Why did NASA ground its shuttle fleet?

Discovery is scheduled to be the first shuttle back into space in spring 2005. Credit: © NASA

NASA decided that no shuttle should fly until the exact cause of the accident was identified. The Columbia Accident Investigation Board (CAIB) was set up to oversee this process. Last year's assessment demanded major changes to shuttle engineering before they could return to flight. It also sought a sea change in NASA's approach to shuttle missions and safety assessments.

Those changes have been monitored by the Return to Flight Task Group, chaired by veteran astronauts Richard O. Covey and Thomas Stafford.

The fleet of three remaining space shuttles, Discovery, Atlantis and Endeavour, has been undergoing intensive refits for the past year.

When will space shuttles fly again?

Whenever the task group says they are safe. The group is due to publish an update on NASA's progress on the first anniversary of the CAIB report.

At present, NASA is aiming to launch Discovery in March or April 2005. The mission, designated STS-114, will be led by Eileen Collins, the only female shuttle commander.

Why did Columbia crash?

"The foam did it," says CAIB member Scott Hubbard, director of NASA's Ames Research Center in Moffett Field, California. A piece of foam insulation covering the bipod fitting, which connects the shuttle to its main fuel tank, fell off during launch, gouging a hole in the craft's left wing. On re-entry, hot air ripped at the hole, weakening the wing and ultimately destroying the craft.

How has NASA addressed the foam problem?

The foam on the bipod fitting, intended to protect it from ice build-up, has been replaced with four heaters. "This is a fix that really gets to the root of the technical problems that caused the loss of Columbia," says Michael Kostelnik, a NASA shuttle administrator.

More than 95% of the foam on the fuel tank was sprayed on by robots, but hard-to-reach places were done by hand. This caused small voids to form inside the foam, weakening its structure. Spraying procedures and safety checks have been improved, and Discovery's main fuel tank should be finished by November. "No engineering solution is ever 100%, but you've got to stop foam coming off the shuttle," says CAIB member Sheila Widnall.

Has NASA made any other modifications?

NASA engineers have installed digital cameras on the outside of the shuttle and the fuel tank to identify damage immediately. Previous missions have carried 35-millimetre film cameras, but images from these could only be studied once the shuttle had returned home. Discovery has had 88 temperature and impact sensors fitted to each wing's leading edge to monitor their health during take-off and orbit. Astronauts should be able to diagnose problems and make minor repairs while the craft is in orbit.

Discovery's wing edges and nose cone have been refitted with carbon-fibre panels that will further protect the craft from the heat of re-entry.

If NASA is focussed on the Moon and Mars, what is the role of the space station? Sheila Widnall , Massachusetts Institute of Technology, Cambridge, and Columbia Accident Investigation Board member

Safety inspections also revealed a problem with the rudder speed brake, which is a small flap used to slow down the vehicle on landing. Its motor had been installed incorrectly in every shuttle, meaning the pilots could lose control during landing. Discovery is having this motor replaced.

But the crash wasn't all down to flawed engineering, was it?

No. To fund other NASA programmes, the shuttle's budget and work-force were cut by 40% during the 1990s. Moreover, decisions about schedules, budget and safety were all made by a single group. With such competing pressures, it proved all too easy to lose the focus on safety. NASA is now instituting a separate engineering authority that will have the final say on whether a mission launches.

The CAIB report said that safety checks were often poorly managed. "The shuttle programme had become comfortable with an operational mindset that treated a developmental vehicle as an operational vehicle, accepting debris strikes as normal, and so on," says Hubbard. This culture is being challenged through increased communication between different areas of NASA, says Hubbard.

What has not been done yet?

Most of the engineering work is finished. Discovery was meant to get another sensor arm with laser systems that would detect any impacts, but engineering problems have put that on hold.

The Return to Flight Task Group's most recent report assessed different options for toughening the entire orbiting craft, such as using new, stronger tiles, but none has yet been implemented.

Will the crew stand a better chance of survival in any future incident?

It depends. There is now a better chance of spotting minor damage incurred during lift-off, but more extensive damage might require a rescue mission. NASA has pledged that the first two shuttle flights next year will also have a back-up shuttle ready to launch a salvage flight if necessary. "I think it's a fair compromise," says Hubbard. "Space exploration will never be risk-free. There's no getting around that."

How long will the shuttle fleet last?

Perhaps as long as ten years. The CAIB insisted that NASA re-certify the shuttles in 2010, which would require all shuttle components to be safety tested to the standard required for completely new vehicles.

NASA has ducked out of the recertification issue so far Sheila Widnall , Massachusetts Institute of Technology, Cambridge, and Columbia Accident Investigation Board member

Widnall says that NASA has not yet sufficiently addressed the re-certification issue. Re-certifying the shuttle will cost billions of dollars, and Widnall believes that is money that NASA will not have. "There's absolutely no way they can come up with an alternative to the shuttle by 2010," she says.

Why does the shuttle have to keep flying?

To finish building the International Space Station (ISS). Other missions such as satellite launches can be carried out using craft not manned by humans, but the ISS requires the combined crew and cargo capabilities of the shuttle.

Experts say it is highly unlikely that the ISS will be finished by 2010. "It turns out that the shuttle decision is a little more fluid than it first appears," says Sherwood Boehlert, chairman of the US government's House Committee on Science. US participation in the ISS is scheduled to end in 2016, but that date is far from firm. After that, the future of the space station looks uncertain. "If NASA is focused on the Moon and Mars, what is the role of the space station?" asks Widnall.

What will replace the shuttle?

Space exploration will never be risk free, there's no getting around that Scott Hubbard , Director of NASA Ames Research Center, California, and Columbia Accident Investigation Board member

It is likely that the space shuttle's crew and cargo functions will be separated in the next generation of space vehicles. Once completed, the ISS will have to be supplied by a craft that does not need a human crew, says Widnall.

At present, a craft is being designed to carry crew to the Moon and Mars, to meet NASA's 'Vision for space exploration', set out by President George W. Bush earlier this year. However, NASA faces a smaller budget allowance for next year. This will leave it with only about 25% of the proposed development costs for the craft and it may struggle to hit the target launch date of 2014.