By Dr David Whitehouse
BBC News Online science editor
An unmanned spacecraft should test ways to deflect a threatening asteroid, two astronauts have told the US government.
How would we deflect it?
Rusty Schweickart and Edward Lu said a mission of this type could be launched to an asteroid in 2015.
In February, Earth was almost placed on alert because of an asteroid then thought to be on an impact course.
Mr Schweickart told a hearing that "the media and the general public realise that asteroids are of more than passing interest".
More than nuclear
Testifying before an investigation into the threat from asteroids to the Earth, Apollo astronaut Russell Schweickart called for a new mission to develop the technologies needed to protect the Earth.
"More and more people are coming to know that some few of these asteroids do not silently pass the Earth, but indeed crash in, largely unannounced.
"On the rare occasions when this happens they can wreak havoc of a magnitude unprecedented in human history."
Schweickart: "Address the threat"
He pointed out that even the small, most frequent events are more powerful than the blast from the most powerful nuclear weapon in the current US nuclear arsenal.
"A known threat that can potentially destroy millions of lives and can be predicted to occur ahead of time, and prevented, cannot responsibly go unaddressed," he said.
Within existing projects
Scientists realise that they lack much fundamental information about asteroids that would hamper them if they ever needed to contemplate nudging one off an Earth-impact course.
To remedy this, Schweickart said the US should "adopt the goal of altering the orbit of an asteroid, in a controlled manner, by 2015".
He added that, in his view, such a mission would not require the development of additional new technologies.
"The key capabilities required are already "in the pipeline" of the existing Prometheus Program. No new Nasa money is required, nor is a change in Nasa's mission called for."
Astronaut Edward Lu, recently back from the International Space Station, backed the audacious plan.
"The first attempt to deflect an asteroid should not be when it counts for real, because there are no doubt many surprises in store as we learn how to manipulate asteroids," he said.
He added that the demonstration asteroid should be large enough to represent a real risk, and the technology used should be scaleable in the future to larger asteroids.
"We are suggesting picking an asteroid of about 200 metres. A 200-metre asteroid is capable of penetrating the atmosphere and striking the ground with an energy of 600 megatons. Should it land in the ocean (as is likely), it will create an enormous tsunami that could destroy coastal cities."
He said that with several decades' warning, the orbital velocity of an incoming asteroid would only need to be altered by a small amount - less than the order of one cm per second - to make it miss the Earth.
"However, this is still a very difficult task since the mass of a 200-metre asteroid is of the order of 10 million tonnes."
Lu agreed that the mission went well beyond the capability of conventional chemically powered spacecraft, meaning that it would require a nuclear powered spacecraft using high-efficiency propulsion (ion or plasma engines).
Edward Lu says the best way to learn is to go there
"Such propulsion packages are currently already under development at Nasa as part of the Prometheus Project. In fact, the power and thrust requirements are very similar to the Jupiter Icy Moons Orbiter spacecraft, currently planned for launch around 2012.
"The best way to learn about asteroids is to go there," he told the US Senate Science Committee.