Already $1 billion over budget and two years behind schedule, NASA's $4.5 billion James Webb Space Telescope has taken yet another shot from the Government Accountability Office.
The congressional watchdog agency said last week that managers of the Maryland-based project have moved forward with development of the observatory before all its new technologies are ready, and without enough contingency funds to handle unexpected problems.
Unless more money is set aside for contingencies, the report stated, "the program's ability to resolve issues, address program risks areas and accommodate unknown problems is very limited," and the planned 2013 launch date is "not viable."
NASA officials replied that they have already adopted some GAO recommendations and have complied with others. Matt Mountain, director of the Space Telescope Science Institute in Baltimore, which is advising the program, said the observatory remains on track for a 2013 launch.
"I've been through these kinds of reports before in other big projects," he said. "If they'd said, `There's a terrible cost overrun you guys didn't see,' I'd be worried. But they haven't."
"It's an expensive program that deserves to be scrutinized heavily," he said. But "the project has had so many independent reviews over the years, I feel it's pretty robust."
The Webb telescope's development is managed by NASA's Goddard Space Flight Center in Greenbelt, with scientific guidance from the space telescope institute under a $162 million contract.
NASA is regarded as likely to select the Baltimore-based institute, with headquarters on the Johns Hopkins University's Homewood campus, to manage Webb's scientific operations after launch.
Serious delays or cancellation of the Webb project could threaten the future of the operation, which employs 370 people in Baltimore. About two-thirds now work with the Hubble Space Telescope, whose life expectancy hinges on whether NASA approves another manned servicing mission. And that, in turn, hangs on the performance of a troubled space shuttle fleet.
Scientists conceived Webb as a successor to Hubble but with a revised mission - to use infrared wavelengths to peer deeper into space and time than any telescope before it, and to seek out planets circling other stars.
Unlike Hubble, which orbits only a few hundred miles above the Earth, Webb would fly a million miles into space, where it would be inaccessible to human repairmen.
Last year, the Webb project was rattled by word that it had gone $1 billion over budget and faced at least a year's delay in its original 2011 launch date. It has since slipped another year, to 2013.
To cut costs, NASA asked Webb managers to consider design changes. They included a reduction in the diameter of the observatory's main mirror from about 21 feet to 13 feet - and elimination of an important infrared instrument.
Astronomers argued that Webb's primary goal - to study the earliest galaxies that formed after the Big Bang - would be impossible to achieve with the smaller mirror.
Mountain said subsequent discussions saved both items. Engineers found cost savings through other simplifications and the elimination of a less critical instrument.
The GAO report acknowledged that Webb managers have made some planning and accounting improvements - but not enough.
For example, work is proceeding even though several innovative technologies - including an umbrella-like sunshade the size of a tennis court - have not been fully tested.
Investigators noted that project engineers are still modifying the instrument's basic design so Webb will fit on the European Ariane rocket they chose for the launch. And, they noted, there is no facility in the country big enough to test the completed observatory under conditions simulating outer space.
The GAO also said the program's contingency account - 1.5 percent of its budget from 2006 to 2010 and 19 percent overall - was too small. The agency said a program of Webb's complexity should set aside 25 percent to 30 percent for contingencies.
"From a budget perspective, the re-baselined program is not viable for a 2013 launch," the report stated.
Mountain said he sees "a little bit of a disconnect" between the GAO's conventional project expectations and the unconventional demands of building a one-of-a-kind, cutting-edge scientific instrument.
Webb is "still in the technology development phase," he said, and "on track" for the 2008 date prescribed by NASA policy for final approvals.
More important, he said, Webb's riskiest component - its 18-segment folding mirror - is being manufactured, and the telescope's key detectors are finished. "If we didn't have the mirrors and didn't have the detectors, I'd feel differently about it." He said Webb's contingency budget is in line with NASA guidelines. It's purposely weighted toward the end of the project.
"What we're trying to do is avoid the Hubble error," Mountain said, referring to the post-launch discovery that Hubble had a crippling mirror flaw that was missed during ground testing.
Once Webb is built, he said, "we're going to do a full end-to-end test in a big cold chamber to make sure all the mirrors are working. ... We put a lot of the contingency in the last phase, where the risk is."
The entire finished spacecraft will be tested as a unit, he added. Only the sunshade is too big to be checked out separately. "Northrop Grumman has made a lot of them, and they all worked very well," he said.
As for the Ariane, he said, "everybody on the project is fully aware that [Webb] must fit the volume and mass constraints of the Ariane, and nobody's doubting we will."