In January 1999, NASA's Mars Polar Lander roared away from Earth on a bold mission to explore a unique region of the Red Planet. The spacecraft was to gently set itself down near the border of Mars' southern polar cap, the first ever spacecraft to study the distant world's polar environment.
That December, after months of crossing interplanetary space, Mars Polar Lander was in the final minutes of slowing itself down, ready to make a self-controlled touch down. It was never heard from again.
Nobody knows for sure exactly what occurred at journey's end.
The loss of the Mars Polar Lander became a detective story that pitted photo analysts at a super-secret spy agency against NASA experts in a debate over the overall condition of the lost-to-Mars probe.
It's a saga of light and dark pixels, egos and professional courtesy, and a report that never saw the light of day, until now.
On Dec. 3, 1999, the Mars Polar Lander was plunging through Mars' atmosphere, headed for a soft landing on the planet's south polar region. Once safely down, the probe was to establish radio chat with Earth and begin months of scientific work.
Attached to the Mars Polar Lander were a pair of small hitchhiking devices, the Deep Space 2 Mars Microprobes — Scott and Amundsen — which were to be ejected at high altitude to fall and penetrate beneath the Martian surface. They too failed to phone home.
Following more than a month of attempts to bring Polar Lander back from the dead, NASA declared the mission a failure. The loss spurred several intense studies, both internal and external to NASA.
Those assessments led to a "most probable cause" for the mishap, according to NASA.
Spurious signals when the trio of lander legs deployed during descent are thought to have given a false indication to onboard smarts of the spacecraft. It fooled itself into thinking it had landed, although it was high above Mars.
The result, according to blue-ribbon study groups: a premature shutdown of the spacecraft's engines and the destruction of the lander when it fell onto the planet. In this scenario, the probe would have been destroyed as it smacked into the surface at 50 miles per hour (22 meters per second), reported a Jet Propulsion Laboratory special review board.
The lander was not equipped to advise Earth controllers what its step-by-step situation was as it zoomed in for a touchdown.
Without any entry, descent and landing telemetry data, there was no way to know whether the lander reached the terminal descent propulsion phase. If it did reach this juncture, it is almost certain that premature engine shutdown occurred, investigators concluded.
Enter the spy analysts
The polar environment was more severe than the landing sites of previous missions. Far less was known about this exotic territory, but Mars Polar Lander was billed as an "exciting and significant step" in Mars research.
Mars Polar Lander's job was to focus primarily on Mars' climate and water. It carried an instrument suite to further scientific understanding of the climate history of the planet. It was outfitted with a robot arm, capable of digging into Mars in a search for near-surface ice.
In an early attempt to find the spacecraft, overhead search imagery of the Polar Lander's projected landing site was acquired by the Mars Orbiter Camera system, carried by NASA's Mars Global Surveyor, which had been orbiting the planet since 1997.
Both JPL as manager of the Polar Lander mission, as well as Malin Space Science Systems, the primary contractor/operator of the Mars Orbital Camera system, conducted additional imagery scans to look for the lander.
But locating Polar Lander, or pieces of a wrecked spacecraft, proved inconclusive. Even if Polar Lander sat on the surface intact, it would have been tough to detect. The camera system was right at the very limits of its abilities to clearly spot Polar Lander hardware.
At NASA's request, a team from the National Imagery and Mapping Agency — recently renamed the National Geospatial-Intelligence Agency — carried out a detailed search of Polar Lander's primary landing area, utilizing the Mars Orbital Camera's images and an array of high-tech analytical equipment.
Why NIMA? The agency is both a combat support as well as national intelligence agency whose mission is to provide timely, relevant and accurate geospatial intelligence, or GEOINT, in support of our national security. The agency is an acclaimed leader in describing, assessing and visually depicting physical features on Earth. In short, it makes use of such hush-hush tools as spy satellites.
The NIMA Mars sleuthing work was led by Ivar Svendsen, who had 27 years of experience in imagery analysis, but who has since passed away.
Svendsen was joined in the search for Mars Polar Lander by James Salacain, who had at the time chalked up some 15 years of specialized duty in support of the national imagery community.
Hunt for evidence
NIMA's task was straightforward: Use its imagery exploitation skills and techniques for locating and making out small human-made objects in terrestrial imagery — but this time apply that handiwork to Mars.
The NIMA experts began studying dozens of Mars Global Surveyor surface shots of the most probable Polar Lander touchdown site and surrounding area. They were on the hunt for evidence of the lander, and other associated hardware — such as a cast-off descent aeroshell and parachute. These objects, in theory, would be barely detectable by Mars Global Surveyor.
Also part of NIMA's investigation was gauging the different types of materials used on Mars Polar Lander hardware, including detailed looks at the reflective properties of the lander's solar panels.
In early and late 2000, the NIMA Mars surface scanning work was done. Analysis of the search findings was wrapped up in early 2001.
Details of the unclassified report to NASA were provided to Space.com by the National Geospatial-Intelligence Agency.
Three candidate sites
The NIMA team identified three candidate sites that had "pixel returns" appearing to match the expected signatures of the lander and its associated hardware. A pixel is the smallest discrete component of an image. The greater the number of pixels per inch, the greater the resolution.
A central feature, tagged as Site 2, was assessed as the possible location of the Mars Polar Lander itself. This double "bright-spot" signature could very well be an upright lander, sitting on the surface with its solar panels in the deployed position, the NIMA experts reported.
Another pixel return, called Site 1, was subjected to intense scrutiny. The NIMA analysts believed that signature may well represent the lander's backshell, a protective cover that encased the robot probe during atmospheric entry. In every image of Site 1, there appears to be an object visible on the surface that is brighter than the background. This object is located nearly two miles (3 kilometers) up range from the possible Polar Lander landing zone.
At Site 1, attempts to discern a large, 20-foot (6-meter) diameter white parachute that was to remain attached to the backshell proved problematic. The interaction of the parachute lying on the surface could be causing that signature to become indistinct, the search team concluded.
Lastly, a bright pixel at Site 3 may be indicative of the presence of human-made materials, the NIMA researchers stated. This locale includes a possible high-velocity impact site, including what appears to be ground scarring, leading up to a glint. That glint could be the MPL heat shield, the analytical team surmised.
The bottom line to the NIMA assessment: The Mars Polar Lander failure likely occurred late in the spacecraft's rocket engine-powered descent phase, or perhaps even after landing.
NIMA's findings about the fate of Mars Polar Lander were surprising to NASA.
NASA, in turn, reviewed the NIMA story — a nicely bound report, one that was complete with lots of Mars Global Surveyor imagery, other color pictures, drawings, circles and arrows throughout.
According to a source familiar with the report, and taking into account expert advice about the inner workings of Global Surveyor's Mars Orbital Camera system, NIMA got it "embarrassingly wrong."
The suspect pixels probed by NIMA were identified as electronic noise in the Mars Orbital Camera hardware. The NIMA experts didn't detect Mars Polar Lander, the source said, "they detected noise."
On March 26, 2001, a joint NASA/NIMA release was issued by the space agency. It saluted NIMA's investigative skills, underscoring the principal challenges in locating the missing lander. One major problem being that the Mars Polar Lander is only somewhat larger -- about six and a half feet across -- than the smallest objects the Mars Global Surveyor's camera can see on the surface of Mars.
Furthermore, NASA made it clear that it had its own "alternative view" of NIMA's findings. It noted that "these features could be noise introduced by the camera system, so further work between NASA and NIMA will be conducted to address differences of interpretation."
At NASA, the report was deep-sixed. The space agency did not want to be in a position of seeing NIMA embarrassed, the source said. "The space agency didn't want to look like it was invalidating the work or claiming to invalidate the work of our nation's premier spy agency."
Case closed? Not by a long shot.
Resolving the mystery
In a December 2002 article in Geospatial Intelligence Review, the two NIMA analysts, Svendsen and Salacain, remain steadfast about their observations. The signatures at the three sites studied "appeared to be reflected light or glints" from some portion of the Mars Polar Lander entry, descent and landing system and/or the lander itself, they asserted.
The NIMA team does note that spurious camera noise cannot be ruled out. However, "the coincidental appearance of spurious noise within the MPL primary landing site that also happened to emulate MPL-like imagery signatures was considered unlikely."
In a postscript to their work, the NIMA researchers underscored the fact that NASA's Mars Reconnaissance Orbiter, set for takeoff in 2005, is built to take very-high-resolution snapshots of the planet's surface. Those MRO images "may help finally resolve the mystery of what actually happened to the MPL," they explained.
"If not, the MPL mystery may have to patiently await a final and definitive investigation by a future visiting astronaut on-site inspection team from Earth," the NIMA experts concluded.
The People's Camera
Onboard the Mars Reconnaissance Orbiter will be the HiRISE (High Resolution Imaging Science Experiment). This super-powerful camera will reveal small-scale objects in the debris blankets of mysterious gullies and details of geologic structure of canyons, craters, and layered deposits. And it could also take long shots at finding Mars Polar Lander.
HiRISE should be able to resolve objects a little smaller than 3.3 feet (1 meter) diameter, said Alfred McEwen, the principal investigator of HiRISE at the University of Arizona in Tucson. "But the Martian surface is littered with meter-scale objects, so we could image the lander but not be able to distinguish it from a boulder," he told Space.com.
The camera also has color imaging capability, so perhaps a bump with an anomalous color could be detected, McEwen said. "Or maybe there's a strewn field, perhaps including a few pieces we could detect as pixels with anomalous colors," he added.
McEwen said he was not enthusiastic about taking on such a search for Mars Polar Lander, unless there are specific locations that are strong candidates. "But if NASA wants us to make a more extensive search, then we will certainly cooperate," he said.
HiRISE has already been dubbed "The People's Camera." The science community and the broader public as a whole are encouraged to participate in HiRISE targeting and data analysis. Anyone may submit suggested image targets, as discussed on a HiRISE Web site: http://hirise.lpl.arizona.edu/HiRISE/public.html.
Tracking down the final resting spot for Mars Polar Lander offers a form of technological closure, said Steven Jolly, chief engineer for Mars Reconnaissance Orbiter at the Denver-based Lockheed Martin Space Systems Co. The aerospace firm also built Mars Polar Lander, with Jolly then serving as the company's flight operations lead for the mission.
"For all of us that worked on that spacecraft, we'd love to know what went wrong," Jolly said. "All of us wished that there was confirmation of the NIMA thoughts. That would tell us an awful lot about the whole entry, descent, and landing sequence," he said.
Jolly told Space.com that spacecraft engineers wrestled with lots of scenarios that might have led to Mars Polar Lander going deaf, dumb and blind.
For one, the craft could have failed to deploy an antenna, the only direct-to-Earth link. Then there's the view that it touched down on the side of a hill, tipping over and also negating radio communications. There were even intriguing but never substantiated signals that looked like coherent utterances from the lander.
Clearly, the post-mortem "probable cause" for Mars Polar Lander's failure must be taken into account too. Those flaws singled out by investigative groups, Jolly said, are being solidly addressed in the Phoenix Scout mission — a lookalike lander scheduled to head for Mars in 2007.
But there remain those that think the lander might have survived a premature engine shutdown and free-fall "crunchdown" on Mars. "It's quite possible that the lander is in exactly the condition that's been postulated by NIMA," Jolly said. "So it's not out of the realm of possibility that it could be viewable."
"We owe it to the American public and to our own conscious to probably take a look with the Mars Reconnaissance Orbiter," Jolly said. "But I don't think I would conduct a search. That would take too much in the way of precious resources that ought to be applied to the science that MRO represents. But certainly a few attempts should be made."
Jolly admitted that there is still residual hope that someday Mars Polar Lander will be found on the planet, sitting there fairly intact.
But if this turns out to be the case, Jolly said that raises a key question: "Why the heck didn't it work?"