[meteorite-list] Supercomputer Takes on a Cosmic Threat from Asteroid Impact

From: Ron Baalke <baalke_at_meteoritecentral.com>
Date: Wed Jun 14 17:51:48 2006
Message-ID: <200606142149.OAA01295_at_zagami.jpl.nasa.gov>

http://www.space.com/businesstechnology/060614_asteroid_computer.html

Supercomputer Takes on a Cosmic Threat
By Leonard David
space.com
14 June 2006

A super-powerful computer has simulated what it might take to keep Earth
safe from a menacing asteroid.

Researchers have utilized the number-crunching brainpower of Red Storm - a
supercomputer at Sandia National Laboratories in Albuquerque, New
Mexico. Red Storm, a Cray XT3 supercomputer, is the first computer to
surpass the 1 terabyte-per-second performance mark - a measure that
indicates the capacity of a network of processors to communicate with
each other when dealing with the most complex situations - in both
classified and unclassified realms.

The massively parallel computing simulations have modeled how much
explosive power it would take to destroy or sidetrack an asteroid that's
got Earth in its cross-hairs.

Radar measurements

For the computer runs, asteroid 6489 Golevka was chosen. Golevka isn't
going to hit the Earth, explained Mark Boslough, a Sandia scientist and
asteroid threat analyst. This particular asteroid was used as a "proxy"
because solid geometry data about the object existed, he said.

Since its discovery in May 1991 by astronomer Eleanor Helin, asteroid
Golevka has been repeatedly radar-scanned. It is roughly .33 mile (one
half-kilometer) across, but tips the scales at about 460 billion pounds
(210 billion kilograms), according to asteroid experts at the Jet
Propulsion Laboratory in Pasadena, California.

The Golevka asteroid has been a particular object of interest since
2003. That's when NASA scientists discovered its course had changed.

Keeping tabs on Golevka has helped pin down the Yarkovsky Effect - a
miniscule amount of force produced as the asteroid absorbs energy from
the Sun and re-radiates it into space as heat. Over time - lots of it -
that force can have a big effect on an asteroid's orbit.

Deflection and disruption

Boslough said the actual geometry from radar measurements of asteroid
Golevka were used in the computer simulations.

"Of course we don't know the internal structure so we had to assume
something," Boslough said. He and his colleagues tried both
heterogeneous and homogenous simulations, but selected the uniform
strength and density for the high-resolution demonstration mainly for
simplicity.

The researchers applied the Keep It Simple Stupid (KISS) principle of
avoiding unnecessary complications...don't try the hardest thing first,
Boslough added.

In general terms, several findings stood out in Red Storm computations
that might be useful for future planetary defense systems.

Boslough first noted that there are two "end-member strategies" in the
Golevka work:

    * Deflection: Keeping the asteroid in one piece and changing its
      trajectory to miss the Earth; and

    * Disruption: Blowing it to smithereens and making sure all the bits
      miss the Earth.

"There are a range of in-between options," Boslough told SPACE.com, "but
the deflection end of the spectrum is much more realistic." On a
kiloton-per-kiloton basis, small, shallow explosions are much more
effective for moving the asteroid than large, deep ones.

Bruce Willis: encore coring

One demonstration simulation - 10 megatons at the center of mass of the
object - is the most spectacular "end member" of the range that the
research team explored - but is also the least likely scenario, Boslough
explained. "It also neglects a fundamental problem of how you would get
the device inside an asteroid."

Unlike Bruce Willis and his team drilling into the core of an asteroid
in the 1998 movie Armageddon and planting a nuclear bomb, that scenario
just doesn't seem likely, Boslough said.

Playing out Golevka's hypothetical demise even on a super-fast computer
took longer than the movie. Sandia's half-second, billion-cell
simulation of a 10-megaton explosion at Golevka's center took 12 hours
to run on 7,200 processors of Red Storm.

The supercomputer is a product of a partnership between Cray Computers
Inc. and Sandia National Laboratories, developed for the Advanced
Simulation and Computing program of the Department of Energy's National
Nuclear Security Administration (NNSA) laboratory.

Low-yield, high payoff

The Red Storm computational output provided useful insights.

In particular, Boslough said, was the realization that using multiple,
low-yield, deflecting explosions is much better than using one
high-yield device.

"There are many advantages" to this approach, Boslough observed. "For
one, you don't need to rendezvous with the asteroid and drill a hole or
otherwise place a devise. You can set it off as a surface
burst. Contrast the time it takes to 'land' something on the surface of
an asteroid - like NASA's Near Earth Asteroid Rendezvous (NEAR)
spacecraft - to how long it takes just to get there, like NASA's Deep
Impact," he said.

You want to solve the problem quickly, Boslough said, "even if we know
about an impact decades in advance - the public perception will be that
time is of the essence."

If asteroid deflection is the game plan, there's need to avoid
accidental breakage.

A low yield blast lessens the volume of material that is subjected to
the highest tensile and shear stress, reducing the likelihood that the
object will come apart.

"If you do break the asteroid, you want to make sure none of the big
pieces hit the Earth," Boslough said. "Multiple low-yield bursts over an
entire hemisphere [of the asteroid] would reduce the likelihood that
anything big would get left behind on the impact trajectory."

Backup plan

The fact that you can get a low-yield device to a menacing object fast
also means that you are more likely to have a second chance, Boslough
noted. That equates to a viable "backup plan", he added, for other, more
elaborate, expensive, and time-consuming methods.

"When you are saving the Earth, it's good to have a plan B. I suspect
that if a Near Earth Object (NEO) were confirmed to be on an impact
trajectory, public opinion would demand fast action and this would
become plan A, if it wasn't already," Boslough said.

Boslough said that follow-on work regarding defending Earth from NEOs is
slated. Specifically on tap is delving into momentum transfer for a
variety of assumed asteroidal and cometary materials and structures.
Received on Wed 14 Jun 2006 05:49:31 PM PDT


Help support this free mailing list:



StumbleUpon
del.icio.us
reddit
Yahoo MyWeb