBC News' Jay Barbree, who has spent 50 years as a space correspondent, was struck by NASA Administrator Michael Griffin's guest column in Florida Today this weekend, headlined "Time to Retire the Shuttles."

In the wake of the 2003 Columbia tragedy, investigators warned that the inherent risks of the shuttle design - including the lack of a reliable escape system and the vulnerability to flying debris - were so great that the fleet had to be retired "as soon as possible."

Citing that warning, Griffin said that the shuttles "should be retired after fulfilling our commitments to our partners from Canada, Europe, Japan and Russia by completing assembly of the international space station." For the five years or so between the fleet's retirement and the debut of the shuttle's successor, NASA would have to rely on Russian spaceships to send crews to the station and back. That situation may be "unseemly," but it's less risky and less expensive than the alternative, Griffin said.

How risky would it be to keep flying the shuttle? In an e-mail, Barbree said Griffin's answer gave him pause:

"Griffin warns, 'With knowledge gained since the loss of Columbia, we estimate there is a one in 80 chance of losing a crew during any single shuttle launch.'

"The NASA administrator goes on to write, 'If we were to conduct 10 additional launches prior to retiring the shuttle, there would be about a one in 8 chance another crew would be lost. These are sobering odds - one reason why the Columbia Accident Investigation Board recommended the shuttle be replaced as soon as possible.'

"Whoa, Nellie!

"If I read this correctly, NASA feels the odds are a one-in-8 chance the agency will lose a shuttle during the 10 flights to be flown before the shuttle is to be grounded Sept. 30, 2010. If this is true, why is NASA flying these missions, even if the USA has contractual obligations with international partners to complete the construction of the international space station?

"On the drawing boards are America's next-generation rockets and spaceships, named Ares and Orion. Experts say they will be 30 times safer - one loss in 2,100 flights.

"Should NASA ground the worn and dangerous space shuttles now, and move ahead with the much safer rockets Ares and the spaceship Orion?

"A sobering question."

Does the math make sense? The 1-in-8 statistic is roughly correct, assuming that each launch truly does pose a 1-in-80 risk of catastrophe. If you make that assumption, the precise figure would be an 11.8 percent risk of encountering at least one fatal mission in any set of 10. That's somewhere between 1-in-8 and 1-in-9. Obviously, the more flights you consider, the higher the risk of suffering a catastrophe at least once. For example, the risk rises to 72 percent if you launch 100 more flights.

Combining probabilities in that way doesn't change the risk for any individual flight. The next shuttle flight, scheduled for launch on Nov. 14, would carry a 1-in-80 risk, as would the flight after that, and the flight after that, and the flight after that. (It's the same with dice: You have a 1-in-36 chance of rolling snake-eyes for each individual throw, but a roughly 1-in-4 chance of rolling snake eyes at least once in 10 throws.) 

The question NASA has to ask itself is whether each mission between now and the fleet's retirement is worth a 1-in-80 risk of losing the crew. Griffin is saying it's worth that risk for finishing the space station (and fixing the Hubble Space Telescope). But in his view, it's not worth that risk for transporting crew members back and forth - particularly if there's a Russian alternative.

NASA is required by law to lay out the risks and the costs associated with extending shuttle operations by as little as a year, or as much as six years. Griffin's guest column in Florida Today shed light on the risk question, and shuttle program manager John Shannon provided further information on the cost question today.

The Orlando Sentinel quotes Shannon as saying that the price tag would be at least $2 billion extra per year - a burden that he said would be "disastrous" unless Congress boosted NASA's budget by that amount. What's more, keeping the shuttle program running while working on the Orion-Ares system would jam up operations at NASA's Kennedy Space Center in Florida as well as the Michoud Assembly Facility in Louisiana, Shannon said.

Bottom line: Extending the shuttle program could be messier than the candidates think. There might be enough leeway to fly one more mission to deliver a $1.5 billion particle-physics experiment known as the Alpha Magnetic Spectrometer to the space station. But if NASA's current management has its way, there'd still be four years or so when the space agency is dependent on other people's rockets.

If you were Obama or McCain, what would you do with this information? Would you want to keep the shuttles in business for a couple of years longer, go with NASA's retirement plan, or pull the plug early? Feel free to add your comments below.

I'll be helping out with msnbc.com's Election Day coverage, so the next Cosmic Log posting will be on Wednesday - and it will likely be related to the election. Don't forget to vote!