by Rod Pyle
Mars 3, USSR—Successful landing on Mars, then operated fifteen seconds on the surface before failure
1973
Mars 4, USSR—Missed Martian orbit, flew by into deep space
Mars 5, USSR—Lasted nine days in Martian orbit before failure
Mars 6, USSR—Some data returned before crashing on Mars
Mars 7, USSR—Lander separated early and missed Mars entirely
1988
Phobos 1, USSR—Lost contact in space
Phobos 2, USSR—Lost contact in space
1992
Mars Observer, US—Lost contact in space
1996
Mars 96, Russia—Launch failure
1998
Nozomi (Planet-B), Japan—Failed to orbit Mars
Mars Climate Orbiter, US—Erroneous aerobraking maneuver, broke up in Martian atmosphere
1999
Mars Polar Lander, US—Crashed on Martian surface
Deep Space 2 (affiliated with MPL), US—Crashed on Martian surface
…and so forth.2 Surveying the list, one might wonder (a) why the Russians bother to continue trying at all and (b) what the heck they did wrong. Analysis of the once-competing US and USSR programs yields mixed conclusions, but it is apparent that Mars exploration has been largely an American game. And as noted, even JPL has its troubles.
December 11, 1998: many folks at work around the United States were just beginning to ponder Christmas vacation activities. But at JPL, most eyes were on the launch of the Mars Climate Orbiter, or MCO. A successful launch would result in Martian orbit late in the following year, and reams of new data about the weather and surface conditions on Mars. MCO was also to act as a relay for the next landers scheduled to set down on Mars. Anticipation was high.
Then, on January 3, 1999, the counterpart to MCO, the Mars Polar Lander, departed Cape Canaveral also bound for Mars. It would descend to the surface of Mars shortly after MCO and begin tandem operations. Life was good.
The ghoul awoke, smelling opportunity, and in late September 1999 reached out and wrapped his taloned hand around the Mars Climate Orbiter. On the twenty-third, communication was lost between JPL and the spacecraft. This was at a critical time when the craft was to begin aerobraking maneuvers to slow it down and circularize its orbit, as the Mars Global Surveyor had successfully done before it. The probe entered the Martian atmosphere at an improper angle and, so far as is known, broke up upon encountering the denser-than-expected air.
The ghoul retreated into the darkness, its simple task complete. Hearts were broken on Earth, and a light rain of thin metal parts burned up high in the Martian sky.
JPL mourned the Mars Climate Orbiter…and looked forward to recovery with a successful Mars Polar Lander.
The ghoul smiled a knowing smile.
JPL and NASA scrambled, meanwhile, to determine the cause of the failure. The press releases were somewhat terse (but straightforward) at first, as there was not a lot of information available to the authors: “Early this morning at about 2 a.m. Pacific Daylight Time the orbiter fired its main engine to go into orbit around the planet. All the information coming from the spacecraft leading up to that point looked normal. The engine burn began five minutes before the spacecraft passed behind the planet as seen from Earth. Flight controllers did not detect a signal when the spacecraft was expected to come out from behind the planet.”3
“We had planned to approach the planet at an altitude of about 150 kilometers [93 miles]” said Richard Cook, project manager for the Mars Surveyor Operations Project at JPL. “We thought we were doing that, but upon review of the last six to eight hours of data leading up to arrival, we saw indications that the actual approach altitude had been much lower: it appears that the actual altitude was about 60 kilometers [37 miles]. We are still trying to figure out why that happened.”4
Huh? The probe went into a low and fatal orbit, in this day and age? After all the Apollo flights, after the many successful Mars orbiters? After the complex Viking missions, for god's sake? After aiming spacecraft at planets throughout the solar system, swinging around four or five of them to reach Jupiter, Saturn, Neptune, and Uranus? Is this the same JPL that tracked Pioneer 10 all the way out of the solar system and into deep space beyond?
Lamentably, yes. And it was a simple human error that precipitated the deadly mistake.
A few days later, the painfully inward-looking announcement came out: “The root cause of the loss of the spacecraft was the failure of translation of English units into metric units in a segment of ground-based, navigation-related mission software…. [T]he failure review board has identified other significant factors that allowed the error to be born, and then let it linger and propagate to the point where it resulted in a major error in our understanding of the spacecraft's path as it approached Mars.”5
The announcement went on to elaborate on a multilevel failure. Too few personnel, a new management design, the handoff of the spacecraft from the design, build, and launch team to a new, multi-mission operations team. In short, JPL was trying to do too much with too little. It had always been known for this ability, but had crossed a threshold. And a large part of this was, quietly, placed at the feet of NASA top leadership and the quest for “faster, better, cheaper.”
But there was one more Mars-bound mission currently en route. It was the Mars Polar Lander, and its success would redeem the lab's reputation. Perhaps the ghoul's appetite had been sated, for now.
On January 3, 1999, the lab tried again. Mars Polar Lander left the cape aboard a Delta II rocket to begin the long coast to Mars. All went well as the craft left the influence of Earth and headed off on a complex trajectory toward Mars. The lander, which measured ten feet by four feet, would descend to the Martian polar area to search for water. Two parasitic craft were affixed, Deep Space A and Deep Space B. These were impactors, and they would descend ahead of the lander to hard-impact the surface and do science work of their own, penetrating up to a yard into the icy soil. The mission would be one for the record books, and could revolutionize our view of the Martian polar regions.
The probe carried the usual array of cameras, a laser-sounding instrument that could detect aerosols in the atmosphere, a robotic arm with digging scoop, a gas analyzer (with eight tiny ovens) to determine amounts of oxygen, water, carbon dioxide, and other life-bearing elements, and a microphone to send home the sounds of Mars. It was a wonderful package for exploring an exiting environment. And attached to it was abundant excitement, and that indefinable attribute of human endeavor, hope.
On December 3 of the same year, the craft entered the Martian atmosphere. All was going well. The lander plummeted into the thin veil of Mars, soon to deploy a large parachute to slow the fall. The usual communication blackout occurred as it screamed through the thickening air, headed for the planned soft landing…
And the controllers waited, and waited. And they waited some more. All too soon, it was apparent that something had gone wrong.
Increasingly anxious attempts were made to establish communication with the lander. The Mars Global Surveyor orbiter was pressed into service in an attempt to photograph the planned landing zone and find the craft, but to no avail.
Subsequent analysis by the Failure Review Board came up with a likely hypothesis. The vibrations emitted by the deployment of the landing legs may have caused the onboard computer to assume that the craft had landed, and then it did what it was supposed to do—shut down the descent engines. Unfortunately, the ship would still have been well over one hundred feet above the frozen ground, and if the hypothesis is correct, would have slammed into the cold soil at deadly speed.6
Whatever the actual events, the mission was concluded in a rude and depressing fashion. Repeated communication attempts failed to rouse the lander, and the mission was assumed to be lost. Coming on the heels of the loss of the ill-fated Mars Climate Orbiter, it was another black eye for unmanned space exploration, and the blame fell to JPL. Tears were shed, heads hung low, and the team me
mbers who had been so fastidiously assembled to operate this exciting mission were prematurely disbanded and sent off to other projects or back to their home institutions. The Mars Polar Lander joined the annals of lost spacecraft, and the ghoul licked its lips once more.
It would be irrational to feel cursed, but more than one “JPL'er” could not completely abandon the idea. But a more relevant assessment of the root cause of this failure, coming just under three months after the previous debacle, prompted Thomas Young, the chairman of the Mars Program Independent Assessment Team, to proclaim that the program “was underfunded by at least 30 percent.” Insufficient staffing, insufficient testing, and insufficient review had taken a fatal toll once again.7
And, once again, “faster, better, cheaper” had proved to be anything but.
You don't hear the ghoul mentioned at JPL much anymore, though it can still incur a nervous chuckle when mentioned. More to the point, when discussing some of the failures from the past, are memories of poor decisions from the top and questionable implementation within the ranks. Like any large bureaucracy, NASA has angels and a few demons scattered throughout. But on the whole, the agency (and while managed by Caltech, JPL is a part of NASA) does amazing work with ever-tightening budgets. If the failure rate even began to approach that of the early days—the 1960s—public (and more to the point, the dreaded congressional) outcry would probably slam the lid closed on the entire operation. As it is, funding is desperately hard to come by. But the men and women of JPL soldier on, driven not by stratospheric salaries or visions of corporate power and grandeur, but by the desire—no, the need—to investigate the great darkness beyond, to discover the mechanisms of the universe, to set foot—whether flesh or robotic—onto new worlds and find the microcosmos within. And so, despite the setbacks, JPL moves forward—with a revamped management structure and a revised rule book—to explore. And the ghoul will have poor hunting in the new millennium.
But he will be there…waiting.
Mars and Earth both have elliptical orbits, one inside the other, so at varying times (about every two years), Mars and Earth are in “opposition,” when they make their closest approach to one another. Due to this, there are favorable launch opportunities every two years for Mars-bound spacecraft. This is a driving force behind scheduling for the Mars program at JPL, because a delay of even a few weeks in development, planning, building, or testing can cause a mission to be delayed for two more years. It is a headache common to all the “Martians” at the lab, as they often refer to themselves.
With the turn of the millennium, JPL and NASA were still recovering from two very embarrassing mission failures. Internal and external reviews had illuminated many failings within the development and the management structure of the program, and these revelations were hard to swallow. Heads rolled, teams were restructured, management methods were reevaluated, testing procedures were strengthened, hardware was reexamined, and perhaps most important, budgets were scrutinized. “Faster, better, cheaper” was deemed, arguably, to have been a fallacy and was quietly retired, though money was still a tremendous challenge.
And throughout, the Mars exploration program moved forward. Incredibly, even with all the shuffling and restructuring, a few missions stayed on track. And the next one to launch, the relatively inexpensive Mars Odyssey, was ready to go in 2001.
The spacecraft was built by Lockheed Martin. The aerospace contractor had proved to be a capable and willing partner, unusually cooperative in unmanned space efforts, an arena where government/contractor relations can get sticky. Previous partnerships with NASA and JPL had gone well.
The primary goal of Mars Odyssey would be to search for water from orbit. To do this, the spacecraft would carry two principal instruments: one, called THEMIS (Thermal Emission Imaging System), would image the planet in infrared, allowing scientists to map Mars in temperatures instead of traditional visual wavelengths. The infrared images could then be aligned with images taken in visible light, and the correlation of visible landforms with areas that radiated stored heat at night would provide a better understanding of the mineralogical makeup of much of the planet. Toward this end, Mars Odyssey also was capable of taking traditional images as well.
The other was called HEND (High Energy Neutron Detector), which would identify elements in the Martian environment, specifically in the first few inches of soil. Here, planetary scientists would be looking for hydrogen, an indicator of water and water ice. Between the two instruments, it was hoped to clarify where the moisture might be on Mars and in what concentrations, among other things.
A third instrument was called MARIE (Mars Radiation Environment Experiment), which would measure radiation in the Martian orbital path. This was not only of interest in strict science terms, but would also assist in the planning of eventual crewed missions to the planet.
The overall spacecraft was about the size of an upright refrigerator, with a boom extended out one end (which held the gamma-ray spectrometer sensor) and solar panels out to two sides.
Mars Odyssey left Earth on April 7, 2001, aboard a Delta II rocket and successfully headed off toward the Red Planet. As the rocket sped toward Earth orbit, the distance from Earth to Mars was about seventy-eight million miles, but due to the course the spacecraft would follow to reach the its goal, Mars Odyssey would travel over 285 million miles.
Once within Mars's wispy embrace, aerobraking was again used to circularize the lopsided orbit inherent in missions utilizing a smaller rocket (the technique saved almost 450 pounds of fuel, which is a huge amount in Mars-bound launches). After the braking rocket fired, dropping Mars Odyssey into that lopsided orbit, aerobraking continued for almost three months until the ellipse became a circle that was proper for surface mapping to begin. This required not only circularizing the orbit but also adjusting it to an almost north-south orientation, also known as a polar orbit, which has been used for most post-Viking orbital missions. With this, the planet rotated perpendicular to the orbit of the spacecraft, allowing it to repeatedly photograph almost every part of the surface as the planet slowly turned below.
And then, in late February 1992, Mars Odyssey got to work. The major risk milestones had been passed, and things seemed to be going well. The folks at JPL breathed a bit easier. Mars Odyssey, it seemed, would not disappoint.
You see, lessons had been learned since the multiple failures of the 1990s. Most of the flight systems (i.e., most things likely to fail or malfunction) were now redundant and had backup units. It was almost as if JPL had taken a page from the manned spaceflight playbook, in which as many systems were doubly and triply redundant as possible.1 This was, in some ways, now applied to Mars Odyssey.
The brain of the spacecraft was a radiation-hardened version of Apple's Macintosh® processor of the time, the Motorola Power PC® chip. With 128 megabytes of RAM and three megabytes of other storage, it was hardly a powerhouse but would do the job.
As data moved from here out to the rest of the spacecraft (and back), an ingenious parallel design was used. Computer cards enabled specific tasks, and were placed in double rows downstream from the processor, allowing for 100 percent backup capability. The only parts of the computer not backed-up were the main processor and a (then staggering) one gigabyte storage card for imaging.
Finally, the flight software, which was carried onboard as opposed to the commands later sent up from Earth, had more sophisticated “fail-safe” routines written in and had frequent self-checks. If something went wrong, the craft would immediately enter a “safe” mode and begin troubleshooting the situation. This had always been a part of deep-space software, but had been beefed up following the reviews of recent losses.
By late March 2002, Mars Odyssey was sending back images that were then posted online almost immediately by the JPL team. It was the second Mars mission to offer the public such immediate feedback, and it had been pioneered by the impressive online presence of the Mars Pathfinder mission. The images were spectacular even in their raw
state, and doubly so once enhanced.
And in an ongoing quest, Mars Odyssey was looking for life on Mars.
Of course, as an orbiter, it did not have the luxury of a dirt scoop and an analytical lab like Viking did, nor the ground-based mobile capabilities of Pathfinder. But with the growing understanding of the nature of Mars, the Odyssey team had been able to focus its instruments on the investigation of water present on Mars. Where water was found, past or present, there could be life, past or present.
The strategy was to look at the surface, and shallow subsurface, environment of Mars for both hydrogen, indicative of water, and various mineral types, indicative of past water. The probe would also be able to spot hot springs now suspected to possibly be on Mars. With patience, a holistic picture of the planet would eventually come to light. The machine simply had to function long enough to allow these results to emerge…and in this, Mars Odyssey would shine like none before or since.
By April 2002, Mars Odyssey had already allowed planners to select landing sites (from many dozens of candidates) for the upcoming Mars Exploration Rovers, which would leave Earth in just a few months. The images being studied were critical to making an informed choice, and the results had been immediate and gratifying. There was a high level of confidence in the selected sites.
Concurrently, visible light images had been combined with earlier shots from Mars Global Surveyor to examine some gullies that appeared suspiciously like drainage channels. It was soon realized that a few had been caused by recent melting of water snow. This was a breakthrough, as it provided proof that water was still “running” on Mars, something long suspected but until now unproved. The key idea was that ice was melting underneath snow packs, and the mass of ice above prevented the water from flash-evaporating in the thin atmosphere as it flowed out to create the gullies.
On Earth, a few little erosion channels might not cause even a moment of excitement, as we are used to seeing such things constantly and easily in our own dynamic ecosystem. But these gullies had first been observed on Mars in 2000 from Mars Global Surveyor images, after countless hours of painstaking investigation of thousands of pictures. They seemed to occur only on the colder, pole-facing “shadow” side of some craters, and as such, indicated that a special set of circumstances was contributing to their formation. This colder, shaded location allowed snow to accumulate and remain across an entire Martian season, so that melting could occur gradually, allowing the water to seep out and create the features observed. Here was the explanation, the “smoking gun,” that so many had been waiting for. For a Mars scientist, it was nirvana.
