Sun-synchronous orbits
Aug. 18th, 2005 11:38 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
mmcirvinUntil recently, one of the most solid arguments skeptics of anthropogenic global warming had was that some measurements of warming in the troposphere didn't seem to match the predictions of numerical models. The best-supported such discrepancy came from Spencer and Christy's analysis of measurements taken by the Microwave Sounding Units on NOAA's Polar Operational Environmental Satellites. Spencer and Christy found almost no warming at all in the lower troposphere, where the models predicted warming.
These instruments weren't designed to detect long-term climate trends, and the analysis is very subtle and tricky. Most climatologists didn't regard this discrepancy as outweighing all the evidence in favor of their climate models, and probably figured that there was just some systematic effect here that nobody had yet figured out. But it was still an unsettling open problem.
Recently, however, Spencer and Christy corrected their analysis online, a paper by Mears and Wentz in Science explains what was wrong with the original numbers, and now the calculated trend is well in line with the climate model predictions, if closer to the low end of the range. In an fascinating (if slightly gloating) guest post on Tim Lambert's blog, Scott Church explains what the trouble was in some detail. These satellites are Sun-synchronous: they're in almost-polar orbits arranged to always present the same angle to the Sun (just what that angle is depends on the satellite), so that they always pass over the lit and unlit hemispheres of the Earth at the same (solar) time of day and night. To do this the satellite needs to maintain its orbit with active maneuvers as the Earth goes around the Sun over the course of a year. But this process is not exact, and the deviations need to be corrected for in order to get accurate long-term temperature trends, because the time of day that the satellite is looking at will gradually change. When applying this correction, Spencer and Christy had made a sign error. (I can sympathize, since an epidemic of sign errors was one of the things that drove me from manual calculations to computer algebra when I was doing my thesis work.)
I'd just been thinking about Sun-synchronous orbits as a result of my second visit to the National Air and Space Museum's Udvar-Hazy Center near my parents' home in Virginia. The last time I went there, the Space Hangar wasn't open yet; now it's finished, so I could go in and walk around its hulking centerpiece, the Space Shuttle test orbiter Enterprise. There were ropes around it, but I could get close enough to see something like the view that Steve Robinson must have had when he went to do his little in-orbit repair job on Discovery.
Anyway, I'd always heard that the size and shape of the Shuttle, which added much to the expense of the project, were largely determined by classified Defense Department demands. It turns out it's pretty well known what those demands actually were. The DoD wanted to use the Shuttle to launch and retrieve spy satellites. These things are huge; in the early 1970s, when the Shuttle was being designed, they'd have taken their pictures on film and dropped them into the atmosphere in reentry capsules that were snagged by airplanes while descending by parachute. Now, most low-orbit satellites that are designed to look at the Earth in some way are in Sun-synchronous orbits. In the case of spy satellites, they'd probably be orbiting so as to cross the day hemisphere sometime in the morning or afternoon, in order to get nice distinct shadows and relief from the sun angle.
Defense planned to launch the Shuttle into these Sun-synchronous, more or less polar orbits from a launch facility at Vandenberg Air Force Base near Lompoc, California, and they wanted it to be able to make one orbit and come right back to Vandenberg. I suppose that would give it a polar "abort once around" capability in case of trouble happening late in the ascent, but from the degree of insistence on this, I get the impression that they actually imagined the Shuttle doing one-orbit missions, maybe so that it could get home before anyone figured out what was going on. But while the Shuttle is making this one orbit the Earth is turning, so to return to California it would have to jog something like 1500 miles to the east when it comes around again. For that degree of cross-range maneuvering during reentry, the Shuttle had to be a delta-wing design, rather than various straight-wing and wingless designs that NASA was also considering.
The Shuttle never launched from Vandenberg and never flew in a polar orbit, and it certainly never flew a one-orbit mission. On the other hand, this cross-range capability does give them a lot of flexibility in planning alternate and emergency landing sites. But I do wonder if the delta-wing design contributed to the Columbia accident, whose mark is still visible on the Enterprise in the chunks missing from the leading edges of its wings, taken by NASA to help with the Columbia investigation.
These instruments weren't designed to detect long-term climate trends, and the analysis is very subtle and tricky. Most climatologists didn't regard this discrepancy as outweighing all the evidence in favor of their climate models, and probably figured that there was just some systematic effect here that nobody had yet figured out. But it was still an unsettling open problem.
Recently, however, Spencer and Christy corrected their analysis online, a paper by Mears and Wentz in Science explains what was wrong with the original numbers, and now the calculated trend is well in line with the climate model predictions, if closer to the low end of the range. In an fascinating (if slightly gloating) guest post on Tim Lambert's blog, Scott Church explains what the trouble was in some detail. These satellites are Sun-synchronous: they're in almost-polar orbits arranged to always present the same angle to the Sun (just what that angle is depends on the satellite), so that they always pass over the lit and unlit hemispheres of the Earth at the same (solar) time of day and night. To do this the satellite needs to maintain its orbit with active maneuvers as the Earth goes around the Sun over the course of a year. But this process is not exact, and the deviations need to be corrected for in order to get accurate long-term temperature trends, because the time of day that the satellite is looking at will gradually change. When applying this correction, Spencer and Christy had made a sign error. (I can sympathize, since an epidemic of sign errors was one of the things that drove me from manual calculations to computer algebra when I was doing my thesis work.)
I'd just been thinking about Sun-synchronous orbits as a result of my second visit to the National Air and Space Museum's Udvar-Hazy Center near my parents' home in Virginia. The last time I went there, the Space Hangar wasn't open yet; now it's finished, so I could go in and walk around its hulking centerpiece, the Space Shuttle test orbiter Enterprise. There were ropes around it, but I could get close enough to see something like the view that Steve Robinson must have had when he went to do his little in-orbit repair job on Discovery.
Anyway, I'd always heard that the size and shape of the Shuttle, which added much to the expense of the project, were largely determined by classified Defense Department demands. It turns out it's pretty well known what those demands actually were. The DoD wanted to use the Shuttle to launch and retrieve spy satellites. These things are huge; in the early 1970s, when the Shuttle was being designed, they'd have taken their pictures on film and dropped them into the atmosphere in reentry capsules that were snagged by airplanes while descending by parachute. Now, most low-orbit satellites that are designed to look at the Earth in some way are in Sun-synchronous orbits. In the case of spy satellites, they'd probably be orbiting so as to cross the day hemisphere sometime in the morning or afternoon, in order to get nice distinct shadows and relief from the sun angle.
Defense planned to launch the Shuttle into these Sun-synchronous, more or less polar orbits from a launch facility at Vandenberg Air Force Base near Lompoc, California, and they wanted it to be able to make one orbit and come right back to Vandenberg. I suppose that would give it a polar "abort once around" capability in case of trouble happening late in the ascent, but from the degree of insistence on this, I get the impression that they actually imagined the Shuttle doing one-orbit missions, maybe so that it could get home before anyone figured out what was going on. But while the Shuttle is making this one orbit the Earth is turning, so to return to California it would have to jog something like 1500 miles to the east when it comes around again. For that degree of cross-range maneuvering during reentry, the Shuttle had to be a delta-wing design, rather than various straight-wing and wingless designs that NASA was also considering.
The Shuttle never launched from Vandenberg and never flew in a polar orbit, and it certainly never flew a one-orbit mission. On the other hand, this cross-range capability does give them a lot of flexibility in planning alternate and emergency landing sites. But I do wonder if the delta-wing design contributed to the Columbia accident, whose mark is still visible on the Enterprise in the chunks missing from the leading edges of its wings, taken by NASA to help with the Columbia investigation.