A Trip to Mars, by Way of Chile
WITH HIS DUSTY BOOTS planted firmly in pinkish rubble, David Wettergreen gazes out across the vast, brutal wasteland and tries to imagine that he is an astronaut landing on Mars.
It is not so far-fetched. This desiccated plateau stretching across 40,000 square miles in northern Chile is the closest thing on earth to the surface of the Red Planet. Reddish sand littered with volcanic rock stretches endlessly into the distance. Barren mountains shrouded in ice caps poke up against a frigid cobalt sky. The air is so dry it hurts to breathe.
Mr. Wettergreen’s partner doesn’t mind, though. Zo?« is short, wide, and a bit dirty, but can tackle conditions that cause even the most intrepid explorers to cower. That is, as long as Zo?«’s batteries don’t run too low.
Zo?« is a 400-pound robot with four wheels and a set of cameras that serve as eyes. Mr. Wettergreen, a robotics engineer at Carnegie Mellon University, in Pittsburgh, has carted the rover to this place to take advantage of the site’s otherworldly attributes.
Like Mars, the Atacama Desert has no visible surface water, with parts of the desert receiving no rainfall for decades. The Atacama’s dry atmosphere and its high altitude, ranging from 2,200 to 12,500 feet above sea level, leave it vulnerable to a barrage of ultraviolet rays. The desert soils are also particularly high in oxidants, which rapidly break down organic material.
Such a lethal combination has led some scientists to describe the Atacama as the most lifeless place on earth. (www.justyou.co.uk)
Mr. Wettergreen wants to test that claim. He is leading a three-year project that seeks to demonstrate that the desert is a patchwork of varied ecosystems, some more barren than others, which provide important clues about life’s capacity to adapt in the harshest of environments.
“We’re saying that the Atacama is not this lifeless place, but rather there are different areas where an organism can survive,” he says. He is surveying a satellite map of the current research area, a particularly arid patch of desert about 130 miles southeast of the coastal mining town of Antofagasta.
Zo?« is here to do the dirty work for him.
The project, which is backed by a $3.9-million grant from NASA, uses the autonomous rover, which is similar to those currently exploring Mars, to probe for microscopic organisms in vast stretches of desert. In the process, the researchers are developing technology that may be used to test for life on Mars.
“Given the amazing conditions at which things survive on earth ??” deep under ice, inside mines,” says Mr. Wettergreen, “who knows where things could hang on?”
MATCHING CONDITIONS
Mr. Wettergreen is a rugged 40-year-old veteran of expeditions to some of the world’s most remote places. As an associate research professor at the university’s Field Robotics Center, he has sent a rover to sample microorganisms living in the Antarctic ice sheet. His research has also taken him to Mount Spur, an active volcano in Alaska, and to the Arctic Circle in Canada.
While other scientists have looked for life in the Atacama, Mr. Wettergreen’s team is the first to do so using a rover ??” a method particularly relevant to future missions to other planets. The robot is designed to navigate rough terrain and to traverse long distances, while conducting complicated experiments.
“When you get into these environments where, like on Mars, you are searching for a needle in the haystack, you need to be able to go a long way to search for the needle,” says Kim Warren-Rhodes, a NASA scientist who is the lead ecologist on the project. She is a member of a science team, based in Pittsburgh, which directs the rover’s operations remotely. The group is headed by Nathalie A. Cabrol, an astrobiologist working at NASA’s Ames Research Center and the SETI Institute, a nonprofit research center searching for life in outer space.
Each morning the science team sends detailed instructions by satellite to the rover ??” named for the Greek word for “life” ??” after reviewing the data collected the previous day. The method is modeled on the system used to communicate with the pair of robots that have been rolling around on Mars since January 2004 as part of the Mars Exploration Rover Mission.
The Atacama team tries to Emit itself to the constraints of the Mars mission, in which scientists are currently communicating with the rovers across a distance of 43 million miles. “The focus on Mars is very heavy,” says Shmuel Weinstein, a research biologist at Carnegie Mellon who is also part of the team. For example, he says, Zo?« is programmed to send a maximum of 150 megabytes of data each day ??” roughly the amount transmitted by the Mars rovers ??” although it is capable of sending hundreds of times as much.
The university has a long track record of developing technology linked to planetary exploration. Part of the computer software that enables the current Mars rovers to navigate autonomously and to view the landscape using stereo vision was developed at Carnegie Mellon’s Robotics Institute. The university also serves as a training ground for future NASA engineers; many of the members of the current Mars mission are Carnegie Mellon graduates.
“The important thing in being exposed to robotics at CMU is that you get to work on real systems, not just toy things in a lab,” says Chris Leger, a rover driver on the current Mars mission who earned his Ph.D. in robotics from Carnegie Mellon in 1999. “That kind of practical exposure is good for developing robots for space.”
Raymond E. Arvidson, a geologist who is deputy principal investigator on the Mars mission, agrees.
“These guys are the advance guard. They are carving the technology envelope for us,” says Mr. Arvidson, who directs the Earth and Planetary Remote Sensing Laboratory at Washington University in St. Louis. “Mars would be a much more difficult target to find and identify life than in the Atacama Desert,” he cautions, “but that’s OK because they’re prototyping the approaches.”
TESTING NEW GEAR
The Atacama project is also pioneering the use of an onboard fluorescence imager, an instrument that uses fluorescent light and special dyes to detect signs of life at the microscopic level. The instrument, which was developed at the university’s Molecular Biosensor and Imaging Center with a $900,000 grant from NASA, allows researchers to detect the likely presence of organisms. A ground team then collects samples and sends them back to a laboratory in Pittsburgh for confirmation.
During the project’s second field season in 2004, the rover’s fluorescence imager detected lichens and microscopic bacteria at two sites in the desert, according to team members. And testing at three separate sites this year has supported those findings, they say.
“The samples we saw light up ??” every one of those did show bacterial growth at some level,” says Mr. Weinstein, the research biologist, who was involved in developing the onboard fluorescence imager. “That says bacteria was there.”
The instrument, which is located on the underside of the rover, sprays the ground with special dyes that bind to nucleic acids, proteins, lipids, or carbohydrates ??” the building blocks of life. The instrument detects their presence by shining a florescent light on the ground, causing any of the biological molecules to glow in a distinctive color. The same instrument looks for chlorophyll, a key molecule in plants, lichens, and some bacteria.
The real test takes place back in the lab, however, where scientists view ground samples under a microscope and test them for DNA. Only then can they be sure that what they are seeing is actually evidence of life.
“In the Atacama, where it is very dry, or on Mars, you’re getting some weird soil chemistry that we’re not used to seeing,” says Ms. Warren-Rhodes, the project’s chief ecologist. She says the technology is still in the testing phase, but that its ability to detect life in the field is “a big improvement” over conventional methods, which involve collecting samples at random and then testing them in a lab.
Indeed, it was the failure of such conventional methods to detect life in the driest parts of the Atacama that led to current project. In the late 1990s, Chris McKay, a NASA astrobiologist, published reports suggesting that parts of the Atacama were utterly lifeless. Chilean researchers challenged those results, arguing that if you probed on a small-enough scale, you could find living organisms.
“That’s what got us excited about the desert,” says Mr. Wettergreen. He notes that the Viking landers performed some simple tests on Martian soil in 1976 and found no evidence of organic matter. But those results are no longer considered definitive.
“At first it seemed that there was no metabolic activity, but maybe it was just below the level of detection,” says Mr. Wettergreen as he drives his Toyota pickup at breakneck speed across the desert.
He is scouting for a site to launch the rover for its final round of field experiments in the three-year project, which was conducted in separate, several-month seasons. In all, the scientists tested six sites that ranged from foggy coastal areas and mud flats to the desert’s bone-dry interior. At each location, the researchers alternated mechanical tests of the rover’s navigational capacity with intensive scientific experiments probing for signs of life.
A FANCY RIDE
The six-and-a-half-foot-wide rover looks like a jacked-up go-cart, with mountain-bike tires and solar panels on its back. A set of high-resolution cameras is mounted on a five-foot pole in the front, resembling an insect’s antenna. Behind its removable fiberglass sides, the rover houses a dizzying array of circuitry, navigation equipment, and scientific instruments. The design is the result of 10 years of prototype testing at Carnegie Mellon’s Robotics Institute. (A less advanced rover, Hyperion, was used during the first year of the project in 2003 before being replaced by Zo?« in 2004.)
Zo?« operates almost entirely on sunlight collected by its six-foot-wide solar panels, but it reverts to rechargeable lithium-ion batteries early and late in the day. The rover can choose alternative paths to avoid obstacles and to plot a route beyond the horizon ??” technology that would be useful in exploring other planets.
The two rovers currently on Mars, Opportunity and Spirit, can only go as far as they can see in a straight line. They can cover a maximum distance of 300 yards in a day, while Zo?« frequently travels more than 10 miles at a stretch. And when confronted with a difficult obstacle, the Mars rovers are forced to stop and wait for new commands.
“It costs a lot to keep scientists and rover engineers running a mission. So you want robots to be more autonomous so you need less people,” Mr. Wettergreen explains, halting the truck at a bowl-shaped area of desert, rimmed by sand dunes.
With future Mars expeditions in mind, he forces himself to imagine where the pilot of a mission would deploy the rover. The terrain must be varied, but without insurmountable obstacles. He also has to consider the needs of the team back in Pittsburgh, which is interested in collecting a wide array of samples to compare different desert ecosystems.
“One of the main hypotheses that we’re testing is whether mobility is important, or whether it’s better to sit in one place and analyze things to the billionth,” he says. “Our hypothesis is that by testing many microhabitats, it bears more fruit.”
He stops the truck to examine another possible launch site further from the final base camp, at the Mina Guanaco, a semi-operational gold mine where the scientists are staying in relative luxury in prefabricated miners’ cabins. The other sites were so remote that the researchers were forced to camp in the desert, going without showers for weeks.
The Atacama’s harsh climate makes for hard living conditions, particularly since the team’s field seasons have coincided with the tail end of the Chilean winter. The temperature frequently dips well below freezing. Gusts of wind kick up swirling clouds of dust. And the ultraviolet rays are so intense that the scientists have to lather themselves in sunblock every few hours.
SECRET LOCATION
By the time they arrive at the final test location, however, the daytime temperature is a balmy 50 degrees. Mr. Wettergreen and four other robotics engineers are getting ready to transport Zo?« from its “garage” at the mine to Site F, a dramatic stretch of desert flanked by snowcapped mountains. The team checks dozens of components on the rover, removes the solar panels, and hoists the rover onto the roof of Mr. Wettergreen’s truck. By midafternoon, Zo?« is safely settled in the sand and ready to take a panoramic shot of the site, for use by the science team in Pittsburgh. (The rover’s exact location is kept secret from the remote science team, which is forced to analyze the data using small bits of information ??” again, in an attempt to re-create the challenges of a mission on another planet.)
It is a rare moment of quiet for Mr. Wettergreen’s base team. The engineers spend a few hours uploading data and filing reports by satellite.
Their work is being monitored round the clock by several researchers who are conducting parallel ethnographic studies of the project. “We’re trying to understand how the science team makes sense of the rover and the data, how they think about the rover, and how that affects how they command the rover,” says Pamela J. Hinds, an associate professor of management science and engineering from Stanford University. She is working in tandem with a Carnegie Mellon graduate student, Karen Stubbs, who is studying the remote science team back in Pittsburgh.
A third researcher, Roxana Wales, is conducting a separate analysis of the project that focuses on human-robot interaction. Working with a moving robot is “a totally new process” that takes scientists some time to master, says Ms. Wales, a NASA ethnographer. She recently worked with scientists on the Mars mission to help them develop a common language among themselves for dealing with the rovers.
“Sometimes, it’s just about helping them recognize big issues because they’re so involved in the trees, they don’t see the forest,” she says, over a breakfast of egg sandwiches in the rickety trailer that serves as the camp’s dining hall.
Minutes later, the researchers set out to find Zo?«, their trucks kicking up plumes of pinkish dust as they careen across the desert.
The haste turns out to be pointless, however. The wind is gusting up to 58 miles per hour, and the scientists are forced to spend a frustrating six hours in their vehicles waiting for conditions to improve. While the rover is designed to withstand high winds, the fluorescence imager needs to be relatively still to take photos and to spray its dye on the target.
Mr. Wettergreen waits until the wind has calmed to about 30 miles per hour, then gets out of his truck and flips the start-up switch under Zo?«. The rover responds by emitting a screeching sound like a chainsaw in need of oil. It takes Chris Williams, a mechanical engineer, another half-hour to figure out the problem. He discovers that the fluorescence imager got knocked out of position during the bumpy truck ride. Finally, around 3 p.m., Zo?« gets started on her task of the day: taking fluorescence tests at several nearby sites.
LIFE ON EARTH
So far, the results have exceeded the team’s expectations, turning up evidence of life in 5 percent to 10 percent of the test sites. Zo?« has also completed this year’s goal of traveling 120 miles autonomously.
Mr. Wettergreen now plans to apply for funds to extend the project for a fourth year, to experiment with a one-meter coring drill to probe for life beneath the desert surface. While it is unusual for NASA to support a project for more than three years, Mr. Wetter-green argues that the success of the research so far merits making an exception.
“I think we’re doing good work,” he says, as he struggles to fix another problem on the rover, which has inexplicably begun driving in circles.
It is the latest in a long list of mechanical mishaps. The most serious ??” which falls into the realm of what Mr. Williams jokingly calls “Mars irrelevant” ??” occurred while transporting Zo?« between Sites D and E, when the rover was jolted so badly that it broke an axle. The fluorescence imager was also damaged, and the ground team had to send back to the United States for replacement parts, delaying their research by several days.
Mr. Wettergreen is not easily discouraged. He drops onto his back in the sand and starts dismantling the rover’s drive train. After another half-hour, he has discovered the source of the steering problem: a dead battery that controls the front wheels.
“Never a dull moment,” he says, laughing at the absurdity of so much going wrong in a single day. But he notes that one of the advantages of working on earth instead of Mars is that engineers can take apart the rover and fix it. “People ask when Zo?« is going to Mars and the answer is never,” he says, watching with relief as the rover finally speeds off straight into the desert.
Mr. Wettergreen and his team did not build the rover as a prototype, but rather as a way to explore techniques for detecting life. If all goes well, the technology they have developed may one day lift off on a trip bound for the Red Planet, where the howling winds and the desolate landscape won’t seem so foreign.