PASADENA, Calif. -- A recent NASA report found the Deep Space 2 microprobes were unfit for launch but sent to Mars all the same, where they vanished December 3, 1999 along with their mothership, the Mars Polar Lander.

But Sarah Gavit, the project manager for the tiny, tenacious, twin spacecraft at NASA’s Jet Propulsion Laboratory (JPL), disputes that finding, saying the mission team had full confidence they would succeed in their derring-do enterprise.

"I couldn’t have stood up in front of the director of the lab and my review board said, ‘We are ready to launch,’ if I didn’t think we were ready," Gavit said.

A NASA-commissioned report that assessed its Mars exploration program, sparked by the losses of the Polar Lander, Deep Space 2 and Mars Climate Orbiter missions, came down especially hard on the experimental microprobes.

The report’s final verdict? "Microprobes not ready for launch."

"I thought that statement was kind of harsh," said Sarah Gavit, who spearheaded the $29 million project to send the two probes hurtling -- on purpose -- into Mars. "We had pretty dang good confidence we were there."

The report, authored by a committee led by former NASA official and Lockheed Martin executive Thomas Young, found a litany of inadequacies plagued the Deep Space 2 mission, including a lack of preflight testing.

A second report, which explored the most likely causes for the loss of the Polar Lander and Deep Space 2 spacecraft, also found the lack of sufficient testing harmed the probes’ chance of survival.

The crux of both reports’ conclusions was the mission’s decision to forgo what NASA calls a system-level test. That test would have involved engineers running a model of the spacecraft through a close approximation of what the two probes would actually encounter during the mission itself.

Instead, engineers tested the experimental spacecraft in a piecemeal fashion -- a decision backed by the mission team and NASA management, Gavit said.

"It wasn’t ready because we didn’t do a system-level test -- I think that’s a fair criticism, but at the time it looked like it was the right thing to do," said John Casani, a JPL veteran who authored the second report.

Gavit said the mission team chose to skip the system-level test because they felt confident that the spacecraft could survive landing on Mars, something already borne out over dozens of impact tests.

Also, severe budget constraints left the team with only one test version of the spacecraft, which engineers would have had to destroy to assess its health after a system-level test. By preserving the model, the mission could carry out further tests on more problematic areas, like its miniature suite of electronics, Gavit said.

"We had a decision to make. We really felt it was better spent on the things we were worried about," Gavit said.

Neither of the two reports found a smoking-gun reason for the probes’ failure, leaving it a mystery that probably will go unsolved forever.

Casani laid out four possible scenarios in his report, including the probes’ failure to pierce the martian soil and battery or electronics failure.

Gavit said she favors the former explanation.

"It’s possible they hit something and couldn’t survive," Gavit said. "I think that’s a lot of it."

She added that the lack of a single likely reason -- unlike the case of the Polar Lander, apparently felled by a sole line of bad computer code -- was a positive.

"To me that says we did a good job," said Gavit during in an interview in her JPL office, strewn with test models of the twin probes. "For any spacecraft to be put under this scrutiny and not to find a single thing is a compliment."