Last night I observed evidence of a large, damaged and gyrating, geosynchronous satellite. A priori, this could be a shot by any of the space powers, either a test against one of its own unneeded satellites, or an act of war against someone else's. Two successful such tests, one by China and one by the U.S., have been publicized recently.
At ~00:00 +/- 00:10 Central Daylight Time March 26, 2008, I observed (from lat 42.09N, long 93.46W) intermittently for ~20 min., by eye and with 8x40 binoculars, a flashing light NE of Tau Leonis and later E of Beta Virginis. I took about 6 one-second 35mm photos, and about 2 approx. 20-sec time exposure photos, at f/2 with a 50mm focal length (25mm aperture)(not telephoto) lens, on a tripod, on Kodak 200 Gold (C-41 film)(unfortunately my film was somewhat outdated). The nearer of two mercury vapor farm lights, I turned off at the pole, but the farther light involved rusty fuses near bare wires, so I didn't try to turn it off. The moon wasn't up yet, but there is light pollution in that direction, due to Ames, and due to a large well-lit feedlot, the county seat, and Des Moines (I own no filters). One of the time exposures included the flash; I think it was the only snapshot that did capture the flash. Aiming the camera was by feel, due to the dimness of stars in the region. I never saw the flash through the viewfinder, but saw it over the top of the camera when the shutter was open.
I took the film for development today. The photo shop proprietor said he lacked the equipment to push-process it, but I told him to proceed anyway. I warned him that it was an astronomical photo of a satellite flash, and not to worry about the other photos on the roll. He said this was something with which he was familiar because he had taken such photos himself in years past.
I timed about six cycles by counting seconds; they always were ~6 sec. The flash lasted no longer than ~1/30 sec. This would be consistent with a < 2-degree light cone and a < 700-mile wide track crossing N. America. The magnitude and color varied. The brighter flashes tended to be blue-white, and the dimmer, orange. As near as I could tell, there was some positive correlation both of color and brightness. That is, the brighter bluish and dimmer reddish flashes generally didn't seem to alternate. I estimated the brightest flashes as equal to Sirius, in my notes last night, but by more considered judgment today would estimate magnitude 0.0; calculation from this suggests a reflecting surface 30 feet in diameter, assuming 100% albedo.
I think my times and positions yield a sidereal speed estimate which is accurate within a factor of two; the path (relative to the stars, which of course move) was approx. 5 deg in 20 minutes, eastward, more parallel to the ecliptic (which was about 2 deg south of the flashes) than to the equator (which also is nearby, farther south), from approx 172 to 177 ecliptic longitude.
The flashes could not be seen after about 00:20 +/- 00:10. This would have been about the time of passage into Earth's umbra.
A balloon with a flashing light hardly could be higher than 30 mi., thus hardly could be farther than 50 mi. Even if my sidereal movement estimate were wrong by as much as a factor of two, that would imply only 5 deg / 20 minutes of time at 50 mi, i.e., 12 miles per hour. This probably is too slow for east-west winds at high altitude.
Cloudy skies are predicted for my region tonight. This phenomenon should be visible from elsewhere in N. or S. America, where skies are clear (and where the track of the reflected flash lies, if it is narrow).
Though many satellites rotate, it would seem inefficient for the large surface areas - the solar panels and dish reflectors - to spin. A 10 rpm spin for a 30 ft diameter object would imply 0.5g centrifugal force, a considerable structural strength and weight requirement.
Update March 28: the photo shop man didn't make prints of the relevant negatives because he saw nothing on them. I see nothing on them either, though I'll look again with a microscope. The stars in the region were, at best, barely bright enough to see in the viewfinder.
Update April 2: last night was very clear; I looked at about the same time and place, but didn't see it.
