Robert Zimmerman has a disturbing (though not surprising, at least to me) piece at Space Daily, which reports that NASA did no analysis in support of its original decision to cancel the planned Shuttle flight to repair Hubble, and ignored more viable options in favor of its misguided robotic gambit:

NASA historian Steven Dick gave a presentation at the American Association for the Advancement of Science annual meeting in Washington, in which he described the process by which that decision was made and revealed that, in fact, no formal risk analysis had been completed.

Dick had interviewed all of the NASA officials who had been involved in the decision to cancel the shuttle mission to the Hubble, a discussion that came to a head in December 2003 when those officials had been working on NASA's fiscal year 2005 budget.

According to Dick's interviews, risk was the major factor in the discussion, but the officials decided a formal risk analysis was unnecessary. Instead, Dick noted, "The decision was made (by O'Keefe) based on what he perceived was the risk."

In other words, O'Keefe canceled the Hubble mission solely on his gut feeling of the situation. So, the only way NASA can provide the House Science Committee's requested copy of that risk analysis from December 2003 is to recreate it after the fact.

I had always suspected this. I think that Sean O'Keefe was good for the agency, in terms of starting to get the books straightened out (a task that's by no means complete), and starting to restructure it for the end of the Cold War, but I also think that he lost his nerve after having to stand on the tarmac and tell those families that their loved ones weren't coming home two years ago. He simply didn't want to have to risk doing that again. And that's fine, but if so, he was no longer the man for the job, and perhaps didn't step down soon enough, because it clearly adversely influenced the decision he made a year later. Spaceflight is inherently risky, and if we can't accept that, as either a NASA administrator or a nation, then we have no business doing it.

And as Zimmerman concludes, that's really what's so disturbing about that decision, in terms of its potential implications for the future:

For NASA and the American space program, this increasingly untenable position is beginning to have a serious political cost. By refusing to reconsider their decision and reinstate the shuttle servicing mission to Hubble, NASA is undercutting its ability to persuade Congress to give it money to build spacecraft to fly humans back to the moon.

As Rep. Mark Udall, D-Colo., noted during those same science committee hearings, "If we're unwilling to take the risks to go to Hubble, then what does that say about (our willingness to mount) a moon and eventual Mars mission?"

Or as Boehlert remarked, "In a budget as excruciatingly tight as this one, NASA probably should not get as much as the president has proposed."

Unless President George W. Bush appoints a new NASA administrator with the courage to reverse the Hubble decision, he is going to find it increasingly difficult to persuade Congress - or anyone else, for that matter - that NASA has the wherewithal to handle his ambitious space initiative.

But it goes beyond the risk aversion. If the story is true, the changing stories and lack of data after the fact bring back memories of the Goldin years, in which some said that NASA stood for "Never A Straight Answer." That was something that O'Keefe was supposed to fix, not contribute to, and it may take a further investigation with some mea culpas and credible recommendations for avoiding this sort of thing in the future, in order for NASA to gain the confidence needed, from both Congress and the public that still wonders why it's about to lose one of the few NASA programs with genuine widespread support.

The fact that NASA did no formal risk analysis before canceling the Hubble servicing mission has been obvious for quite awhile now. If they had done the risk analysis, they would have released it. At the very least, they would have done a much better job of publicly defending their decision. There are some aspects of a shuttle mission to Hubble that are in fact riskier than an ISS mission that NASA has never mentioned publicly, such as the abort options over the Atlantic. Why didn't NASA mention them? Probably because the agency had no internal document that clearly listed them.

In fact, two weeks ago, during a hearing before the House Science Committee on the FY06 budget, Acting Administrator Gregory was specifically asked by both the ranking Democrat and the chairman of the committee to provide the committee with their risk analysis for the Hubble decision. Gregory was non-committal about the request. But the very fact that it is one year later and Congress still does not have a risk analysis document is proof that it never existed in the first place.

Risk analysis does not seem to be NASA's strong suit. You might think that would lead to competence at crisis management.

I think O'Keefe made a good start a reforming NASA. There are people, though, who think he did real damage as well with the emphasis on bean counting and adding administrative layers. I'm not currently an aerospace insider so it's a bit harder for me to tell.

I must observe that O'Keefe came into the agency as an outsider. The agency has massive problems -- just read the Columbia report for starters. I do know NASA needs to develop a more open culture. Some insiders recognize that now.

I think O'Keefe was spooked by Columbia -- not just the accident, but by the fact that the agency he was leading had such problems. Budgets out of control are one thing -- failure at technical activities are another.

I'll be most interested to see what the new administrator is like. If he's from the military, I'd like to see him more like Eisenhower than Patton.

O'Keefe openly embraced the Columbia report, which gave him much credibility. It seemed like he "got it" and understood that his agency needed to change.

Then he did a lot of damage to himself and his agency's credibility by the way he made the Hubble decision. He pulled the plug on the agency's most popular program--one of the few projects where scientists, human spaceflight advocates, and the general public all agree. And then he failed to justify his decision or explain it very well. His response when people asked "Why?" was essentially to reply "Because I say so." But the problem with this response is that after Columbia, nobody buys the "because I say so" argument from NASA anymore. So O'Keefe earned himself credibility, then kissed a lot of it away.

And that decision continues to hurt NASA. Because now it is trying to justify the Vision for Space Exploration in the same terms of "because we say so."

Risk aversion is an institutional and political problem. However if you were to poll the American public today I think you would find their sensitivity to this has been greatly exaggerated. In taking a poll you would also find that a majority of people would prefer the Hubble mission be completed without this silly repair kit or viable back-up rescue plan. We have been at war for many months and that has a very sobering effect on a population.

Terry,

That's a good point. I know I was more than a little irritated at the Hubble decision. Hubble should be fixed in my view. I've also heard a few rumors that O'Keefe doesn't listen. I don't know O'Keefe at all. Quite a few people describe me as a good listener. Still, though, there are several individuals I basically ignore. Why? Because I've decided they are either untrustworthy or so delusional that what they report can't be trusted.

Next week I'm going to be up on the Hill doing some of that citizen lobbying stuff. I'll see what I can find out.

Sometimes when faced with a multiple guess question. The first guess is the best answer. I say dump Hubble because its a relic of the Shuttle era. The Shuttle is a vehicle that needs to be pushed in obsolescence as fast as possible in my opinion. One less place for it to go is one less reason to keep it around. We have ground based telescopes that have approached visible light sensitivies and have exceeded Hubble's IR capabilities already. We need a space telescope with more aperture and more IR sensitivity to unravel the mystery's of the cosmic dark ages shortly after the big bang.

Reiter wrote:
"We have ground based telescopes that have approached visible light sensitivies and have exceeded Hubble's IR capabilities already."

Actually, this is not really true. Hubble had two primary strengths--wide field visible and UV. The UV collection cannot be done from the ground. Although astonomers can do amazing things with adaptive optics in the visible spectrum, this capability is limited to point sources, whereas Hubble can gather wide fields. Further, the upgrades that were planned for the shuttle servicing mission would have restored Hubble's tremendous lead over other telescopes. So it is a little unfair to compare what it can do now with near-future ground-based telescopes instead of what it would be capable of doing if NASA serviced it.

But the best answer comes from the astronomy communty itself, and they recently spoke thru a National Academy of Sciences report saying to save Hubble. They said this not because they love the shuttle, but because they believe that Hubble is an important scientific instrument.

What Mr. Simberg missed in his comments was that it is hard to say that NASA is "risk averse" when they never even performed a proper risk analysis in the first place. They made a gut decision, and risk was probably only one part of it.

Now this raises an interesting question--if they are indeed as risk averse as Mr. Simberg and NASA historian Steven Dick assert, why did NASA not also cancel the remaining ISS flights? Or cancel at least some of them? It has always been hard to square canceling a single shuttle flight to Hubble with the decision to conduct 28 shuttle flights to ISS.

Although I believe that risk was an important consideration in this decision, there were most certainly other considerations as well. Certainly one of them must have been a desire to clear away any distractions besides completing ISS. So getting rid of a unique shuttle flight helped do that. Another consideration was undoubtedly cost--NASA could save a few hundred million dollars (and additional schedule pressure) by canceling this mission.

And finally, there was the consideration of how the public and the press and the Congress would respond. Clearly NASA leadership (primarily O'Keefe) banked on there being little opposition to the cancelation. They thought it would be a one-day story and that would be it. They certainly never thought that they would still be getting raked over the coals about it a year later. (And you can bet that NASA will get more bad publicity when Hubble finally dies, and more bad publicity when NASA finally de-orbits it. The result is that the triumphant story of Hubble will always end with the bad chapter about how the agency treated it in the end.)

What I find so puzzling is that they thought that this would be such a small deal that they never even bothered to prepare for it. They have never prepared a good response to the criticism. They almost acted as if they did not have to do so. That seems to say a lot about the leadership, including O'Keefe--he thought that his decision never required justification.

I have a slightly related question about the Hubble. If NASA is going to send a robotic booster to de-orbit the Hubble, couldn't the same booster technology be used to transfer the Hubble into an orbit more friendly to a shuttle service mission? If the shuttle does get into trouble during repair mission the crew could still abort to the ISS.

I assume that the Hubble design criteria dictates the precise orbit that it is in. Moving it to an ISS-compatible orbit would likely degrade it's capabilities to some degree, but it would still be a lot more useful than a hunk of metal at the bottom of the sea. Also, I also assume the booster could attach and detach from the Hubble in a non-destructive manner.

Another alternative would be to use the booster to park the Hubble in a higher/safer orbit for a decade or so in the hope that commercial space ventures can eventually salvage the telescope and upgrade/restore it to working order. The Hubble would be a hell of a salvage prize and I'm fairly certain the scientific establishment could redirect funds to pay for a reduced cost private mission in the future. If NASA was to ceed ownership of the Hubble to a private salvage mission, rental of the Hubble back to the scientific community could perhaps even make such a mission a profitable venture.

Turning the Hubble into the ultimate prize for a private space venture seems much more worthy and inspiring to me than shooting billionaires into orbit.

I'd wondered about Mike Thompson's suggestion too.

But my main concern is that this reveals a startling (to me anyway) flaw in the NASA corporate culture.

I happen to believe that O'Keefe is right, that Hubble should be replaced, not repaired. But as a sometime project manager, there's no way on God's Earth that I'd let my Intuition become Holy Writ, I'd always get it formally checked (even if only a 2-day documented study to show that the Risk is orders of magnitide greater than acceptable - and if the study shows it isn't that bad, then we'd need a full-blown Risk Analysis.).

That this wasn't done when Congress was breathing down NASA's neck shows political incompetence at the least. Worse, it more than hints of a corporate culture where decisions From On High aren't always checked, just because they appear plausible.

The UV collection cannot be done from the ground.

Yes, that is the one area in which Hubble will always do better than a ground-based scope.

Now this raises an interesting question--if they are indeed as risk averse as Mr. Simberg and NASA historian Steven Dick assert, why did NASA not also cancel the remaining ISS flights? Or cancel at least some of them? It has always been hard to square canceling a single shuttle flight to Hubble with the decision to conduct 28 shuttle flights to ISS.

Because of the perception that the Hubble mission was quite a bit more risky than an ISS mission, because there was no safe haven there if they ran into another Columbia-like problem.

I don't think that risk was the only issue, but I do think it was the dominant one.

Risk aversion -- or, more properly, risk perception aversion -- is also apparent in the decision to send a deorbiting module to HST. This move makes no economic sense whatsoever, if you look at the cost vs. expected casualties and damage on the ground.

I'm troubled that even such simple decisions apparently cannot be made rationally.

The need of a safe haven is a red herring. Shuttle did most of its flights when there wasn't an ISS and having the ISS in there did not save the astronauts last time.

I agree that there should have been a small short study, possibly followed by full-blown risk analysis.

...having the ISS in there did not save the astronauts last time.

Of course it didn't. Because it wasn't "there." Coumbia was in a 28.5 degree orbit (the same inclination as Hubble, and just as impossible to get to ISS from).

In 100 shuttle flights, there was 1 launch accident and 1 reentry accident. If having the "safe haven" of the ISS reduces the risk of a reentry accident to zero, then the risk of an ISS mission is 1%, while the risk of a Hubble mission with no backup shuttle at the ready is 2%. This is all just rough estimating, of course, but a Hubble mission is in no way "orders of magnitude" more risky than an ISS mission.

28 ISS missions implies a 28% chance of another fatal shuttle accident and loss of orbiter. What is the sense in finishing the ISS - and then abandoning it shortly thereafter - at such a high risk, while dumping Hubble because a 2% risk is "too high"? ISS appears to me to be producing nothing of value to justify the cost and effort, and is without doubt the most boring space "mission" of all time.

I come down to the basic question, "If you are afraid of going to Hubble, how can you ever go to the moon or Mars?" There is no reason to believe than a new moon/Mars vehicle will be any safer than the shuttle, and the missions are inherently more complex and risky.