Photograph courtesy MRSS and NASA/JPL
Read Caption

Dust avalanches typically a few hundred to a thousand meters long tumble toward the center of an impact crater on Mars. What sets them off? Some say a gust of wind, but some think trace amounts of water can play a role. Slope streaks are found only in areas where surface temperatures rise just above the freezing point of water; the increase of pressure when water vaporizes might trigger the slides.

Photograph courtesy MRSS and NASA/JPL

Mars—Planet Ice

Read a National Geographic magazine article about Martian landscapes and get information, facts, and more about Martian ice.

It's dustier than the road to death, drier than Dorothy Parker's martini, colder than the devil's kiss. Like much of Mars, the butterscotch plain is inhospitable, empty, ancient, and, when it comes down to it, pretty dull. But a few hundred meters to the south, over a shoal of low, uneven hummocks, the landscape changes. The hummocks reach a crest that curves smoothly to the east and rises sharply to the southwest. Beyond it the land falls away in a steep slope. This is the rim of a crater. Its floor—some 800 meters (2,625 feet) below—is ridged and rippled in strange, concentric forms. It looks like ground within which something has recently stirred. Or is still stirring.

And there is more evidence of movement: A smooth layer that once covered the slope has come loose and crept toward the rippled floor. Roundheaded gullies slice into that same smooth layer, while strange tongues of who-knows-what lap at the slopes beneath them.

Landscapes like this are changing the way geologists look at Mars. They've long been fascinated by the planet's distant past. Now they're getting ever more excited by the mysterious processes shaping its present—thanks in large part to the planet's apparent iciness. Martian ice is not a novelty in itself; for years geologists have expected to find it frozen into the soil at mid and high latitudes. The excitement comes from a growing suspicion that the ice doesn't just sit there but has a dynamic role to play. That it moves from place to place around the globe. That it reshapes the textures of the surface. And that it may sometimes produce fleeting traces of liquid water.

According to this view, the smooth drapery slumping into the crater is dirty ice, or possibly very icy dirt, that flows like an earthly glacier. If you could clamber a few hundred meters around the slope, or bounce down it like a low-gravity scree walker, you could find out for yourself. If you had the right instruments, a little bit of that layer could teach you a lot about Mars. You could scoop out the ice and hold it in your hands. You could watch as it vaporized into the bitterly cold, terribly thin air.

But for now the closest anything or anyone from Earth can get to this tantalizing landscape in Mars's southern highlands is when one of the growing fleet of spacecraft orbiting the planet passes some 400 kilometers (250 miles) overhead. Three spacecraft are scheduled to land on Mars—one on December 25, 2003, and two others to follow this month—but they're headed for very different destinations. The first craft, the British-led Beagle 2, piggybacked on Europe's Mars Express orbiter, will hunt for chemical signs of past—or even present—life in the soil and atmosphere. Two rovers sent by NASA, Spirit and Opportunity, will mount small-scale geologic surveys of sites that may bear traces of water billions of years old.

The overarching theme of NASA's current Mars exploration is "follow the water," because where there is water, there might possibly be life. The landers will carry out that mission, but to achieve any of their scientific goals, the spacecraft have to land safely. With almost no ability to maneuver as they descend through the atmosphere, the craft must be aimed toward smooth, flat landing sites the size of small countries. And because the landers need warmth and use solar power, these sites need to be near the Martian equator. That rules out the southern crater with the gullies on its rim and the strange slumping stuff within. In fact, it rules out about 95 percent of the surface of Mars.

Happily, the three landers are not the whole story. The latest generation of orbiters is revolutionizing the study of Mars from space, discerning ever smaller surface features and analyzing terrain with increasingly sophisticated tools. Cameras on NASA's Mars Global Surveyor (MGS) and Mars Odyssey watch the planet day and night in visible and infrared light. Other sensors on Odyssey detect gamma rays and neutrons radiated from the minerals below, revealing to those versed in nuclear physics the abundance of different elements, such as hydrogen and iron.

When Mars Express reaches the planet, it will add yet more instruments to those already aloft, including a multispectral imaging spectrometer that can identify minerals using visible and infrared wavelengths, and a radar that may detect ice and liquid water below the surface. Also due to arrive in December is Japan's Nozomi orbiter, designed to study Mars's atmosphere.

As they beam a continuous stream of fresh data to Earth, the orbiters evoke a thrilling new picture of Mars—and a genuinely puzzling one: While searching for answers to old mysteries, the orbiters are bombarding us with new ones. Hugh Kieffer, a Mars veteran at the United States Geological Survey (USGS) who has been studying the planet for almost four decades, tells his younger colleagues they are in a "period of maximum confusion." New data—whole new types of data—are accumulating faster than researchers can make sense of them. The result is something like an optical illusion. Contradictory images of Mars seem to flicker in and out of focus in the mind's eye.

"Although Mars is supposed to be the god of war, the planet is much more like a prima donna," says Nathalie Cabrol, a planetary geologist at NASA's Ames Research Center. "When you think you have it right, Mars always has a surprise."

The man whose imagination first took him to the crater rim on the edge of the butterscotch plain is Philip Christensen, a geology professor at Arizona State University in Tempe. Christensen leads the team responsible for designing and building the Thermal Emission Imaging System (THEMIS), the instrument on board Odyssey that generates pictures of the surface in visible and infrared wavelengths. When THEMIS sent back its picture of that crater and its strange lining in the early summer of 2002, Christensen had what every scientist yearns for—a eureka moment. For the first time he thought he could see how Mars's mysterious gullies might form.

Two years earlier geologists Michael Malin and Ken Edgett had spotted what looked like recently carved gullies in images from the Mars Orbiter Camera (MOC), which flies aboard MGS. Malin, who designed MOC and set up the company that operates it, and Edgett, who actually targets the camera, saw gullies up to a couple of kilometers long and a few tens of meters wide on crater rims, canyon walls, and other inclines all over Mars's midlatitudes.

The shape and freshness of the gullies made Malin think that they had been carved by running water within the past few million years—and possibly much more recently. He soon convinced the rather more skeptical Edgett, and the two went public with their theory in June 2000. The announcement, made during a big press conference at NASA headquarters in Washington, D.C., sent an immediate shock wave through the field: It required scientists to reexamine their view of the planet.

It's not the first time an old Mars has been killed by new data. Until Mariner 4 zipped past in 1965, taking pictures as it went, earthbound astronomers had imagined a place at least a little like home, though much colder and with thinner air. But Mariner 4's snapshots of Mars showed a heavily cratered surface far more moonlike than earthlike. Readings showed the atmosphere—composed of almost pure carbon dioxide—was even thinner than had been thought, exerting just one percent of the pressure at sea level on Earth. A new Mars was born, an inert, barren place with a rime of carbon dioxide frost. On this Mars there was no room for life.

Then in the 1970s this Mars too was killed. Orbiting spacecraft—first Mariner 9 and then the two Viking missions—showed there was much more to the Martian surface than craters. Mariner 9 saw volcanoes twice as tall as any on Earth. There were canyons as deep as the Earth's deepest ocean trenches. There were what appeared to be desiccated river systems. There were plains scoured by floods large enough to drain the Mediterranean in a month.

Pictures from these orbiters introduced another new Mars: one that was not just a dead rock but rather a fascinating fossil. In the planet's youth, scientists had theorized, the atmosphere was thick, the greenhouse effect strong, and water flowed through the valleys. But the greenhouse had been temporary; the atmosphere thinned, and the planet cooled. Except for the scraps frozen into the polar caps, its water was lost to space or stored away as subsurface ice. And this all happened billions of years ago. The early Mars had been dynamic; the Viking Mars was more or less inert and—as the Viking landers showed us—lifeless.

That was the Mars the gullies killed. It's not that it vanished, but it has been overlaid, and in places undercut, by findings from far more acute instruments aboard the current orbiters. The gullies Malin and Edgett saw in 2000 were too small to have been picked up by the Viking cameras. Indeed, it was the very fineness of their features that made them look much younger than anything the Vikings had seen. They appeared to be evidence that liquid water could have flowed on the surface in the recent past. And that led to confusion by the bucketful.

Water will stay liquid only if it's warm enough and at high enough pressure. Drop the temperature, and it will freeze; drop the pressure, and it will vaporize. Physics seems to say that Martian midlatitudes are far too cold for liquid water to persist for any length of time at the surface. But the gullies seem to say differently. As a result some researchers suggested the gullies weren't in fact carved by water. Others invoked ice buried near the surface that had somehow melted. Malin and Edgett imagined the gully-carving fluid might come from very salty (and thus hard-to-freeze) aquifers just below the surface.

Like many of his baffled colleagues, Phil Christensen didn't much like any of the explanations. That said, he also hadn't given a great deal of thought to them, as he'd been focused instead on using infrared cameras to detect various sorts of rocks and minerals from a distance. But while watching the THEMIS pictures come in from Odyssey, he got interested in another phenomenon revealed by Malin's MOC: a strange "pasted-on layer" draped over some north-and south-facing slopes in the mid and high latitudes.

Going through the latest THEMIS feedback one day, Christensen remembers, "There was this one 'ah-ha' image: I saw this pasted-on stuff with gullies in it. And that was it." From a distance of 335 million kilometers (208 million miles) or so, THEMIS had lit up a lightbulb over its creator's head. Looking at the picture, he suddenly saw the pasted-on stuff as very dirty snow—not from its color but from the way it was draped on the landscape. When sunlight warmed the dust within it, the snow would melt from the inside out, and the meltwater, kept liquid by the insulation of the snow above, would carve the gullies.

In a little over six months this interpretation made it to the cover of the prestigious scientific journal Nature. One of the researchers who reviewed the paper was Mike Carr, a geologist and Hugh Kieffer's colleague at USGS. Carr directed the team that ran the Viking orbiter cameras and is now a member of the MOC crew. He wrote the book on evidence for water on Mars (literally, it's called Water on Mars) and was one of the first to report about the pasted-on layer when it showed up in the MOC images. When he read Christensen's hypothesis about the gullies, all he had to say to Nature's editors was that he was embarrassed not to have thought of it himself.

But how could there be dirty snowpacks over a large fraction of the planet's surface? Like its liquid form, water ice is stable only within a specific range of temperature and pressure. Drop the pressure on a piece of ice far enough, and it will turn straight into water vapor, a process called sublimation. This is exactly what would happen under today's conditions if you were to put fresh snow at the latitudes where most of the Martian gullies are found. So how could any snow accumulate there? The answer, Christensen and others have suggested, lies in changes in Mars's obliquity—the angle between the planet's axis of rotation and the plane of its orbit.

On Mars, as on Earth, this angle varies over time, responding to the gravitational pull of the other planets, particularly Jupiter. The poles nod up and down, and when they're down they point more directly at the sun, which makes high-latitude summers hotter. The Earth's nodding, though relatively modest, influences the rhythm of our ice ages. Mars nods much more vigorously—think involuntary spasms rather than a measured sign of assent—and the climatic effects could be dramatic. When the poles dip by 35 degrees or so, the residual polar ice caps will no longer be stable in summer. At the same time the lower latitudes will get cooler, and the water vapor that sublimes from the polar caps will fall there as snow. When the poles shift back, the lower latitudes get warmer, and the snow becomes unstable. That's when, according to Christensen, it melts from within to form the gullies.

Mike Carr and Bruce Jakosky of the University of Colorado first suggested, almost 20 years ago, that the planet's nodding could have moved ice from the poles. The reason the idea is now coming to the fore—not just, or first, in Christensen's explanation of the gullies—is that MOC and other instruments orbiting Mars are seeing features that make the concept seem more than just a possibility. The evidence for ice looks real, and it looks recent.

Jack Mustard of Brown University has mapped what he thinks are the remains of a layer of mixed dust and ice over much of the midlatitudes that is related to Christensen's pasted-on material. Computer models of the planet's atmosphere suggest that more than once over the past few million years, tilting of Mars's axis could have moved enough water from the poles to make a layer of snow as much as ten meters (32.8 feet) thick over parts of the planet. Mustard's colleague Jim Head, Nathalie Cabrol at Ames, and Jeff Kargel at USGS, among others, all report seeing the remains of glaciers—or things that look like glaciers—in various places and at a wide range of scales, from features that dominate the flanks of giant volcanoes to the strange flow-like folds beneath some gullies.

No one person agrees with all of this, and some don't like any of it. Malin and Edgett, discoverers of the gullies, are deeply skeptical of Christensen's snow model. But that doesn't mean they think Mars is the static desert of the Viking days. In fact they believe it could be changing on an even faster timescale than that defined by its nodding poles. Their camera, MOC, has revealed "Swiss cheese" holes in the solid carbon dioxide of the southern polar ice cap, holes that grow from year to year. It's possible, they say, that over a century or so, those holes could add enough additional carbon dioxide to the atmosphere to ratchet up the greenhouse effect and trigger subtle changes elsewhere.

The most dramatic evidence for change going on right now, beneath our orbiting eyes, comes from what could be seen as the least dramatic features on Mars: slope streaks. These are just what their name implies— streaks that run down slopes like two-dimensional stains. A commonly accepted explanation is that they are very thin avalanches of dust slipping away to reveal a darker surface beneath.

Norbert Schoörghofer and Oded Aharonson at Caltech have been studying the images of slope streaks sent back by MOC. To their surprise, they found that the number is growing from year to year at what Aharonson calls "an incredible rate." For every hundred existing slope streaks there are seven more every Martian year, making them among the fastest changing surface features on Mars. But the pace of their creation is only part of the story. What seems really remarkable is that their occurrence looks as though it may be related to the presence of traces of water.

Given how they're thought to form, you'd expect slope streaks to turn up in dusty places and on relatively steep terrain. Examining MOC images and altitude measurements from MGS, Schoörghofer, Aharonson, and their co-workers confirmed that the streaky regions were indeed dusty and rich in ups and downs. But something didn't quite fit. Mars offers plenty of slope-rich dusty areas completely free of streaks. What makes the streaky regions different?

On a hunch, the scientists looked at readings from the Thermal Emission Spectrometer on MGS, which records the surface temperature at any given place on Mars. Bingo! The slope streaks appeared in regions where, for some of the year, the temperature at the warmest time of day crept above the freezing point of water. So in places where water might appear, just fleetingly, something seems to destabilize the dust and let it slip. In places where water is not expected to occur, no such thing happens. This doesn't necessarily prove that ice sublimating, or melting and refreezing, triggers the dust avalanches. But water—maybe only a trace—seems a distinct possibility.

From Percival Lowell at the turn of the 20th century to theories based on Viking data in the 1980s and 1990s, the story of Mars has almost always been told as a long planetary diminuendo—a slow sinking from the vibrant to the moribund. Lowell thought Mars was covered with canals because it was drying out and that the canals were an ancient civilization's last-ditch—last megaditch—attempt to stave off the planet's decline into waterless and lifeless senility. From then until now people have maintained that planetary activity—volcanism and crust movement—is driven by internal heat, that a smaller world like Mars will necessarily cool down much quicker than a bigger planet like our own, and that we have therefore missed its glory days.

True as far as it goes: Mars is not the hot-hearted volcano-building place it once was. But if the planet is no longer active in the way Earth is, it's still exquisitely reactive, responding to little orbital pulls and shoves with planet-girdling sheets of dirty water—ice or snow that bring striking changes to its surface. Its capacity for change is no longer buried in the distant past. Instead this change is almost contemporary, sometimes happening on the same 10,000- to 100,000-year timescales as the Earth's ice ages, if not quicker. Mars is no longer a world of endless decline, but one of rhythmic regeneration. That regeneration may include occasional splashes of liquid water, making Mars a more plausible abode of life than it has seemed since the Viking missions.

While the orbiters continue to show us more of the planet's cold, capricious present, the current landers are slated to go to places that tell the story of an older Mars. Beagle 2 is targeted at Isidis Planitia, a large equatorial basin. Spirit and Opportunity are headed for regions where water may have shaped the geology near the surface billions of years ago. According to infrared readings, Opportunity's landing site in Terra Meridiani shows traces of a form of the mineral hematite that, on Earth, is created typically only in the presence of water. Studies by Cabrol and others (including her husband, Edmond Grin) argue that Gusev crater, the landing site for Spirit, was once a basin flooded with water from Ma'adim Vallis, a huge channel that empties into it. The bottom of the basin—the rover's intended landing site—may have been a lake bed more than once as long-lost waters came and went.

It may seem a shame that the rovers aren't going to look at the gullies, or slope streaks, or at sites near possible glaciers, or at the dozens of fascinating places that have cropped up in the past few years. However, it's fortunate that Spirit and Opportunity can get someplace so interesting. The landers don't just need flat plains near the equator, they need low levels of dust, a scarcity of boulders, and gentle winds.

The fact that two sites as intriguing as Gusev and Terra Meridiani seemed safe enough to try for came as a relief to mission scientists. Terra Meridiani, indeed, is almost too good to be true: It's one of the few places on the planet where evidence for any mineral associated with water has been found; it's just about the flattest, smoothest, safest landing site anyone has ever seen; and it's pretty much slap on the equator to boot.

There's no doubt that many Mars scientists, given free rein, would choose other sites (Nathalie Cabrol is an exception—she and Grin have lived, breathed, and dreamed Gusev for years, and the idea of going there lights up her face). But that doesn't make them any less enthusiastic about the rover missions. Given the sites, it's hard to imagine the rovers not making striking discoveries about Mars's distant past. There's even a chance that they could learn something about the icy present too. Some people think there may be subsurface ice right at the equator. Who's to say that some trace of it might not be found?

This is all, in a way, secondary. When asked why he's so excited about the rovers, Phil Christensen's geologist co-worker Jim Rice looks amazed anyone could ask such a thing: "They're going to Mars, man." Equipped with yet more of Christensen's infrared mineralogy-at-a-distance spectrometers, the rovers will be able to spot the most interesting rocks for analysis in a landscape, trundle over to them, and learn what they're made of. It's as close to putting a geologist on Mars as you can currently get, and for everyone involved that's enough.

Spirit, Opportunity, and Beagle 2 can never crack all of Mars's mysteries. Targeted at the equator, they can't examine processes at work in the midlatitudes. And they won't be able to sort out questions of Mars's climate cycles or dig deep enough at suitable locations to rule liquid water in or out. But they will undoubtedly deepen the relationship Earth's scientists have with Mars.

If all goes well, the teams running the rovers will come as close as they can to actually living on the planet. Their lives will follow Martian clocks and calendars. Their questions will be Martian questions, their problems Martian problems. They won't taste the pervasive dust or feel the deep chill of the air—but it's a fair bet that, late in the night, they'll half believe they do.

Will that sense of place matter? Maybe not. A rich stream of data continues to arrive from spacecraft in orbit and will increase further with the addition of Europe's Mars Express this winter. In 2005 NASA's Mars Reconnaissance Orbiter (MRO) will add data from yet more instruments including HiRISE—a camera that will pick out details as small as 30 centimeters (11.8 inches). Somewhere in all this information may lie clues to help scientists cut through much of today's confusion. But in the face of such a torrent of data, it will be helpful to have solid rock to stand on. The landers can provide that sure footing by showing us what a couple of places on the planet are really like. What's more, the wheel-on-rock experience gained from Spirit and Opportunity will make it possible to develop future rovers capable of tackling the most promising new sites the orbiters discover.

But perhaps the most important new frontier in Mars exploration is the sense of time. When MRO gets to Mars, MGS will have been in orbit for nearly ten years, and there are plans for two more NASA missions this decade. As the missions add up, so does something greater than the sum of their individual achievements: a composite record of Mars's three-dimensional surface stretched over the fourth dimension of time. If the key to Mars is the way that it changes, the key to understanding is constant inspection. And surely no planet other than the Earth has ever enjoyed such continuous attention.

You can't help thinking that Galileo, arguably the first planetary scientist, would have approved. Before him, doctrine held that the heavens were perfect and unalterable. Galileo discovered that in the heavens, as on Earth, things didn't stay the same and said that change was everywhere—and that it was good. A changeless Mars, like a changeless Earth, would be so dull as to merit no respect: "If at the time of the flood the waters which covered [the Earth] had frozen, and it had remained an enormous globe of ice where nothing was ever born or ever altered or changed," he wrote, "I should deem it a useless lump in the universe."

Today Mars looks a lot more like a globe of ice than it ever has before. But it also looks like something shucking that ice away, something moving on, something undergoing change. Whatever else Mars turns out to be, it won't be a useless, changeless lump in the universe.