On Sept. 19 NASA Administrator Mike Griffin revealed the agency’s new plan for implementing the president’s Vision for Space Exploration.

The plan has significant positive and negative features.

On the positive side, it recognizes the need for the development of a true heavy-lift launch vehicle (HLV), and takes concrete steps to preserve the shuttle industrial infrastructure necessary to produce such a vehicle. It does this by initiating development of a medium-lift launch vehicle using shuttle technology.

The importance of this cannot be overemphasized. An HLV is absolutely necessary to enable human exploration of the Moon or Mars, and it was a measure of former NASA Administrator Sean O’Keefe’s unfitness for his position that he was willing to promote a clearly unworkable quadruple launch/quadruple rendezvous lunar architecture for the purpose of justifying the abandonment of that capability. Dr. Griffin has reversed that position and backed his policy with action, and that is excellent.

Another strong feature of the plan is its decision to develop and employ a methane/oxygen rocket engine for Lunar ascent. Methane/oxygen is a far more storable propellant combination than hydrogen/oxygen and offers much better performance than conventional hypergols, making it a much better choice than either for lunar ascent and return propulsion.

More importantly, however, methane/oxygen is the easiest propellant combination to synthesize out of the Martian atmosphere, and some could be made out of lunar base waste products as well. The choice of this propellant therefore shows good system engineering sense with thoughtful consideration of the best way to select Lunar mission technologies that will be most useful in enabling human exploration of Mars as well.

On the more problematic side is the decision to develop such a large Crew Exploration Vehicle (CEV). While a large CEV certainly enables larger crews and greater comfort, it will cost more to develop, produce and launch than a smaller capsule. Furthermore, because of its excessive mass, the large CEV makes direct return lunar missions impossible, thus mandating a lunar orbit rendezvous mission architecture. This, in turn, will require the costly development and production of lunar excursion modules, and impose return rendezvous phasing complications that could hamstring the operations of a lunar base, especially if surface stays greater than two weeks are desired.

Another cause for concern is the decision to launch the CEV after the HLV delivers the rest of the mission components to orbit. The HLV’s cargo will include stages employing cryogenic liquid hydrogen/oxygen propellant, and this propellant will start to boil away immediately after launch. Thus for the mission to succeed, the CEV must be launched on time, within a few weeks at most of the prior flight, without fail.

Otherwise, the billion-dollar class HLV launch and cargo will have to be written off. This situation will put great pressure on managers to launch despite warnings, thereby putting crews at risk. Moreover, NASA’s record of achieving on-time crewed launch to date is very poor. Unless it is radically improved, this aspect of the plan will have to be abandoned.

That said, the plan is an enormous improvement over its predecessors. One only has to compare it to the psychedelic NASA mission architecture of 2002, which called for supporting Lunar exploration from a LaGrange point space station supplied by giant cycling nuclear electric spaceships, or the nonsensical O’Keefe quadruple-launch/quadruple rendezvous lunar mission plan of 2004, in order to breathe a huge sigh of relief.

The previous NASA plans were pure nonsense . This one is real engineering. Finally, we have a plan that could actually work.

There is, however, a deeper problem with the plan than the engineering concerns noted above. That is, that while preserving the HLV infrastructure, the plan relegates the development of an HLV to a subsequent administration. In consequence, for the next 13 years, NASA will continue to send crew after crew up and down to low Earth orbit, at a cost of some $70 billion, for no justifiable purpose whatsoever.

Both Adm. Harold Gehman, the chairman of the Columbia Accident Investigation Board, and Dr. Griffin have made the point that if we are to accept the costs and risks of human spaceflight, we should be undertaking missions that are worthy of those costs and risks. But for the next 13 years, we will continue not do so.

To paraphrase St. Augustine, NASA is now saying “Lord, make me a destination-driven space agency, but not yet.”

In saying this NASA is, in fact, acting in accord with the Bush Vision for Space Exploration, as enunciated in January 2004. That policy however, was formulated by a White House which lacked a competent NASA administrator to advise it. Now, however, that we have a qualified NASA administrator, this policy needs to be revisited and reformulated.

Let us review the consequences of blindly following the mediocre vision scriptural document of January 2004.

That document was a compromise between those who wanted a destination-driven space program, and those who did not. Therefore, in accord with the bargain reached, NASA would be allowed to continue to fly shuttle missions for the rest of the decade, after which the destination driven program could begin.

But does this make any sense? The only really time-critical shuttle mission is Hubble Space Telescope repair. This is indeed a truly important mission, and it should be flown with dispatch, as it is without question worthy of the 2-percent risk to crew that any shuttle mission must entail. But the rest of the shuttle manifest is devoted to space station construction, and these cargos could be delivered much more expeditiously by the HLV that NASA needs to develop to reach the Moon anyway.

Griffin’s HLV design will be able to deliver 125 metric tons to low Earth orbit. The shuttle can only deliver 20 tons. With a single launch then, the HLV will be able to deliver as much payload as the shuttle program can during a year — and that’s during a good year.

Compared to current shuttle launch rates, which will have managed only one flight between February 2003 and February 2006, (at a cost of $15 billion), the HLV will be able to launch in an afternoon everything the shuttle program would be able to launch for the next 18 years.

Operating the shuttle program for the next five or six years to deliver a few space station payloads early will cost us $30 billion. All that money could be saved simply by shutting the shuttle down after Hubble repair, and shifting the shuttle program funds over to immediate development of the HLV and the other Lunar exploration hardware elements. We could then use the HLV to complete the space station and reach the Moon by 2012 instead of 2018.

In the wake of Hurricane Katrina and the financial burdens it will impose on the nation, gratuitously wasting $30 billion of the taxpayers’ money in order to dogmatically fulfill an old scriptural document is unacceptable. The new NASA architecture is a good plan for implementing a flawed policy. We need a good policy. We have real talent at NASA now, and we should make use of it to revise the policy itself.

Robert Zubrin, an astronautical engineer, is president of the Mars Society (www.marssociety.org), and author of The Case for Mars (Simon and Schuster, 1996), Entering Space (Tarcher Putnam, 1999), and Mars on Earth (Tarcher Penguin, 2005).