Out-of-This World Data Pose Riddle for Mars Team

The Pathfinder mission to Mars treated millions of awe-struck viewers, readers and Web surfers to astonishing views of our neighboring planet. But at the same time, Pathfinder opened the door to perhaps an equally mysterious and fascinating world: behind-the-scenes science.

Science, as the late Carl Sagan pointed out, is a strange business, requiring its practitioners to be two contrary things at the same time: complete skeptics, believing nothing that isn’t proven beyond doubt, and wild dreamers, willing to risk everything on a hunch and a prayer. It is equal parts guesswork and high precision, looking for hidden patterns and inventing clever gadgets, rational Sherlock Holmes-style sleuthing and a game of Kick the Can.

Scientists need all these tools because unraveling the stories hidden with rocks and atoms and stars is rarely straightforward. They don’t always know what they’re looking at, or how exactly to interpret alien worlds through the equally alien senses of their instruments.

“What we really want to do is figure out the story,” said geologist Jeff Johnson of the U.S. Geological Survey. “When you’re getting down so much data at once,” the story can take some surprising turns.

Take Pathfinder’s camera for example--the Imager for Mars Pathfinder, or IMP. Researchers got a huge surprise when it stood up to its full 5 1/2-foot height, and suddenly, rocks and boulders looked smaller. A robotic take on “Honey, I shrunk Mars.”

“That was amazing,” said camera scientist Peter Smith.

But IMP can’t look at Mars from, say, three feet up or 12 feet up, much less get a bird’s-eye view. Deputy Project Manager Brian Muirhead, for one, would love to know what the landing site looks like from directly above. “I’d like to know how we got to where we are,” he said. “If I could look down from above, I could see where it bounced, where it spun sideways off a rock.”

*

Another thing IMP can’t do is wander around in its landscape.

“We get a lot of information by moving through a scene,” Smith said. “Without that information, you really don’t know how to interpret what you see.”

Before the July 4 landing, Smith and his colleagues practiced “seeing” on Mars for months in the copycat Martian landscape--”the sandbox”--at the Jet Propulsion Laboratory. To determine how well they could see through IMP’s eyes, the scientists would stage a Martian scene in the sandbox, shifting around rocks and terrain, sometimes sticking in tricks, like a philodendron leaf.

The camera would take pictures and the scientists would try to interpret them. But even in the sandbox, “The IMP images were never quite what I thought [they would be],” Smith said.

Other instruments “see” things human senses can’t. The Alpha Proton X-ray Spectrometer on Sojourner, for example, can “sense” the elements that make the soil and rocks. By contrast, Muirhead said, a person “can’t put his finger down on something and tell what it’s made of.”

So one of the primary problems scientists face is in translating--from human view to instrument view and back again.

Another is “noise”--an ever-present distraction in trying to extract data from the messy outside world. If a human looks at a scene, the brain automatically erases extraneous objects or features, such as the motion of its own head, distortions of size due to distance, and changing lighting conditions.

IMP doesn’t automatically account for variations in lighting at different hours of the Martian day, however. If it is trying to get the color of a rock to find out what minerals it’s made of, it could be fooled by changes in brightness.

Sorting out all these sources of “noise” on the camera could well take a year, Johnson said. “We haven’t had time to go over all the diffuse lighting conditions, to find out what causes a particular wiggle, and whether it’s a mineral or whether it’s noise.”

Sojourner’s sensitive “nose” has an even bigger noise problem. Its Alpha Proton X-ray Spectrometer (APX) determines the elements in a piece of rock by shooting subatomic particles at the rock, then keeping track of the radiation and particles that bounce back.

But how does it know which particles are coming from the sample and which are stray cosmic rays and other “noise” from its surroundings?

To sift noise from signal, the APX keeps its nose to the ground for more than 10 hours at a time, taking the same data over and over again. This is what people do when trying to hear a very faint phone message in a noisy room--play it repeatedly until repetition sorts the message from the chaos in the background.

“If you take 10 minutes [with the APX], you measure mostly noise,” Johnson said. “If you take 10 hours, you get mostly signal.”

*

The most frustrating part, Johnson said, is that each instrument “provides only a glimpse.” The X-ray spectrometer looks at the individual atoms that are the smallest components of matter. The IMP camera’s color vision provides information about minerals--huge conglomerations of atoms. But on Pathfinder, no instrument can look at molecules--the in-between level that would tell the researchers exactly what minerals Mars was made of.

That will have to wait for missions to come.

No matter how sharp the scientists’ instruments, the data they obtain is useless unless it can be woven into a story. The bits of information “have to form a narrative that makes sense,” said Matthew Golombek, project scientist.

Geologists, for example, tell stories about rocks and landscapes. They look at an unidentified standing object and figure out whether it is a piece of granite or a dinosaur bone, whether it is from Japan or Mars, whether it has been melted or frozen and when and how many times.

How do they do it? Elementary.

Just as all life (as we know it) is made from the element carbon, all rocks are made from the element silicon--a heftier member of the same chemical family. And just as carbon atoms grab onto other atoms to form the long and complex molecules of life, silicon atoms grab onto whatever happens to be in the neighborhood to form the complex minerals in rocks. So in part, geologists can deduce the history and origins of rocks by looking at the ingredients silicon grabs onto.

They also get clues about when and where the rock was formed by deducing how it was “cooked.”

Just as refried beans look and feel very different from boiled beans, rocks change dramatically, depending on whether they are melted down in volcanoes or folded under the ocean floor or distilled like wine. Indeed, the first rock studied by Pathfinder--Barnacle Bill--contained material that had been distilled many times, leading scientists to conclude that a lot more cooking was going on in the Martian crust than previously thought.

In the meantime, engineers are piecing together the story of what actually happened to the spacecraft as it hit the ground, attached to its rig of rockets, parachutes and air bags.

Scientists have learned a great deal by watching things fall ever since Newton connected the falling apple with the falling of the moon in its orbit around the Earth. Einstein said the happiest moment in his life was when he realized that a person falling off a building would experience no gravity, but would perceive himself floating in space.

Indeed, JPL engineer Rob Manning said Pathfinder didn’t feel a gravitational thing when it was grabbed by Martian gravity--proving once again that Einstein was right.

But all the theories, equations and deep understandings of how gravity behaves was not enough to tell engineers exactly how to make an air bag to put a spacecraft safely on Mars.

One reason is simply that real life is a lot more chaotic than the physicist’s equations. And the combination of Pathfinder’s parachute, back shell and lander turned it into a classic chaotic system--a series of coupled pendulums.

This, among many other unknown factors, meant the engineers could not exactly predict how the Pathfinder would fall. So they made a fake Martian landscape and dropped the air bags many dozens of times. They tried dropping the whole rig out of helicopters. They put it in wind tunnels.

They even tried using the fastest supercomputers to test virtual air bags hitting virtual Martian surface. But it wasn’t all that helpful. “Supercomputers cannot simulate the dynamics of thread,” Manning said.

In the end, they went back to the “old-fashioned way [of testing the air bag system],” he said. “We dropped it until it stopped failing.”

Over the next few months and years, the researchers at JPL and elsewhere will continue to learn a great deal about gravity and Mars and how to see and hear in foreign landscapes. They will continue to come up with more things they want to know.

“That’s why people get frustrated with science,” Johnson said. “We’re always coming up with more questions.”