In a chandeliered ballroom at the Embassy Suites Hotel in Arcadia, a roomful of men and women are seeking the perfect piece of real estate but rejecting nearly everything they see.
Some sites are too cold, others too windy, and most aren't sunny enough. A few are just plain boring, and a handful turned out to have too many triangular-shaped rocks.
It's never easy to find the perfect place to land. It's even harder when you're looking on Mars.
"The good news is, it's Mars," said Jet Propulsion Laboratory geologist Matt Golombek. "The bad news is, there's not much to choose from."
In May and June, NASA engineers plan to hurl two spacecraft toward Mars -- brawny 400-pound rovers dubbed "robot geologists." They are on a mission to find out whether water needed to sustain life once existed on the planet's surface.
But first they need a place to land. Although Mars is about half the size of Earth and has no pesky oceans to avoid, much of the planet is out from the start.
The landers must touch down in a thin strip along the equator where there is enough sunlight to keep solar arrays juiced. Within this band -- less than 10% of the planet -- huge swaths with yawning chasms, steep volcanic slopes and sudden windstorms are off-limits. Of the 185 sites first chosen, only four remain.
For its landing cushioned by air bags, each spacecraft requires an ellipse of land the size of Cape Cod that's free of stiff winds and lander-crushing hazards such as snaggletoothed rocks.
The stakes for the $800-million mission are high. Only three other spacecraft have successfully landed on Mars, the major focus of NASA's robotic space exploration program. The recent loss of the space shuttle Columbia, along with the loss of two Mars spacecraft in 1999, are reminders of the perils of space. Two-thirds of all missions sent to Mars by the United States and Russia since 1960 have failed.
In choosing a site, tension bubbles between the project's engineers, who are responsible for making sure the landers don't crash, and the project's scientists, who are eager to get to the most geologically promising parts of the planet: steep-sided volcanoes, chasms that dwarf the Grand Canyon and wind-whipped fields of boulders that may hold evidence of a watery past for the Red Planet.
"It's always the same," said Mike Carr, a geologist for the U.S. Geological Survey in Menlo Park, Calif., who has participated in nearly every Mars mission to date. "Scientists want to go to places that are interesting, and engineers want to go to places that are safe."
In 1976, Carr helped choose landing sites for the Viking missions. Back then, the landing site team had nothing to go on except fuzzy, 1960s-era images.
Luckily, the two Viking spacecraft had the luxury of orbiting to check out the neighborhood before they touched down. It was a good thing they did. The landing site that Carr and others had chosen from Earth was full of potentially lethal rocks.
"We looked at the landing site and were horrified," Carr said.
The new Mars Exploration Rovers cannot orbit first; their landing sites must be chosen in advance. The same was true of Pathfinder, the lander and small rover that reached Mars in 1997.
In that project, a three-person crew had only low-resolution pictures from the Viking missions to work with. "The engineers were saying, 'I don't want any rocks bigger than the chair I'm sitting on,' and all I could see were things larger than a football stadium," said Golombek, a leader of the team selecting new landing sites.
Because two spacecraft now orbit Mars -- Mars Odyssey and Mars Global Surveyor -- the landing team has tens of thousands of pictures showing the Martian landscape in detail. Some would say excruciating detail.
Instead of peering at hazy images, Golombek can order crisp photos of areas of Mars that have never been seen by human eyes. A team of 50 is helping analyze them all. "It's the first time science has been used to triage landing sites," he said.
Amid the stretches of chaotic terrain, massive volcanoes and deep canyons as long as the United States, scientists can see hazardous rocks about the size of a football huddle.
But geologists want to land close enough to rocks to study them. With an average speed of about half an inch per second, the six-wheeled rovers are expected to travel just one-third of a mile in their three-month life spans. Scientific success depends on their landing very close to something interesting.
Many of the new pictures show potential geological pay dirt: sedimentary rock that appears to have been laid down in water. Others show what look to be ancient lake beds or channels carved by floods. An infrared instrument has detected the presence of the mineral hematite, which typically forms in water.
They also show many threats. As one might expect from a lifeless, frozen planet scoured by ferocious dust storms, racked by the solar system's largest volcanic eruptions and pummeled by meteors for billions of years, hazards are everywhere.
In his JPL office, Tim Parker has spent thousands of hours piecing together high-resolution maps in search of "fresh, scary craters" and "pathologically shaped rocks" that could disable the landers. The maps are so data-rich they leave Parker's powerful desktop computer sluggish. But he won't stop until the decision on where to land is completed next month.
"Scientists are like baby birds," he said. "You feed them, and they still want more."
The gigabytes of data, and their many interpretations, have made for lots of quibbling -- and very long meetings, like the Arcadia gathering in January. It produced a set of rankings now dubbed "the Consumer Reports guide to landing sites."
Landing engineer Mark Adler's favorite sites are those with the least wind shear, not necessarily ones with the best rocks, or even with any rocks at all. While geologists describe rocks on Mars as "the planet's history books" and "treasure troves," landing engineers call them "scary," "pointy" and "bad."
The air-bag landing system, a similar but improved version of what was used on the Pathfinder mission, is tough enough to withstand quite a beating. But the wrong rock in the wrong place could prove disastrous.
"The only time there's no such thing as a bad rock," Adler said, "is after we land."
Because of engineering concerns, sites favored by geologists have been cut one by one.
One discard is Athabasca Vallis, an area just north of the equator cut with deep channels that indicate catastrophic floods may have once roared through. The terrain appears too rocky for landing.
Some sites have been harder to let go of than others.
Phil Christensen, a Mars geologist from Arizona State University, flashed a wall-sized image of the rejected Melas Chasma in the Arcadia meeting room even after the group was admonished by those in charge of the selection process to "shed a little tear" and get on with it.
"I couldn't resist showing the old Melas site," Christensen said to the crowd as he lingered on images of the area's stacked rock layers. "It's fabulous."
The canyon, with 6-mile-high walls that would dwarf our Grand Canyon, has alluring layers of sedimentary rock at the bottom that may have formed by ancient lakes. "Everyone was just drooling to get the rovers there," Golombek said.
But stiff horizontal canyon winds could have wreaked havoc on a lander's parachute. The planet's most interesting places, said Mars meteorologist Mark Richardson of Caltech, "tend to be associated with scary winds."
The finalists are:
* Meridiani. This site, a favorite of scientists and engineers, is flat with little wind and few rocks. Geologists are excited because the area contains large amounts of hematite.
* Isidis. This site, a flat plain at the base of the Martian Highlands, has little wind or large rocks. It may contain rocks that tumbled from the mountainous highlands but is not considered of high scientific interest.
* Elysium. This site, similar to Isidis, is favored by engineers because it has even less wind.
* Gusev Crater. This massive crater may once have been filled with a lake. Channels show where water might have entered and exited. But it is windy and rocky and has steep terrain.
In April, NASA officials will pick the two final landing sites. The landers, to be launched in May and June, are expected to reach Mars in January 2004.
The Meridiani site seems a shoo-in because it is safe and has the best shot of producing evidence of surface water.
The real question is where the second lander will go. The smart money is on Gusev Crater. It has the potential of helping prove that water existed on the planet and features a wide variety of terrain. And its dramatic crater walls, carved channels and expansive views are sure to please the masses back on Earth.
The other two sites, though wind-safe, have all the allure to geologists of a trip to Iowa. "Elysium," Christensen said, "doesn't have anything going for it."
The team is working to determine whether Gusev is safe for landing, given its slopes, rocks and winds. If deemed safe, it has a strong chance of being selected. If not, one of the two "boring" sites likely will be chosen, scientists said.
Whichever site is chosen, JPL engineers are working overtime to ensure a safe landing.
They have been conducting "drop tests" to improve the landing systems should the craft encounter any of those scary, pointy or bad rocks. Mock landers have been pushed out of helicopters onto the harsh desert floor, buffeted in wind tunnels and dropped onto specially designed ramps festooned with the nastiest rocks science can find.
"We test and debug software until we drop," said JPL engineer Rob Manning. "That's another kind of drop test."