A new technique for finding wet exoplanets got a field test when astronomers pretended to be aliens.
"If you were on another planet, you’d look at Earth and say, 'That looks like the most interesting planet around that star,'" said Nicolas Cowan, a grad student at the University of Washington and lead author of the study. "Any critter with half a brain can look at Earth and say, 'That’s the one that looks different.' The question is how to quantify what it is that makes it look interesting."
Astronomers used a telescope aboard the Deep Impact spacecraft — which crashed a probe into a comet in 2005 and is on its way to another — to stare at Earth for two separate 24-hour periods. They tracked the changes in light and color that crossed the Earth's surface as it rotated, and connected them back to continents and oceans. The results will be published in the August issue of Astrophysical Journal.
Though the spacecraft was only 30 million miles away from Earth, light years closer than the nearest extrasolar planet, it was far enough to blur out the distinctive features of the Earth's surface.
"It’s like watching a movie if you had really poor eyesight," said co-author Eric Agol of the University of Washington. "You’ll see the screen getting lighter and darker, you might see different colors, but it wouldn’t give you very much information about what’s going on on the screen."
Once the team had data on the color variations of the blurry dot's surface, they used a mathematical analysis to pick out which colors were the most important.
"The technique is largely used for pattern recognition," Cowan said. "It's trying to fake human intuition for a computer."
They found that some areas of the Earth are reflective at long, or red, wavelengths, and others are reflective at short, or blue, wavelengths. When Cowan and his colleagues mapped the red and blue zones and compared them to a map of the Earth, the red areas corresponded to continents and the blue ones lined up with oceans.
The same technique could pick out oceans and continents on other blurry dots orbiting other stars.
"We’d like to be able to tell when we’re looking at an extrasolar system, are we looking at something potentially habitable and even more interestingly, something inhabited," said co-author Timothy Livengood of NASA Goddard Space Flight Center. "So we look back at our one example, ourselves, to see if our models are any good."
"If you wanted to do that by hand it would’ve been cheating, because we know that Earth has oceans," Cowan said. "We were trying to see if the data could support that without imposing our own ideas about what the Earth actually looks like."
The technique has a number of limitations. Because it averages colors over a north-south slice of the planet as it rotates, it can only detect changes from east to west. This is great for telling Africa from the Atlantic, but would miss things like lakes at the poles or planets made entirely of oceans. Earth is the only planet we know of with distinct oceans and continents, and we don't fully understand their origins. Extrasolar planets may have to be remarkably Earth-like for this method to find them.
"People run into this problem a lot in astrobiology," Cowan said. "We’re looking for a planet that has life on it, and because Earth is the only one we know of, we end up looking for planets just like Earth. That’s probably a little narrow-minded. We’re being conservative, saying 'Let’s start with the one example that we know.'"
Still, the technique is a much quicker way to get a broad idea of what a planet looks like than other standard methods. Taking a full spectrum of the planet and looking for "fingerprints" of individual molecules can take months. This method takes one of the target planet's days. Livengood said the technique could be used to quickly pick out planets to observe in greater detail, making missions more efficient.
"If we can minimize the amount of time spent observing, we’ve just made it a much more practical mission," he said.
Knowing the technique works could help shape the next generation of telescopes, and perhaps even speed up the search for other Earths.
"People assumed that the cool science was only going to happen when we had really enormous spectrographs up in space to image the atmospheres," Cowan said. "Turns out we can leapfrog that."
"It’s very interesting and promising, but it’s a step in a very complex and challenging goal," said astronomer Eric Ford at the University of Florida. "It’s another step along the way to figure out how we’ll be able to study these planets and what we’ll be able to learn and what types of observations we need to make to learn it. I’m sure it will motivate further work."
Images: NASA, Nicolas B. Cowan, and The EPOXI Team
See Also:
