Zoë pauses during her search for life in the Atacama desert.©Carnegie Mellon University

An autonomous robot has found life in one of the most lifeless places on Earth: the Atacama desert in northern Chile, thought to be a close analogue of Mars's arid surface.

"Our life detection system worked very well, and something like it may ultimately enable robots to look for life on Mars," says Alan Waggoner, one of the expedition team members from Carnegie Mellon University in Pittsburgh, Pennsylvania. The team say this is the first time an autonomous robot has identified life in the Atacama.

The four-wheeled droid, called Zoë, found colonies of bacteria and lichens in two different parts of the desert, which has the least amount of organic material anywhere in the world.

Scientists back in Pittsburgh sent commands to guide Zoë's exploration each day, but she relied on her own cameras and internal sensors to navigate the tough terrain. As she looked for signs of life, fellow researchers in the desert followed to check her results. "There is not a single example of the rover giving a false positive," says Edwin Minkley, a biologist on the Carnegie Mellon team.

The scientists' expedition to Atacama in September and October 2004 was part of NASA's astrobiology programme. They presented their results this week at the 36th Lunar and Planetary Science Conference in Houston, Texas.

Dye for life

Zoë looks for life by detecting fluorescence from biological molecules such as chlorophyll. She can also spray dyes on to the ground that light up when they bind to chemicals such as the nucleic acids found in DNA, or the amino acids in proteins. The robot carries a camera on her underside to take pictures of the finds.

Zoë can cope with steep slopes and rugged terrain.© Carnegie Mellon University

Although fluorescence sensing is standard technology in the lab, sunlight easily disrupts the measurements. This is a problem for any martian rover that relies on solar power, as it can only explore in the daytime. So Zoë's dyes are probed with high intensity flashes of light and her camera opens only at the precise moment of fluorescence.

"Our fluorescent imager is the first such system to work in the daylight while in the shade of the rover," says Gregory Fisher, an imaging scientist on the project. Even better, the samples do not have to be scooped into an analysis chamber, and this speeds up Zoë's experiments. "Other testing methods require considerably more sampling or are less sensitive," Fisher says. The rover can spot patches of just a few thousand bacteria, adds Waggoner.

To the limit

Atacama is a popular test area for martian life probes, because intense ultraviolet light and strong oxidants in the soil quickly break down organic molecules, just as on Mars. The signs of life that Zoë found actually lie below the detection limits of NASA's two Viking landers that arrived on Mars in 1976, but several teams are extending such limits.

For example, Richard Mathies, a chemist at the University of California, Berkeley, has recently tested Atacama soil in the laboratory using his own team's amino-acid detection system (see 'Life Chip ready for 2009 Mars mission' ). Their 'life chip' could sense concentrations of amino acids down to ten parts per billion, more than 1,000 times the sensitivity of the Vikings' system¹.

Zoë has certainly proved successful, says Alison Skelley, one of Mathies's team. But she points out that some of the sites the robot visited have visible lichens on rocks. "In the sites we visited these obvious signs of life were not present," she says, making them a much more challenging proving ground for their instrument.

Zoë returns to Atacama in August for her toughest assignment yet: to autonomously cover 50 kilometres over two months, sampling the soil as she goes. 

Carnegie Mellon University, Pittsburgh

• References

  1. Skelley A. M., et al. Proc. Natl. Acad. Sci. USA , 102. 1041 - 1046 (2005). | Article | PubMed | ChemPort |