From: NASA Blogs
Posted: Friday, August 29, 2008
NASA has the best weather forecasters in the world. Both the Spaceflight Meterology Group - who produce landing weather forecasts for all the shuttle landing sites all around the world - and the USAF 45th Weather Squadron - who produce the launch weather forecasts for all the launch operations at the Kennedy Space Center and Cape Canaveral Air Force Station -- are the best. Not only do they make forecasts every day and every hour of the year, but they actually check to see if the forecast was accurate, and keep score. And their forecasts are really micro-forecasts -- exactly what will happen at the launch pad or at the runway; not a general area forecast, which is much tougher.
America has lost a number of space launch vehicles over the years due to poor weather decision -- Atlas/Centaur-76 is the case study example -- and the weather community is striving to make sure that we never lose another.
I've talked about the shuttle landing before, and you know it is the toughest decision that a Flight Director ever makes. The shuttle gets only one shot and it has to be right. The FD goal is to provide the shuttle commander acceptable weather -- not perfect or the shuttle would never land. The decision has to be made an hour and a half before landing. There have been some long waits between the deorbit burn and landing.
Early on, the shuttle was supposed to have jet engines so, among other reasons, it could fly multiple approaches or divert to different runways. However, the weight of the orbiter in the design phase kept growing, from 150,000 pounds empty to 200,000 pounds and more. And the ops guys kept asking (as they always do) "what if the engines don't start during entry, don't we have to protect against that situation?" So fairly early on, the jet engines, the fuel tanks, and all that stuff got deleted from the design. So the orbiter is the world's heaviest and only hypersonic glider. One shot at landing is all the commander gets.
The auto landing capability that was built into the shuttle is not perfect. It could work, if necessary, but engineering analysis shows that there are more times than we would like where the auto landing system would fail. On a commercial jetliner, this is accommodated by an auto go-around feature. But then the shuttle . . . well, see the paragraph above.
So the commander flies the vehicle on final through touchdown and rollout. Visual cues are very important even though there are electronic navigation aids, a head-up display, and all sorts of redundancy. But being able to see the runway and the PAPI lights are mandatory. With the steep approach required by the brick like flying qualities of the shuttle, a minimum altitude of 8,000 feet to see the PAPI lights is required (for end of mission in the daylight -- different rules apply to other situations). (OK, PAPI stands for Precision Approach Pilot Indicator -- four lights which show white at a certain angle and red if you are below that angle).
Mother Nature, of course, makes this difficult. There are these things called "clouds". One meterologist called them "hydro-meteors". Occasionally these "clouds" have been known to obscure the navigation aids. Since there are almost always some clouds in the sky below 8,000 feet, long studies were undertaken to determine how many clouds below the limit constitute a violation. See what I mean by giving the commander acceptable - not perfect - weather? After many approaches in the training aircraft, long arguments, a decision was finally reached: if the clouds obscured 50% or more of the area, that was no-go, if it was less than 50% then it is go because the commander could see the "runway environment".
STS-53 was a "classified" mission but what happened at landing is not secret. In fact several things happened on that landing so it could be grist for more than one story, but today I'm going to talk about the weather. We wanted to land at KSC - saves the ferry flight with its expense and risk and it saves a week or more of time. So we started looking at KSC.
On the particular day in question there was a front coming through Florida - not unusual. Associated with this were low overcast (100%) clouds - at 3,000 feet the bases of the clouds would clearly violate the landing rules, if they got to the Shuttle Landing Facility. On the satellite photos it looked like there was a ruler running diagonally across the peninsula: north and west were socked in and south and east were crystal clear. Weather at the SLF all during the deorbit preparation was perfect: no wind, crystal clear skies, beautiful. Just that threat. The inexorably approaching line of low overcast clouds that would block the commander's view of the field and the all important PAPIs until the last minute -- actually last 20 or so seconds -- before landing. Much too late. Very scientifically our weather forecasters plotted the approach and predicted that the clouds would cover the SLF about a half hour before orbital mechanics would allow the shuttle to arrive there.
So we turned our attention towards Edwards AFB in the high desert of California. Perfect weather there. Virtually no clouds, winds were reasonably low, everything was good to go. Except one cloud. There was one cloud hovering over the PAPI lights on the approach and its base was 3,000 feet. The astronaut pilot flying reconnaissance in the Shuttle Training Aircraft reported that all was go with the exception of that one cloud.
Well, it met our criteria; clearly less than 50% of the area was obscured; in fact probably 90%+ of the area was crystal clear. And in an hour and a half, there is no way that cloud would remain there. I gave a GO for deorbit. The astronaut in the STA had a conniption: in his opinion it was clearly no go. We had a short discussion (see above). I reiterated my GO to the crew. The deorbit burn happened on time for the EDW landing.
Want to guess what happened in the next hour?
In defiance of all the known laws of nature, the front threatening Florida stalled out well northeast of KSC. The KSC center director and his technical staff were later shown on the runway at what would have been the landing time pointing up at clear blue sky. It would have been a perfect day to land in Florida.
In defiance of all the known laws of nature, the one cloud in the sky at Edwards air force base not only did not move but stayed exactly where it was and grew a bit. Still technically within limits, it totally obscured the line of sight for the PAPIs and the runway threshold.
As the Commander later debriefed "we never saw the PAPIs or the runway until we broke out at 3,000 feet. Somebody has some 'splaining to do"
That would be me.
The landing was a good one; all the electronic navigation aids backed up the visual ones and the commander touched down on speed and on distance. Hey, any landing you can walk away from is a good one. If you can fly the machine again, as the saying goes, it was a great landing.
I stand by my initial statement, we have the best weather forecasters in the world. But you can't fool Mother Nature. No wonder Flight Directors get gray or bald . . .
// end //